Search is not available for this dataset
article
stringlengths 4.36k
149k
| summary
stringlengths 32
3.35k
| section_headings
sequencelengths 1
91
| keywords
sequencelengths 0
141
| year
stringclasses 13
values | title
stringlengths 20
281
|
|---|---|---|---|---|---|
To identify novel susceptibility loci for Crohn disease ( CD ) , we undertook a genome-wide association study with more than 300 , 000 SNPs characterized in 547 patients and 928 controls . We found three chromosome regions that provided evidence of disease association with p-values between 10−6 and 10−9 . Two of these ( IL23R on Chromosome 1 and CARD15 on Chromosome 16 ) correspond to genes previously reported to be associated with CD . In addition , a 250-kb region of Chromosome 5p13 . 1 was found to contain multiple markers with strongly suggestive evidence of disease association ( including four markers with p < 10−7 ) . We replicated the results for 5p13 . 1 by studying 1 , 266 additional CD patients , 559 additional controls , and 428 trios . Significant evidence of association ( p < 4 × 10−4 ) was found in case/control comparisons with the replication data , while associated alleles were over-transmitted to affected offspring ( p < 0 . 05 ) , thus confirming that the 5p13 . 1 locus contributes to CD susceptibility . The CD-associated 250-kb region was saturated with 111 SNP markers . Haplotype analysis supports a complex locus architecture with multiple variants contributing to disease susceptibility . The novel 5p13 . 1 CD locus is contained within a 1 . 25-Mb gene desert . We present evidence that disease-associated alleles correlate with quantitative expression levels of the prostaglandin receptor EP4 , PTGER4 , the gene that resides closest to the associated region . Our results identify a major new susceptibility locus for CD , and suggest that genetic variants associated with disease risk at this locus could modulate cis-acting regulatory elements of PTGER4 .
Crohn disease ( CD ) is a chronic relapsing inflammatory disorder of the intestinal tract , described for the first time in the 1920s [1] . Lifetime prevalence has increased to current estimates of ∼0 . 15% in Caucasians . The precise environmental causes underlying this rise remain essentially unknown , but familial clustering and twin studies clearly identify an inherited component to predisposition . More than ten susceptibility loci have been identified by linkage and/or association studies and convincing causative mutations have been reported , particularly in CARD15 [2–3] . As known loci don't fully account for the genetic risk for CD we performed a genome-wide association scan ( GWA ) to contribute to the identification of additional susceptibility loci .
Genotype data from the Illumina HumanHap300 Genotyping Beadchip [4] were obtained on 547 Caucasian CD patients from Belgium and compared to genotypes for 928 healthy controls from Belgium and France . Genotype call rates were >93% for all individuals included in the study . Of the total 317 , 497 SNPs available , 15 , 046 with genotyping success rate of less than 96% or deviating from Hardy-Weinberg proportions in controls ( Fisher's exact test p ≤ 10−3 ) were eliminated from further analysis as it is known that less reliable markers generate spurious associations . For the remaining 302 , 451 SNPs , we compared allele frequencies between cases and controls as outlined in Methods . Figure 1 shows the 10 , 000 most significant p-values obtained across the human genome . Regions on Chromosomes 1 , 5 , and 16 harbored clusters of markers with suggestive evidence of association at significance levels between 10−6 and 10−10 . The significance of tests of association with these markers remained within this range after controlling for possible effects of population structure using a backwards stepwise regression [5] . The strongest association was found with markers of the IL23R gene on Chromosome 1 , which has recently been identified as a novel CD susceptibility locus in a case-control and family-based association study of Caucasian and Jewish cohorts [6] . In our data , two markers of the IL23R gene , rs11209026 and rs11465804 , gave the most significant association signals ( p < 10−9 ) . Rs11209026 corresponds to an Arg381Gln substitution in IL23R while rs11465804 is intronic and in strong linkage disequilibrium ( LD ) with the former marker . A marker within the CARD15 gene on Chromosome 16 , which is the first susceptibility gene to have been identified in CD [3] , also showed suggestive evidence of association ( rs5743289; p < 10−6 ) . We also examined the results of the GWA with respect to other previously reported susceptibility loci , including OCTN [7] , DLG5 [8] , TNFSF15 [9] , and ATG16L1 [10] . None of these obtained a similar level of significance for association in our study . Genotyping our cohorts for other SNPs at these loci that are reported in the literature to be associated with CD did not improve the signals , with the exception of rs224188 corresponding to a Thr-to-Ala substitution within ATGL16L1 ( p < 2 × 10−4 ) , thus providing confirmation of this novel susceptibility locus for the first time [10] . On Chromosome 5p13 . 1 , we identified a region of approximately 250 kb that contained six markers with p < 10−6 in the association test ( Figure S1 ) . This region has not previously been reported as a CD susceptibility locus . We selected ten markers from the regions of IL23R and 5p13 . 1 for confirmation genotyping in up to 1 , 266 additional Caucasian CD patients and 559 additional controls . The IL23R locus was included in the confirmation genotyping as it had not yet been reported at the time of our study [6] . The associations at these two loci were clearly replicated with p-values as low as 4 . 2 × 10−7 at the IL23R and 3 . 7 × 10−4 at 5p13 . 1 ( Table 1 ) . In the combined data from the GWA and replication studies , we obtained p-values as low as 2 . 2 × 10−18 at IL23R and 2 . 1 × 10−12 at the 5p13 . 1 locus . In addition , we genotyped trios with non-affected parents for the same SNPs to perform a transmission disequilibrium test ( TDT ) . The ten SNPs were typed on 137 trios with affected offspring included in the case-control study , while two of the 5p13 . 1 SNPs were typed on an additional 291 independent trios , also originating from Belgium . Significant over-transmission of the associated alleles were found at both loci , thus providing additional confirmatory evidence in support of the IL23R1 and 5p13 . 1 susceptibility loci ( Table 1 ) . To further characterize the novel 5p13 . 1 locus , we genotyped a subset of 1 , 092 CD patients and 374 Belgian controls for 111 markers ( average interval: 2 . 3 kb ) spanning the 250-kb segment . We determined the most likely linkage phase for each individual using PHASE [11] , and used the corresponding haplotype frequencies to quantify the level of LD between all marker pairs . The 250 kb encompass five clearly delineated LD blocks , the central one ( block III ) being the largest and spanning 122 kb ( Figure 2A ) . We first performed single-marker association analyses . The strongest effects were observed within the 122-kb block III with several SNPs yielding p-values <10−5 . p-values <10−3 and 10−4 were observed in flanking blocks II and IV , respectively ( Figure 2B ) . We then performed haplotype analysis of the region spanned by blocks II to IV . For block III , 20 haplotypes accounted for 93% of the observed chromosomes . These could be grouped in three clades comprising respectively six ( IIIA ) , six ( IIIB ) , and two ( IIIC ) haplotypes , plus a group of six haplotypes that apparently originated from various recombination events . Likewise , evaluation of block II revealed three clades ( with respectively two [IIA] , three [IIB] , and two [IIC] haplotypes ) and two recombinant haplotypes , while block IV was characterized by two clades with two ( IVA ) and one ( IVB ) haplotype respectively . We compared the clade frequencies in cases and controls at intervals bounded by ancestral recombination events ( Figure 2C ) . In agreement with the results of the single-marker analysis , the most significant associations were found in block III followed by IV and II . To verify whether the entire 5p13 . 1 effect could be attributed to block III ( i . e . , the effects observed for blocks II and IV would be mere echos of the block III effect ) , we performed a multi-variate analysis as described in Methods . The clade effects of blocks II and IV conditional on the effect of block III and vice versa remained significant ( p ( II|III ) = 0 . 023 , p ( III|II ) = 0 . 0004 , p ( IV|III ) = 0 . 003 , and p ( III|IV ) = 0 . 026 ) , suggesting that multiple variants in the region may jointly account for the observed effect on CD . Commonly occurring recombinant haplotypes in blocks II and III caused local drops in significance , thus suggesting that causal variants lie outside the corresponding subsegments ( Figure 2C ) . No known genes or CpG islands were found within the region of association on 5p13 . 1 after examination with the Ensembl and UCSC genome browsers . The region has an average GC content of 38% , and an excess of interspersed repeats given GC content ( 58 . 36% versus 42 . 3% ) , which is mainly due to an excess of LINE1 ( 33 . 05% versus 19 . 6% ) and LTR elements ( 15 . 36% versus 7 . 70% ) [12] . It contains 98 highly conserved elements [13] . It is part of a 1 . 25-Mb gene desert between DAB2 ( 850 kb distally from the block ) and PTGER4 ( 270 kb proximally from the block ) . Interestingly , several of the genes flanking the region have been implicated in pathogenesis of CD , or are related to genes that have been implicated in the disease . These include a member of the caspase recruitment domain family ( CARD6 ) , three complement factors ( C6 , C7 , and C9 ) , and—most notably—the prostaglandin receptor EP4 ( PTGER4 ) , which resides closest to the group of disease associated markers ( Figure S1 ) . One hypothesis is that the disease-associated region contains cis-acting regulatory elements that control the expression levels of the causal gene ( s ) located in the vicinity , and that the causal variants modulate the activity of these elements . As a first step to test this , we studied the effect of SNPs in the disease-associated region on the expression levels of neighboring genes . To that end we exploited a database of genome-wide gene expression ( Affymetrix HG-U133 Plus 2 . 0 chips; http://www . affymetrix . com ) measured in EBV-transformed lymphoblastoid cell lines from 378 individuals genotyped with the Illumina HumanHap300 Genotyping Beadchip ( W . Cookson , unpublished data ) . Remarkably , eight of the 26 Illumina markers spanning 264 kb coinciding precisely with the CD-associated region yielded p-values <2 × 10−3 for PTGER4 ( Figure 2B ) . Three of the markers influencing PTGER4 expression are located in block III ( rs16869977 , rs10512739 , and rs6880934 ) . The first two are tagging the IIIBa sub-clade ( Figure 2C ) , while the third one is in complete LD with it ( IIIA + IIIBa ) . The corresponding SNPs did not show convincing evidence for association with CD . Two strongly associated SNPs ( D′ = 0 . 84 ) located respectively in block IV ( rs4495224 ) and V ( rs7720838 ) showed the most significant effect on PTGER4 expression and were also associated with CD ( Table 1 ) . The rs4495224 A and rs7720838 T risk alleles were associated with increased PTGER4 expression . Although these results must be treated as preliminary , they tend to support the hypothesis that the disease-associated polymorphisms may be related to the expression levels of one or more genes in the region . CD is the most common form of inflammatory bowel disease , the other being ulcerative colitis ( UC ) . We genotyped a cohort of 246 Belgian UC patients ( Caucasians ) for IL23R ( rs11209026 ) , ATG16L1 ( rs2241880 ) , and the novel 5p13 . 1 locus ( rs4613763 ) . Consistent with published results [6 , 10] we found a significant association for IL23R ( p = 1 . 2 × 10−3; odds ratio: 2 . 51 ) but not for ATG16L1 ( p = 0 . 78 ) . There was no effect of the novel 5p13 . 1 locus on UC ( p = 0 . 54 ) . While additional studies will be needed to exclude completely a role in UC , these results suggest that the principal susceptibility effects of the 5p13 . 1 locus are for CD . The restriction to CD risk observed for ATG16L1 and the 5p13 . 1 locus is similar to that found for CARD15 [3] . We herein describe the localization of a novel major susceptibility locus for CD on 5p13 . 1 by GWA . The region of strongest association coincides with a gene desert devoid of known protein-coding genes . The observed effect may be mediated by as-yet unknown transcripts mapping within the region . As a matter of fact , limited numbers of spliced and unspliced ESTs originating from the HT1080 fibrosarcoma cell line or medulla ( e . g . , BG182136 and BG184600 ) map to the region . An alternative explanation , however , is that the disease-associated region contains cis-acting elements controlling the expression of more distant genes . We provide evidence in support of this hypothesis by demonstrating that genetic variants in the CD-associated region differentially regulate the expression levels of PTGER4 , the closest known gene located 270 kb proximally . PTGER4 is a strong candidate gene for CD , as it is known that knock-out mice develop severe colitis upon dextran sodium sulphate treatment , unlike mice deficient in any of the seven other types of prostanoid receptors . Increased susceptibility to colitis is also observed in wild-type mice administered an EP4-selective antagonist , while EP4-selective agonists are protective [14] . We observe in particular that the CD susceptibility allele at marker rs4495224 is associated with increased PTGER4 transcript levels in lymphoblastoid cell lines . This finding establishes a direct link between disease susceptibility and PTGER4 expression , although the direction of the effect apparently contradicts the results in knock-out mice . Detailed studies of the effect of genetic variants in the disease-associated region on PTGER4 expression in different tissues and of a possible connection between PTGER4 levels and CD susceptibility are certainly needed and work towards that goal is in progress . The hypothesis that the 5p13 . 1 CD-susceptibility locus operates by modulating PTGER4 expression levels could—at least in theory—be tested by replacing the corresponding murine sequences with the human orthologous variants and quantitatively complement the murine knock-out allele [15] . Our results suggest that the 5p13 . 1 effect on CD could result from the combined action of multiple susceptibility variants . Extensive sequencing of the most common haplotypes in the region of association is being conducted towards their identification .
Genotyping for the whole genome scan was performed on a Illumina HumanHap300 Genotyping Beadchip ( http://www . illumina . com ) [4] . Genotyping of individual SNPs was performed on an ABI7900HT Sequence Detection System using TaqMan MGB probes from Pre-designed SNP Genotyping or Custom TaqMan SNP Genotyping assays ( Applied Biosystems , http://www . appliedbiosystems . com ) . Association analyses were conducted using Fisher's exact test ( whole genome scan ) or chi-squared tests of independence ( confirmation analysis ) . We applied the logistic regression method of Setakis et al . [5] to test for the possible effect of population structure on the most significant association results . The 110 control markers included in the logistic regression had 100% genotype success rate with minor allele frequency >30% , and no two markers were within 20 Mb of one another . To test for an effect of block I conditional on the effect of an adjacent block II , we compared the proportion of I haplotype clades nested within a given II clade ( for instance , proportion of IA , IB and IC within IIA ) between cases and controls by chi-squared . Chi-squared values ( and degrees of freedom ) were summed across II clades to yield an overall ( I|II ) test statistic . The database genome-wide expression analysis data was provided by W . Cookson ( Imperial College , London , United Kingdom ) . Briefly , expression data were generated from RNA extracted from EBV-transformed cells from 378 genotyped offspring in nuclear families . Annotations for individual transcripts on the Affymetrix arrays were extracted from the Affymetrix NetAffx database ( http://www . affymetrix . com ) . Data from the gene expression experiment was normalized together using the RMA ( Robust Multi-Array Average ) package [16 , 17] to remove any technical or spurious background variation . An inverse normalization transformation step was also applied to each trait to avoid any outliers . A variance components method was used to estimate heritability of each trait using Merlin-regress ( RandomSample option ) [18 , 19] . For PTGER4 , we obtained a mean quantitative expression value of −0 . 017 and a variance of 0 . 722 while the heritability estimate for PTGER4 estimated using the sibship data was 0 . 844 . Association analysis was applied with Merlin ( FASTASSOC option ) . We estimated an additive effect for SNPs and tested its significance using a score test that adjusts for familiality and takes into account uncertainty in the inference of missing genotypes . | Individual susceptibility to many common diseases is determined by a combination of environmental and genetic factors . Identifying these genetic risk factors is one of the most important objectives of modern medical genetics , as it paves the way towards personalized medicine and drug target identification . Recent advances in SNP genotyping technology allows systematic association scanning of the entire genome for the detection of novel susceptibility loci . We herein apply this approach to Crohn disease , which afflicts an estimated 0 . 15% of the people in the developed world and identify a novel susceptibility locus on Chromosome 5 . A unique feature of the novel 5p13 . 1 locus is that it coincides with a 1 . 25-Mb gene desert . We present evidence that genetic variants at this locus influence the expression levels of the closest gene , PTGER4 , located 270 kb away , in the direction of the centromere . PTGER4 encodes the prostaglandin receptor EP4 and is a strong candidate susceptibility gene for Crohn disease as PTGER4 knock-out mice have increased susceptibility to colitis . | [
"Abstract",
"Introduction",
"Results/Discussion",
"Materials",
"and",
"Methods"
] | [
"homo",
"(human)",
"genetics",
"and",
"genomics"
] | 2007 | Novel Crohn Disease Locus Identified by Genome-Wide Association Maps to a Gene Desert on 5p13.1 and Modulates Expression of PTGER4 |
Aedes aegypti , the principal vector of dengue fever , have been genetically engineered for use in a sterile insect control programme . To improve our understanding of the dispersal ecology of mosquitoes and to inform appropriate release strategies of ‘genetically sterile’ male Aedes aegypti detailed knowledge of the dispersal ability of the released insects is needed . The dispersal ability of released ‘genetically sterile’ male Aedes aegypti at a field site in Brazil has been estimated . Dispersal kernels embedded within a generalized linear model framework were used to analyse data collected from three large scale mark release recapture studies . The methodology has been applied to previously published dispersal data to compare the dispersal ability of ‘genetically sterile’ male Aedes aegypti in contrasting environments . We parameterised dispersal kernels and estimated the mean distance travelled for insects in Brazil: 52 . 8m ( 95% CI: 49 . 9m , 56 . 8m ) and Malaysia: 58 . 0m ( 95% CI: 51 . 1m , 71 . 0m ) . Our results provide specific , detailed estimates of the dispersal characteristics of released ‘genetically sterile’ male Aedes aegypti in the field . The comparative analysis indicates that despite differing environments and recapture rates , key features of the insects’ dispersal kernels are conserved across the two studies . The results can be used to inform both risk assessments and release programmes using ‘genetically sterile’ male Aedes aegypti .
Dengue , an arbovirus , has seen recent re-emergence and spread on a global scale [1] and is now responsible for an estimated 390 million infections annually [2] . The vector of dengue is the Aedes mosquito , with Ae . aegypti and Ae . albopictus responsible for the majority of disease transmission [3] . The release of ‘genetically sterile’ male Aedes mosquitoes has been demonstrated to be a valuable additional tool by which the vector can be controlled [4] . Understanding the ability of the released ‘genetically sterile’ insects to disperse , and their behaviour whilst doing so , is an important step in designing robust , efficient and effective releases . Attaining adequate coverage of released sterile insects across a given area is a major operational challenge of a sterile insect control effort [4] . Knowledge of the distribution of dispersal distances of released insects will improve our ability to target releases , obtain required coverage densities , confidently predict the potential spatial range of a release and is a key element for the assessment of risk . Independently conceived by Petersen in 1896 and Lincoln in 1930 [5] , mark-recapture , capture-recapture or mark-release-recapture studies ( hereafter referred to as MRR ) have since become a key set of ecological methods . MRR allows inference to be drawn about a number of important ecological factors including the estimation of population size and quantification of dispersal and survival . The methods have been used across a diverse range of species , from whales [6] to fruit flies [7] , and have seen extensive use in mosquito ecological studies . Analysis of the location of recaptured marked insects with respect to the release point allows inference to be made about the dispersal of the released insects . A number of MRR studies have been performed with the aim of assessing the dispersal ability of both lab [8–11] and ‘genetically sterile’ [12] strains of male Ae . aegypti . However , these studies often document only the mean distance travelled ( MDT ) [8–12] or range [11 , 12] of dispersal of the released insects . Common measures of range are the flight range 50% and flight range 90% ( FR50 and FR90 respectively ) which are estimates of the distance within which 50% or 90% of all insects are expected to disperse [12 , 13] . The MDT and flight range are intuitive summary measures but do not characterise dispersal well when the distribution of dispersal distances is positively skewed with a long tail . For greater insight , a better understanding of the distribution of dispersal distances can be obtained by incorporating dispersal kernel theory , popular in studies of population spread [14] and seed dispersal [15] , into a generalised linear model ( GLM ) framework [16 , 17] . Dispersal kernels represent the distribution of dispersal distances over the whole flight range . They can take a wide range of forms with the flexibility to represent dispersal for a diverse range of species [18] . This study attempts improve the characterisation of the dispersal ability of ‘genetically sterile’ male Ae . aegypti mosquitoes using data from large-scale MRR experiments carried out at an urban field site in Brazil . The analysis facilitates the quantification of dispersal through the parameterisation of a dispersal kernel for the released insects . Many summary measures of interest relating to dispersal may be drawn from such a kernel . To enable a comparison of both the biological outcomes and methodology employed , the analytical methods are also used to re-analyse published data on the dispersal of ‘genetically sterile’ male Ae . aegypti at an uninhabited forested site in Pahang , Malaysia [12] . The dispersal ability of the ‘genetically sterile’ insects was previously analysed in the Malaysian study using methods detailed in Morris et al . ( 1991 ) [13] and evaluated the MDT to be 52 . 4m ( 95% CI: 41 . 6m , 61 . 4m ) [12] . The aims of the re-assessment of dispersal ability are: i ) to assess the robustness of the estimate of dispersal from Brazil data to habitat and locational heterogeneities , ii ) to explore potential differences in dispersal behaviour between sites and iii ) to assess the applicability of the dispersal kernel method in comparison with more common approaches to estimating and quantifying dispersal .
Before establishment of the ‘genetically sterile’ male Ae aegypti line in the mass rearing laboratory and subsequent open releases , regulatory approvals were obtained from the appropriate Brazilian national regulatory body: the Brazilian National Biosafety Technical Commission ( CTNBio ) . Releases were preceded by community engagement with consent and support obtained from regional ( Bahia health secretary ) and local community leaders ( Town Mayor , health secretary and vector control authorities ) . Prior to sampling , informed consent was received from the landowners . The field site is located in Itaberaba , a suburb of the city of Juazeiro , Bahia , Brazil ( Latitude: -9° 26' 59" , Longitude: -40° 28' 53" ) ( Fig 1A ) . The site is located in a semi-arid part of Brazil and consists mainly of low-socioeconomic status residential housing . The majority of houses at the study site were single-story brick and concrete buildings with unscreened windows . The habitat across the sampled region was a homogenous urban environment . The ‘genetically sterile’ line used was OX513A and was reared according to methodologies given in Carvalho et al . ( 2014 ) [19] . A total of 19 , 164 ‘genetically sterile’ male Ae . aegypti formed three releases . Individuals from each release were marked with the same coloured fluorescent powder ( www . dayglo . com ) . Release 1 ( red release ) and release 2 ( blue release ) were performed on 21 February 2011 and consisted of 5 , 349 and 5 , 465 individuals released from points 1 and 2 ( Fig 1B ) respectively . Release 3 ( yellow release ) was performed on 25 February 2011 with 8 , 350 individuals released from point 1 ( Fig 1B ) . Aspiration sampling was used to recapture marked adults . Sampling was conducted using locally made battery powered hand-held aspirators . After obtaining consent from the respective property owner , each building was sampled for a set period of 15 minutes . Sampling locations were distributed across the study site ( Fig 1B ) and sampling was conducted for up to nine days post release ( S1 Data ) . Sampling locations were chosen by randomly selecting one household from each of the 47 ( 60m by 60m ) grid squares each day ( with the exception of day 5 for the red and blue release and day 1 for the yellow release where , for logistical reasons , the number of households sampled was lower ) . Each mosquito collected was assessed to determine the i ) origin ( ‘genetically sterile’ or wild as indicated by the presence/absence of fluorescent powder respectively ) , ii ) sex and iii ) genus ( Aedes or non-Aedes ) . Weather variables ( daily maximum temperature and maximum humidity ) were recorded from a local weather station ( situated approximately 10 . 2km north-west of Itaberaba ) . A secondary analysis of the MRR data from a previously published study in Malaysia [12] was undertaken to obtain a comparative second estimate of the ‘genetically sterile’ insect’s dispersal ability and associated density kernel . In this study a total of 6 , 045 marked ‘genetically sterile’ males were released at a forested site in Pahang , Malaysia . For a detailed description of the study site and MRR methods please see Lacroix et al . ( 2012 ) [12] . All multivariable analyses were performed within a GLM framework . In instances where the number of recaptures is small relative to the total releases , the Poisson regression model may be used as an appropriate approximation [20] . We assume that the count response variable ( recaptures ) is Poisson distributed with mean μ and variance μ Yi∼Pois ( μi ) . ( 1 ) The response must be ≥ 0 . Therefore a log link function is used to link the mean to the explanatory variables ln ( μi ) =xiβ , ( 2 ) where xi β is a linear predictor xiβ=β0+β1xi1+…+βpxip , ( 3 ) where β denotes the unknown parameters to be estimated and xi , the explanatory variables . Parameter estimates were obtained by maximising the log-likelihood ( l ) of the data: l ( β|Y ) =∑i=1n ( yilnμi−μi−ln ( y ! ) ) =∑i=1n ( yixiβ−exp ( xiβ ) −lnyi ! ) . ( 4 ) In the situation where the response variable is overdispersed ( variance>>mean ) a Poisson GLM , where the variance is assumed to equal the mean , would be misspecified . In this instance , the negative binomial GLM , detailed below , may be used [21–23] . We assume that the count response variable follows a negative binomial distribution . A Poisson model is used for the count , conditional on the mean value , Zi , which is assumed to have a gamma distribution , with mean , μi , and constant scale parameter , θ Yi∼Poisson ( Zi ) , Zi∼gamma ( μi , θ ) . ( 5 ) Therefore the expected value of Y and the variance of Y are as follows E ( Yi ) =μi , Var ( Yi ) =μi+μi2θ ( 6 ) The mean response , μi , may be linked to a linear combination of explanatory variables using the log link function ( Eq ( 2 ) ) and linear predictor ( Eq ( 3 ) ) . Parameters were estimated by maximising the log-likelihood ( l ) of the model: l ( β , θ|Y ) =∑i=1i=nθ ( ln ( θ ) −ln ( θ+μi ) ) +ln ( Γ ( θ+yi ) ) −ln ( yi ! Γ ( θ ) ) −yi ( ln ( μi ) −ln ( θ+μi ) ) , ( 7 ) where the gamma function , Γ , is Γ ( n ) = ( n−1 ) ! ( 8 ) Considerable inconsistencies abound regarding different interpretations of the term ‘dispersal kernel’ [18 , 24 , 25] . Two kernel definitions , often used interchangeably , are i ) the probability density function ( pdf ) of the dispersal distance of each disperser . Referred to as the distance kernel [18] or the distance pdf [25] and ii ) the density of probability of a given bearing and dispersal distance from the source . Referred to as the location kernel [18] or the density pdf [25] . We adopt the terminology of Cousens et al . [25] , henceforth referring to kernel type 1 as the distance pdf and type 2 as the density pdf . Both kernel types are true pdfs , integrating to 1 ( the density pdf is integrated over the whole 2d space ) . Both kernel types are closely related . The distance pdf can be derived by multiplying the density pdf by 2πd where d is the distance from the source [18] ( assuming radial symmetry ) . Examples of these kernel types are shown in Fig 2 . The density pdfs , assuming radial symmetry , used in this analysis are defined by the following functions Negativeexponentialkernel=12πa2e ( −da ) a>0 , ( 9 ) Exponentialpowerkernel=b2πa2Γ2be ( −dbab ) a , b>0 , ( 10 ) where d is the distance ( metres ) , a and b are kernel parameters and Γ the gamma function ( Eq ( 8 ) ) [18] . The negative exponential kernel is characterised by the exponential power kernel when b = 1 . The associated MDT functions are NegativeexponentialMDT=2a ( 11 ) ExponentialpowerMDT=a ( Γ3bΓ2b ) . ( 12 ) Estimates of FR50 and FR90 are made by assessing the cumulative distribution of the distance pdf at the 50% and 90% levels . The outcome variable was the number ( count ) of marked ‘genetically sterile’ male Ae . aegypti recaptured . Potential explanatory variables included in the Brazil analysis were: i ) a spatial measure which could either be the measured distance ( m ) between release and recapture or the density as calculated by a parameterisation of a given density pdf , ii ) the number of days post release , the effect of which is assumed to be linear , iii ) the number of wild Aedes species collected iv ) the number of non-Aedes wild mosquitoes collected , v ) the maximum temperature ( °C ) on the day of collection , vi ) the maximum humidity ( % ) on the day of collection and vii ) the directional quadrant , North , South , East or West ( relative to release point ) that the collection was made in . Due to the relatively low recapture number , data from all three MRR experiments were combined for analysis . For the Malaysia analysis the outcome variable was the number ( count ) of marked ‘genetically sterile’ male Ae . aegypti recaptured . Potential explanatory variables included in the analysis were: i ) a spatial measure which could either be the measured distance ( m ) between release and recapture or the density as calculated by a parameterisation of a given density pdf , ii ) the number of days post release , the effect of which is assumed to be linear iii ) the number of wild Aedes species ( specifically: aegypti , albopictus and togoi ) collected , iv ) the number of wild Culex collected and v ) a categorical variable indicating if the recapture location was uphill or downhill from the release site . Three models were evaluated to compare different transformations of the distance explanatory variable . Model 1 incorporated all explanatory variable including distance , Model 2 all explanatory variables including distance density ( negative exponential kernel ) and model 3 all explanatory variable including distance density ( exponential power kernel ) . For model 1 the full model was fitted using maximum likelihood techniques , utilising the GLM and Negative binomial GLM function of the statistical software package R [26] with the MASS package [21] . All explanatory variables were included in the initial model as well as an interaction term between distance and day post release . Model selection by minimising the Akaike information criterion ( AIC ) , a penalised likelihood score , was then performed using the MASS package [21] . The AIC is calculated as AIC=2k−2ln ( L ) ( 13 ) where k is the number of parameters and L the maximised likelihood value . For models 2 and 3 fitting was performed using the following process . First , the distance density was estimated using the assigned kernel . The GLM was then fitted using the same process as for model 1 , as a function of the transformed distance explanatory variable . These steps were then optimised over the kernel parameter space allowing identification of the optimal combination of explanatory variables and kernel parameters as indicated by the AIC . The best overall model was judged to be the one with the minimum AIC value . For comparison , the estimated survival of released insects in the Brazil study predicted using the GLM was also calculated using a non-linear regression approach [27] , that was also used in the original analysis of the Malaysia data [12] . Following model estimation , 95% confidence intervals were calculated for the maximum likelihood kernel parameter estimates using the profile likelihood method . The maximised log-likelihood with respect to β , α and b , i . e . that corresponding to the maximum likelihood estimates ( MLEs ) of β , α and b , is defined as l ( β^ , a^ , b^|Y ) . First kernel parameter a was increased or decreased in small increments whilst β was held at the MLE ( β^ ) and kernel parameter b was optimized conditional on β^ and the assumed value of α ( a0 ) , giving the log-likelihood: l ( β^ , a0 , b˜|Y ) , ( 14 ) where b˜=MLE ( b|a0 , β^ ) . Secondly , kernel parameter b was increased or decreased in small increments whilst β was held at the MLE ( β^ ) and kernel parameter α was optimized conditional on β^ and the assumed value of b ( b0 ) , giving the log-likelihood: l ( β^ , a˜ , b0|Y ) , ( 15 ) where a˜=MLE ( a|b0 , β^ ) . After each change the log-likelihood of the model was recalculated and a corresponding test statistic assessed . For example , evaluating for a , the G2 statistic was calculated G2=2 ( l ( β^ , a^ , b^|Y ) −l ( β^ , a0 , b˜|Y ) ) . ( 16 ) The G2 statistic was compared to the χ2 distribution ( with 1 degree of freedom ) for the ( 1-α ) percentile . Thus for 95% confidence intervals the critical G2 value is 3 . 84 . The log-likelihood surface was calculated with respect to kernel parameters of the optimal model for exploration and visualisation of the parameter space for both the Brazil and Malaysia analyses .
Recaptures for the three MRR experiments are summarised in Table 1 . The locations of recaptures are shown in Fig 3 . The mean count of recaptured marked males ( the response ) , per sample , per day was 0 . 077 ( variance = 0 . 73 ) . Over the recapture period the maximum daily temperature ranged between 25 . 4°C-34 . 6°C and the maximum relative humidity between 66%-92% . A summary of model performance using the untransformed- and transformed-distance explanatory variable is shown in Table 2 . Combining all available data ( from the red , yellow and blue releases ) the best fitting model ( lowest AIC ) incorporated the exponential power kernel . The maximum likelihood exponential power density pdf has an associated MDT of 52 . 8m ( 95% CI: 49 . 9m , 56 . 8m ) , FR50 of 52 . 4m ( 95% CI: 50 . 6m , 54 . 7m ) and FR90 of 83 . 0m ( 95% CI: 74 . 8m , 93 . 9m ) . The MLE kernel , log-likelihood surface and examples of kernels drawn from 95% CI parameter values are shown in Fig 4 . For all three models the distance or distance density and the number of days post release were strongly associated with recapture number . There was no evidence of an interaction between distance ( untransformed or transformed ) and the number of days post release in any of the models considered . Assuming no emigration , the estimated mortality rate of released insects of 0 . 62 ( 95% profile likelihood CI: 0 . 84 , 0 . 42 ) would equate to a mean average lifespan of 0 . 62−1 = 1 . 61 days ( 95% CI: 1 . 19 days , 2 . 38 days ) . Estimates of the mean average lifespan calculated using the non-linear regression approach [27] were 1 . 00 days ( 95% bootstrapped CI: 0 . 63 days , 1 . 48 days ) . Other significant explanatory variables were the number of non-Aedes mosquitoes recorded from the sample and the maximum humidity . There was evidence of a lack of radial symmetry in dispersal from the release point as the quadrant explanatory variable was also associated with recapture number . A summary of the parameter estimates from the optimal model , using the exponential power transformation of distance as an explanatory variable is shown in Table 3 . The MRR performed in Malaysia was associated with consistently higher recaptures than the MRR experiments in Brazil . Of 6 , 045 released ‘genetically sterile’ males 3 , 034 ( 50 . 2% ) were recaptured over the 15-day course of the experiment . The count of recaptured , marked males ( the response ) was very overdispersed ( mean = 2 . 6 per sample per day , variance = 523 ) , and therefore a negative binomial GLM was fitted . A summary of the model performance using the untransformed and transformed distance explanatory variable is shown in Table 4 . Again , the optimal model , as determined by AIC , used the exponential power density pdf , although the negative exponential density pdf produced only marginally inferior fit ( AIC = 668 . 0 and 668 . 4 for the exponential power and negative exponential kernels respectively ) . The MLE exponential power pdf estimates a MDT for the ‘genetically sterile’ release of 58 . 0m ( 95% CI: 51 . 1m , 71 . 0m ) , FR50 of 51 . 8m ( 95% CI: 47 . 9m , 58 . 7m ) and FR90 of 105 . 7m ( 95% CI: 86 . 5m , 141 . 1m ) . The MLE kernel , log-likelihood surface and examples of kernels drawn from 95% CI parameter values are shown in Fig 5 . The coefficient summaries from the negative binomial model using the exponential power transformed distance explanatory variable are shown in Table 5 . This second analysis again indicates that distance is an important significant predictor of the expected count of recaptures . The number of days post release was also significantly associated with recapture number . Assuming no emigration , the estimated mortality rate of released insects of 0 . 46 ( 95% profile likelihood CI: 1 . 03 , 0 . 13 ) would equate to a mean average lifespan of 0 . 46−1 = 2 . 17 days ( 95% CI: 0 . 97 days , 8 . 85 days ) . Unlike the analysis for Brazil there was evidence of an association between the distance and the number of days post release explanatory variables . For a direct comparison the distance pdf and density with respect to distance for the optimal kernels estimated from the Itaberaba and Malaysia MRR experiments have been overlaid ( Fig 6 ) .
An in-depth analysis of the dispersal ability of released ‘genetically sterile’ male Ae . aegypti mosquitoes in the field has been conducted . The primary analysis , of MRR data from Brazil , indicates distance from the release point to be an important predictor of the expected number of recaptures . The relationship between the recapture number and the distance from the release point is highly non-linear . The regression model performed optimally when an exponential power dispersal kernel was used to transform distances . The analysis methodology was also used to re-analyse MRR from Malaysia , where again an exponential power kernel provided best model fit . The optimal Brazil GLM performed well , explaining around half of the variation observed in the data . Transformed distance and the number of days post release were the most influential , highly significant predictors of recapture number . The decline in numbers temporally after release is considered to be predominantly due to the effect of mortality . The effect of emigration from the study area can be important [28] and would further reduce numbers but was assumed to be small due to the size of the study area and low recaptures at the periphery . Distance was highly significantly correlated with recapture number in all models considered . The exponential power dispersal kernel provided the optimum model fit . This kernel parametric form is slightly more flexible than the negative exponential . The kernel produced showed a high and consistent level of dispersal from 0-33m from the release site . After this point the density falls fairly steeply , reaching very low levels shortly after 100m ( FR90 = 83 . 0m ) , indicating that coverage decreases quickly at increasing distances more than 33m from a release point . The MDT estimated using the best-fit kernel parameters was 52 . 8m ( 95% CI: 49 . 9m , 56 . 8m ) . This is consistent with a number of published field studies of male Ae . aegypti dispersal which estimate mean distance travelled to range from 10m to 100m ( Table 6 ) . There are however , a limited number of studies of male Ae . aegypti dispersal as focus has been on the biting females . It is however , important to note that for skewed distributions of dispersal distances the MDT as a measure of central tendency should be interpreted with some caution . The number of non-Aedes mosquitoes was significantly positively correlated with the number of recaptured ‘genetically sterile’ mosquitoes . This explanatory variable is thought to be a proxy for the house-attractiveness or accessibility of a house to mosquitoes; large numbers of other mosquito species may indicate that the household is situated in a favourable location or particularly amenable or attractive to mosquitoes . Clustering of Aedes mosquitoes at the household level is a commonly observed phenomenon [29–31] . An alternative explanation could also lie in differences in the abilities of operators to sample mosquitoes . The humidity and directional quadrants were significantly associated with the number of recaptured ‘genetically sterile’ mosquitoes but had small effect sizes . It would be expected that any directional differences are attributable to site- and time-specific heterogeneities in terrain , habitat type , wind direction or other external factors [32–34] . Humidity was positively associated with recapture number , which could be due to an increased tendency for Ae . aegypti to seek shelter with increasing humidity [35] . These explanatory variables must be interpreted with some caution as there is the potential for selection by AIC to overfit models [36 , 37] . The covariates may therefore be included in the final models despite their relatively small influence on model fit . One limitation of this study was the low number of recaptures in the Brazil dataset . For this reason there was little power to analyse individual releases separately , necessitating the analysis of the combined datasets and the assumption that influences not accounted for would be similar across all three releases . The proportion of individuals recaptured may have been improved by a more targeted , or higher intensity , sampling effort or increased survival of the released individuals . Alternatively , the absolute number of recaptures could have been increased with larger release numbers . We have assumed throughout that aspiration will not remove all fluorescent dust from the marked individuals [38] . The dataset could also have been further improved with household-specific monitoring of climatic factors to give greater resolution to observations on the relationship between meteorological variables and recapture numbers . The standard errors of the GLM coefficients ( Tables 3 and 5 ) were estimated conditional upon the MLEs for kernel parameters a and b whilst the confidence intervals for the kernel parameters were obtained conditional upon the MLE for the GLM coefficients . Thus the reported standard errors are likely to be smaller than if we had been able to compute the unconditional standard errors and confidence intervals for all of the parameters . Analysis of the residuals indicated little residual spatial autocorrelation with perhaps the exception of some under-estimation of recapture numbers at further distances ( >150 m ) , potentially indicating the influence of long-distance dispersers [39] , although , due to the small number of recaptures , this is difficult to verify . The optimal GLM associated with the Malaysia data also explained approximately 50% of the variation observed in the recapture data . For the optimal model only two covariates , number of days post release and distance , plus their interaction term were included . The optimal model again used the exponential power dispersal kernel , with a corresponding MDT estimate of 58m ( 95% CI: 51 . 1m , 71 . 0m ) that corroborated the MDT estimate of 52 . 4m ( 95% CI: 41 . 6m , 61 . 4m ) from the previously published analysis of these data [12] . The FR50 estimate of 51 . 8m ( 95% CI: 47 . 9m , 58 . 7m ) was substantially different from the previously published estimates of 16 . 2m ( 95% CI: 10 . 5m , 22 . 5m ) , a product of the different underlying models for dispersal with respect to distance used in each analysis . This deviation further highlights the potential benefits of more accurately characterising dispersal behaviour . The number of days post release covariate , as expected , was significantly negatively associated with recapture number . The associated average lifespan of 2 . 17 days ( 95% CI: 0 . 97 days , 8 . 85 days ) was in good agreement with the previously published estimate of 2 days ( 95% CI: 1 . 8 days , 2 . 2 days ) . The coefficient indicated a smaller effect size than seen in the Brazil MRR data , implying improved survival , less emigration or a combination of the two for individuals in the Malaysia releases . The recapture rate in the Malaysia MRR experiment was very high , approximately 50% of all individuals released were recaptured . This may bias the results and could violate the underlying assumption that the negative binomial distribution approximates proportions when recapture numbers are small relative to the release size . The lack of significant interaction between the number of days post release and distance in the Brazil data provides evidence for a single main dispersal event on release ( the probability of travelling a given distance is not influenced by the number of days post-release ) . In the Malaysia analysis the significant interaction term indicates a more continual dispersal process over time , possibly due to the lack of favourable ( urban/peri-urban ) habitat across the whole range of the study site at this location . However the influence of the interaction term on predicted recaptures is very small; the majority of released individuals have died ( or emigrated ) before the interaction term becomes influential . For both locations the majority of recaptures are predicted spatially and temporally close to the release location and date respectively . There is published evidence to support either the occurrence of a single dispersal event [40–42] or a more continuous dispersal process upon release [8 , 9 , 11] . For experiments carried out in different habitats , on different continents , the estimated dispersal kernels were very similar ( Fig 6 ) . The Brazil dispersal kernel is slightly fatter tailed ( larger b parameter ) but in general there is evidence for a degree of consistency in the dispersal ability of ‘genetically sterile’ male Ae . aegypti across a range of environments . Consistent dispersal may facilitate more generalised release procedures for sterile insect releases across a range of release locations and scenarios . Accurately measuring and assessing the dispersal of released ‘genetically sterile’ male Ae . aegypti in the field is a vital component necessary to optimise vector control using these genetically sterile individuals . A successful control program using ‘genetically sterile’ male Ae . aegypti would maximise ‘genetically sterile’ insect density over the target area . Knowledge of the released insects’ ability to disperse is vital in predicting their density with respect to specific release points or routes . An ability to predict the coverage of dispersed individuals will facilitate the design and implementation of more efficient control and monitoring programs in the future . | Vector control using releases of sterile insects is a well-known approach . ‘Genetically sterile’ male Aedes aegypti have been developed and released in a modern realisation of the sterile insect technique . Released engineered males seek out and mate with wild females , with the resultant offspring dying before they reach maturity . Control of a wild vector population can therefore be achieved by maintaining sustained releases of sterile males whilst ensuring sufficient distribution and coverage of released males across the target area . In order to efficiently plan releases of these , individuals’ detailed knowledge of how they disperse in the field is required . We present an analysis of the dispersal of these engineered male Aedes aegypti using data from field experiments in Brazil . Our results provide detailed information on how the mosquitoes disperse over their potential flight range . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [] | 2015 | Dispersal of Engineered Male Aedes aegypti Mosquitoes |
Mitotic and cytokinetic processes harness cell machinery to drive chromosomal segregation and the physical separation of dividing cells . Here , we investigate the functional requirements for exocyst complex function during cell division in vivo , and demonstrate a common mechanism that directs anaphase cell elongation and cleavage furrow progression during cell division . We show that onion rings ( onr ) and funnel cakes ( fun ) encode the Drosophila homologs of the Exo84 and Sec8 exocyst subunits , respectively . In onr and fun mutant cells , contractile ring proteins are recruited to the equatorial region of dividing spermatocytes . However , cytokinesis is disrupted early in furrow ingression , leading to cytokinesis failure . We use high temporal and spatial resolution confocal imaging with automated computational analysis to quantitatively compare wild-type versus onr and fun mutant cells . These results demonstrate that anaphase cell elongation is grossly disrupted in cells that are compromised in exocyst complex function . Additionally , we observe that the increase in cell surface area in wild type peaks a few minutes into cytokinesis , and that onr and fun mutant cells have a greatly reduced rate of surface area growth specifically during cell division . Analysis by transmission electron microscopy reveals a massive build-up of cytoplasmic astral membrane and loss of normal Golgi architecture in onr and fun spermatocytes , suggesting that exocyst complex is required for proper vesicular trafficking through these compartments . Moreover , recruitment of the small GTPase Rab11 and the PITP Giotto to the cleavage site depends on wild-type function of the exocyst subunits Exo84 and Sec8 . Finally , we show that the exocyst subunit Sec5 coimmunoprecipitates with Rab11 . Our results are consistent with the exocyst complex mediating an essential , coordinated increase in cell surface area that potentiates anaphase cell elongation and cleavage furrow ingression .
Cytokinesis results in the physical separation of two daughter cells . Immediately prior to the initiation of cytokinesis , cells also begin to elongate along the spindle axis , concomitant with the anaphase spindle elongation that helps drive chromosomal separation . To achieve such a fundamental remodeling of shape and topology , cells martial multiple cytoskeletal and membrane trafficking pathways . Contraction of an equatorial actomyosin ring is required for inward progression of the cleavage furrow , and a further abscission process operates to fully separate the incipient daughter cells into two distinct membranous structures . In addition , processes that regulate membrane trafficking events are also required for successful cytokinesis [1–3] . Previous studies demonstrated that Drosophila male meiotic cells represent a sensitive system for identification of cellular components that contribute to cytokinesis [4] . Genes that regulate central spindle function , contractile ring assembly , phosphoinositide composition , and exocytic trafficking have all been identified through mutations that disrupt male germline cytokinesis . Trafficking proteins that are required for cytokinesis include the Conserved Oligomeric Golgi Complex complex ( COG ) subunits Cog5 and Cog7 , the Rab11 GTPase , the Syntaxin 5 ER-to-Golgi vesicle-docking protein , the endosomal Arf6 GTPase , the phosphatidylinositol 4-kinase IIIβ Four Wheel Drive ( Fwd ) , the TRAPPII complex subunit Brunelleschi , and phosphatidylinositol 4-phosphate [PI ( 4 ) P] effector GOLPH3 [5–14] . However , the final proteins in these exocytic pathways that may direct membrane addition at the cell surface have remained unidentified . Spatial specificity of vesicle trafficking occurs through the targeting of exocytic vesicles at defined membrane sites by tethering complexes such as the exocyst complex [15 , 16] . The eight subunits of the exocyst ( Sec3 , Sec5 , Sec6 , Sec8 , Sec10 , Sec15 , Exo70 , and Exo84 ) were originally identified based on their role in polarized secretion in Saccharomyces cerevisiae [17] and were subsequently shown to form a complex that is highly conserved from yeast to mammals [18–23] . We have previously demonstrated that the Exo84 subunit of the exocyst complex mediates apical epithelial identity in Drosophila [24] . Other groups have shown that members of the Drosophila exocyst are required for membrane addition and expansion in developing oocytes and neurons , in photoreceptor cells and during embryonic cellularization [25–31] . Additionally , the exocyst complex has been shown to be required for cell abscission at the end of cytokinesis in mammalian tissue culture cells [32–35] . Here , we demonstrate that funnel cakes ( fun ) and onion rings ( onr ) encode the exocyst proteins Sec8 and Exo84 , respectively . We show that dividing spermatocytes mutant for either onr or fun display an exceptionally early defect in progression of the cleavage furrow and fail to accumulate Rab11 and Giotto/Vibrator at the cell midzone . Quantitative analysis suggests that rather than disrupting gross membrane addition to the cell surface , these mutations specifically affect a trafficking pathway required for both anaphase cell elongation and cleavage furrow ingression .
fun and onr were identified in a screen for mutations that disrupt cytokinetic events in male germline cells [4] . Previous characterization of fun and onr revealed that these mutations do not affect central spindle or F-actin ring formation in dividing spermatocytes . Nonetheless , in fun and onr mutants , cytokinesis fails at an early stage [4] . The funz1010 mutation was mapped to the 83C1;83C4 interval on chromosome III in the region of the Sec8 gene . Deficiency mapping revealed that funz1010 failed to complement Df ( 3R ) Exel6145 for the male sterility and cytokinesis defects ( Fig 1A and 1B ) . Two lines of evidence indicate that funz1010 is an allele of Drosophila Sec8 , which encodes a protein with 35% identity to human and mouse Sec8 proteins and 19% identity to the S . cerevisiae Sec8 protein ( S1 Fig ) . First , a 6 . 6 kb genomic transgene containing the predicted Sec8 coding region , 1 . 0 kb of upstream promoter sequence , and 1 . 9 kb of downstream sequence fully rescued the cytokinesis defects in fun mutant male germline cells ( Fig 1B and 1D ) . Indeed , 100% of onion-stage spermatids from funz1010/Df ( 3R ) Exel6145 males bearing a single copy of the rescuing transgene possess a wild type 1:1 ratio of nuclei to nebenkern ( n = 102 ) , compared with 0 . 8% in males of identical genotype devoid of the transgene ( n = 125 ) . Additionally , DNA sequencing of the Sec8 gene in funz1010 mutant males revealed a C to T mutation resulting in replacement of a conserved Serine residue by Phenylalanine at position 322 of the predicted 985 amino acid polypeptide ( S1 Fig ) . Together , these results provide clear evidence that funz1010 represents a mutation in the Sec8 gene . Remarkably , while funz1010 disrupted functioning of the Sec8 exocyst subunit , the onr mutation from the same phenotypic class of mutants [4] was previously shown to affect the Exo84 exocyst subunit [24] . In short , the onrz4840 allele possesses a nonsense mutation that is predicted to generate a truncated protein containing 581 of 672 amino acids [24] . Consistent with this , a 4 . 5 kb genomic transgene containing the predicted Exo84 coding region , 1 . 5 kb of upstream promoter sequence , and 1 kb of downstream sequence fully rescued cytokinesis defects in onr mutant male germline cells ( Fig 1C and 1E; 98 . 2% of onion-stage spermatids from onrz4840/Df ( 3R ) Espl3 hemizygous males bearing a single copy of the rescuing transgene exhibit a wild-type 1:1 ratio of nuclei to nebenkern ( n = 112 ) , compared to 0% in onr hemizygotes devoid of the transgene ( n = 101 ) . Localization of Sec8 protein was analyzed in primary spermatocytes from larval testes fixed with either methanol-free formaldehyde ( Fig 2A ) or methanol/formaldehyde ( S2 Fig ) . Staining of interphase primary spermatocytes with anti-Tubulin and anti-Sec8 antibodies revealed that Sec8 protein was diffuse throughout the cytoplasm and enriched at the plasma membrane ( Fig 2A ) . In dividing spermatocytes , in addition to localization at the plasma membrane , Sec8 was enriched in a broad cortical area at the cell equator and excluded from the poles ( Fig 2A ) . During mid-telophase and late telophase , Sec8 protein accumulated at the cortex , near the ingressing furrow membrane ( Fig 2A ) . Analysis of larval testes from transgenic animals expressing a GFP-Exo84 fusion protein revealed that , similar to Sec8 , Exo84 appeared diffuse in the cytoplasm during interphase and became enriched in the furrow region during early telophase ( S3 Fig ) . Analysis of dividing cells stained for Tubulin and Drosophila Sec5 revealed that Sec5 was enriched in small puncta at the astral microtubules and concentrated at the furrow region in telophase ( Fig 2B ) . Previous data showed that onr and fun mutations exhibited normal F-actin ring formation and central spindle assembly in dividing spermatocytes [4] . However , in mid to late telophase spermatocytes from onr and fun mutants , F-actin rings appeared poorly constricted and the central spindles were less dense than in wild type . Imaging of wild-type primary spermatocytes expressing myosin II regulatory light chain fused to GFP ( Sqh-GFP , [5] ) revealed that dividing spermatocytes ( n = 9 ) assembled Sqh-GFP rings during anaphase that underwent full constriction within 20 minutes ( Fig 3A and 3B; S1 Movie ) . In contrast , in dividing spermatocytes from either funz1010/Df ( 3R ) Exel6145 ( n = 8 ) or onrz4840/Df ( 3R ) Espl1 ( n = 8 ) , Sqh-GFP rings underwent minimal constriction accompanied by furrow regression and contractile ring rupture during the time of observation ( Fig 3A and 3B and S2 Movie ) . In examining onr and fun mutant cells , we observed that dividing spermatocytes did not appear to lengthen along the spindle axis as much as wild-type cells do prior to cytokinesis . This elongation during anaphase may identify a time when a critical increase in surface area is initiated . To examine this quantitatively in an unbiased fashion , we developed a computational approach to segment cell boundaries and volumes . Dividing primary spermatocytes from wild-type and mutant males expressing PLCδ-PH-GFP [37] , a plasma membrane marker , and β-Tub-GFP [38] , a spindle and microtubule marker , were imaged by spinning disc microscopy ( Fig 4A–4C ) . Image sets were acquired with XY resolutions of 0 . 166 microns per pixel and a Z-layer spacing of 1 micron every 60 seconds . Cells were then segmented using an automated 3D seeded watershed algorithm ( Fig 4D–4F; S4–S6 Movies ) . From these voxelized representations of the cells , we computed a number of parameters that describe cellular geometries as male germline cells divide . Cell volume was computed as the sum of the voxel volumes , while surface area was computed as the sum of the areas of the exposed voxel surfaces . To quantify ingression of the furrow , we used the convex hull volume ratio ( CHVR ) . For a set of points in 3D space , the convex hull is the smallest convex spatial body spanned by a subset of the points that contains all the points of the set , i . e . , the smallest convex envelope . The CHVR is defined as the convex hull volume divided by the actual segmented volume ( schematically depicted in S4F Fig ) . By definition , the convex hull volume can be greater or equal to the actual volume , greater when concavities are present and equal when fully convex . Thus for an ellipsoid or sphere the CHVR = 1 . For an idealized example of two perfect equal sized spheres touching at a point the CHVR = 1 . 25 . Therefore , the CHVR provides a quantitative global measure of the amount of ingression . The behaviors of wild-type cells were very consistent ( Fig 4G–4J ) . Wild-type volume did not change significantly during cytokinesis ( Fig 4H , p = 0 . 5297 when comparing wild-type cells at t = 0 to t = 25 min ) . Cytokinesis is therefore dependent on an increase in surface area . For the idealized geometry of a sphere dividing into two spheres of half the volume , the increase in surface area is approximately 26% . Our wild-type data are in good agreement with this percentage increase ( 26 . 1% ) , and the peak rate of increase is approximately 63 μm2/min . The average aspect ratio increased by just over 51 . 8% , and the average CHVR increased by 23% over the course of 25 minutes ( Fig 4I and 4J ) . In contrast , onrz4840 mutant cells had a brief period where surface area temporarily increased at a peak rate of 5 . 0 μm2/min , and surface area increased by 1 . 3% over 25 minutes ( Fig 4G ) . In funz1010 mutant cells , the peak rate of surface area increase was 3 . 0 μm2/min , a rate similar to onr mutants but over 20 times slower than wild type , and the total percent increase over 25 minutes was 2 . 0% ( Fig 4G ) . Intriguingly , cell volume and surface area were nearly identical in wild type , onr mutant , and fun mutant cells prior to the start cell division , suggesting that there is not a general blockade of plasma membrane trafficking in onr and fun mutants ( Fig 4G and 4H ) . This also further suggests that directed trafficking specifically during anaphase cell elongation and cytokinesis may be an essential mediator of cell shape change . An essential requirement for onr and fun function during anaphase cell elongation and cytokinesis can also be observed by directly examining the aspect ratio and the CHVR in these two mutants . In both mutants , the aspect ratio initially displayed a slight increase but peaked at 1 . 5 in onrz4840 mutants and at 1 . 4 in funz1010 mutants before it then started to decline ( as compared to 2 . 4 in wild-type cells ) . Similarly , cleavage furrow progression was disrupted in onr and fun mutant cells . Intriguingly , ingression of the cleavage furrow failed almost immediately in spermatocytes lacking onr or fun function ( Fig 4B” , 4C” and 4J ) . During this process , the average CHVR reached a peak of 1 . 015 in onr mutants . Thus , on average , the volume of the ingression furrow was at most 1 . 5% of the cell volume . In fun mutant cells the CHVR peaked at 1 . 018 . These results suggest that , in vivo , Exo84 and Sec8 function is required for a core set of cell shape changes that occur during cell division . Several mutations in membrane trafficking components have been shown to disrupt the structure and/or the number of Golgi stacks in interphase primary spermatocytes [12 , 13 , 39] . To test whether onr and fun are required for Golgi organization in these cells , we used the Golgin Lava lamp ( Lva ) as a marker to examine the structure and distribution of the Golgi by immunofluorescence [40] . This analysis revealed defects in both the size and the number of Golgi stacks in onr and fun mutants ( Fig 5A and 5B ) . Since surface area addition was defective in onr and fun mutant cells , and Golgi architecture was also disrupted , we analyzed the ultrastructure of spermatocyte cells by transmission electron microscopy ( TEM ) to determine if internal membrane compartments are altered . Intriguingly , onr and fun mutant cells displayed large accumulations of cytoplasmic membranes ( Fig 6B and 6C ) . Indeed , parafusorial and astral membranes appeared enlarged , fragmented and vacuolated in fun and onr mutant dividing spermatocytes ( Fig 6B and 6C , 6E–6F , 6H and 6I ) . Additionally , Golgi compartments were bloated and vacuolated when fun ( 10/14 Golgi bodies , or 71% ) or onr ( 10/10 Golgi bodies , or 100% ) functions were disrupted ( Fig 6K and 6L ) , as compared to wild type ( 1/15 Golgi bodies , or 7% ) . Moreover , the extent of cisternal stacking within the Golgi was vastly reduced and the cisternae appeared disrupted by the vacuolated regions , potentially explaining the apparent fragmentation of the Lva signal in fun and onr mutant spermatocytes . Additionally , as Lva marks cis Golgi compartments [7 , 40] , these results suggest that the expansion and bloating may preferentially affect medial or trans Golgi compartments . These results are consistent with a failure in vesicle trafficking to the cell surface required to mediate cell remodeling and elongation during anaphase and cytokinesis . As Rab11 has been shown to be essential for cytokinesis during male meiotic divisions [10] , we examined Rab11 behaviors in cells in which exocyst function has been compromised . Rab11 localization was abnormal in fun and onr mutant dividing spermatocytes ( Fig 7A ) . In wild type , Rab11 was enriched in puncta at the cell poles during anaphase and telophase ( n = 38; Fig 7A ) and accumulated at the cleavage furrow during mid-telophase . By contrast , in ana-telophase spermatocytes from funz1010/Df ( 3R ) Exel6145 mutants , Rab11 was enriched in few puncta at the cell poles and failed to concentrate into a tight band at the midzone ( Fig 7A ) . In all the telophase cells from funz1010/Df ( 3R ) Exel6145 mutants ( n = 30; Fig 7A ) , Rab11 appeared enriched in a broad midzone area . Localization of Rab11 in onrz4840/Df ( 3R ) Espl3 dividing spermatocytes also appeared diffuse at the midzone and excluded from the cell poles ( n = 27; Fig 7A and 7B ) . Localization of Rab11 was also examined in dividing spermatocytes simultaneously stained for Rab11 and the furrow membrane marker anillin ( Fig 7B ) . In wild-type telophase cells , Rab11 and anillin colocalized at the cleavage furrow ( n = 32 ) . In telophase cells from both onrz4840/Df ( 3R ) Espl3 ( n = 28 ) and funz1010/Df ( 3R ) Exel6145 ( n = 24 ) mutants , anillin and Rab11 failed to co-localize at the equatorial cortex ( Fig 7B ) . Rather , anillin formed a large ring at the equatorial cortex , consistent with defects in contractile ring constriction , and Rab11 accumulated at the midzone . In addition , onr and fun were also required for normal localization of phosphatidylinositol transfer protein Giotto/Vibrator ( Gio/Vib , [9 , 12 , 41]; Fig 8 ) . In wild-type anaphase and early telophase spermatocytes , Gio was enriched at the endoplasmic reticulum ( ER ) derived membranes that comprise the astral and parafusioral membrane arrays ( Fig 8 , [41] ) . In wild-type early ( n = 23 ) and late telophases ( n = 30 ) , Gio also concentrated at the cleavage furrow ( Fig 8 ) . In early telophases from funz1010/Df ( 3R ) Exel6145 ( n = 24 ) and onrz4840/Df ( 3R ) Espl3 ( n = 28 ) mutants , Gio was diffuse throughout the cells and failed to accumulate to the astral and parafusioral membrane arrays or to the cleavage furrow ( Fig 8 ) . Gio localization remained diffuse in late telophases from fun ( n = 26 ) and onr ( n = 28 ) , ( Fig 8 ) . The onr and fun mutants interacted genetically with Rab11 mutants . Heterozygosity for fun dramatically increased the frequency of cytokinesis failures caused by homozygosity for the weak Rab11 allele Rab1193Bi , indicating a strong genetic interaction . funz1010 Rab1193Bi/+Rab1193Bi males raised at 25°C exhibited a 7-fold increase in the percentage of multinucleate spermatids relative to testes from Rab1193Bi/Rab1193Bi single mutants ( Fig 9A and 9B ) . In addition , although Rab1193Bi and Rab1193Bi/Rab11E ( To ) 11 transheterozygotes were viable , as were funz1010/ funz1010 flies , funz1010 Rab1193Bi/ fun z1010 Rab11E ( To ) 11 double mutants died mostly at early larval stages . Examination of testes from rare escaper larvae of genotype funz1010 Rab1193Bi/ fun z1010 Rab11 E ( To ) 11 revealed that 13 . 9% of spermatids exhibited more than four nuclei per mitochondrial derivative , indicating a dramatic increase in cytokinesis failures during the gonial divisions that precede meiosis ( Fig 9B ) . Rab11 also interacted genetically with onr . onr z4840 Rab1193Bi double mutants died in early larval stages , as did individuals that were homozygous for onr z4840 and transheterozygous for Rab1193Bi/Rab11E ( To ) 11 . To test whether Rab11 associated with the exocyst complex proteins Sec8 and Exo84 encoded by onr and fun , we performed co-immunoprecipitation ( Co-IP ) experiments using testis extracts . Immunoprecipitation by GFP-trap revealed that Sec8-HA co-precipitated with GFP-Exo84 , consistent with the two proteins being subunits of the exocyst complex . Although we did not detect Rab11 in the precipitates from lysates of testes expressing GFP-Exo84 , we could demonstrate biochemical interaction between Rab11 and Sec5 when YFP-Rab11 proteins expressed in adult testes , were immunoprecipitated with antibodies against GFP ( Fig 9E ) . Sec5 co-immunoprecipitated with both YFP- tagged wild type Rab11 and Rab11Q70L proteins , but only weakly with Rab11S25N .
The evolutionarily conserved octameric exocyst complex has been proposed to tether exocytic vesicles to specific sites on the plasma membrane and to regulate the SNARE complex during vesicle fusion [17 , 42 , 43] . A role for the exocyst in cell division was originally described in both budding and fission yeast where the exocyst proteins localize at the cleavage site and are required for vesicle trafficking during cytokinesis [19 , 44] . Here we provide evidence that the exocyst complex is required for the major cell shape changes that occur in dividing animal cells during anaphase and telophase . Through automated computational analysis of live Drosophila spermatocytes , we have shown that membrane addition correlates specifically with onset of anaphase cell elongation and that membrane addition peaks during early stages of cytokinetic furrow ingression in wild-type cells . Spermatocytes carrying mutations in the Exo84 or Sec8 proteins display a greatly reduced rate of surface area growth specifically at anaphase and cytokinesis , indicating a requirement for exocyst complex function in guiding plasma membrane expansion and remodeling in dividing cells . In agreement with this hypothesis , TEM analysis of onr and fun spermatocytes showed a massive build up of cytoplasmic astral membranes in dividing cells and altered Golgi architecture in interphase primary spermatocytes , suggesting that defective vesicular trafficking through these membrane compartments may result in reduced membrane material for the surface area increase required during anaphase cell elongation and cytokinesis . Indeed , proper localization of the Rab11 GTPase and PITP Gio to the cleavage site required wild-type Exo84 and Sec8 function . In cultured mammalian cells , the exocyst is required late in cytokinesis for final resolution of the intercellular bridge [32–34] , yet Sec5 and Exo84 are enriched in the cleavage furrow during early telophase [45] . Our data provide evidence for an early requirement for the exocyst during cytokinesis . Time-lapse analysis of spermatocytes undergoing anaphase and telophase showed that fun and onr mutations did not prevent recruitment of myosin II light chain at the cell equator . However , the Sqh rings assembled in the exocyst mutants underwent minimal or no constriction and failed to mediate cleavage furrow invagination . This is consistent with our previous characterization of fun and onr mutants , which revealed defects in F-actin ring contraction [4] . Failure to assemble functional contractile rings accompanied by early cleavage furrow regression also characterize Drosophila mutants in other vesicle trafficking components , including the COG complex subunits Cog5 and Cog7 [7 , 12] , the ortholog of the yeast TRAPP II ( trafficking transport protein particle II ) TRS120p subunit [11] , the PI4K Fwd [5] , the Arf6 and Rab11GTPases [8 , 10] , and GOLPH3 [13] . Defects in myosin II rings and incomplete furrow ingression were also observed in Dyctiostelium discoideum clathrin null cells [46] . Additionally Drosophila S2 cells depleted of syntaxin 1 displayed defective actin rings [47] . These observations suggest the existence of a close interplay between contractile ring dynamics and membrane trafficking at the cleavage furrow [7 , 10 , 11 , 48] . It has been proposed that altered membrane addition at the cleavage furrow would impair plasma membrane remodeling at the furrow and physically obstruct the contraction of the actomyosin ring [9 , 10] . In addition , transport of exocytic vesicles and their fusion with the furrow membrane might also be necessary to target structural components of the contractile apparatus or factors that regulate its constriction . In agreement with this , live imaging of actin and endocytic vesicles in cellularizing Drosophila embryos has suggested a model in which F-actin and vesicles are transported as a unit to the furrow site as F-actin-associated vesicles [49] . Interestingly , several studies have reported that Rab11 protein binds to two distinct exocyst complex subunits , Sec5 and Sec15 [29 , 50–53] . We have shown that Sec5 coimmunoprecipitates with Rab11 from Drosophila testis extracts , suggesting that these proteins may form a complex in spermatocytes . Furthermore , we have demonstrated that subcellular localization of Rab11 protein depends on onr and fun and that Rab11 genetically interacts with both onr and fun . Remarkably , immunofluorescent analysis of telophase spermatocytes from fun mutants revealed that Rab11 accumulated in a broad cortical area , suggesting that Rab11-containing vesicles failed to reach the cleavage furrow plasma membrane . Together , these results indicate that exocyst complex proteins cooperate with the Rab11 GTPase in directing vesicle trafficking required for proper cytokinesis . In agreement with this idea , ultrasensitive live-imaging of fluorescently-tagged Sec8 in cultured mammalian cells revealed that this protein moves to the cell cortex on vesicles that preferentially contain Rab11 , and that Sec8 remains with these vesicles until SNARE mediated fusion at the furrow [54] . Our results also indicate that a common membrane trafficking pathway may link anaphase cell elongation and cytokinesis . Previous studies have shown a fundamental connection between cell size and the extent of anaphase elongation [55] , suggesting that limits in cell size and available surface area may dictate the degree to which elongation of the spindle at Anaphase B can occur . Our data also demonstrate that cell volume is conserved throughout anaphase and cytokinesis . This implies that , due to geometric constraints , cell surface area must increase as the cell adopts an elongated shape . Consistent with this , surface area addition fails in cells mutant for onr or fun , and anaphase cell elongation is also disrupted . A small change in aspect ratio is still observed in onr and fun mutant spermatocytes , which might indicate that a limited reservoir of excess membrane/elasticity exists in the plasma membrane at the beginning of anaphase elongation . Alternatively , this may result from residual exocyst function in the hypomorphic onr and fun alleles . Interestingly , previous work has also shown that cells with lengthened chromosomes undergo anaphase elongation to a greater degree , suggesting that there may , in turn , be an instructional cue from the spindle to the elongation machinery [56] . An additional component to anaphase elongation is the contribution of actin-dependent cortical stiffness . Recently , it has been shown that a PP1-Sds22-Moesin pathway is required for cortical polar relaxation and that excess rigidity can inhibit anaphase elongation and spindle function [57 , 58] . It therefore appears that exocyst-dependent membrane trafficking may function along with cytoskeletal regulation to direct cell elongation during division . Initiation of cleavage furrow ingression occurs within a few minutes ( 6 . 6±1 . 1 minutes , n = 8 ) of the start of anaphase elongation . This tight juxtaposition in time of both anaphase elongation and cytokinesis suggests that these two processes may be poised to take advantage of similar cell shaping and membrane trafficking mechanisms . As discussed above , the requirement for targeted membrane addition during cytokinesis is well-established [3 , 59 , 60] . The conservation of volume that we observed throughout our quantitative measurements indicates that , similar to the geometric requirements imposed on anaphase elongation , surface area must increase as the cell divides into two daughter cells . Our data support this approximate 26% predicted total increase in surface area , and illustrate that surface area addition peaks early in cytokinesis , consistent with findings from a study on Arf6 function in spermatocytes [8] . We further observed that this increase in surface area initiated at anaphase elongation and continued as cytokinesis progressed . Surface area addition was disrupted in onr and fun mutant cells and cytokinesis failed almost immediately on initiation . These results are consistent with a shared requirement for exocyst-dependent trafficking in anaphase cell elongation and cytokinesis . It may also be that essential guidance factors or components of the ingression machinery are dependent on membrane delivery to the cleavage furrow . As Rab11 has been implicated in guiding central spindle function [61] , an interesting aspect for future studies will be to further examine the relationship between central spindle function and exocyst-dependent membrane delivery in directing the profound cell shape changes that occur in cell division . It is also intriguing to note that exocyst function is required during plant cytokinesis [62 , 63 , 64] , suggesting a potentially ancient connection between membrane trafficking pathways and cell division .
A 6 , 563bp BamHI-XbaI genomic fragment was subcloned from BACR02L23 into pCasper4 . Sec8 was the only complete predicted open reading frame in this genomic fragment . Transgenic stocks expressing this transgene were crossed to funz1010 and assayed for rescue of cytokinesis defects . To generate the GFP-Exo84 construct , the EGFP coding sequence was fused in frame to the amino terminus of the full-length cDNA corresponding to Exo84 and cloned into the pCaSpeR4 under the control of α-tubulin promoter ( as described in [13] ) . GFP-Exo84 was crossed into the onr background to test for phenotypic rescue of male sterility and cytokinesis failures . Flies were maintained at 25°C by standard procedures . y w and Oregon-R were used as wild-type controls . onrz4840 is synonymous with onr142-5 and corresponds to Z4840 in the Zuker viable collection , while funz1010 is synonymous with fun145-27 and Z1010 in the Zuker collection [4] . Time-lapse imaging was conducted with Sqh-GFP [36] , PLCδ-PH-GFP [37] , and β-Tub-GFP [38] . Df ( 3R ) Espl3 and Df ( 3R ) Exel6145 ( Bloomington ) uncover onr and fun , respectively . The Rab11E ( To11 ) and Rab1193Bi mutant strains were described previously [10] . Rab11-GFP was a gift from R . S . Cohen [65] . Strains carrying the UASp-YFP-Rab11 transgenes [66] were obtained from the Bloomington Drosophila Stock Center . Flies carrying the UASp-HA-Sec8 transgene were a gift from T . L . Schwarz ( Harvard Medical School ) . Bam-Gal4 [67] was used to drive expression of YFP-Rab11 from the UASp-YFP-Rab11 transgenes and HA-Sec8 from the UASp::HA-Sec8 transgene . Cytological preparations were made with testes from third instar larvae or adults . To visualize GFP-Exo84 or Rab11-GFP , larval testes were fixed in 4% methanol-free formaldehyde ( Polysciences , Warrington , PA ) , as previously described [7] . Following fixation , testes were incubated with GFP-Booster ( ChromoTek ) diluted 1:100 in phosphate-buffered-saline ( PBS ) , as described in [14] . To visualize α-Tubulin and either Sec8 or Sec5 , larval testes were dissected in PBS ( Sigma-Aldrich ) and transferred into a drop ( 4 μl ) of PBS containing 4% methanol-free formaldehyde placed on a coverslip . Preparations were kept at room temperature for two minutes before gently squashing on an inverted slide . They were then fixed for an additional 5 minutes before immersing in liquid nitrogen . After removing the coverslip , preparations were immersed in PBS for five minutes and permeabilized in PBS containing 0 . 1% Triton-X ( PBT ) for 10 minutes at room temperature and washed in PBS 0 . 1% Tween-20 for 20 minutes before incubation with primary antibodies diluted in PBT containing 3% BSA . To visualize α-Tubulin and Sec8 of cells shown in S2 Fig , larval testes were dissected in 0 . 7% NaCl and transferred into a drop of PBS containing 0 . 5% Triton for two minutes . Testes were then transferred to 4 μl of PBS containing 3 . 7% formaldehyde on a coverslip , gently squashed on an inverted slide and fixed for ten minutes before immersing in liquid nitrogen . After removing of the coverslip , samples were immersed for 20 minutes in cold methanol ( -20°C ) and in PBS containing 0 . 1% Triton for 20 minutes at room temperature . For immunostaining with other antibodies , preparations were fixed using 3 . 7% formaldehyde in PBS and then squashed in 60% acetic acid as previously described [10] . Monoclonal antibodies were used to stain α-Tubulin ( 1:300; Sigma-Aldrich , T6199 ) and Sec5 ( 1:30; [25] ) , gift from T . L . Schwarz ( Harvard Medical School ) . Polyclonal antibodies were as follows: rabbit anti-Lva ( 1:500; [40] ) , gift from O . Papoulas ( University of Texas at Austin ) ; rabbit anti-Gio ( 1:2000; [41] ) , guinea pig anti-Sec8 ( 1:250; [51] ) , gift from Ulrich Tepass ( University of Toronto ) ; rat anti-Rab11 ( 1:200; [65] ) , gift from R . S . Cohen; rabbit anti-anillin ( 1:1000; this study ) . Secondary antibodies were Alexa 555-conjugated anti-rabbit IgG ( 1:300 , Life Technology ) , FITC-conjugated anti-mouse/anti-rat IgG ( 1:20 , Jackson ImmunoResearch ) , Alexa 555-conjugated anti-guinea pig IgG ( 1:300 , Life Technology ) . All incubations with primary antibodies ( diluted in PBT containing 3% BSA ) were performed at 4°C overnight . Incubations with secondary antibodies were performed at room temperature for 50 minutes . After immunostaining , all preparations were mounted in Vectashield mounting medium with DAPI ( Vector Laboratories ) to stain DNA and prevent photobleaching . Images were captured with a charged-coupled device ( CCD camera , Photometrics Coolsnap HQ ) , connected to a Zeiss Axioplan epifluorescence microscope equipped with an HBO 100-W mercury lamp and 40X and 100X objectives . The number of Golgi stacks per cell was calculated manually , by analyzing images of G2 spermatocytes at S5 stage stained for Tubulin , Lva and DNA . The size of Golgi bodies was measured using Image J software ( NIH; http://rsbweb . nih . gov/ij/ ) by manual demarcation with a limiting polygon and calculation of its area ( see also [13] for the procedure ) . Time-lapse imaging of PLCδd-PH-GFP and β-Tub-GFP was performed on a spinning disk confocal microscope from Zeiss and Solamere Technologies Group with 63x/1 . 4NA objectives . Germline cells were imaged after dissection and placement in Voltalef 10S oil . Live imaging was performed using exposure settings of 250 msec and 4D image sets were acquired every 60 seconds with a Z-step of 1 micron . Images were edited using Adobe Photoshop . Larval testes expressing Sqh-GFP were dissected and prepared for time lapse using the protocol described previously [13] . Meiotic divisions were analyzed with a Zeiss Axiovert 20 microscope equipped with a 63X , 1 . 25 NA objective and a filter wheel combination ( Chroma Technology Corp . ) . Images were collected at 1-minute time intervals with a CoolSnap HQ camera ( Photometrics ) controlled by MetaMorph software ( Universal imaging ) . Eleven fluorescent optical sections were captured at 1 μm Z-steps and maximally projected using MetaMorph software . We performed a 3D seeded watershed algorithm using the MATLAB image processing toolbox . For the first frame of each movie , we manually initialized the seeds separately in each Z-layer to construct a single 3D seed ( see S4A–S4D Fig ) ; using the seed , the 3D watershed algorithm was applied on the 3D Gaussian filtered image stack ( σx = σy = 1 pixel = 0 . 166 μm , σz = 0 . 2 pixels = 0 . 2 μm ) . For each subsequent frame , the new seeds were then generated automatically by eroding the results of the watershed segmentation from the last frame , with occasional manual intervention , e . g . , to ensure that the seeds masked off any bright features inside the cell , such as spindles . We defined the aspect ratio as the length of the major axis divided by the length of the minor axis ( see S4E Fig ) . We determined the major axis length in 3D by finding the maximum distance between any pair of surface positions of the cell . As the minor axis length , we used the diameter of the larger sphere-like lobe of the cell , which we computed through a 3D distance transform on the 3D binary image of the cell . With these definitions for the major and minor axis length , a perfect sphere will have an aspect ratio of 1 , and two just-touching spheres of equal radius will have an aspect ratio of 2 . Since cells are frequently ‘deformed’ due to mechanical contact with neighboring cells , they generally don’t approximate perfect spheres , so that aspect ratios are frequently >1 before the initiation of division , and can reach values >2 during division . The convex hull of a cell is the smallest convex volume that fully contains the segmented volume of the cell on the inside ( see S4F Fig ) , i . e . it represents the segmented cell volume with all the concave regions next to the cleavage furrow ‘filled in’ . We defined the convex hull volume ratio ( CHVR ) as the convex hull volume of the cell divided by the actual segmentation volume of the cell , which is thus a volume-based measure of furrow ingression . For a cell without concavities , the CHVR will be equal to one; conversely , when concavities are present , the CHVR will increase with the relative volume of the concave areas . Thus , a CHVR value of 1 . 1 means that the volume of the concavity is equal to 10% of the segmented cell volume . For reference , the CHVR of two touching spheres of equal radius ( an idealization of two daughter cells in contact after division ) is 1 . 25 . In order to average time-courses of multiple experiments for a given condition , and to effectively compare wild-type , onr , and fun conditions with each other , cell shape measurements have to be aligned to a common ‘reference’ time point that represents the initiation of cytokinesis . While the onr and fun mutants do not undergo significant rate changes in volume , surface area , or CHVR that could provide useful fiduciary markers for temporal alignment , we observed that the mutants still undergo a distinct initial increase of their aspect ratio–i . e . , they show a small but significant elongation , even in the absence of effective furrow ingression . We used an automated algorithm to identify this ‘shoulder’ point of the aspect ratio in each individual cell trace ( see samples in see S4G Fig ) , and used it as a reference time point ( representing t = 0 ) for subsequent temporal alignment . Mathematically , the reference time point is the first time point at which the slope of the forward 10 min time window increases by 20% ( wild type ) or 60% ( mutants ) relative to the backward 10 min time window . This automated alignment was in excellent agreement with manual alignment . A similar inflection point in CHVR was used to determine the start of cytokinesis ( arrowhead marker in Fig 4G–4J ) . Testes for transmission electron microscopy were prepared using a protocol modified from [68] . Briefly , testes from third instar larvae and 0–3 day-old adults were dissected in ice-cold phosphate buffer ( PB ) ( pH = 7 . 4 ) and immediately transferred into ice-cold Trump’s fixative , where they were kept for 2h . Samples were post-fixed with 1% OsO4 for 1 hour , rinsed and dehydrated with an acetone series and embedded in Quetol-Spurr or Epon resin . Images were acquired with a JEOL JTE141011 ( JEOL , Peabody , MA; The Hospital for Sick Children Electron Microscopy Facility ) and were processed with Adobe Photoshop . GST-full-length Drosophila Rab11 was expressed in BL21-CodonPlus [DE3] cells ( Invitrogen ) and purified using HiTrap affinity columns ( GSTtrap FF , and GSTtrap HP columns , GE Healthcare ) operated with AKTA 900 Fast Protein Liquid Chromatography as previously described [13] . Polyclonal antisera were raised against the purified GST-Rab11 protein . Polyclonal anti-anillin antibodies were raised against the N-terminal 270 amino acids of anillin , following the procedure described in [69] . Immunization was carried out at Agro-Bio Services ( www . agro-bio . com ) using standard procedures . The anti-GST-Rab11 and anti-anillin antisera were depleted of anti-GST antibodies and affinity-purified against either GST-Rab11 or GST-anillin before use in immunoblotting . Co-IP experiments from testes expressing GFP- or YFP-tagged proteins were performed using GFP Trap-A kits beads purchased from ChromoTek ( Planegg-Martinsried ) , as previously described [12] . For the experiment in Fig 9D , testes expressing GFP-Exo84 and HA-Sec8 or HA-Sec8 alone were used as controls . Samples were separated on Mini-PROTEAN TGX precast gels ( Bio-Rad ) and blotted to PVDF membranes ( Bio-Rad ) . Membranes were blocked in Tris-buffered saline ( Sigma-Aldrich ) with 0 . 05% Tween-20 ( TBST ) containing 5% nonfat dry milk ( Bio-Rad; Blotting GradeBlocker ) for 3–4 hours at room temperature followed by incubation with primary and secondary antibodies diluted in TBST . Primary antibodies used for immunoblotting were as follows: rat monoclonal anti-GFP , ( 3H9; 1:1000; ChromoTek ) , mouse anti-Rab11 ( 1:1000; this study ) , rat anti-HA ( 1:1000; Roche ) . HRP-conjugated secondary antibodies ( GE Healthcare ) were used at 1:5000 . After incubation with the antibodies , blots were washed in TBST and imaged using an ECL detection kit ( GE Healthcare ) . | The cell shape changes that underlie cell division are some of the most fundamental changes in cell morphology . Here , we show that a common membrane trafficking pathway is required for both the cell lengthening that occurs during anaphase , and the physical separation of a cell into two equal daughter cells . We measure and define the periods of surface area increase during cell division in Drosophila male germline cells , and demonstrate that subunits of the exocyst tethering complex are required for this process . Invagination of the cleavage furrow fails at an early stage in exocyst mutant spermatocytes , suggesting that membrane addition is part of the initial ingression mechanism . In the absence of exocyst complex function , vesicular trafficking pathways are disrupted , leading to enlarged cytoplasmic membrane stores , and disruption of Golgi architecture . In addition , a vesicular Rab protein , Rab11 , biochemically associates with the exocyst complex subunit Sec5 . These results suggest that remodeling of the plasma membrane and targeted increases in surface area are an active part of the fundamental mechanisms that permit eukaryotic cell division to occur . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2015 | Exocyst-Dependent Membrane Addition Is Required for Anaphase Cell Elongation and Cytokinesis in Drosophila |
African trypanosomes cause disease in humans and livestock , generating significant health and welfare problems throughout sub-Saharan Africa . When ingested in a tsetse fly bloodmeal , trypanosomes must detect their new environment and initiate the developmental responses that ensure transmission . The best-established environmental signal is citrate/cis aconitate ( CCA ) , this being transmitted through a protein phosphorylation cascade involving two phosphatases: one that inhibits differentiation ( TbPTP1 ) and one that activates differentiation ( TbPIP39 ) . Other cues have been also proposed ( mild acid , trypsin exposure , glucose depletion ) but their physiological relevance and relationship to TbPTP1/TbPIP39 signalling is unknown . Here we demonstrate that mild acid and CCA operate through TbPIP39 phosphorylation , whereas trypsin attack of the parasite surface uses an alternative pathway that is dispensable in tsetse flies . Surprisingly , glucose depletion is not an important signal . Mechanistic analysis through biophysical methods suggests that citrate promotes differentiation by causing TbPTP1 and TbPIP39 to interact .
Eukaryotic developmental events are a response to single or multiple external cues . Commonly , the existence of multiple cues ensures that cells do not embark prematurely on a developmental process that could damage their viability or fitness [1] . Additionally , the presence of multiple cues can lower the threshold at which cells respond to differentiation signals or refine their response , with inputs from distinct signalling pathways co-operating to generate a specific developmental outcome ( e . g . [2] , [3] ) . In this way , quite sophisticated perception mechanisms can contribute to ensure an appropriate and timely developmental response when cells encounter conditions where differentiation is the optimal survival response to a changing environment . Although cell type differentiation is most studied in the programmed specialisation of metazoan cells as they form tissues or adapt for particular functions in the body , unicellular organisms can also undergo development in response to external signals . Exemplary of this are the differentiation responses of vector-borne parasites . These undergo development within distinct environments in their mammalian host , as well as during colonisation of their arthropod vectors to ensure their transmission [4] , [5] . Among the best studied of these are the kinetoplastid parasites , representing the earliest branching extant eukaryotes [6] that are responsible for a range of tropical diseases such as visceral and cutaneous Leishmaniases ( caused by different Leishmania spp . ) , American trypanosomiasis ( ‘Chagas' disease’ , caused by Trypanosoma cruzi ) and Human , and Animal , African trypanosomiasis ( HAT , AAT , caused by Trypanosoma brucei [7] ) . For both Leishmania and T . cruzi several signals have been discovered that can trigger life-cycle differentiation including low temperature , pH balance [8] and , most recently in Leishmania , iron availability [9] . For African trypanosomes , citrate-cis aconitate ( CCA ) is routinely used in differentiation studies [10] . In T . brucei the optimal response to CCA requires the generation of a transmissible life cycle stage in the blood called ‘stumpy’ forms [11] , which are non-dividing and show partial adaptation for conditions in the midgut of tsetse flies , the parasite's vector . For example , mitochondrial activity is elevated in the insect forms of the parasite compared to the proliferative bloodstream ‘slender’ forms , which are able to meet their energy demands through their metabolism of blood glucose through glycolysis alone [12] . Stumpy forms are sensitive to CCA through their expression of members of a surface carboxylate transporter family , called PAD ( Proteins Associated with Differentiation ) proteins [13] . These proteins transport CCA and at least one family member , PAD2 , is elevated by the temperature reduction associated with passage from a homoeothermic to poikilothermic environment , sensitising parasites to physiological concentrations of CCA [14] , [15] . The molecular details of the development from slender to stumpy forms are not well understood , although laboratory adapted lines that have lost the capacity to become stumpy ( ‘monomorphic lines’ ) can generate stumpy-like forms upon exposure to cell permeable cAMP and AMP analogues [16]–[18] , or through TbTORC4 depletion [19] . A key molecular regulator of trypanosome differentiation to tsetse midgut procyclic forms is the tyrosine phosphatase TbPTP1 [20] . This enzyme acts as an inhibitor of parasite differentiation such that inactivation of the enzyme , or reduction of its levels by RNAi , elicits spontaneous differentiation in the absence of any trigger . Recently , a substrate of TbPTP1 was identified as a serine/threonine DxDxT class phosphatase , TbPIP39 [21] . This molecule is dephosphorylated on tyrosine 278 by TbPTP1 and thereby held inactive , this inhibition being reinforced because TbPIP39 itself promotes the activity of TbPTP1 . In the presence of CCA , however , the activation of TbPTP1 by TbPIP39 is reduced , such that TbPIP39 becomes phosphorylated and differentiation is stimulated [21] , [22] . This appears to be mediated via trafficking to the glycosome , a peroxisome-like organelle in trypanosomes that is the site of glycolysis and several other metabolic activities [23] . As in other kinetoplastid parasites , a number of other stimuli of trypanosome differentiation have been reported in addition to CCA , although the use of different cell lines and developmental forms has complicated interpretation of their efficacy or physiological relevance . These reported stimuli include ( i ) mild acid conditions [24] , ( ii ) exposure of the parasite surface to limited protease digestion [25]–[27] and ( iii ) a reduction of glucose mediated through either the use of glucose depleted media [28] , or exposure to phloretin , an inhibitor of glucose uptake [29] . Of these , both mild acid and protease treatment induce adenylate cyclase activity [30] and are restricted to transmissible stumpy forms , since bloodstream slender cells are not viable after exposure to these conditions [31] , whereas glucose depletion has only been evaluated in monomorphic slender forms . Here we have systematically investigated the importance of distinct stimuli of trypanosome differentiation , their physiological relevance and , in the case of CCA , mechanistic basis . Firstly , we used the phosphorylation of TbPIP39 as a marker for operation of the citrate-dependent signalling cascade to determine whether a single , or multiple transduction pathways can elicit differentiation . This revealed that independent pathways for the initiation of trypanosome differentiation exist , one ( stimulated by CCA and mild acid ) operating via the TbPTP1-TbPIP39 signalling pathway , whereas an alternative stimulus , protease treatment , signals via an alternative route . Analysis of the contribution of each pathway in tsetse flies supports a physiological role for TbPTP1-TbPIP39 signalling , with insect trypsin activity not being required to stimulate parasite differentiation signal in vivo . Thereafter , we investigated the interaction between TbPTP1 and TbPIP39 and found that this is dependent upon the presence of the citrate differentiation signal , revealing a hitherto unexpected interplay between these molecules .
To date , a number of stimuli have been reported to induce trypanosome life cycle differentiation , but these have been analysed in different cell lines and under varying experimental conditions . Our identification of TbPIP39 as a downstream substrate of the CCA-responsive regulator TbPTP1 [21] provided tools to systematically investigate the pathways through which the development of trypanosomes is signalled . Therefore , we carefully analysed , in parallel , trypanosome differentiation stimulated via four reported triggers: cis-aconitate ( CA ) , mild acid , pronase treatment or glucose deprivation using phloretin , a glucose transport inhibitor . Initially the efficacy of each pathway was evaluated such that bloodstream stumpy forms were exposed to 6 mM cis-aconitate , were incubated for 2 hours at pH 5 . 5 , were treated with 4 units pronase from Streptomyces griseus for 10 minutes or were incubated in the presence of 100 µM phloretin . In each case , differentiation was monitored by flow cytometry for the expression of the developmental surface marker EP procyclin . Figure 1 shows EP procyclin expression for stumpy or procyclic cells ( Figure 1A ) or stumpy cells incubated in the absence of any trigger ( Figure 1B ) , demonstrating that stumpy cells do not express EP procyclin , although a reversible weak procyclin expression can be detected after incubation , reflecting cold-related expression of this marker [14] . However , when exposed to cis-aconitate ( Figure 1C ) , mild acid ( Figure 1D ) or pronase ( Figure 1E ) , the stumpy forms underwent effective differentiation into procyclic forms , with the expression of EP procyclin being evident after 2 hours , and becoming progressively stronger throughout the time course of the experiment . For the mild acid ( Figure 1D ) and pronase treatment ( Figure 1E ) there was some heterogeneity in procyclin expression due to stress associated killing or damage to some cells , this being absent for the CA treatment . In contrast , phloretin treated cells did not express EP procyclin ( Figure 1F ) . To confirm that the phloretin treatment was effective , pleomorphic slender cells were incubated for 2 days in HMI-9 in the presence or absence of phloretin , or in the presence of cis-aconitate ( Figure 2 ) . Phloretin treatment arrested the growth of the treated cells within 24 hours , as expected if glucose uptake were prevented ( Figure 2A ) . Nonetheless , these cells did not outgrow as differentiated forms when incubated in procyclic form medium nor did they express EP procyclin ( Figure 2B ) , contrasting with CA treated cells . Furthermore , incubation of stumpy forms in a more physiologically relevant medium [32] containing low glucose ( ∼0 . 5 mM final concentration ) did not stimulate their differentiation , unless cis-aconitate was also included ( Figure S1 in Text S1 ) . Hence , neither pleomorphic slender or stumpy forms differentiated in response to phloretin and stumpy cells were not stimulated by low glucose medium , demonstrating the glucose depletion is not an effective differentiation stimulus . Having confirmed cis-aconitate , mild acid and pronase treatment as effective stimuli of differentiation , we monitored the phosphorylation of TbPIP39 under each condition using a phospho-specific antibody directed against tyrosine 278 residue in the protein . This is the site of TbPTP1 phosphatase activity [21] , such that tyrosine 278 phosphorylation is diagnostic for the activity of this signal transduction pathway . Figure 3 shows the level of TbPIP39 , detected using a polyclonal antibody to the protein , and the level of phosphorylated TbPIP39 detected using the tyrosine 278-phosphospecific antibody . As expected , untreated cells showed no evidence of TbPIP39 phosphorylation , whereas the level of phosphorylated TbPIP39 progressively increased throughout differentiation in the CA treated samples , with this being phosphorylated . Similarly , mild acid exposure generated phosphorylated TbPIP39 , indicating activity of the TbPTP1/TbPIP39 signalling pathway under that treatment regimen . Interestingly , however , no phosphorylated TbPIP39 was detected after pronase treatment , despite the effective induction of differentiation in these cells as determined by EP procyclin expression ( see Figure 1E , from the same experiment ) . As expected , exposure to phloretin did not induce TbPIP39 phosphorylation . These results indicated that CA and mild acid exposure of stumpy cells induces differentiation via TbPTP1/TbPIP39 , whereas pronase treatment stimulates an alternative pathway that does not generate phosphorylated TbPIP39 . To test this hypothesis we exploited a pleomorphic T . brucei AnTat1 . 1 TbPIP39 RNAi cell line that generates effective and inducible TbPIP39 depletion in stumpy forms [21] . Our prediction was that TbPIP39 depletion would reduce the efficiency of differentiation for those stimuli ( CA , mild acid ) that operate via this signalling pathway , whereas there would be no effect upon pronase treatment . Figure S2 in Text S1 shows Western blots indicating the level of TbPIP39 in stumpy forms in which RNAi against the TbPIP39 transcript had been induced , or not , by provision of doxycycline to the drinking water of infected mice . In each case there was a clear evidence of TbPIP39 depletion in those samples where RNAi was induced in the host animals this being particularly evident as the cells underwent differentiation , whereupon the levels of TbPIP39 normally increase ( see Figure 3 and [21] ) . To assess the differentiation efficiency of the cells , EP procyclin expression was monitored at 4 h after exposure to each trigger , before the outgrowth of differentiated cells and the consequences of the low levels of TbPIP39 that remain after RNAi induction complicate the analysis [21] . Figure 4A shows that untreated cells did not differentiate , with only the weak cold-related expression of procyclin detected , as also seen in Figure 1B . When triggered with CA ( Figure 4B ) TbPIP39 RNAi reduced the differentiation efficiency , and a similar response was also observed when cells were treated with mild acid ( Figure 4C ) . Upon pronase treatment there was a population of unstained cells , representing undifferentiated slender cells ( also seen in Figures 4A–C ) and dead or damaged cells generated by the pronase treatment . However , when analysing EP procyclin expressing cells , either induced or uninduced to deplete TbPIP39 , approximately equivalent differentiation efficiency was observed ( Figure 4D ) , consistent with TbPIP39-independent differentiation . The same outcome was observed in four independent experiments , with TbPIP39 depletion reducing differentiation triggered by CA ( p<0 . 001; general linear mixed model ) and mild acid ( p<0 . 001 ) but not pronase ( p = 0 . 4270 ) ( Figure 4E ) . We conclude that two independent pathways can stimulate differentiation , one initiated by CA or mild acid , that acts via TbPIP39 phosphorylation , and one , stimulated by pronase , that does not . We analysed the relevance of the different signalling pathways in vivo by assaying tsetse infections when protease activity was blocked using inhibitors . Thus , batches of tsetse flies were fed with trypanosomes in horse serum either in the presence or absence of 1 mg/ml of soybean trypsin inhibitor ( STI ) , a treatment predicted to block the insect midgut trypsin-like activities that comprise a major digestive component of the tsetse midgut [33] and which have been reported to stimulate differentiation in vivo [34] , [35] and in vitro [26] . Confirming the efficacy of inhibition , an analysis of 30 tsetse extracts revealed that midgut trypsin/chymotrypsin activity was reduced 84%–100% in three replicate experiments ( P<0 . 01 ) , irrespective of whether trypanosomes were included in the serum meal ( Figure 5A ) . Stumpy forms were then fed to flies in horse serum in the presence or absence of STI and differentiated cells analysed in extracted midguts 4 hours after feeding . Replicate stumpy samples were also exposed to CA in culture and particular care was taken to ensure that tsetse fed and in vitro prepared samples were processed under identical conditions in order to eliminate effects attributable to cold-induced procyclin expression [14] . In each case , since neither flow cytometry nor automated fluorescence analysis proved possible due to the debris present in the tsetse midgut extracts , we analysed EP procyclin expression using a careful visual scoring system , categorising the labelling of cells as ‘bright’ ( representing a homogenous , bright , EP signal detected on the whole cell ) , ‘faint’ ( a faint and/or non-homogeneous EP signal detected on the whole cell , with a brighter flagellum and/or uneven , punctuated staining pattern ) or ‘negative’ ( no EP signal detected ) . Figure 5B demonstrates that under these conditions , EP procyclin was strongly expressed on ∼90% of cells from tsetse midguts regardless of the presence or absence of protease inhibitor , matching in vitro differentiation in the presence of CA ( representative images are shown in Figure 5C and Figure S3 in Text S1 ) . Untreated samples showed only faint EP procyclin expression associated with cold shock during sample processing . Furthermore , similar numbers of cells were observed in STI treated and untreated midgut extracts , eliminating the possibility that only those few cells able to differentiate had survived in the midguts of the flies fed trypsin inhibitor . We conclude that blocking protease activity in the tsetse midgut does not prevent differentiation . Having demonstrated that inhibiting trypsin activity in the bloodmeal did not prevent the differentiation of stumpy forms , we repeated the analysis using the pleomorphic TbPIP39 RNAi line to determine the influence of inhibiting the TbPIP39 signalling pathway alone , or in combination with trypsin inhibition . Specifically , parasites were grown in mice provided , or not , with doxycycline in their drinking water to induce TbPIP39 gene silencing during the infection . Aliquots of the resulting cells were then fed to tsetse flies either in the presence or absence STI , such that in individual analyses either or both proteolytic activity and cellular TbPIP39 was depleted . Gut extracts were isolated 4 h after feeding , and the differentiation of the midgut parasites analysed by immunofluorescence for the expression of EP procyclin as before , with at least 1000 cells being scored for each sample from a total of 3 replicate experiments ( Figure 6 ) . Confirming the data in Figure 5 , trypsin inhibition alone did not inhibit differentiation ( p = 0 . 5706; Wilcoxon Mann Whitney Rank sum test ) . However , with TbPIP39 depletion , the percentage of weakly stained cells increased from approximately 17 . 9% ( 17 . 9% , 17 . 8% + or – STI , respectively ) for the uninduced samples to 25 . 4% ( 24 . 3% , 26 . 5% + or – STI , respectively ) , and overall differentiation was reduced , weakly but significantly ( p = 0 . 0073 ) , even when a Bonferroni correction was used to account for the number of tests ( thereby setting the significance threshold at 0 . 008 ) . Interestingly , however , when both RNAi was induced and proteases were inhibited , the efficiency of differentiation was more significantly reduced ( p<0 . 0009 ) with respect to untreated samples . Although the observed effects were expected to be minor , given the transient inhibition of differentiation upon TbPIP39 depletion in vitro , these analyses indicated that inhibition of differentiation is observed in vivo upon TbPIP39 depletion , with a minor contribution when trypsin activity is also inhibited . The in vivo experiments supported the function of the TbPTP1/TbPIP39 pathway in differentiation , but the technical limitations with these studies highlighted that more detailed analysis of these signalling events required in vitro analysis . Therefore , we sought to gain mechanistic understanding of the interaction between TbPTP1 and TbPIP39 using biochemical methods . Initially , we investigated whether residues in TbPIP39 predicted from homology modelling to comprise part of a citrate-binding pocket [21] ( Figure S4 in Text S1 ) would influence the regulatory cross talk between TbPIP39 and TbPTP1 , whereby TbPIP39 activates TbPTP1 , unless in the presence of citrate [21] . To analyse this , recombinant TbPIP39 was generated in which either the aspartic acid at position 57 was mutated to alanine ( generating DxAxT; “PIP39 D mutant” ) , both aspartic acid residues in the DxDxT motif were mutated ( generating AxAxT; “PIP39 DD mutant” ) or the TV motif at position 63 and 64 was mutated to AA ( “PIP39 6364 mutant” ) ( Figure S4 in Text S1; the respective recombinant proteins are shown in Figure 7A ) . The residues at position 57 , 63 and 64 are each predicted to be involved in citrate binding whereas mutating both aspartic acid residues is known to render the enzyme catalytically inactive [21] , [36] . The wild type ( Figure 7B ) and TbPIP39 mutants ( Figure 7C–E ) were each then tested for their ability to enhance TbPTP1 activity and the sensitivity of this to citrate inhibition . Figure 7 demonstrates that each of the TbPIP39 mutants ( TbPIP39 D , TbPIP39 DD , TbPIP39 6364 ) enhanced the activity of TbPTP1 at equivalent levels to the wild type TbPIP39 , regardless of their individual activity . However , this was not sensitive to the presence of citrate for any of the mutants , contrasting with the wild type protein where there was a citrate-dependent decrease in TbPTP1 activity ( P<0 . 0072 ) . Unlike citrate , isocitrate ( which does not act as a differentiation trigger ) did not generate a decrease in activity when wild type TbPIP39 was used ( Figure 7F ) . This demonstrated that the regulatory cross talk between TbPTP1 and TbPIP39 depended upon the integrity of the citrate binding residues in TbPIP39 . Having examined regulatory interaction between the molecules we examined their biophysical interactions in the presence or absence of citrate using surface plasmon resonance . Specifically , TbPTP1 protein was immobilised and covalently stabilised on a nitrilotriacetic acid ( NTA ) chip and then interactions with wild type or mutant TbPIP39 tested in the presence or absence of citrate ( Figure 8 ) . In the absence of citrate , no interaction between TbPTP1 and TbPIP39 ( wild type or any of the TbPIP39 mutants ) was observed ( Figure 8A , Figure S5 in Text S1 ) . In contrast , wild type TbPIP39 showed interaction with TbPTP1 in the presence of citrate , consistent with a 1∶1 binding stoichiometry , the predicted half-life for the complex being around 10 seconds ( on rate: 0 . 015 µM-1 . s-1; off rate: 0 . 068 s-1 ) ( Figure 8B , 8C , Figure S5 in Text S1 ) . Interestingly , when the mutant forms of TbPIP39 were tested in combination with TbPTP1 by surface plasmon resonance , each also showed a citrate-dependent interaction with TbPTP1 , matching that of the wild type TbPIP39 ( Figure 8D–F ) . Moreover , isocitrate , which does not stimulate differentiation [26] , [37] , also promoted the interaction between TbPTP1 and wild type TbPIP39 ( on rate: 0 . 014 µM-1 . s-1; off rate: 0 . 064 s-1 ) indicating that the interaction is not contributing to the specific differentiation stimulus ( Figure S5 in Text S1 ) . When analysed by isothermal calorimetry neither the wild type nor any of the TbPIP39 mutant proteins was able to bind citrate independently of TbPTP1 ( Figure S6 in Text S1 ) . Overall , these experiments indicated that individual mutation of specific residues in TbPIP39 predicted to be involved in citrate binding did not prevent the citrate-dependent interaction of TbPIP39 with TbPTP1 detected by SPR and that this interaction was also generated by isocitrate , which is not a differentiation trigger . In conclusion , the ability of citrate to reduce the phosphatase activity of TbPTP1/TbPIP39 was lost upon mutation of the predicted citrate-binding residues in TbPIP39 . Nonetheless , regardless of the integrity of these residues , these molecules continue to interact in a citrate ( or isocitrate ) dependent-manner . Based on these analyses we propose that citrate causes TbPIP39 and TbPTP1 to bind to one another , at least in vitro , but that citrate also independently blocks the ability of TbPIP39 to enhance TbPTP1 activity . Hence , in vivo , the transport of citrate in the bloodmeal by PAD proteins would stimulate the initiation of parasite differentiation through the co-ordinated inhibition of TbPTP1 and activation of TbPIP39 .
When ingested in a tsetse fly bloodmeal trypanosomes rapidly initiate differentiation in order to adapt to their new environment . Our earlier studies have demonstrated that the CCA differentiation signal is transduced via PAD proteins through a phosphatase-signalling cascade , whereby TbPTP1 is inactivated and TbPIP39 becomes phosphorylated and activated [20]–[22] . Here we have used these components to investigate the molecular basis of differentiation initiated by CA and to demonstrate that independent signalling pathways can operate to stimulate development . Firstly , the availability of phospho-specific TbPIP39 antibodies and pleomorphic RNAi lines targeting TbPIP39 allowed the relationship between different differentiation signalling pathways to be investigated . This demonstrated that CA and mild acid are triggers that both result in the phosphorylation of TbPIP39 on tyrosine 278 , with differentiation via these signals being inhibited by RNAi against TbPIP39 . In contrast , pronase stimulates differentiation effectively but this does not generate phosphorylated TbPIP39 nor does RNAi against this molecule inhibit it . In mammals , protease activated small G protein coupled receptors , PARs , operate to regulate signalling events [38] , but conventional G-protein signalling is missing in trypanosomes [39] . Hence , although cleavage-mediated activation might also contribute to the initiation of differentiation in trypanosomes , the molecule responsible is likely to be novel , potentially at the cell surface or within the flagellar pocket membrane . Our findings reveal the presence of independent differentiation signalling pathways in trypanosomes ( Figure 9A ) , with activation by different stimuli converging downstream of TbPIP39 . Supportive of the existence of independent signalling pathways , reporter assays for the activation of EP procyclin expression in bloodstream forms have demonstrated enhanced expression when both CA and proteases were used compared with either stimulus alone [26] . Our in vivo experiments supported this observation , since the depletion of TbPIP39 by RNAi reduced differentiation , the significance of this being increased in the presence of protease inhibitors . Although the reduction of TbPIP39 by RNAi is incomplete , significantly reducing the magnitude of the effect observed in vivo , trypsin/chymotrypsin-type protease inhibition was almost complete when monitored in the midgut of treated flies . Hence , although trypsin activities have been reported [33] and proposed to be involved in the transformation of trypanosomes in vivo [34] , [35] and in vitro [26] , our experiments demonstrate that inhibiting this protease activity alone does not prevent the initiation of parasite differentiation in tsetse flies . Unfortunately , a complementary analysis of differentiation in the complete absence of TbPIP39 via a gene knockout and add-back approach is not possible in parasite lines capable of generating stumpy forms . In contrast to the above reported triggers we did not observe any induction of differentiation for either bloodstream stumpy forms or bloodstream slender forms when parasites were exposed to phloretin or low glucose conditions . Phloretin blocks glucose uptake by the trypanosome and has been reported to induce the expression of a subset of procyclic specific transcripts in monomorphic bloodstream forms [29] . An effect of glucose depletion has also been reported earlier , whereby glucose-depleted medium generated the outgrowth of differentiated procyclic forms from a monomorphic cell culture [28] . Although these findings seem consistent with the rapid loss of glucose from the tsetse blood meal , they may also reflect a monomorphic cell line specific response , or the outgrowth of a small , differentiated subpopulation when cells are subcultured into procyclic form media . Alternatively , phloretin may induce transformation of monomorphic slender forms toward more stumpy-like forms , since stumpy forms elevate the expression of some procyclic form transcripts including procyclins in preparation for differentiation in the tsetse fly or in vitro [29] . Nonetheless , overall our results eliminate glucose depletion alone as an efficient differentiation trigger for transmissible stumpy forms , making the physiological relevance of this trigger questionable . Through biochemical and biophysicial analyses , our experiments also revealed that TbPTP1 and TbPIP39 interact in a citrate or isocitrate dependent manner , this being retained even when individual residues proposed to be important in citrate binding within TbPIP39 are mutated . However , mutation of these residues prevented the citrate-dependent reduction of TbPTP1/TbPIP39 activity . These results contrast with a model where citrate binding to TbPIP39 would prevent its interaction with TbPTP1 [21] , [22] , potentially through steric hindrance within the catalytic region . Instead , our results show that the presence of citrate can allow TbPTP1 and TbPIP39 to physically engage , at least in vitro , potentially in a substrate trapping type interaction although this is not specific for citrate or the predicted citrate binding capacity of TbPIP39 . Hence , in a bloodmeal , TbPTP1 could be sequestered by interaction with TbPIP39 such that excess unbound TbPIP39 could become stably phosphorylated by its , as yet uncharacterised , kinase thereby promoting differentiation . Furthermore , these experiments revealed that the predicted citrate binding residues in TbPIP39 are required for the regulatory cross talk between TbPIP39 and TbPTP1 but not their specific interaction , suggesting that citrate has distinct ‘dock’ and ‘block’ activities on the TbPIP39/TbPTP1 complex , with only the ‘blocking’ function being specific for citrate ( Figure 9B ) . The integration of the combined signals of temperature reduction , CCA reception and protease attack of the parasite surface could ensure a strong differentiation response . Given their sensitivity to citrate/cis-aconitate and resistance to protease attack [40] , these responses will be limited to the transmissible stumpy form , with slender cells in the blood meal being rapidly killed . The role of multiple signal inputs that converge to drive differentiation is characteristic of several developmental systems including in fungi [41] , cell type differentiation in Drosophila development [42] , [43] and in mammalian bone formation [44] , as well as in arthropod-borne parasites [5] , [8] . In parasites , this stringent control would ensure that the initiation of an irreversible developmental programme occurs only under the correct environmental conditions , avoiding the risk of initiating an inappropriate and lethal differentiation response whilst still in the mammalian host .
Animal experiments in this work were carried out in accordance with the local ethical approval requirements of the University of Edinburgh and the UK Home Office Animal ( Scientific Procedures ) Act ( 1986 ) under licence number 60/4373 . Bloodstream and procyclic form trypanosomes were cultured in vitro in HMI-9 [45] medium or SDM-79 [46] medium respectively . T . brucei AnTat1 . 1 slender and stumpy parasites were obtained 3 and 6 days post infection , respectively , and purified by DEAE chromatography . TbPIP39 RNAi lines were described in [21] . For the initiation of differentiation the following conditions were used: Recoded wild type ( wt ) and ExExT ( DD ) synthetic pHD451TbPIP39 constructs [21]were used to produce mutants predicted to have reduced ability to bind citrate ( Figure S4 in Text S1 ) . A commercial site-directed mutagenesis kit ( Stratagene ) was used with the mutagenesis Primer 1 and Primer 2 ( Table S1 ) to produce TbPIP39D57A ( TbPIP39D ) and Primer 3 and Primer 4 ( Table S1 ) to produce t63A v64A ( TbPIP39 6364 ) . Each of these pHD451TbPIP39 constructs ( TbPIP39wt , TbPIP39D , TbPIP39 6364 and TbPIP39DD ) were reamplified with recoded TbPIP39 specific Primer 5 and Primer 6 ( Table S1 ) and integrated into the pGEX4T1 ( GE Healthcare Life sciences ) protein expression vector for recombinant protein production . Expression and purification of His tagged TbPTP1 and GST tagged TbPIP39 ( TbPIP39wt , TbPIP39D , TbPIP39 6364 and TbPIP39DD ) were performed as described in [21] . Phosphatase activity was measured by monitoring the TbPTP1 ( 0 . 01–1 µg ) catalyzed hydrolysis of pNPP to p-nitrophenol [20] . Methods for protein expression and purification for biophysical analyses are described in the Supplementary data . SPR measurements were performed on a BIAcore T200 instrument ( GE Healthcare ) . Ni2+-nitrilotriacetic acid ( NTA ) sensor chips , 1-ethyl-3- ( 3-diaminopropyl ) carbodiimide hydrochloride ( EDC ) and N-hydroxysuccinimide ( NHS ) were purchased from GE Healthcare . Pure His-PTP1 was immobilized and covalently stabilized on an NTA sensor chip . Briefly , following Ni2+ priming ( 60 s of 500 µM NiSO4 in 10 mM HEPES , pH 7 . 5; 150 mM NaCl; 0 . 05% P20; 50 µM EDTA; 20 mM MgCl2 , at 5 µl . min−1 ) dextran surface carboxylate groups were activated by injection of 20 µl of 0 . 2 M EDC; 50 mM NHS at 5 µl . min−1 . Protein ( between 10 nM and 100 nM ) in 10 mM HEPES , pH 7 . 5; 150 mM NaCl; 0 . 05% P20; 50 µM EDTA; 20 mM MgCl2 was captured and covalently stabilized on the surface to between 100 and 300 RU by injection at 30 µl . min−1 . Following attainment of the desired RU signal a brief injection of 10 mM HEPES , pH 7 . 5; 150 mM NaCl; 0 . 05% P20; 350 mM EDTA ( 30 s at 30 µl . min−1 ) was used to remove non-covalently attached protein , followed by quenching of the unreacted succinimide esters by an injection of 20 µl of 1 M H2N ( CH2 ) 2OH , pH 8 . 5 at 5 µl . min−1 . Non-covalently bound proteins were washed off the surface with excess 10 mM HEPES , pH 7 . 5; 150 mM NaCl; 0 . 05% P20; 50 µM EDTA; 20 mM MgCl2 at 100 µl . min−1 . Final protein immobilized levels were between ∼70 and 290 RU . SPR single cycle kinetic titration binding experiments were performed at 25°C , using a 2-fold dilution series , in 10 mM HEPES , pH 7 . 5; 150 mM NaCl; 0 . 05% P20; 50 µM EDTA; 20 mM MgCl2 ( or with buffer supplemented with 2 mM citrate ) at 100 µl min−1 with a 60 seconds contact time and a 60 seconds dissociation time . The sensor surface was regenerated between experiments by dissociating any formed complex 10 mM HEPES , pH 7 . 5; 150 mM NaCl; 0 . 05% P20; 50 µM EDTA; 20 mM MgCl2 ( or with buffer supplemented with 2 mM citrate ) . The apparent equilibrium dissociation constants ( Kd ) were calculated from double reference corrected sensorgrams by global fitting of a 1∶1 binding model , including a mass transport term , using analysis software ( v . 1 . 0 , GE Healthcare ) provided with the BIAcore T200 instrument . Protein expression analyses by western blotting and flow cytometry were carried out according to [21] . Immunofluorescence was carried out according to [13] . For quantitative scoring of differentiation in tsetse midguts the following criteria were used: ‘Bright’: homogenous , bright EP signal detected on the whole cell , ‘Faint’: faint and/or inhomogeneous EP signal detected on the whole cell ( with a brighter flagellum and/or uneven , punctuated pattern ) ; ‘Negative’- no EP signal detected . In all cases illumination and imaging settings were identical . Phase–contrast and Immunofluorescence microscopy images were captured on a Zeiss axioskop2 ( Carl Zeiss microimaging ) with a Prior Lumen 200 light source using a QImaging Retiga 2000RCCD camera; the objective was a Plan Neofluar ×63 ( 1 . 25 NA ) . Images were captured via QImage ( QImaging ) . All tsetse flies were taken from the Glossina morsitans morsitans ( Westwood ) colony at LSTM , which is maintained at 26°C and 65–80% relative humidity , and fed on the supernatant of defibrinated horse blood . Male flies were fed only once , 24–72 hours after eclosion from the pupa . A minimum of 50 males were offered a feed on one of the meals described above for 10 minutes and were subsequently kept at 26–27°C for 4 hours . Flies were then chilled to 4°C and unfed flies removed . Flies with a meal in the gut were subsequently kept on ice until dissection . For the EP expression experiment , the entire midgut from the proventriculus at the anterior end to the Malpighian tubules at the boundary with the hindgut was isolated in vPBS ( 8 g/l NaCl; 0 . 22 g/l KCl; 2 . 27 g/l Na2HPO4; 0 . 41 g/l KH2PO4; 15 . 7 g/l sucrose; 1 . 8 g/l glucose; pH 7 . 4 ) . Dissected guts were then kept on ice in vPBS until all had been dissected . The guts were then gently disrupted using a micropestle in a 1 . 5 ml microcentrifuge tube . The content of the tube was then filtered through a 35 mm nylon mesh ( BD Biosciences , England , #352235 ) to remove large pieces of midgut . The filtrate , which included trypanosomes , fly gut cells and fly gut bacteria was then washed twice in vPBS by pelleting at 836 g for 8 minutes and removing the supernatant . The cell pellet was then resuspended in around 600 µl vPBS and around 100 µl aliquots of the cell suspension was smeared over the surface of a polylysine treated slides and air dried . Horse serum was prepared by filtering through a 0 . 2 µm syringe filter the supernatant of defibrinated horse blood ( TCS Biosciences Ltd . , Buckingham , UK ) . Soybean ( Glycine max ) trypsin inhibitor ( STI , Sigma Aldrich ) was added to some aliquots of the serum at 1 mg/ml . All the serum aliquots were re-filtered and warmed to 37°C . Stumpy enriched populations of AnTat1 . 1 90:13 cells were mixed with 37°C HMI9 media and incubated for one hour at 37°C at 5% CO2 . Cells were pelleted for 8 minutes at 836 g and resuspended in either 2 ml warmed horse serum or 2 ml warmed horse serum containing 1 mg/ml STI . The cells were then used immediately to feed tsetse flies after briefly checking cell viability by microscopy . The experiment was repeated three times , each time using a different batch of horse blood to prepare the serum . For the protease activity assays with T . b . brucei strain TSW196 , aliquots of TSW196 blood from infected rats and mixed with 2 ml horse serum in the presence or absence of 1 mg/ml STI and immediately fed to tsetse flies . Protease activity of midgut extracts was assayed using Chromozym TRY ( Z-Val-Gly-Arg-pNA , Bachem , Switzerland ) largely as described in [47] . | African trypanosomes are important pathogens transmitted by tsetse flies in sub-Saharan Africa . Upon transmission , trypanosomes detect citrate and cis-aconitate in the bloodmeal , this inactivating a negative regulator of differentiation , the tyrosine phosphatase TbPTP1 . One TbPTP1 substrate is another phosphatase , TbPIP39 , which is more active when phosphorylated ( after TbPTP1 inhibition ) and promotes differentiation . These differentiation regulators have provided tools to monitor whether one or more environmental signals are used to initiate trypanosome development and their relevance in vivo . This is important because different studies over the last 30 years have disputed the physiological importance of different signals . Here we have , firstly , compared the efficacy of the different reported differentiation signals , establishing their relative importance . We then monitored TbPIP39 phosphorylation to show that two signalling pathways operate: one signalled by citrate or mild acid , the other stimulated by external protease activity . Thereafter , we showed that , of these different signals , protease activity is dispensable for differentiation in tsetse flies . Finally , we used biophysical methods to investigate how citrate causes TbPIP39 and TbPTP1 to interact , enabling their regulatory cross-talk . These studies have established the importance of different developmental signals in trypanosomes , providing molecular insight into how the development signal is transduced within the pathogen . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2013 | Independent Pathways Can Transduce the Life-Cycle Differentiation Signal in Trypanosoma brucei |
Insect transmission is obligatory for persistently transmitted viruses because the vector insect is the only means of virus spread in nature . The insect midgut is the first major barrier limiting virus acquisition , but the mechanisms by which viruses are able to cross the cell membrane and then infect the midgut epithelial cells of the insect have not been elucidated completely . Here , we found that the outer capsid or nucleocapsid protein ( NP ) of three viruses can interact and colocalize with sugar transporter 6 that is highly expressed in the midgut of Laodelphax striatellus ( LsST6 ) . In contrast , LsST6 did not interact with the NP of rice grassy stunt virus , which cannot be transmitted by the same planthopper . LsST6 not only altered the cellular location of viral proteins and then colocalized with them in the cell membrane , but also mediated the entry of rice stripe virus ( RSV ) particles into Spodoptera frugiperda 9 ( Sf9 ) cells that expressed the heterologous gene LsST6 . We further showed that RSV particles initially bound to the cell membrane of midgut epithelial cells where it colocalized with LsST6 , and then invaded the cytoplasm . When LsST6 expression was knocked down , viral titre , acquisition percentage and transmission efficiency of the treated insect decreased significantly , but virus replication was not affected . This work thus uncovered a strategy by which LsST6 mediates viral entry into midgut epithelial cells and leads to successful transmission by the insect vector .
Many viruses persistently transmitted by arthropods cause serious diseases in plants , animals and humans . More than 76% of plant viruses and 40% of mammalian viruses are transmitted to the hosts by specific arthropods , mainly planthoppers , aphids , mosquitoes , and ticks [1 , 2] . Frequent epidemics of viral diseases in rice , wheat and vegetables are largely attributed to high populations and viral transmission efficiency of the insect vectors [3–6] . Similarly , viruses that cause diseases in humans and animals such as dengue fever , Zika fever and Japanese encephalitis , are vectored by different species of Aedes mosquitoes and are endemic in many areas of the developing world [7–10] . Understanding the virus–insect vector interaction and transmission mechanisms will provide important information on the epidemics of the diseases caused by plant and animal arboviruses and lead to the development of better control strategies . Plant viruses transmitted in a persistent propagative manner and animal arboviruses follow a similar circulative route within their insect vectors . After they are acquired from plant sap or blood ingested by the insect , the virions must first cross the cell membrane of the midgut epithelial cells where the viral particles multiply [11] . They must then leave the midgut to disseminate to other tissues including the salivary glands , from where they can be transmitted to new hosts [12] . During the circulative process , arboviruses must overcome multiple barriers , including the infection and dissemination barriers of the midgut , salivary gland escape barrier , and transovarial barrier [13 , 14] . Previous studies showed that Aedes aegypti cannot be infected by eastern equine encephalomyelitis virus after ingesting viruliferous blood; however , it can transmit this virus after a virus suspension is directly injected into the insect’s abdomen [15] . Many plant viruses can also be transmitted by an insect that is not a natural host after the virus is injected into the hemocoel of the insect [16 , 17] , thus bypassing the midgut infection barrier , the first major barrier that viruses encounter and an important factor limiting virus transmission [1 , 18 , 19] . To overcome the midgut barrier , viruses have evolved different strategies . The entry of rice dwarf virus into cultured cells of its vector insect and of tomato yellow leaf curl virus midgut in its vector Bemisia tabaci , is mediated by clathrin-dependent endocytosis [20 , 21] , whereas southern rice black-streaked dwarf virus ( SRBSDV ) induces the formation of tubules as a vehicle for viral spread in infected epithelial cells of Sogatella furcifera [22] . The small brown planthopper , Laodelphax striatellus ( Hemiptera: Delphacidae ) , is an important vector because it transmits numerous viruses that cause serious diseases of staple crops such as rice stripe virus ( RSV ) , rice black-streaked dwarf virus , maize rough dwarf virus , northern cereal mosaic virus and barley yellow striate mosaic virus [4 , 6 , 23] . These plant viruses infect and replicate in L . striatellus and are retained by the vector insect throughout their life , as are vertebrate-infecting arboviruses [24–26] . In most experiments when feeding on RSV-infected plants , less than 30% of the insects acquire the virus [27–29] . A high affinity line of L . striatellus attained an acquisition level of about 50–60% after 4 days of acquisition feeding on RSV-infected rice plants , but four other lines reached a level of less than 10% after 8–11 days of acquisition feeding [30] . However , once acquired , the virus will replicate and be transmitted by vector insects at a moderate to very high rate [24 , 31] . Although various biotic and abiotic factors affect virus acquisition by the vector insect , the epithelium , intercellular junctions , and basal lamina of the midgut present further barriers to viral entry and dissemination [13 , 32] . The genome of RSV consists of four single-stranded RNAs ( RNA1-4 ) , which can encode at least seven proteins including the major nucleocapsid protein ( NP ) encoded by the ORF at the 5′ half of the viral complementary RNA3 [33] . NP is considered the key viral component for specifically interacting with the vector components and may play an important role in persistent transmission process . In previous studies , 66 proteins ( including LsST6 ) were identified as being able to interact with the NP of RSV . Among these proteins , we chose several proteins according to molecular function and biological pathway to further investigate their function in virus transmission . CPR1 was demonstrated to stabilize the viral particles in the hemolymph [34] , while vitellogenin , the precursor of a yolk protein in the insect , mediates virus entry into the ovary [35] . However , the proteins involved in the ability of RSV to overcome the midgut infection barrier were not identified . Because the sugar transporter Glut1 of human acts as a receptor for human T cell leukemia virus ( HTLV ) infection [36] , and because a membrane protein named sugar transporter 6 of L . striatellus ( LsST6 ) is highly expressed in the midgut , we selected LsST6 for further study . Our results showed it is an essential and key factor for RSV to cross the midgut infection barrier in vector insects .
The full-length LsST6 ( GenBank accession: MG589412 ) , amplified from total RNA of L . striatellus using RT-PCR and 5’RACE , contained a 1470-bp open reading frame ( ORF ) encoding a predicted 489-amino-acid ( aa ) protein , which had 85 . 6% identity with sugar transporter NlST6 in the brown planthopper ( Nilaparvata lugens ) ( S1 Fig ) . LsST6 belongs to the major facilitator superfamily ( MFS ) of membrane transport proteins because it has two symmetrical six-TMS ( transmembrane spanner ) units within a single polypeptide chain and a GRK domain conserved between TM2 and TM3 ( S2 Fig ) . A yeast two-hybrid ( Y2H ) assay based on split ubiquitin was used to verify whether LsST6 interacts in vivo with the NP of RSV and rice grassy stunt virus ( RGSV ) or the p10 of RBSDV and SRBSDV . In nature , L . striatellus can acquire RSV , RBSDV and SRBSDV , but not RGSV . The Y2H results showed that the yeast cells cotransformed with LsST6 and RSV NP , RBSDV p10 or SRBSDV p10 , respectively , grew on the selective plates , but those with RGSV NP did not ( Fig 1A ) . A similar result was obtained in a β-galactosidase activity assay ( Fig 1A ) . Further , we exploited the cells of Spodoptera frugiperda 9 ( Sf9 ) to coexpress LsST6 and the respective viral proteins for in vitro coimmunoprecipitation ( Co-IP ) . LsST6 was labeled with a His tag and viral proteins with Myc tags at their C termini . The result showed that the anti-Myc antibody coimmunoprecipitated LsST6 that was coexpressed with RSV NP , RBSDV p10 or SRBSDV p10 in Sf9 cells , but not with RGSV NP ( Fig 1B ) . All these results demonstrated that LsST6 only interacted with the proteins encoded by viruses that can be acquired by L . striatellus . Sf9 cells were transfected using a recombinant baculovirus that produces LsST6 or different viral proteins . Laser scanning confocal microscopy ( LSCM ) revealed that LsST6 ( green ) was mainly localized in the cell membrane , whereas viral proteins ( red ) localized in the cytoplasm of Sf9 cells ( Fig 2A ) . Interestingly , RSV NP and p10 of RBSDV and SRBSDV moved from the cytoplasm to the cell membrane and colocalized with LsST6 , while RGSV NP remained in the cytoplasm , separated from LsST6 when Sf9 cells coexpressed the respective viral protein and LsST6 ( Fig 2B ) . The result suggested that the expression of LsST6 only changed the position of each interacting viral protein in Sf9 cells , and then colocalized with the protein on the cell membrane of Sf9 cells . The salivary gland , gut , hemolymph and ovary were excised separately from adult insects , then total RNA was extracted from individual tissues to quantify LsST6 by RT-qPCR . The results showed that LsST6 had the highest expression in gut tissues , followed by the hemolymph , salivary glands and ovary ( Fig 3A ) . The alimentary canal of the planthopper , which mainly comprises the esophagus ( es ) , anterior diverticulum ( ad ) , midgut ( mg ) , hindgut ( hg ) and malpighian tubules ( mt ) ( S3A Fig ) , was excised , and then incubated with anti-LsST6 antibody labeled with Dylight 488 ( green ) . The confocal image showed that LsST6 was located in the cell membrane and cytoplasm of the midgut epithelial cells ( Fig 3B ) . The distribution of RSV particles in the alimentary canal over time was also visualized by LSCM using anti-RSV antibody labeled with Dylight 549 ( red ) . At 2 d after a 2-d acquisition access period ( AAP ) , a few RSV particles were observed in several epithelial cells of the midgut . By 4 d after the AAP , RSV particles had replicated and spread to the neighboring epithelial cells , and by 8 d after the AAP , RSV particles were observed in the entire alimentary canal ( S3B Fig ) . Thus , the original RSV infection site was the epithelium of the midgut . Interestingly , at 2 d after the AAP , LsST6 had colocalized with RSV in the cell membrane of epithelial cells ( Fig 3C ) . At 4 d and 8 d after the AAP , viral particles were still colocalized with LsST6 in the cell membrane of the epithelial cells , and some particles had already invaded the cytoplasm , but the fluorescent signals indicative of the viral titre were stronger at 8 d ( Fig 3D and 3E ) . When we used immunoelectron microscopy to examine the virus-infected midgut epithelium , RSV particles also colocalized with LsST6 on the microvilli of cell membranes and cytoplasm ( Fig 4 ) . This evidence strongly suggested that LsST6 is a key factor for enabling RSV entry into the midgut epithelium of the planthopper . RSV particles were added to a liquid culture of Sf9 cells that expressed the heterologous gene LsST6; by 7 h , they had colocalized with LsST6 on the cell membrane of Sf9 cells . Few viral fluorescence signals were found in the cytoplasm by 15 h , but more signals were detected in the cytoplasm by 20 h ( Fig 5A ) . Immunoelectron microscopy also consistently showed the same results; RSV particles colocalized with LsST6 in the cell membrane at 7 h ( Fig 6A ) and were observed in the cytoplasm of Sf9 cells at 15 h and 20 h ( Fig 6B and 6C ) . We then quantified the mRNA level for RSV and LsST6 at various times using northern blots and RT-qPCR . The RSV mRNA level was obviously higher at 20 h than at 7 h when LsST6 was expressed ( Fig 5B ) . The RT-qPCR data also supported the northern blot results: at 15 h and 20 h , the level of viral RNA was 1 . 5 and 3 . 5 times higher , respectively , than at 7 h ( Fig 5D ) , whereas the level of LsST6 mRNA was approximately 1 . 2 and 1 . 3 times higher , respectively , at 15 h and 20 h ( Fig 5C ) . That viral particles passed though the cell membrane , mediated by LsST6 , revealed that LsST6 played a critical role in RSV entry Sf9 cells . When dsRNA of LsST6 ( dsLsST6 ) was injected into third-instar nymphs , LsST6 mRNA level in the midgut had declined by 61% and 87% at 1 d and 2 d compared with the control ( insects injected with dsGFP ) . Subsequently , LsST6 mRNA level remained 20–40% lower than in the control group injected with dsGFP ( Fig 7A ) . We then allowed third-instar nymphs that had been injected with dsLsST6 or dsGFP to feed on RSV-infected plants for a 2-d AAP , then quantified RSV and LsST6 mRNA levels in treated insects after different times using RT-qPCR and northern blots . The RSV mRNA level in insects injected with dsLsST6 was 22% to 35% of that in SBPH injected with dsGFP ( Fig 7C ) , indicating significant interference with LsST6 expression ( Fig 7B ) . The northern blot also showed that the quantity of RSV RNA was evidently lower than in the group injected with dsGFP after 4 and 8 d ( Fig 7D ) . In addition , viral acquisition and transmission by dsLsST6-injected SBPHs decreased by nearly 80% compared with those injected with dsGFP ( Fig 7E , S3 Table ) . Confocal images also showed fewer RSV infection sites and fewer RSV particles in a single epithelial cell than in the control at 2 and 4 d ( Fig 7F ) . Overall , these results demonstrated that RSV initial infection of the midgut epithelial cells was inhibited when LsST6 expression was knocked down . We also injected viruliferous insects with dsLsST6 or dsGFP . Compared with the RSV mRNA level in the insects injected with dsGFP , the level in the dsLsST6-injected insects only decreased by 10–20% at 2 , 4 and 8 d after injection , while the LsST6 mRNA level decreased by 65–70% ( Fig 8A and 8B ) . The northern blot assay showed that the change in RSV mRNA levels followed a similar trend in insects after injection with dsLsST6 or dsGFP ( Fig 8C ) . Confocal images also suggested that the quantity of RSV in epithelial cells was similar to the control ( Fig 8E ) . In addition , two groups of treated viruliferous insects had a similar transmission efficiency ( Fig 8D , S4 Table ) . Therefore , LsST6 has no significant effect on virus replication .
The insect midgut consists mainly of a single layer of epithelial cells , with extensive microvilli on the lumen side and a porous basal lamina on the hemocoel side [37 , 38] . The midgut absorbs the nutrients necessary for insect survival and provides an environment for the development and multiplication of viruses and parasites [39 , 40] . The midgut epithelial cells have been identified as the initial infection site and the first barrier to virus invasion [1 , 13 , 41] . Our results on the distribution of RSV particles in the alimentary canal over time also demonstrated that the midgut epithelial cells of L . striatellus served as the initial infection site of RSV . After successful invasion of the midgut , RSV began its replication process , then spread into neighboring cells . Most viruses invade insect epithelial cells via specific interaction between the structural proteins of the virus and the cell surface receptor complexes in vectors , similar to their infection of host cells [18 , 19 , 42 , 43] . Viral surface components have been well demonstrated to play an important role in virus infection and transmission [44 , 45] , and putative surface receptors including glycans and glycoconjugates for flaviruses have been found in host and insect cells [46 , 47] . A 32-kDa laminin-binding protein and 35-kDa prohibitin that mediate entry of Venezuelan equine encephalitis virus and Dengue virus-2 , respectively , into mosquito cells have been identified [48 , 49] . Membrane alanyl aminopeptidase N has been identified in the pea aphid as responsible for the entry of the pea enation mosaic virus into the aphid gut [50] . Most of these putative receptors in insects were discovered by in vitro interactions; in vivo evidence is still lacking . We thus used cellular and molecular biological techniques to advance our understanding of the interaction between virus and insect vector . Here , we found that LsST6 not only strongly interacts with RSV NP in vitro and in vivo , but also alters the cellular location of NP and then colocalizes with it in the cell membrane of Sf9 cells . Moreover , LsST6 mediates the entry of RSV particles into Sf9 cells that expressed the heterologous gene LsST6 . In the vector insect body , RSV initially binds to the cell surface of midgut epithelial cells where it colocalizes with LsST6 in the cell membrane . When expression of LsST6 was knocked down in healthy insects injected with dsLsST6 , viral titre and acquisition subsequently decreased significantly . Therefore , LsST6 plays an important role in facilitating virus invasion in both Sf9 model cells and the midgut epithelial cells of the vector insect , L . striatellus . Our previous bioinformatic analysis revealed that LsST6 has 85% similarity with NlST6 , a facilitative glucose/fructose transporter in brown planthoppers ( N . lugen ) [51] . They both belong to the major facilitator superfamily ( MFS ) of transporters , which are ubiquitous among organisms and enable the import and export of essential nutrients and ions ( not just sugars ) , the excretion of metabolic end products and deleterious substances and communication of the cells with the environment [52 , 53] . Some members of the MFS are also exploited by viruses to invade host cells . Glut1 , a receptor for HTLV [36] , was recently found to mediate glucose transport , which regulates human immunodeficiency virus ( HIV ) infection in human T cell lines [54] . Glut1 of shrimp is thought to be a putative cell surface receptor for white spot syndrome virus [55] . Feline leukemia virus subgroup C receptor ( FLVCR ) , another member of the MFS , is considered to be the cell surface receptor for feline leukemia virus [56 , 57] . HTLV and HIV infection of host cells are all regulated by Glut1-mediated glucose metabolism , via an increase in Glut1 expression and to a change in the conformation of the protein [53 , 54] . These studies thus provide a precedent for the involvement of another MFS member , LsST6 , in virus invasion of midgut epithelial cells in L . striatellus using a similar transport mechanism , rather than receptor-/clathrin-dependent endocytosis or membrane fusion and/or actin-based tubular structures to overcome the cell barriers . Knock down of the expression of LsST6 in healthy insects of L . striatellus , resulted in a decrease in virus acquisition after they fed on RSV-infected rice plants compared with the control insects with the functional gene . The viral titre in viruliferous insects that were similarly treated by injection with dsLsST6 had no significant changes . These results suggest that LsST6 mediates RSV entry into cells but is not involved in virus replication . On the basis of our results , we propose that LsST6 on the cell membrane of epithelial cells in the midgut mediates RSV invasion during facilitative transport of glucose/fructose from the phloem sap of rice plants across the cell membrane . Interestingly , we found that LsST6 not only interacted with the outer capsid of RBSDV and SRBSDV , but also colocalized with each in the cell membrane of Sf9 cells , but it did not colocalize with NP of RGSV . Because L . striatellus can transmit both RSV and RBSDV but not RGSV , we consider that the two transmitted viruses may also require LsST6 to mediate their entry into midgut epithelial cells . Although the planthopper cannot transmit SRBSDV efficiently , previous evidence has shown that this virus does invade midgut tissues , but it does not spread into the hemolymph or other organs of SBPH [58 , 59] . SRBSDV cannot break through the release barrier of the midgut because it cannot replicate enough to reach the threshold required for further spread , and/or the siRNA antiviral pathway has a direct role in controlling viral dissemination from the midgut [60 , 61] . RGSV cannot invade the midgut epithelium of L . striatellus , because LsST6 did not interact or colocalize with the NP of RGSV . Therefore , LsST6 might specifically mediate initial infection by the numerous viruses that are transmitted by L . striatellus . Based on all the data obtained , we propose a model by which RSV overcomes the midgut infection barrier in vector planthopper . After entering the alimentary canal of the vector insect and arriving in the midgut , intact RSV particles can bind to the midgut epithelial cells , where the NP of RSV interacts specifically with sugar transporter 6 on the cell membrane and is transported into the epithelial cells , where it replicates and finally disseminates to other parts of the vector ( Fig 9 ) . This model should also be applicable to other viruses transmitted by L . striatellus . In conclusion , our results provide direct evidence that LsST6 is essential for RSV invasion of the midgut epithelial cells in its insect vector . The fact that LsST6 can also interact and colocalize with the outer capsid or NP of other viruses transmitted by L . striatellus suggests that numerous arboviruses might use a similar vector protein to invade the midgut epithelium of the insect vector . This key vector protein could be used as a target for blocking virus transmission and lead to a new strategy to control outbreaks of diseases caused by arboviruses .
Nonviruliferous and viruliferous L . striatellus with a high affinity for rice stripe virus were reared on healthy and RSV-infected rice seedlings ( cv . Wuyujing 3 ) , respectively [29 , 34] . Every 3 months , the offspring of viruliferous insects were confirmed as RSV-positive using RT-PCR . Viruses ( RSV , RBSDV and SRBSDV ) -infected rice leaves collected from the field were stored at −80°C in the lab ( 61 ) and RGSV-infected rice plant was provided by Prof . Taiyun Wei ( Fujian Agriculture and Forestry University ) . RSV particles were extracted from RSV-infected rice leaves as described previously [62] , and purified virions were stored at −80°C . The mouse monoclonal anti-RSV antibody was kindly provided by Prof . Jianxiang Wu ( Zhejiang University ) . The rabbit anti-RSV antibody was the kind gift of Yan Huo ( Chinese Academy of Sciences ) . An anti-LsST6 monoclonal antibody against the LsST6 peptide SKGDHNTEAALP was produced by Abmart ( Shanghai , China ) . The following antibodies were obtained from the sources indicated: mouse monoclonal anti-Myc tag ( cat . 66004 , Proteintech ) , rabbit polyclonal anti-His tag ( cat . 2365 , Cell Signaling Technology ) , Dylight 488 goat anti-rabbit IgG ( cat . A23220 , Abbkine ) , Dylight 488 goat anti-mouse IgG ( cat . A23210 , Abbkine ) , Dylight 549 goat-anti-mouse IgG ( cat . A23310 , Abbkine ) , goat anti-mouse IgG+HRP ( cat . 32430 , Thermo ) , goat anti-rabbit IgG+HRP ( cat . 32460 , Thermo ) . Alexa Fluor 633 phalloidin was obtained from Invitrogen and 4’ , 6-diamidino-2-phenylindole ( DAPI ) was purchased from Sigma . The genes including RSV NP , RBSDV p10 , SRBSDV p10 , RGSV NP and LsST6 were amplified using specific primers ( S1–S2 Tables ) and were subsequently cloned into the bait plasmid pDHB1 or prey plasmid pPR3-N ( Dualsystems Biotech ) to generate pDHB1-RSV NP , RBSDV p10 , SRBSDV p10 or RGSV NP or pPR3-LsST6 . Genes including RSV NP , RBSDV p10 , SRBSDV p10 and RGSV NP linked with a Myc tag sequence were cloned into plasmid pFastBac ( Invitrogen ) , while LsST6 was inserted into the BamHI/XbalI sites of pFastBacHTB ( Invitrogen ) vector containing a 6×His tag . Sf9 cells were incubated in Sf-900 III SFM Serum Free medium containing 5% newborn calf serum at 27°C . To confirm any interaction between LsST6 and RSV NP , RBSDV p10 , SRBSDV p10 or RGSV NP , we used the DUALhunter starter kit , a yeast two-hybrid system based on the reconstitution of ubiquitin . A clone of yeast strain NMY51 was selected and incubated in 25 ml yeast peptone dextrose adenine agar ( YPDA ) at 30°C with shaking until the culture reached an OD546 of 0 . 6–0 . 8 . The culture was collected by centrifugation , and the pellet was suspended in 1 . 5 ml water . Then , 1 . 5 μg bait vector plasmid ( pDHB1-NPs/-p10s ) and 1 . 5 μg prey vector plasmid ( pPR3-LsST6 ) were added to 100 μl culture resuspended in 300 μl PEG/ lithium acetate Master Mix ( 50% PEG , 1 M lithium acetate and 125 μl single-stranded carrier DNA ) and incubated in a 42°C water bath for 45 min . Finally , the mixture was collected by centrifugation at 700 × g for 5 min , and the pellet resuspended in 100 μl 0 . 9% NaCl ( wt/vol ) , then plated onto selection plates of DDO ( SD-trp-leu ) and QDO ( SD-trp-leu-his-ade ) medium with 20 mM 3-aminotriazole ( 3-AT ) and incubated for 3–4 days at 30°C . For distinguishing false-positive interactions , the clone grown on DDO was incubated on 1 ml liquid DDO overnight at 30°C with shaking until the culture reached an OD546 of 0 . 5–0 . 8 to check for β-galactosidase activity . The culture was then collected by centrifugation , and the pellet was added to 100 μl of freshly prepared lysis mixture ( 9 . 95 ml of one-step lysis and assay reagent with 50 μl of dye stock solution ) provided in the HTX High-throughput β-galactosidase kit with a brief vortex . Sf9 cells were transfected using Cellfectin II according to the manufacturer’s instructions . Briefly , 2 × 106 cells were added to each well of a 6-well culture dish for at least 30 min . Then 2 μg recombinant bacmid–plasmid DNA encoding LsST6 , RSV NP , RBSDV p10 , SRBSDV p10 or RGSV NP was mixed with 8 μl Cellfectin II and incubated for 20 min . Then each mixture was added to a well of the 6-well culture dish and incubated at 27°C for 5 h . The transfection mixture was then removed and replaced with growth medium . Transient expression was measured 72 h later using LSCM and western blot . Genes including RSV NP , RBSDV p10 , SRBSDV p10 and RGSV NP linked with a Myc tag sequence were cloned into plasmid pFastBac ( Invitrogen ) , while LsST6 was inserted into the BamHI/XbalI sites of pFastBacHTB ( Invitrogen ) vector containing a 6×His tag . Recombinant LsST6 DNA was used to cotransfect Sf9 cells with recombinant RSV NP , RBSDV p10 , SRBSDV p10 or RGSV NP DNA during a 72-h incubation at 27°C , then cultures were collected and lysed in the lysis buffer ( 20 mM Tris-HCl pH 7 . 6 , 150 mM NaCl , 0 . 5% NP-40 , 5 mM EDTA , and complete protease inhibitor cocktail tablets ) for 1 h on ice . After the cultures were centrifuged at 10 , 000 × g for 20 min at 4°C , 50 μl protein A/G plus agarose beads was added to the supernatant for 1 h at 4°C to decrease any nonspecific binding of proteins . Then the supernatant was incubated with 2 μl ( 1 μg/μl ) anti-Myc antibody for 1 h , followed by incubation with protein A/G plus agarose beads at 4°C with end-over-end agitation for overnight . The protein A/G plus agarose beads were washed with washing buffer ( 20 mM Tris-HCl pH 7 . 6 , 150 mM NaCl , 5 mM EDTA ) 5 times , then mixed with 1× loading buffer ( 0 . 08 M Tris pH 6 . 8 , 2 . 0% SDS , 10% glycerol , 0 . 1 M dithiothreitol , and 0 . 2% bromophenol blue ) and finally boiled for 5 min . The cell lysates and IP cultures were separated on SDS-PAGE gels and transferred to nitrocellulose membranes . Membranes were incubated with antibody of anti-His or anti-Myc ( 1:3000 ) for 1 . 5 h at room temperature , then incubated with the secondary antibody-alkaline phosphatase-conjugated goat anti-mouse or rabbit IgG ( 1:5000 ) at 37°C for 1 h . Membranes were then incubated with a chemiluminescent substrate mixture and imaged using the imageQuant LAS 4000 mini biomolecular imager ( GE Healthcare Life Sciences , USA ) . RSV particles ( 1 . 5 μg/μl ) were added to the Sf9 cells transfected by recombinant bacmids LsST6 or empty bacmids at 48 h , then the sf9 cells were rinsed with PBS buffer 3 times and collected at 7 , 15 and 20 h after RSV particles were added to the culture medium . To avoid the influence of residual inoculum , the cells are collected by centrifugation to remove the inoculum . Then they were washed with double-distilled water for 3 times to avoid residual inoculum , finally they were prepared for RNA extraction . The relative quantity of RSV particles RNA was determined using RT-qPCR and DIG-northern blot . A partial sequence of LsST6 and GFP was amplified using specifically designed primers as templates to synthesize dsRNA using the protocol for the T7 RiboMAX Express RNAi System . Approximately 400 third-instar and nonviruliferous nymphs of L . striatellus were injected with 23 nl dsLsST6 ( 2 . 5 μg/μl ) or GFP ( 2 . 5 μg/μl ) using an Auto-Nanoliter Injector ( Drummond , USA ) and then allowed to feed on the RSV-infected rice plant for a 2-day acquisition access period ( AAP ) . At 2 , 4 and 8 days after the AAP , RNA was extracted from 50 insects to quantify the LsST6 and RSV mRNA levels by RT-qPCR and DIG-northern blot , and the midgut tissues were excised from 50 insects for immunofluorescence assay . The remaining insects were tested for RSV by RT-PCR with RSV-specific primers . To identify the influence of injection on RSV titre and transmission in insects , we injected 400 third-instar and viruliferous nymphs with 23 nl dsLsST6 ( 2 . 5 μg/μl ) or dsGFP ( 2 . 5 μg/μl ) and then transferred them to healthy rice seedlings . Fifty insects were collected for RNA extraction and detection at 2 , 4 and 8 days after injection . One hundred insects were transferred to healthy seedlings ( 1/seedling ) for a 10 h inoculation access period , and the seedlings were then grown in a greenhouse for 3 weeks . The infection status of each seedling was assessed by RT-PCR using specific primers for RSV NP . Midguts of the remaining insects were excised for immunofluorescence assay . RT-qPCR and northern blot were used to quantify any changes in RNA levels for RSV and LsST6 , and LSCM was used to visualize RSV particles in the excised midgut of L . striatellus . Each set of experiments was repeated three times . cDNA was synthesized from 1 μg of total RNA using a FastQuant RT Kit according to the manufacturer’s instructions at 4°C for 3 min , 42°C for 15 min , and 95°C for 3 min . The RT-qPCR was performed using a SuperRealPreMix Plus ( SYBR Green ) kit , a reaction volume of 20 μl ( 10 μl of PCR buffer , 0 . 6 μl of each primer [10 μM/μl] , 3 μl of template cDNA , and 5 . 4 μl of DEPC H2O and 0 . 4 μl 50× ROX Reference Dye ) and ABI-7500 thermocycler ( Applied Biosystems ) . The thermocyling program was 94°C for 15 min , followed by 40 cycles of 95°C for 10 s and 60°C for 32 s . Fluorescence was measured at the end of every 60°C extension phase . Beta-actin of SBPHs or ecdysoneless ( ECD ) of Sf9 cells were used for normalization as housekeeping genes in respective experiments . RT-qPCR data were analysed using the Livak method ( 2−ΔΔCt ) [63] . The experiments were repeated 3 times independently . RNA probes for LsST6 and RSV detection were labeled with DIG using the DIG Northern Starter Kit according to the manufacturer’s instructions . Total RNA extracted from SBPH or Sf9 cells was separated in 1 . 2% formaldehyde agarose gel and then transferred to nylon membranes using a vacuum regulator ( Bio-Rad , USA ) for 3–4 h . The membranes were then incubated with the respective RNA probes for 16–20 h in a 65°C hybridization oven , incubated with anti-digoxigenin-AP for 30 min , then in CDP-Star solution for 5 min and imaged with the imageQuant LAS 4000 mini ( GE , USA ) . Sf9 cells previously fixed on cover slips were incubated in 4% ( wt/vol ) paraformaldehyde for 30 min at room temperature . After being washed 3 times with PBS buffer , the samples were then incubated in osmotic buffer ( 2% [vol/vol] TritonX-100 in PBS ) for 15–30 min at room temperature , then incubated with anti-Myc monoclonal antibody ( MAB ) ( 1:400 ) , anti-His MAB ( 1:400 ) , anti-His rabbit polyclonal AB ( 1:400 ) or anti-RSV MAB ( 1:500 ) in PBS containing 3% ( wt/vol ) BSA at room temperature for 1 h and then with goat anti-mouse ( 1:400 ) or goat anti-rabbit ( 1:400 ) secondary antibody labeled with Dylight 488 or Dylight 549 in PBS for 1 h at 37°C after extensive washing with PBS buffer . The nucleus was stained with 50 nM 4′ , 6-diamidino-2-phenylindole ( DAPI ) in PBS at 37°C for 10 min . Midgut tissues excised from the planthoppers were fixed in 4% ( vol/vol ) paraformaldehyde for 2 h at room temperature and incubated in osmotic buffer for 30 min at room temperature . Then the samples were incubated in anti-RSV MAB labeled with Dylight 549 ( red ) or anti-ST6 labeled with Dylight 488 ( green ) MAB for 1 . 5 h at room temperature . All samples were visualized with LSCM ( Zeiss LSM880 , GER ) , and the images saved in ZEN 2011 blue light . Dissected midguts or Sf9 cells were fixed for 2 h in 2% ( vol/vol ) paraformaldehyde and 2% ( wt/vol ) osmium tetroxide in PBS , and after sequential dehydration in 30% , 50% , 70% , 90% , 100% and 100% alcohol , midguts or Sf9 cells were embedded in LR Gold Resin ( cat . 62659 , Sigma ) . Sections ( 70–90 nm ) of the midguts or Sf9 cells were cut using an ultramicrotome ( Leica , GER ) , then blocked for 30 min in blocking buffer . The sections were then incubated at room temperature with the antibodies in the following order: anti-RSV rabbit serum ( 1:300 ) for 1 . 5 h , 10-nm gold-conjugated goat-anti-rabbit IgG for 1 h , anti-LsST6 mouse serum ( 1:50 ) for 1 . 5 h and 5-nm gold-conjugated goat-anti-mouse IgG for 1 h with a wash in distilled water after each antibody incubation . They were then stained in 2% neutral uranyl acetate ( w/v in distilled water ) for 10 min . The sections were viewed with a transmission electron microscope at 80 kV accelerating voltage . Means ± SEM of three independent experiments ( one-way ANOVA , least significant difference test ) were statistically analysed using Prism 6 software ( GraphPad ) ; *P < 0 . 01 was considered statistically significant . | Sap/blood-feeding arthropods are major vectors of viruses that infect plants and vertebrates . Studies on the insect midgut , the first barrier for virus transmission , and its interactions with viruses and parasites are fundamental to understanding the transmission mechanism in vector insects and the epidemics caused by the vectored pathogen . Some putative receptors in arthropods have been discovered by in vitro protein interactions , but in vivo evidence is still lacking . Here , we found that the specific interaction between viral nucleocapsid protein and vector sugar transporter 6 of Laodelphax striatellus ( LsST6 ) determines whether the virus can invade midgut epithelial cells or not . These results provide direct evidence that LsST6 is an essential and key factor in crossing the midgut infection barrier for viruses , especially for RSV . This vector protein may be a promising target for blocking transmission of diverse plant viruses . Our discovery has important implications for better understanding the interaction among host–virus–insect vector and disease epidemics caused by plant and animal arboviruses . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"invertebrates",
"medicine",
"and",
"health",
"sciences",
"viral",
"transmission",
"and",
"infection",
"rna",
"extraction",
"microbiology",
"light",
"microscopy",
"animals",
"epithelial",
"cells",
"viral",
"vectors",
"rice",
"confocal",
"laser",
"microscopy",
"microscopy",
"experimental",
"organism",
"systems",
"confocal",
"microscopy",
"insect",
"vectors",
"plants",
"cellular",
"structures",
"and",
"organelles",
"extraction",
"techniques",
"research",
"and",
"analysis",
"methods",
"infectious",
"diseases",
"grasses",
"animal",
"cells",
"biological",
"tissue",
"disease",
"vectors",
"insects",
"cell",
"membranes",
"arthropoda",
"eukaryota",
"plant",
"and",
"algal",
"models",
"cell",
"biology",
"anatomy",
"virology",
"epithelium",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"species",
"interactions",
"organisms"
] | 2018 | Invasion of midgut epithelial cells by a persistently transmitted virus is mediated by sugar transporter 6 in its insect vector |
Modified vaccinia virus Ankara ( MVA ) is an attenuated double-stranded DNA poxvirus currently developed as a vaccine vector against HIV/AIDS . Profiling of the innate immune responses induced by MVA is essential for the design of vaccine vectors and for anticipating potential adverse interactions between naturally acquired and vaccine-induced immune responses . Here we report on innate immune sensing of MVA and cytokine responses in human THP-1 cells , primary human macrophages and mouse bone marrow-derived macrophages ( BMDMs ) . The innate immune responses elicited by MVA in human macrophages were characterized by a robust chemokine production and a fairly weak pro-inflammatory cytokine response . Analyses of the cytokine production profile of macrophages isolated from knockout mice deficient in Toll-like receptors ( TLRs ) or in the adapter molecules MyD88 and TRIF revealed a critical role for TLR2 , TLR6 and MyD88 in the production of IFNβ-independent chemokines . MVA induced a marked up-regulation of the expression of RIG-I like receptors ( RLR ) and the IPS-1 adapter ( also known as Cardif , MAVS or VISA ) . Reduced expression of RIG-I , MDA-5 and IPS-1 by shRNAs indicated that sensing of MVA by RLR and production of IFNβ and IFNβ-dependent chemokines was controlled by the MDA-5 and IPS-1 pathway in the macrophage . Crosstalk between TLR2-MyD88 and the NALP3 inflammasome was essential for expression and processing of IL-1β . Transcription of the Il1b gene was markedly impaired in TLR2−/− and MyD88−/− BMDM , whereas mature and secreted IL-1β was massively reduced in NALP3−/− BMDMs or in human THP-1 macrophages with reduced expression of NALP3 , ASC or caspase-1 by shRNAs . Innate immune sensing of MVA and production of chemokines , IFNβ and IL-1β by macrophages is mediated by the TLR2-TLR6-MyD88 , MDA-5-IPS-1 and NALP3 inflammasome pathways . Delineation of the host response induced by MVA is critical for improving our understanding of poxvirus antiviral escape mechanisms and for designing new MVA vaccine vectors with improved immunogenicity .
Attenuated poxviruses are currently being developed as vaccines vectors against various infectious diseases including HIV , malaria and tuberculosis [1] . Modified vaccinia virus Ankara ( MVA ) and NYVAC are highly attenuated poxvirus strains due to multiple deletions of viral genes and are replication-deficient in human cells . MVA and NYVAC are immunogenic and safe and have been shown to be excellent vaccine vectors for the expression of foreign antigens . MVA is a leading vaccine candidate for delivery of HIV genes with efficient induction of T-cell mediated immune responses [1]–[3] . Profiling of the immune responses triggered by poxvirus vaccine vectors is critical not only for optimal design of vaccine vectors but also for anticipating potential harmful interactions between naturally acquired or vaccine-induced immune responses against the vaccine target . This is indeed an important lesson learned from the adenovirus type 5 ( Ad5 ) HIV vaccine ( MRKAd5 ) STEP trial . Pre-existing neutralizing antibodies against the Ad5 vaccine vector were found to increase the relative risk of HIV infection [4] , [5] . Hence the need for extensive assessments of vaccine-induced innate and adaptive immune responses to prevent unexpected adverse events . Sensing of invasive pathogens by sentinel innate immune cells is a fundamental feature of the host antimicrobial defense response . Toll-like receptors ( TLRs ) , retinoic acid-inducible gene-I ( RIG-I ) like receptors ( RLRs ) and nucleotide-binding and oligomerization domain ( NOD ) -like receptors ( NLRs ) have recently emerged as central innate sensors of viruses [6] . Virus sensing by TLR occurs at the cell surface and in the endosomal compartment . At the cell surface , TLR2 or TLR4 recognize either DNA ( herpes viruses ) or RNA viruses ( respiratory syncitial , hepatitis C and measles viruses ) . In the endosomal compartment , TLR7 , TLR3 or TLR9 sense single stranded ( vesicular stomatitis virus , Sendai , West Nile and influenza viruses ) and double stranded ( reovirus ) RNA viruses , and DNA viruses ( herpes simplex viruses , cytomegalovirus ) , respectively [7]–[13] . Two members of the cytosolic pattern recognition RLR receptors , RIG-I ( also known as DDX58 ) and melanoma differentiation-associated gene 5 protein ( MDA5 ) ( also known as helicard ) , have been shown to function as sensors of RNA viruses [14]–[19] . RIG-I detects 5′-triphosphate of ssRNAs and short dsRNAs , while MDA5 preferentially recognizes long dsRNAs . NALP3 ( NLRP3 also known as cryopyrin ) is a member of the NLR family which have been involved in the sensing of both DNA ( adenovirus ) and RNA ( rotavirus , Sendai and influenza viruses ) viruses [20] , [21] . NALP3 , ASC and pro-caspase 1 form a multimeric cytosolic molecular complex known as the NALP3 inflammasome that controls the processing of the IL-1β cytokine precursor pro-IL-1β into IL-1β [22] . Sensing of viruses by TLRs , RLRs and NLRs activates intracellular signalling pathways resulting in the expression of pro-inflammatory cytokines and type I interferons that then act on innate immune cells to limit viral replication and promote the adaptive immune response . Here we report that the TLR2-TLR6-MyD88 , MDA-5-IPS-1 and NALP3 inflammasome pathways are the main innate sensors of MVA in the macrophage and that they induce a cytokine response profile characterized by a vigorous chemokine , IFNβ and IL-1β production . Beyond the dissection of the molecular bases of MVA recognition by the innate immune system the present data are likely to help design MVA vaccine vectors with improved immunogenicity .
The profile of innate immune responses elicited by MVA was first examined by RT-PCR and ELISA in a mouse model of poxvirus infection [23] . MVA infection induced a robust innate immune response in peritoneal cells , peritoneal lavage fluid , splenocytes and splenocyte homogenates characterized by the production of pro-inflammatory cytokines ( TNF , IL-1β , IL-6 , IL-12p40 ) , chemokines ( IP-10/CXCL10 , RANTES/CCL5 , MCP-5/CCL12 , MIP-2/CXCL2 ) and type I interferon ( IFNβ ) mRNA and protein ( Figure 1A and B and data not shown ) . Infection of human whole blood with MVA also induced a vigorous innate immune response characterized by an abundant production of chemokines ( IL-8/CXCL8 , MIP-1α/CCL3 and IP-10 ) and less abundant production of pro-inflammatory cytokines ( TNF , IL-1β , IL-6 ) ( Figure 2 ) . Interestingly , MVA was previously shown to down-regulate IL-8 and IL-1β mRNA expression in human monocyte-derived dendritic cells [24] , [25] , suggesting that MVA infection may induce the production of various patterns of cytokine depending upon the cell-type studied . Dissection of the molecular mechanisms of MVA-induced innate immune responses was preformed in PMA-differentiated human THP-1 macrophages and primary human macrophages . Flow cytometry analyses performed with GFP-expressing MVA ( MOI 5 ) indicated that MVA rapidly infected THP-1 cells ( Figure 3A and B ) . More than 60% of cells became GFP positive within 2 h followed by a progressive decline of GFP fluorescence thereafter , which could result either from MVA-induced apoptosis as observed in human HeLa and monocyte-derived dendritic cells [24] , [25] or from the shutting down of protein synthesis through activation of the PKR pathway by MVA [26] . Indeed , the number of apoptotic cells increased from 5% at 6 h to 35% at 24 h post-infection as assessed by annexin V and propidium iodine staining ( data not shown ) . The profile of cytokines and chemokines released by MVA-infected THP-1 cells was analyzed with the Luminex technology . Twenty four h after infection , 12 of the 30 mediators analyzed ( see Materials and Methods ) were detectable in cell-culture supernatants . Similarly to the results obtained with human whole blood ( Figure 2 ) and in agreement with a recent report by Lehmann et al . [27] , MVA induced the production of large quantities of chemokines ( IL-8 , MIP-1α , MIP-1β/CCL4 , MCP-1/CCL2 , RANTES and IP-10 ) . MVA also induced large amounts of IFNβ and of IL-1ra , but small amounts of pro-inflammatory cytokines ( TNF , IL-1α , IL-1β , IL-6 and IL-12p40 ) ( Figure 3C and D ) . Kinetics and patterns of chemokines and type I interferon mRNA expression were similar in MVA-stimulated THP-1 cells and primary human macrophages ( Figure 3E and F ) . We then also examined the production of cytokines and chemokines induced by two other vaccinia virus ( i . e . the attenuated NYVAC strain and the virulent Western Reserve strain ) . When compared to MVA , NYVAC induced low levels of IL-8 , IL-1β and IFNβ and no TNF , IL-6 , MIP-1α , RANTES or IP-10 ( Figure S1 ) . The virulent Western Reserve strain of vaccinia virus was observed to also induce low levels of IL-8 and IFNβ in THP-1 cells , but no IL-1β , MIP-1α or IP-10 ( Figure S2 and data not shown ) . Altogether , these results indicated that the innate immune response induced by MVA in human macrophages was characterized by a powerful chemokine production and a less abundant production of pro-inflammatory cytokines probably related to the attenuation of MVA [28] . In contrast , the NYVAC and Western reserve strains stimulated less powerful chemokine and cytokine responses , that most likely reflect differences in the expression of immunomodulatory genes in the genome of MVA , NYVAC and Western Reserve [24] , [25] . TLRs have been shown to play an important role in the sensing of viruses and in the initiation of the anti-viral host defense response [29] , [30] . Analyses of the TLR repertoire used by the host for sensing of MVA were conducted in bone marrow-derived macrophages ( BMDMs ) isolated from TLR1−/− , TLR2−/− , TLR4−/− , TLR6−/− , MyD88−/− and TRIF−/− mice and the read-out was the expression of IFN-independent chemokine MIP-2 and of IFNβ . MVA-induced MIP-2 production by BMDMs was completely abrogated in TLR2−/− , TLR6−/− and MyD88−/− cells but not in TLR1−/− , TLR4−/− and TRIF−/− cells , which produced amounts of MIP-2 similar to that of wild-type cells ( Figure 4A ) . In contrast , the amount of IFNβ produced by TLR2−/− , TLR6−/− and MyD88−/− BMDMs was similar to that of wild-type cells ( Figure 4B ) , a finding consistent with the notion that activation of the TLR2 pathway is not implicated in the production of type I IFNs . Similar results were obtained with THP-1 cells stably transduced with a lentiviral delivery system expressing a short hairpin RNA ( shRNA ) targeting the expression of the TLR2 gene ( Figure S3 ) . All together , these results indicated that the activation of the TLR2-TLR6-MyD88 pathway was required for the induction of IFNβ-independent chemokines in MVA-stimulated macrophages . Experiments conducted with NYVAC and the Western Reserve strain of vaccinia virus confirmed that TLR2 was required for IL-8 production by THP-1 cells ( Figure S1 and S2 ) . Vaccinia virus penetrates into target cells either by endocytosis or by membrane fusion in a low pH-independent manner [31] . To determine the contribution of endocytosis to MVA-induced intracellular signalling and cytokine production , THP-1 cells were treated with cytochalasine D , an actin-depolymerizing drug that blocks the endocytotic trafficking , or with chloroquine , a lysosomotropic weak base to neutralize the acidic environment of endocytic vesicles . IL-1β and to a lesser extend IFNβ production were inhibited by cytochalasine D and chloroquine treatment . The inhibition was not related to drug toxicity because chloroquine did not affect IL-8 production and cell viability ( Figure 5 and data not shown ) . The reason why the inhibition of cytokine production ( particularly IFNβ ) was only partial after treatment with the inhibitors remains uncertain . The data suggest that additional non-endocytic pathways may play a role in the production of IFNβ . In agreement with a key role for membrane-bound TLR2 for IL-8 induction , the production of IL-8 was not reduced after cytochalasine D or chloroquine treatment ( Figure 5 ) . UV treatment of MVA causing a nearly complete ( i . e . 90% ) inhibition of the expression of the early C6L gene ( data not shown ) did not affect IL-1β , IL-8 and IFNβ production ( Figure 5 ) . Although one cannot completely rule out a contribution of residual viral protein synthesis , these observations support the view that induction of cytokines by MVA is most likely independent of viral gene synthesis [32]–[34] . Overall , endocytosis of MVA was required for IL-1β and IFNβ release suggesting a role for intracellular pattern recognition receptors in the production of these cytokines . The RLR family of cytosolic pattern recognition receptors has been implicated in the sensing of RNA viruses [35] , but very little is known about their role in host response to DNA viruses . Extending the observations by Guerra et al . who noted an increased expression of RIG-I and MDA-5 mRNA in human dendritic cells infected with MVA [24] , we observed that MVA caused a time-dependent increase in RIG-I , MDA-5 and IPS-1 mRNA and protein expression in THP-1 cells ( Figure 6A and B ) . RIG-I and MDA-5 mRNAs rose within 3 h of infection and remained elevated for up to 24 h ( Figure 6A ) . In vivo , MVA up-regulated RIG-I and MDA-5 mRNA levels in peritoneal cells and splenocytes ( Figure S4 ) . When compared to MVA , NYVAC induced lower levels of MDA-5 and , to a lesser extent , RIG-I and IPS-1 mRNA and protein expressions ( Figure S1 and data not shown ) . Using shRIG-I , shMDA-5 and shIPS-1 THP-1 cells ( Figure S5 ) , we then examined whether RIG-I and MDA-5 were involved in MVA-induced IFNβ production . IFNβ and IP-10 mRNA and protein levels were markedly reduced in shMDA-5 and shIPS-1 cells , but not in shRIG-I cells . By contrast , the time-course and magnitude of the IL-8 and IL-1β production was similar in shMDA-5 , shIPS-1 , shRIG-I and control THP-1 cells ( Figure 7A and B ) . Sensing of MVA by the MDA-5/IPS-1 pathway is therefore critical for the production of IFNβ and IFNβ-dependent chemokines in macrophages . In line with these data , the production of IFNβ , but not of IL-8 , was also dependent on the MDA-5/IPS-1 pathway in cells infected with NYVAC and the Western Reserve strain of vaccinia virus ( Figure S1 and S2 ) . IL-1β is a key cytokine of antimicrobial host defenses , whose expression is regulated at a transcriptional and post-transcriptional level [36] . IL-1β is likely to play an important role during poxvirus infection , as suggested by the fact that poxviruses encode for IL-1β decoy receptor and disrupt intracellular IL-1 receptor signalling [37] , [38] . We therefore examined whether activation of the TLR2-MyD88 pathway was implicated in the activation of the IL1b gene . As shown in Figure 8A , up-regulation of IL-1β mRNA was markedly impaired in TLR2−/− and MyD88−/− BMDMs infected with MVA , indicating that activation of the TLR2-MyD88 signalling pathway is critical for transcription of the IL1b gene during MVA infection . Secretion of mature IL-1β p17 in response to endogenous and exogenous danger signals requires the cleavage of the inactive pro-IL-1β precursor by the cysteine protease caspase-1 . Conversion of pro-caspase-1 into caspase-1 is tightly regulated by the NALP3 inflammasome composed of NALP3 , ASC and pro-caspase-1 [22] . To examine the contribution of the NALP3 inflammasome in the production of IL-1β triggered by MVA , we analyzed the expression of pro-IL-1β and IL-1β p17 in THP-1 cells deficient in NALP3 , ASC or caspase-1 [39] . Knocking down of either one of the three components of the NALP3 inflammasome ( i . e . NALP3 , ASC or caspase-1 ) was associated with a massive reduction of mature and secreted IL-1β ( Figure 8B and C ) . Similar results were obtained in THP-1 cells infected with NYVAC ( Figure S1 ) and in NALP3−/− BMDMs infected with MVA ( Figure 8D and E ) . Of note , in THP-1 cells and in BMDMs the expression of pro-IL-1β was unaffected by the absence of either NALP3 , ASC or caspase-1 clearly indicating that NALP3 inflammasome does not itself regulate the transcriptional and translation control of the IL-1β precursor . The NALP3 inflammasome was also dispensable for activation of the IRF3 transcription factor and IFNβ secretion ( Figure S6 ) . Altogether , these data demonstrate that IL-1β production after MVA infection requires a crosstalk between TLR2-MyD88 ( initiation of the transcription and translational of IL-1β ) and the NALP3 inflammasome ( processing of pro-IL-1β into mature IL-1β ) . Poxviruses have been reported to activate the NF-κB , ERK1/2 and JNK pathways in epithelial and fibroblastic cell lines [40]–[43] and IRF3 and IRF7 in dendritic cells [24] , [25] . Having identified the pathogen recognition receptors implicated in macrophage response to MVA ( TLR2-TLR6 , MDA-5 and NALP3 ) , we next examined which downstream signalling pathways are activated for the expression of cytokines , chemokines and type I IFNs . Kinetics studies of NF-κB , ERK1/2 and JNK MAP kinases and IRFs activation were performed in THP-1 cells ( Figure 9A ) . Electrophoretic mobility shift assay revealed that NF-κB nuclear content peaked 3 h after MVA infection . Phosphorylation of the ERK1/2 and JNK MAP kinases was between 1 and 6 h after infection . IRF3 , which is essential for transcription of the IFNB gene , was detected 3 h after infection , peaked at 6 h and rapidly decreased thereafter . IRF7 was detected 3 h after infection and levels remained unchanged for 24 h . Phosphorylation of signal transducer and activator of transcription 1 ( STAT-1 ) , a critical target of IFNβ signalling required for the transcriptional activation of IFNβ-dependent genes , was first detected 3 h post-infection and gradually increased until 24 h ( Figure 9A ) . The functional significance of the increased binding activity of NF-κB and phosphorylation of the IRF3 was confirmed by showing that MVA increased the transcriptional activities of multimeric-κB and IRF3-dependent-IFNβ promoter luciferase reporter vectors in transiently transfected THP-1 cells ( Figure 9B and C ) . Confirming the importance of NF-κB and ERK1/2 in mediating innate immune response to MVA infection , pre-incubation of THP-1 cells with drugs ( i . e . NEMO and U0126 , see Materials and Methods ) selectively inhibiting the NF-κB and ERK-1/2 signalling pathways impaired , albeit to a different extent , IL-1β ( 70% and 65% inhibition ) , IL-8 ( 75% and 72% inhibition ) and IFNβ ( 28% and 42% inhibition ) mRNA expression ( p<0 . 05 for all conditions ) . Therefore , consistent with the fact that several pattern recognition receptors are engaged in the sensing of MVA by the innate immune system , multiple intracellular signalling pathways , including NF-κB , MAP kinases and IRFs were found to be activated upon infection of THP-1 macrophages with MVA . Of note , NYVAC induced very weak induction of intracellular signalling ( i . e . NF-κB , ERK-1/2 , IRF3 and STAT-1 ) and low levels of cytokines and IFNβ when compared with MVA ( Figure S1 ) which is likely due to the expression of different patterns of immunomodulatory genes by these two poxviruses [24] , [25] .
Analyses of pattern recognition receptors engagement by poxviruses are essential for improving our understanding of the pathogenesis of this important class of DNA viruses and for designing new viral vaccine vectors with improved immunogenicity . Dissection of the molecular bases of innate immune responses elicited by the attenuated poxvirus MVA strain in human macrophages revealed a critical role for TLR2-TLR6-MyD88 , MDA-5-IPS-1 and NALP3 inflammasome pathways in the production of chemokines , IFNβ and IL-1β . These observations provide novel information on MVA recognition by sentinel innate immune cells and highlight the existence of potential differences between attenuated and non-attenuated poxviruses in the engagement of or recognition by innate sensors . Up to now the retinoic acid-inducible gene-I-like receptors ( RLR ) RIG-I and MDA-5 had been viewed as master cytosolic sensors of RNA viruses [29] . However , recent observations suggested a role for the RLR pathway in the recognition of DNA viruses . Mouse embryo fibroblasts deficient in IPS-1 displayed reduced induction of IFNβ in response to MVA lacking the E3 protein [44] . Adenovirus and HSV1 have also been shown to replicate at much higher titers in RIG-I mutant than in RIG-I wild-type human hepatoma cell lines [45] . Moreover , microarray analyses revealed that RIG-I and MDA-5 expression was upregulated in human monocyte-derived dendritic cells infected with MVA [24] . Here we also showed that MVA caused a strong up-regulation of RIG-I , MDA-5 and IPS-1 , yet only MDA-5 and IPS-1 were found to mediate MVA-induced IFNβ and IFNβ-dependent chemokine production by macrophages ( Figure 10 ) . As anticipated , transcriptional activation of IFNb and IFNb-dependent chemokine genes was associated with the activation of IRF3 and IRF7 and STAT-1 . To the best of our knowledge this is the first demonstration of a direct role for MDA-5 in innate sensing of a DNA virus . Moreover , the MDA-5/IPS-1 pathway was also implicated in the production of IFNβ by macrophages infected with the NYVAC and the Western Reserve strains of vaccinia virus ( Figure S1 and S2 ) . RIG-I has been shown to be involved in the induction of TNF and type I IFN by myxoma poxvirus in human macrophages [46] . Yet , silencing of MDA-5 was associated with a small ( about 25% ) but clear reduction of macrophages response to myxoma virus suggesting that both RIG-I and MDA-5 were implicated , albeit to various degree , in innate immune response to myxoma virus . The nature of the component ( s ) of DNA viruses activating the RLR pathway remains to be identified . Obvious candidate molecules include , envelope or core proteins , early mRNA and DNA itself . Unless RLR engagement is used primarily to the virus own benefit , it is likely that poxviruses have developed antiviral escape strategies interfering with the host RLR antiviral defense pathway . In line with this assumption , the dsRNA binding protein E3 of vaccinia virus has been reported to inhibit IPS-1 signaling , IRF3 phosphorylation , cytokine and IFNβ production [47]–[49] . Should inhibitors of the RLR pathway be identified in the MVA genome , gene deletion might provide an opportunity to generate new MVA vaccine vectors with increased immunogenicity . In addition to RLR , profiling of the cytokine response induced by MVA in the macrophage revealed a key role for the heterodimeric TLR2-TLR6 complex and the adapter protein MyD88 in the production of IFNβ-independent chemokines ( such as IL-8 , MIP-1α , MIP-1β and MIP-2 ) ( Figure 10 ) . Innate immune recognition of the vaccinia virus has also been shown to depend on TLR2 and MyD88 [32] . The present observation is one of the few examples of viral recognition mediated by TLR2 heterodimers . Recognition of human cytomegalovirus has been shown to be mediated by a TLR2-TLR1 heterocomplex and that of hepatitis C virus by either TLR2-TLR1 or TLR2-TLR6 [50] , [51] . The facts that TLR2 is expressed at the cell surface and that the inhibition of endocytosis or UV-irradiation of MVA did not affect IL-8 production by macrophage suggest that a component of the MVA envelope or a core protein is responsible for the activation of the TLR2-TLR6-MyD88 pathway . However , the nature of the viral component likely to serve as ligands for these TLR2-TLR1/TLR6 heterodimers has so far remained elusive . Other TLRs have also been implicated as innate sensors of poxviruses . Ectromelia virus , the causative agent of mousepox , was shown to be recognized by mouse dendritic cells in TLR9 dependent manner [33] . In contrast , responses of dendritic cells to MVA was both TLR9-dependent ( up-regulation of CD40 ) and TLR9-independent ( up-regulation of CD69 and production of IFNα and IL-6 ) [33] , [52] . Although we did not perform experiments with TLR9-deficient macrophages in the present study , the data obtained with MyD88 deficient cells clearly rule out the implication of TLR9 in MVA-induced IFNβ and IFNβ-dependent chemokines . However , we cannot exclude the involvement of TLR9 in the production of IFNβ-independent chemokines . Finally , in a mouse model activation of TLR3 contributed to the pathogenesis of Western Reserve vaccinia virus [53] . In contrast , experiments conducted with TRIF-deficient macrophages clearly showed that the production of chemokines and IFNβ induced by MVA was TLR3-independent in the present study . Taken together these observations demonstrate that TLRs may exert a two-sided role in poxvirus infections acting on the one hand as key initiators of the host anti-poxvirus defense response and on the other hand as important mediators of viral pathogenicity and tissue damage . The other important intracellular innate immune sensor of microbial products and endogenous molecules is the NALP3 inflammasome that controls the processing and maturation of the cytokines IL-1β and IL-18 [22] . Here we show that MVA is a potent activator of the NALP3 inflammasome and of IL-1β release by macrophages . IL-1β and IL-18 are key mediators of the host antimicrobial defense response and several lines of evidence suggest that these cytokines are likely to play an important role in host defenses against poxvirus infections . For example , the B15R gene of the vaccinia virus encodes an IL-1β decoy receptor blocking the activity of IL-1β and IL-18 and inactivation of B15R gene reduces the virulence of the vaccinia virus [38] , [54] . Furthermore , poxviruses release IL-18 binding proteins inhibiting IL-18 activity and vaccinia viruses A46R , A52R , N1L and , K1L gene products have been shown to disrupt the IL-1 receptor intracellular signaling pathway at multiple levels [37] , [55] . Interestingly , we observed that MVA stimulated the release of large amounts of the IL-1 receptor antagonist by macrophages ( Figure 3C ) adding further support to the view that IL-1 is an important target of the poxvirus antiviral escape strategy . Finally , consistent with the notion that the NALP3 inflammasome plays an important role in host defenses against poxviruses , several inhibitors of caspase-1 and ASC , like CrmA ( cowpox virus ) , M13L-PYD ( myxoma virus ) and PYD-only ( shope fibroma-virus ) have been identified in the genomes of several poxviruses [56]–[58] . Crosstalks between TLRs and NLRs have been demonstrated to occur in the course of bacterial infections , such as between TLR5 and the IPAF inflammasome after exposure to flagellated bacteria or the flagellin protein itself [59]–[61] . To the best of our knowledge , however , the present data provide the first demonstration of a crosstalk between the TLR and NLR pathways in the context of a viral infection ( Figure 10 ) . While TLR2 and MyD88 were necessary to induce IL-1β mRNA expression ( Figure 8A ) , the NALP3 inflammasome was absolutely required for the processing of pro-IL-1β and IL-1β secretion ( Figure 8B and C ) . Dual activation pathways coupling MVA recognition to IL-1β may provide the host with an increased capacity of fine tuning of its cytokine response . In summary , the present data show that the TLR2-TLR6-MyD88 , MDA-5-IPS-1 and NALP3 inflammasome pathways exert both specific and coordinated functions in the sensing of MVA infection and in the regulation of cytokine , chemokine and IFNβ responses ( Figure 10 ) . After the unfortunate failure of the adenovirus type 5 HIV vaccine STEP trial due to issues related to natural immunity against this virus , the attenuated MVA and NYVAC strains of poxvirus have become attractive vaccine vectors against HIV/AIDS . Arguments supporting the use of MVA and NYVAC as vaccine vectors include excellent immunogenicity and safety profiles and limited pre-existing immunity to poxvirus in the population at risk of HIV infection due to the abandon of vaccine campaigns after the eradication of smallpox in the 1970s . The present findings are therefore likely to provide important information relevant to the study of the pathogenesis of poxvirus infections , the understanding of antiviral escape mechanisms of poxvirus and may help to design new vaccine vectors with increased immunogenicity .
All animal procedures were approved by the Office Vétérinaire du Canton de Vaud ( authorizations n° 876 . 5 , 876 . 6 , 877 . 5 and 877 . 6 ) and performed according to our institution guidelines for animal experiments . Eight to ten-week-old female BALB/c and C57BL/6 mice were purchased from Charles River Laboratories ( L'Arbresle , France ) and were acclimatized for at least one week before experimentation . MyD88−/− , TRIF−/− , TLR1−/− , TLR2−/− , TLR4−/− , TLR6−/− and NALP3−/− C57BL/6 mice have been described previously [62]–[68] . Mice were bred and housed in specific pathogen free conditions . The human monocytic THP-1 cell line ( American Type Culture Collection , Manassas , VA ) was cultured in RPMI 1640 medium containing 2 mM L-glutamine , 50 µM 2-mercaptoethanol , 100 IU/ml of penicillin , 100 µg/ml of streptomycin ( all from Invitrogen , San Diego , CA ) and 10% heat-inactivated FCS ( Sigma-Aldrich , St . Louis , MO ) . THP-1 cells differentiated into macrophages by treatment with 0 . 5 mM phorbol 12-myristate 13-acetate ( PMA , Sigma-Aldrich ) for 24 h were used in all experiments except those for reporter gene analyses . THP-1 cells stably expressing control , NALP3 , caspase-1 and ASC shRNA have been described previously [69] , [70] . THP-1 cells expressing TLR2 , IPS-1 , MDA-5 and RIG-I shRNA were generated using lentiviruses expressing hairpins directed against TLR2 , IPS-1 and MDA-5 ( 5 for TLR2 , 5 for IPS-1 , 2 for MDA-5 and 5 for RIG-I ) produced with the second-generation pMD2-VSVG and pCMV-R8 . 91 packaging plasmids as described previously and cultured in the presence of 5 µg/ml puromycin [71] . The sequence of the hairpins selected that gave the best targeting of TLR2 , IPS-1 , MDA-5 and RIG-I were AAACCCAGGGCTGCCTTGGAAAAG , CAAGTTGCCAACTAGCTCAAA , CCAACAAAGAAGCAGTGTATA and AAACCCAGGGCTGCCTTGGAAAAG , respectively . Levels of expression of targeted genes were analyzed by real-time PCR using specific oligonucleotides ( Table S1 ) and the most efficiently silenced THP-1 subsets were selected for further studies ( i . e . cell lines #1 in Figure S2 ) . Peripheral blood mononuclear cells from healthy donors ( recruited by the Blood Center , Lausanne , Switzerland ) were purified by Ficoll-Hypaque density gradient ( GE Healthcare , Uppsala , Sweden ) . Macrophages were obtained by culturing adherent PBMCs cells for 6 days in RPMI 1640 with Glutamax . Bone marrow-derived macrophages ( BMDMs ) isolated from wild-type , TLR1−/− , TLR2−/− , TLR4−/− , TLR6−/− , MyD88−/− and TRIF−/− mice were cultured for 7 days in IMDM ( Invitrogen ) containing 50 µM 2-mercaptoethanol and monocyte-colony stimulating factor to obtain BMDMs . All media were supplemented with 10% FCS , 100 IU/ml of penicillin and 100 µg/ml of streptomycin . In selected experiments , cells were stimulated with 100 ng/ml Salmonella minnesota ultra pure LPS ( List Biologicals Laboratories , Campbell , CA ) , 10 µg/ml polyinosine-polycytidylic acid ( poly ( I∶C ) , Invivogen , San Diego , CA ) , 1–10 µg/ml S-[2 , 3-bis ( palmitoyloxy ) - ( 2RS ) -propyl]-[R]-cysteinyl-[S]-seryl-[S]-lysyl-[S]-lysyl-[S]-lysyl-[S]-lysine×3 CF3COOH ( Pam2CSK4 ) or N-Palmitoyl-S-[2 , 3-bis ( palmitoyloxy ) - ( 2RS ) -propyl]-[R]-cysteinyl-[S]-seryl-[S]-lysyl-[S]-lysyl-[S]-lysyl-[S]-lysine×3 HCl ( Pam3CSK4 ) lipopeptides ( EMC microcollections , Tuebingen , Germany ) , or treated with 50 µg/ml of anti-IFNβ antibodies ( BioLegend , San Diego , CA ) , 2 µM cytochalasine D , 100 µM chloroquine ( Sigma-Aldrich ) , 10 µM SB203580 ( p38 inhibitor ) , 10 µM U0126 ( MEK1/2 inhibitor ) or 50 µg/ml NEMO-binding domain binding peptide ( IkB kinase inhibitor ) ( Calbiochem-Novabiochem , Nottingham , UK ) . MVA and NYVAC were cultured in chicken embryo fibroblasts and WR in HeLa cells . Viruses were purified by two sucrose cushions and titrated on BHK-21 and BSC-40 cells as previously described [24] , [72] . Cells were infected with MVA , NYVAC or WR at various multiplicities of infection ( MOI 1 , 5 or 20 pfu/cell ) . After 1 h of contact with cells , the virus inoculum was removed and fresh medium added to the cultures . Cell-culture supernatants and cells were collected at different time points after infection and processed for flow cytometry , Luminex technology , ELISA , RNA extraction , and Western blot analyses . In selected experiments , MVA suspension ( 0 . 2 ml in 24-well plates laid on ice ) was irradiated by a 15-min exposure to a 365-nm UV bulb at a distance of 4 cm . UV-irradiation caused a 90% inhibition of the expression of C6L early gene as determined by RT-PCR using oligonucleotides ( 5′-3′ sense and antisense at position −19541/−19503 and −19071/−19090 in MVA019L ) AACTGCAGAAATGAATGCGTATAATAAAGCCGATTCGTTTTCTTTAGAG and CGGGATCCTTACTTGTCATCGTCGTCGTTCTTGTAGTCCSTGTTTAGGAAAAAAfAAATATC . MVA did not propagate in THP-1 cells as demonstrated by the absence of infective viral particles in cell-culture supernatants collected 24 h after infection ( data not shown ) . For whole blood assay , 100 µl of heparinized whole blood collected from 3 healthy volunteers were diluted 5-fold in RPMI 1640 medium containing MVA ( MOI 1 ) and incubated for 24 h at 37°C in the presence of 5% CO2 . Samples were centrifuged , and cell-free supernatants were stored at −80°C until cytokine measurement . For in vivo studies , 2×107 PFU of MVA in 1 ml phosphate-buffered saline ( PBS ) were injected intraperitoneally into BALB/c mice . After 12 h , a peritoneal lavage was performed . The supernatant obtained after centrifugation of the lavage fluid was collected for cytokine measurement by ELISA whereas the cell pellet was processed for gene expression analysis by RT-PCR . Spleens were collected from the same animals to quantify cytokine protein and mRNA expression levels . To follow cell infection , THP-1 cells were infected ( MOI 5 ) with a GFP-expressing mutant MVA , whereas all other experiments used wild-type MVA . The percentage of GFP-positive THP-1 cells was measured 0 , 2 , 4 , 6 , 12 and 24 h after infection . MVA-induced cell apoptosis was determined 6 h and 24 h post-infection using the Annexin-V FITC apoptosis detection kit according to manufacturer's recommendations ( BD Biosciences , Erembodegem , Belgium ) . Acquisition and analysis were performed using a FACS Calibur ( BD Biosciences ) and FlowJo 8 . 5 . 3 software ( FlowJow , Ashland , OR ) . A screening of mediators produced by MVA-infected THP-1 cells was performed with the human cytokine Bioplex assay ( Bio-Rad , Hercules , CA ) using the Luminex technology ( Luminex Corporation , Austin , TX ) available at the Cardiomet Mouse Metabolic Evaluation Facility , Center for Integrative Genomics , University of Lausanne , Lausanne , Switzerland . Thirty mediators were tested: TNFα , IL-1α , IL-1ra , IL-1β , IL-2 , IL-4 , IL-5 , IL-6 , IL-7 , IL-8 , IL-10 , IL-12p40 , IL-12p70 , IL-13 , IL-15 , IL-17 , IFNγ , RANTES , IP-10 , MIP-1α , MIP-1β , MCP-1 , eotaxin , fractalkine , TGFα , EGF , VEGF , GM-CSF , G-CSF and sCD40L . The concentrations of human IL-1β ( Bender MedSystems , Vienna , Austria ) , IL-8 , ( BD Biosciences ) , IP-10 , MIP-1α ( R&D ) and IFNβ ( PBL Biomedical Laboratories , Picataway , NJ ) in whole blood assay and cell-culture supernatants were measured by ELISA . TNF and IL-6 concentrations were measured by bioassay as described elsewhere [73] . Mouse IL-1β , MIP-2 ( R&D ) and IFNβ were quantified by ELISA ( Biomedical Laboratories , Picataway , NJ ) . Total RNA was isolated from THP-1 cell lines , human monocytes/macrophages , peritoneal cells and splenocytes using the RNeasy kit ( Qiagen , Hombrechtikon , Switzerland ) . Reverse transcription of 1 µg of RNA was performed using the ImProm II RT System kit ( Promega , Dübendorf , Switzerland ) . Quantitative PCR was performed with a 7500 Fast Real-Time PCR System ( Applied Biosystems , Rotkreuz , Switzerland ) using the Power SYBR Green PCR Master Mix ( Applied Biosystems ) and primer pairs listed in Table S1 . All samples were tested in triplicates . Amplifications consisted of a denaturation step at 95°C for 15 sec and an annealing/extension step at 60°C for 60 sec , with the 9600 Emulation mode . For each measurement , a standard made of successive dilutions of a reference cDNA was processed in parallel . Gene specific expression was expressed relative to the expression of HPRT in arbitrary units ( A . U . ) . Gene specific over HPRT ratios were validated using the house-keeping gene ACTB ( human studies ) or Gapdh and Actg1 ( mouse studies ) . THP-1 cells were seeded at 5×104 cells per well in 24-well plates . The following day , cells were transiently transfected with 700 ng of multimeric κB site [73] and IFNβ promoter [74] luciferase reporter vectors together with 70 ng of a Renilla luciferase control vector ( Promega ) using jetPEI™ transfection reagent ( Polyplus-transfection SA , Illkirch , France ) . Twenty-four h after transfection , cells were infected with MVA . Luciferase and Renilla luciferase activities were measured 24 h latter using the Dual-LuciferaseTM Reporter Assay System ( Promega ) . Results were expressed as relative luciferase activity ( the ratio of luciferase to Renilla luciferase activity ) . THP-1 cells were washed with ice cold PBS and lysed for 5 min at 4°C with the M-PER Mammalian Protein Extraction Reagent ( Pierce Biotechnology Inc , Rockford , IL ) . Reaction mixtures were centrifuged 5 min at 14'000 rpm . Protein concentration of supernatants was determined using the bicinchoninic acid protein assay ( Pierce Biotechnology ) . Cell-lysates were electrophoresed through 12% ( w/v ) polyacrylamide gels and transferred onto nitrocellulose membranes ( Schleicher & Schuell , Keene , NH ) . Membranes were incubated with antibodies directed against RIG-1 , MDA-5 , IPS-1 ( Apotech Corporation , Epalinges , Switzerland ) , cleaved IL-1β , total- and phospho-p44/42 ( ERK1/2 ) , and -JNK MAP Kinases , phospho-IRF3 ( Cell Signalling Technology , Danvers , TX ) , caspase 1 ( Santa Cruz , Santa Cruz , CA ) , phospho-STAT-1 ( BD Biosciences ) , IRF7 ( Zymed , San Franciso , CA ) and tubulin ( Sigma ) . After washing , membranes were incubated with horse radish peroxidase ( HRP ) -conjugated secondary antibody ( Pierce ) . Signals were revealed using the ECL Western blotting Analysis System ( GE Healthcare ) . Nuclear extracts were prepared and analyzed by EMSA [73] . Briefly , protein concentration of cell extracts was measured using the Bradford-dye assay ( Bio-Rad ) . Two µg of nuclear extracts were incubated for 15 min at room temperature with a radio-labeled consensus NF-κB probe ( Santa Cruz ) . Reaction mixtures were electrophoresed through 6% non-denaturing polyacrylamide gels . Gels were dried and exposed to X-ray films . Supershift experiments using anti-p65 antibody ( sc-109 , Santa Cruz ) were performed as previously described [75] ( data not shown ) . Comparisons among treatment groups were performed by two-tailed paired Student's t-test . p values less than 0 . 05 were considered to indicate statistical significance . | Modified vaccinia virus Ankara ( MVA ) is a highly attenuated , replication-deficient , poxvirus currently developed as a vaccine vector against a broad spectrum of infectious diseases including HIV , tuberculosis and malaria . It is well known that robust activation of innate immunity is essential to achieve an efficient vaccine response , and that poxviruses have developed numerous strategies to block the innate immune response . Yet , the precise mechanisms underlying innate immune sensing of MVA are poorly characterized . Toll-like receptors ( TLR ) , RIG-I-like receptors ( RLR ) and NOD-like receptors ( NLR ) are families of membrane-bound and cytosolic sensors that detect the presence of microbial products and initiate host innate and adaptive immune responses . Here , we report the first comprehensive study of MVA sensing by innate immune cells , demonstrating that TLR2-TLR6-MyD88 , MDA-5-IPS-1 and NALP3 inflammasome pathways play specific and coordinated roles in regulating cytokine , chemokine and interferon response to MVA poxvirus infection . Delineation of the pathways involved in the sensing of MVA by the host could help designing modified vectors with increased immunogenicity , which would be of particular importance since MVA is considered as a leading vaccine for HIV/AIDS vaccine following the recent failure of an adenovirus-mediated HIV vaccine trial . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"infectious",
"diseases",
"virology/vaccines",
"immunology/immune",
"response",
"immunology/innate",
"immunity",
"infectious",
"diseases/hiv",
"infection",
"and",
"aids",
"infectious",
"diseases/viral",
"infections",
"immunology",
"virology/immune",
"evasion",
"virology/effects",
"of",
"virus",
"infection",
"on",
"host",
"gene",
"expression",
"immunology/immunity",
"to",
"infections",
"virology/host",
"antiviral",
"responses"
] | 2009 | Innate Immune Sensing of Modified Vaccinia Virus Ankara (MVA) Is Mediated by TLR2-TLR6, MDA-5 and the NALP3 Inflammasome |
The continual rise of asthma in industrialised countries stands in strong contrast to the situation in developing lands . According to the modified Hygiene Hypothesis , helminths play a major role in suppressing bystander immune responses to allergens , and both epidemiological and experimental studies suggest that the tropical parasitic trematode Schistosoma mansoni elicits such effects . The focus of this study was to investigate which developmental stages of schistosome infection confer suppression of allergic airway inflammation ( AAI ) using ovalbumin ( OVA ) as a model allergen . Moreover , we assessed the functional role and localization of infection-induced CD4+Foxp3+ regulatory T cells ( Treg ) in mediating such suppressive effects . Therefore , AAI was elicited using OVA/adjuvant sensitizations with subsequent OVA aerosolic challenge and was induced during various stages of infection , as well as after successful anti-helminthic treatment with praziquantel . The role of Treg was determined by specifically depleting Treg in a genetically modified mouse model ( DEREG ) during schistosome infection . Alterations in AAI were determined by cell infiltration levels into the bronchial system , OVA-specific IgE and Th2 type responses , airway hyper-sensitivity and lung pathology . Our results demonstrate that schistosome infection leads to a suppression of OVA-induced AAI when mice are challenged during the patent phase of infection: production of eggs by fecund female worms . Moreover , this ameliorating effect does not persist after anti-helminthic treatment , and depletion of Treg reverts suppression , resulting in aggravated AAI responses . This is most likely due to a delayed reconstitution of Treg in infected-depleted animals which have strong ongoing immune responses . In summary , we conclude that schistosome-mediated suppression of AAI requires the presence of viable eggs and infection-driven Treg cells . These data provide evidence that helminth derived products could be incorporated into treatment strategies that specifically target suppression of immune responses in AAI by inducing Treg cells .
Over the last century and in strong contrast to third world countries , Western populations have shown a consistent rise in autoimmune disorders ( e . g . Crohn's disease ) and allergic conditions such as asthma [1] . Indeed , allergic asthma is the most common disease in industrialized countries , with a prevalence of up to 29% [2] . It is a potentially life-threatening illness since severe allergic responses within the respiratory airways elicit inflammation and swelling that results in varying breathing difficulties , e . g . dyspnea , tight chest , cough or bronchospasm [3] . Further clinical features of bronchial asthma include inflammatory cell recruitment and impaired lung function . Moreover , several defined immune-response characteristics can be assessed including levels of IgE specific for the allergic agent and the development of Th2 responses upon challenge with the allergen . Epidemiological studies have clearly shown that the observed increase in allergy prevalence in Western countries is not reflected in the developing world which instigated the concept of the Hygiene Hypothesis . In essence , this hypothesis states that due to well-established sanitation and vaccination procedures , the overall reduction in common Th1-inducing ( bacterial and viral ) infections has resulted in a decreased ability to immunologically counterbalance Th2-driven diseases . This has created a conundrum , since a large body of epidemiological evidence and research has established that helminth diseases , which are strong Th2 inducers themselves , actually protect against developing allergic responses [4] . At first glance , the idea of protective cross-reactive mechanisms is surprising since worms drive eosinophilia and IgE production: hallmarks of pathology inducing factors in asthmatic disease . Nevertheless , after consolidating the findings of 30 independent epidemiological surveys , studying the influence of geohelminth infections on allergy prevalence , the “Parasites in Asthma Study Group” concluded that protective effects are dependent on the worm species , age , state of infection ( chronic versus acute ) and parasite burden [5] . Consequently , despite their Th2 inducing capacity , helminth infections are now incorporated into the “Expanded Hygiene Hypothesis” . Interestingly , the blood fluke Schistosoma mansoni was one of the parasites found to have a protective effect . More than 250 million people in 74 tropical and subtropical countries are chronically infected with this trematode , which has life-stages that pass through both the skin and lung of the definite host . During the course of the disease an immune homeostasis eventually evolves that is supported by long-lasting immunomodulatory mechanisms and potentially deviates other responses . Worm development , pathology and immune responses , including the switch from Th1 to Th2 upon egg expulsion , are parallel to those seen in man and studies in mice have shown the ability of schistosomes to decrease autoimmune and allergic diseases [6]–[10] . These manipulative strategies are directed through immune cell populations such as Foxp3+ regulatory T cells ( Treg ) or B regulatory cells [11] . Treg are essential for controlling unwarranted responses to “self-antigens” [12] and during schistosomiasis this cell population increases within the CD4+ T cell compartment in a homeostatic fashion . Moreover , Foxp3+ Treg maintain granuloma development , the main cause of morbidity and develop a unique genetic signature [13] , [14] . Using murine models of allergic airway inflammation ( AAI ) , Treg in general have been shown to control overt allergic responses [15] , [16] and appear to be required in mediating protection elicited via schistosome infection [17]–[19] . Here we evaluate in detail which life-cycle stage of the worm confers protection and assess the capacity of Foxp3+ Treg induced during infection to suppress allergic airway disease by depletion Foxp3+ Treg cells in the molecularly defined DEREG ( Depletion in Regulatory T cell ) mouse model [20] .
This animal study was conducted in accordance with an application to perform in vivo experiments ( license number AZ . 55 . 2 . 1 . 54-2532-147-08 ) and was approved by the local government authorities Bezirksregierung Oberbayern . Animals were housed at the Institute of Medical Microbiology , Immunology and Hygiene ( MIH ) , Technische Universität München , Germany , in accordance with the German Tierschutzgesetz ( German animal protection laws ) and the EU guidelines 86/809 . Wildtype BALB/c female mice ( 6–8 weeks old ) were purchased from Harlan ( Borchen Germany ) . DEREG C57BL/6 mice were bred in house at the MIH . Infections with a Brazilian strain of S . mansoni were instigated with the injection of 90 cercariae per mouse and were performed as depicted in Figure 1A–E . Praziquantel ( PZQ ) was obtained from Bayer Healthcare , Leverkusen , Germany and was administered orally at a dose of 100 mg/kg body weight over 5 consecutive days during the 6th week of infection ( Figure 1D ) . S . mansoni-infection was confirmed through visible granuloma development in liver sections and egg burden in the liver following KOH digestion following standard techniques [21] . In the PZQ experiments , Masson's stained liver sections were also used to determine the percentage of viable eggs . In Figure 1E , gfp+Foxp3+ Treg in C57BL/6 DEREG mice were depleted by i . p . application of 1 µg diphtheria toxin ( DT ) purchased from Merck KGaA , Darmstadt , Germany and was dissolved in endotoxin-free PBS ( PAA Laboratories GmbH , Linz , Austria ) [20] . On the days depicted in Figure 1 , intraperitoneal ( i . p . ) sensitizations were performed using 10 µg of grade VI ovalbumin ( OVA ) , ( Sigma , Deisenhofen , Germany ) emulsified in 1 . 5 mg alum ( aluminium hydroxide ( Al[OH]3 ) ) ( Sigma ) . Subcutaneous injections of OVA ( 10 µg ) ( Figure 1A ) were administered without alum at the back of the neck . Control groups were injected with PBS . Mice were challenged over three consecutive days with 10 µg of grade V OVA using the Pari-Master ( PARI GmbH ) aerosolic nebulizer [16] . AAI parameters were analyzed 72 hours after the last challenge . Since age has been shown to influence the intensity of developing AAI in mice [22] , we limited possible age-bias by ensuring that all groups of mice were age-matched at the onset of the experiment . Moreover , they were housed for the entire experimental period under the same conditions . Following challenge , airway responsiveness to methacholine ( MCh ) ( Sigma ) was determined in individual mice using the Flexivent system ( SCIREQ , Montreal , QC , Canada ) . Following anaesthesia with Ketanest ( Inresa Arzneimittel GmbH , Freiburg , Germany ) and Rompun ( Bayer Health Care , Leverkusen , Germany ) mice were paralyzed with Esmeron ( N . V . Organon , Oss , Netherlands ) . The trachea was then intubated with a 1 . 2 mm tracheal cannula and the lungs mechanically ventilated at a respiratory frequency of 150 breaths per min , a tidal volume of 10 ml/kg and a positive end-expiratory pressure of 3 ml H2O . After exposing mice to aerosolized PBS to retrieve the baseline value , bronchoconstriction was induced by increasing the concentration of MCh . Resistance was recorded over 1 minute intervals using a standardized inhalation maneuver ( SnapShot-150 ) [23] . To obtain bronchoalveolar lavage ( BAL ) cells , the lungs of individual mice were rinsed with 1 ml of PBS containing proteinase inhibitor cocktail tablets ( Roche Diagnostics Mannheim , Germany ) . The resulting fluid was weighed and centrifuged at 230 g for 5 minutes at 4°C . Cells were then re-suspended in PBS containing 2% FCS and then centrifuged ( 400 rpm for 5 minutes ) onto glass slides using the Shandon Cytospin 3 Centrifuge ( Thermo Scientific , Hamburg , Germany ) . After overnight drying slides were stained using the Diff-Quick staining kit ( Medion Diagnostics , Langen , Germany ) . Cell differentiation was determined microscopically . OVA-specific IgE levels were measured in the sera of individual mice . In brief , 96-well ELISA plates ( Nunc , Langenselbold , Germany ) were coated overnight ( 4°C ) with 1 µg per well of OVA grade V diluted in aqua dest . containing NaHCO3 and Na2CO3 . After washing and blocking in 50 mM Tris solution containing 3% BSA , sera was diluted in blocking buffer ( 1∶200 to 1∶100 , 000 dilutions ) and a standard of mouse α-OVA IgE antibody ( Biozol , Eching , Germany ) was applied in two-fold serial dilutions and incubated overnight ( 4°C ) . Subsequently , plates were washed and α-mouse IgE biotinylated detection antibody ( Biozol ) was applied and incubated for 2 h ( RT ) . After further washing streptavidin-horseradish peroxidase ( HRP ) conjugate ( R&D Systems GmbH , Wiesbaden , Germany ) was added and plates were incubated for 30 min ( RT ) in the dark . After a final washing step , TMB substrate ( BD , Heidelberg , Germany ) was applied and reactions were stopped with 2MH2SO4 . Finally , ODs were determined at 450 nm using the Sunrise ELISA microplate reader ( Tecan , Crailsheim , Germany ) . The concentration of the samples was then calculated according to the standard curve . Paraffin embedded sections ( 3 µm ) from the left lungs of individual mice were stained with PAS ( Periodic acid-Schiff ) . Lung tissue inflammation was microscopically determined by the degree of cell infiltration around the basal membrane of bronchi or vessels , which was graded on a scale from 1 to 4 . In brief , a value of 1 was assigned for occasional cuffing with inflammatory cells , a value of 2 was assigned for a thin layer ( two to three cells thick ) of inflammatory cells , a value of 3 was assigned when bronchi or vessels were surrounded by a thick layer of four to five inflammatory cells and a value of 4 was assigned when bronchi or vessels were surrounded by a layer of more than five inflammatory cells . Per lung section , a mean inflammation score was determined . Immunohistochemistry of CD3+Foxp3+ Treg was performed as previously described [14] . In brief , lung sections were deparaffinized and heated in sodium citrate buffer ( pH6 . 0 ) for 2 min at high pressure . After washing and blocking , slides were incubated with a goat polyclonal antibody ( Ab ) against the C terminus of the Foxp3 protein ( ab2481 , dilution 1∶50; Abcam Limited , Cambridge , U . K . ) . Sections were then exposed to a biotin-conjugated rabbit anti-goat and the EnVision peroxidase kit ( Dako-Cytomation ) . After a further incubation with the second Ab against CD3 ( clone UCHT1 , 1∶25; DakoCytomation ) cells were visualized using the alkaline phosphatase–anti-alkaline phosphatase method [24] . In a blind fashion , Foxp3+ T cells were quantified in individual lung sections within a 1 mm2 area using a Zeiss microscope ( Axio Lab . A1 , Oberkochen , Germany ) at 20× magnification . Intracellular staining of Foxp3+CD4+ T cell populations was performed on erythrocyte-depleted mediastinal lymphnode cells ( medLN ) . Prior to staining , Fc receptors were blocked using anti-CD16/CD32 ( eBiosciences , Frankfurt , Germany ) . Thereafter , cells were stained with APC-conjugated anti-CD4 mAb ( eBiosciences ) . Intracellular Foxp3 levels were detected using the PE-conjugated anti-Foxp3 mAb staining kit according to the manufacture's instruction ( eBiosciences ) . Fluorescence was analyzed using a flow cytometer and software from BDbiosciences ( Heidelberg , Germany ) . The presence or absence of egfp+ cells in DEREG mice was also detected by flow cytometry in peripheral blood samples from individual mice [20] . Erythrocyte-depleted cell suspensions ( 2×105 ) from individual lung lymph node ( LLN ) or individual spleens were re-stimulated in vitro for 72 hours with 10 µg/ml OVA ( Grade VI ) or with 20 µg/ml soluble egg antigen ( SEA ) prepared from schistosomal eggs as previously described [13] . Cytokine content in the culture supernatant was then determined using ELISA in accordance with the manufacturer's instructions ( eBiosciences ) . Statistical differences were analyzed using GraphPad Prism 5 software ( San Diego , CA , USA ) . Parametrically distributed data were analyzed using unpaired t-tests or one-way ANOVA .
Although humans are the main definite host of S . mansoni in many regions , this helminth also infects rodents . Moreover , immunological and pathological responses observed in humans are mirrored in mice . Worms require 5–6 weeks to mature and enter patency which is classified as the production of eggs by fecund females [25] . Previous studies have shown that an ongoing infection dampens AAI in C57BL/6 mice [7] . Using the Th2 biased BALB/c mouse strain [26] , [27] we aimed to differentiate which stage of infection provided a protective environment and performed three different infection/allergy scenarios which are depicted in Figures 1A–C . To confirm the onset of asthma numerous parameters were assessed including cell infiltration levels in the BAL ( Fig . 2A–F ) , lung inflammation ( Figures 2G–I and S1A–D ) and AHR ( Figure S1E ) . First , to observe whether patently-infected BALB/c mice were protected against the development of AAI , mice were sensitized on days 37 , 51 and 58 post-infection ( Figure 1A - PiP: patent-infected-protection ) . Sensitizations were administered either i . p . in the presence of alum or s . c . in the absence of adjuvant . Mice sensitized and challenged during patent infection ( Inf/OVA ) had significantly decreased cell infiltration ( Figure 2A ) and eosinophil numbers ( Figure 2D ) when compared to non-infected OVA groups ( OVA ) , indicating that S . mansoni infections in BALB/c mice protect against allergy development . These findings were also observed using the adjuvant-free model ( Figure 2A right and D right ) . Within the lung , decreased inflammation was only observed in Inf/OVA groups that had been sensitized with alum ( Figure 2G ) . Figures S1A–D show representative histological sections from the different groups of mice in these PiP experiments . As shown , when compared to OVA groups ( Figure S1A ) , inflammation was less severe in Inf/OVA groups ( Figure S1B ) . Control groups that were ( Inf . ) or were not infected ( PBS ) and only exposed to aerosolic OVA upon challenge did not present any signs of inflammation or cell infiltration ( Figures 2A–I and S1C and D ) . Airway resistance was also measured in the PiP investigation studies and as with the other measured parameters , Inf/OVA mice had resistance levels that were comparable to the control groups ( Figure S1E ) . With regards to immunological responses , OVA-specific Th2 ( Figures S2A and B ) and regulatory ( Figure S2C ) recall responses of draining lymph node cells were significantly dampened in Inf/OVA groups and this was regardless to the sensitisation route . Interestingly , OVA-specific IgE levels in the sera were elevated in Inf/OVA mice that were sensitised i . p . but not s . c . ( Figure S2D ) . Next , we assessed whether protection occurred when infected mice received sensitization rounds within the first 21 days of infection ( prepatency ) but were challenged during patency ( Figure 2B - PP: prepatent ) . Interestingly , although leucocyte and eosinophil infiltration was still significantly dampened ( Figures 2B and E respectively ) , within this experimental scenario , no protective influence on lung inflammation could be observed in the Inf/OVA group ( Figure 2H ) . Finally , experiments were performed in which AAI was induced and assessed during prepatency i . e . the first five weeks of infection ( Figure 2C - EP: early phase ) . Here , S . mansoni infection elicited no protection since cellular infiltration and lung inflammation scores were equal to OVA groups ( Figures 2C , F and I ) . Parasitological assessment of the S . mansoni groups , in all infection scenarios , showed no differences between granuloma size or egg burden in the liver ( data not shown ) . Eggs were not detected in the EP experiments since at week 4–5 , adult female worms are not yet fecund . In another set of experiments we investigated the development of AAI upon curing a patent S . mansoni infection ( Figure 1D ) . In brief , groups of infected mice were treated with praziquantel ( PZQ ) during the 6th week of infection . After a further 2 or 6 weeks , OVA sensitization commenced , thus , analysis of AAI was performed after either 12 or 18 weeks post infection ( or 6 and 10 weeks post PZQ-therapy ) . Interestingly , infected PZQ-treated groups of mice ( Inf/OVA/PQ ) analysed after 12 weeks but not 18 weeks post-infection were still significantly protected from cellular infiltration ( Figures S3A–D ) but not inflammation ( Figures S3E and F ) . As shown in Figure S3G , egg counts in the liver were reduced after 10 weeks of PZQ therapy . This was also confirmed in the amount of released eggs in the stool since none were detected on the days of analysis ( data not shown ) . Moreover , upon assessment of individual liver sections , there was a significant reduction in viable eggs in the liver after 10 weeks of PZQ treatment ( Figure S3H ) . As mentioned above , previous studies have shown that Treg play a role in helminth-mediated immunomodulation during AAI . However , questions pertaining to their redistribution into the lymphatics or relevant organ tissues remain unanswered . Thus , we analysed the levels and distribution of Foxp3+ T cells in the draining lymph nodes ( LLN ) of mice from the PiP experiments ( Figure 2A ) . Figure 3A shows that the absolute cell counts in OVA treated mice , regardless of infection , is higher than in PBS groups . No significant differences could be observed between OVA and Inf/OVA groups . With regards to the percentage of Foxp3+ T cells in the CD4+ T cell compartment , these regulatory cells were elevated in Inf/OVA mice but not in OVA groups alone ( c . f OVA and Inf/OVA in Figure 3B ) . Interestingly , Treg numbers were high in the Inf . group as well . Within the lung however , numerous CD3+Foxp3+ T cells could be identified in sections from OVA mice ( Figure 3C ) but not Inf/OVA groups ( Figure 3D ) . These sections were comparable to those sampled from the PBS groups ( Figure 3E ) . These impressions were verified through further quantification of the Treg numbers in the lung sections of individual mice . As shown in Figure 3F , CD3+Foxp3+ Treg were significantly higher in OVA but not Inf/OVA group . To address whether S . mansoni-mediated protection of AAI was mediated by Treg we infected DEREG mice , which allow the specific depletion of egfp+Foxp3+ cells at specified time-points through the administration of DT [20] . Initially we performed such depletion experiments in BALB/c DEREG mice but due to their higher sensitivity to DT application and schistosome infection , Treg depletion during AAI development resulted in high mortality rates ( over 90% , data not shown ) . Therefore , we continued the experiments with DEREG mice on a C57BL/6 background . As depicted in Figure 1E , Treg were depleted in groups of infected and non-infected mice during sensitization . As observed before within the PiP studies described above , Inf/OVA groups had significantly lower levels of leucocytes and eosinophils when compared to OVA groups ( c . f . bars 1 and 4 in Figures 4A and B ) . However , depletion of Treg in the infected group ( Inf/OVADT ) prevented the previously observed suppressive effects ( c . f . bar 3 with bars 1 and 4 in Figures 4A and B ) and levels of cellular infiltration even equalled those of the OVADT groups . In association with the current literature [16] , non-infected OVA groups , depleted of Treg , also showed increased leucocyte and eosinophil infiltration ( c . f . bars 1 and 2 in Figures 4A and B ) . With regards to inflammation score depletion of Treg abrogated the suppressive effect of schistosome infection ( Figure 4C c . f . bars 3 and 4 and Figures 4E and G ) . Moreover , Treg depletion led to enhanced lung inflammation when compared with the OVA group ( Figures 4C c . f bar 1 with 2 and 3 and Figures 4D , 4F and 4G ) . Immunologically , both Treg depleted groups showed significantly higher levels of OVA-specific IgE in the sera ( Figure 4H ) and OVA-specific IL-5 upon recall of splenocytes cells ( Figure 4I ) . Interestingly , schistosome-specific responses in the infected groups were also significantly increased in the Inf/OVADT groups ( Figure 4J ) . Moreover , these mice showed lower egg burden in the liver ( Figure 4K ) . These data correlate to our earlier studies in which we depleted Tregs in S . mansoni infected mice by administering an anti-CD25 antibody . There , we also observed elevated schistosome-specific Th2 responses and strongly decreased egg numbers in the livers of Treg-depleted S . mansoni infected animals [13] . During these experiments , we controlled for the depletion of Treg by observing levels of CD4+egfp+ T cells in peripheral blood . Figure S4A shows a representative comparison of CD4+egfp+ T cells in infected ( left ) and non-infected ( right ) mice on the day before depletion and 3 days thereafter . As shown in the bottom panel of flow cytometry images , no egfp+ cells are visible after depletion . Levels were measured again following asthma induction ( Figure S4B ) . Here , the percentage of Foxp3+ Treg in infected-depleted mice was approximately 50% lower than in non-infected depleted mice ( 4 . 7% vs 8 . 24% ) . This was observed in all the experimental mice ( Figure S4C ) and indicates that the re-establishment of Treg in infected mice is slower than in non-infected controls and provides a possible explanation as to why infected mice are no longer suppressed from AAI development .
Schistosome infections result in immune responses that are tightly controlled by various host mechanisms including cell types such as Treg and Breg [13] , [28] or elicited through the release of immunosuppressive schistosomal antigens [29] . The latter is thought to generate bystander immune responses against unrelated antigens or allergens . Several studies have addressed the role of helminth infections using the classical ovalbumin model of AAI . With regards to schistosome-allergy studies both protective and aggravated responses have been noted [7] , [19] , [30] . For example , Smits et al . showed that only chronically infected C57BL/6 mice ( 12–16 weeks ) were protected from AAI , whereas in our study protection in BALB/c mice was already observed at 9 weeks of infection . Our studies required the induction of AAI at this time-point since in contrast to C57BL/6 mice , BALB/c mice have an accelerated disease progression and higher mortality rate . Our findings are further supported by the studies from Pacifico et al . who reported a protective effect in BALB/c mice , also at 9 weeks of infection , using a dose of OVA and alum that was equivalent to that employed in our study [19] . This is in contrast to Mangan et al . , who used double the amount of OVA and alum and interestingly , they reported aggravated airway hyper-responsiveness and enhanced OVA-specific Th2 responses in acute and chronically schistosome-infected BALB/c mice [30] . Perhaps such strong immunizations can no longer be counterbalanced by the schistosome infection . These opposing studies clearly highlight that the different protocols used in the field trigger different immune responses and have to be considered when interpreting and comparing the results . Interestingly , in the Mangan study infections with male schistosome worms resulted in reduced airway responses which was mediated through B cells and IL-10 but not Treg . Hypothetically therefore , worm-induced B cells and egg induced Treg could both contribute to suppression of AAI . Another factor which has to be stringently addressed is age , since it is known that age effects the development of AAI in mice [22] . To ensure that our findings were not due to an age-bias , all mice were age–matched at the onset of experiments and housed under the same conditions . Evidently , S . mansoni-infected mice had reduced AAI when sensitized and challenged during patent infection and moreover , Inf/OVA mice still had dampened responses when challenge but not sensitization occurred in the patent phase . However , no protection was observed in Inf/OVA mice when both sensitization and challenge were performed in the prepatent phase ( EP studies ) or six weeks after praziquantel treatment . Taken together these results clearly highlight the requirement of patency for infection-mediated AAI suppression . Interestingly , our anti-helminthic treatment experiments revealed that some level of protection against AAI remained when sensitization commenced two weeks after treatment . At this time-point although many tissue-residing eggs are still viable , most of the worms have died and only stunted , non-egg producing worms are left over [31] , [32] . Complete abrogation of the protective effects was only observed in those experiments analyzed on the 18th week of infection , when sensitization began 6 weeks after PZQ treatment . Here it is considered that the proportion of viable eggs has strongly decreased and our findings here coincide with these data [13] . In association , a human study demonstrated that schistosome-infected individuals displayed increased levels of HDM-specific IgE and low responses to skin prick tests [33] . However , following anti-helminthic treatment these patients presented increased HDM skin reactivity [34]–[36] indicating that in order to maintain a suppressive milieu an ongoing patent infection is required . The patently-induced protection was also independent of the OVA-application route and the presence of alum . OVA-specific IgE levels in infected mice were increased when OVA was injected i . p . with alum , which has been reported to drive the induction of the disease independently from mast-cell mediated early phase reactions [37] , [38] . OVA-specific IgE levels remained unaltered when alum was applied subcutaneously which indicates that the suppression of AAI occurs independently of IgE levels . Thus , we conclude that the suppressive effects mediated by schistosome infection are not due to defective OVA priming which supports earlier findings [30] . Concerning the possible underlying mechanisms that lead to dampened allergic responses in schistosome infected animals , we and others have shown previously that Foxp3+ Treg expand homeostatically during the course of infection and this phenomenon already starts when the first SEA-specific Th1 responses can be detected around the 5th week of infection [13] . The first hint that Treg might be involved in schistosome-mediated protection of AAI came from the finding that when compared to OVA groups , Foxp3+ Treg were significantly elevated in the LLN of Inf/OVA and Inf alone groups of mice . However , whereas numerous CD3+Foxp3+ Treg were present in the lung tissues of OVA groups very few were detectable in Inf/OVA mice . Therefore , we consider that during schistosome infection , regulation of lung infiltrate upon challenge occurs at the draining LLN and not within the lung tissue itself . It will be interesting to decipher how Treg in Inf/OVA mice are retained in the LLN and whether these are indeed infection-induced Foxp3+ T cells . In our previous studies we showed that infection-induced Treg but not Treg from naive mice suppress SEA-specific CD4+ T cell effector responses and develop a unique gene expression profile that includes upregulation of granzyme B and anti-inflammatory molecules such as SLPI ( secretory leucocyte peptidase inhibitor ) [14] . This change of phenotype could account for their selective suppressive activity since it was shown , for example , that activated Treg cells can actually kill B cells in a granzyme-B-dependent manner [39] . It is now discussed that Treg adapt their mode of immune suppression according to the altered requirements found under inflammatory conditions in comparison to those in the steady-state [12] . The role of these molecules in Treg-mediated suppression of allergies in vivo will be an interesting field of future research . To assess whether Treg played a role in the development of AAI during infection we employed the DEREG mouse model which allows depletion of Treg at the experimenter's desired time-points [20] , [40] . This model ensures that all Foxp3+ T cells are depleted regardless of their CD25-coexpression and also circumvents the criticized application of anti-CD25 antibodies in which recently activated effector T cells are targeted as well [20] , [41] , [42] . BALB/c mice have a higher morbidity due to schistosome infection and additional DT application further increased mortality rates . Thus , we changed to the DEREG mice on a C57BL/6 background , which tolerated our experimental procedure [20] . Interestingly , Treg depletion during OVA-sensitization in C57BL/6 DEREG mice resulted in a loss of protection in S . mansoni infected mice . Furthermore , cellular infiltration levels and pathology scores were higher in Inf/OVADT groups when compared to control OVA groups . These findings confirm other studies that have shown the requirement for Treg in AAI [16] and that Foxp3+ T cells down-modulate airway eosinophilia but not AHR and IgE levels in a model of SIT-induced tolerance [43] . Indeed , OVA-specific IgE levels were enhanced in both depleted groups indicating a role of Treg in initiating Th2 responses . In addition , schistosome-specific immune responses were elevated in Inf/OVADT mice and alongside the strongly reduced egg burden this confirms our previous findings in schistosome-infected mice treated with anti-CD25 antibody [13] . In addition , we noted a delayed reconstitution of expanding Treg in Inf/OVADT mice compared to OVADT mice after Treg depletion . Thus , the removal of Treg during a chronic infection , in contrast to the steady-state or immunization situation , probably gives rise to more pronounced and quicker Th responses against schistosome antigens . This might favor immune reactions against OVA antigens and eventually results in decelerated or unbalanced Treg reconstitution . Collectively , by comparative experiments using different protocols for AAI our data support recent findings that schistosome infection can indeed suppress allergic airway responses and that this suppression requires patency and the continuous release of eggs . Furthermore , this suppression is partly mediated by expanding Foxp3+ Treg cells within the draining lymph nodes of the lung . We therefore propose the concept that patent infection inhibits the effector phase of AAI through schistosome egg-derived factors and not worm or schistosomula antigens . An in-depth analysis of egg components and their potential to drive Treg expansion or induction could therefore have potential therapeutic value . | Infections with schistosomes , such as S . mansoni , S . japonicum and S . haematobium , are considered a major public health concern . Morbidity arises through granulomatous responses to eggs that become trapped in infected tissues . Interestingly , schistosomes belong to the group of helminths that have been shown to reduce allergy or autoimmunity . Indeed , the evidence provided by epidemiological surveys and experimental animal models has been so overwhelming that such helminths are now included in the Hygiene Hypothesis . However , since helminths provoke immunological responses that are similar to those seen in allergy ( increased eosinophilia and IgE ) it is suggested that additional mechanisms dampen such allergic responses . Helminth-induced regulatory T cells ( Treg ) are considered a component of these modulatory networks . Using an allergic airway inflammation model , we have elucidated that schistosome-mediated protection requires patency , that is , active egg production from fecund female worms . In addition , protection was shown to be mediated by infection-induced Treg . Interestingly , in endemic countries it is usually individuals with strong patent infections that show reduced allergic prevalence . Thus , further research into the immunomodulatory capacity of schistosome-egg derived factors may elucidate novel drug candidates or enhance treatment strategies to reduce allergic responses on the cellular level . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"medicine",
"infectious",
"diseases",
"schistosomiasis",
"immune",
"cells",
"t",
"cells",
"allergy",
"and",
"hypersensitivity",
"immunology",
"biology",
"parasitic",
"diseases",
"immunomodulation"
] | 2013 | Schistosoma mansoni-Mediated Suppression of Allergic Airway Inflammation Requires Patency and Foxp3+ Treg Cells |
Melioidosis , a fatal infectious disease caused by Burkholderia pseudomallei , is increasingly diagnosed in tropical regions . However , data on risk factors and the geographic epidemiology of the disease are still limited . Previous studies have also largely been based on the analysis of case series data . Here , we undertook a more definitive hospital-based matched case-control study coupled with spatial analysis to identify demographic , socioeconomic and landscape risk factors for bacteremic melioidosis in the Kedah region of northern Malaysia . We obtained patient demographic and residential information and clinical presentation and medical history data from 254 confirmed melioidosis cases and 384 matched controls attending Hospital Sultanah Bahiyah ( HSB ) , the main tertiary hospital of Alor Setar , the capital city of Kedah , during the period between 2005 and 2011 . Crude and adjusted odds ratios employing conditional logistic regression analysis were used to assess if melioidosis in this region is related to risk factors connected with socio-demographics , various behavioural characteristics , and co-occurring diseases . Spatial clusters of cases were determined using a continuous Poisson model as deployed in SaTScan . A land cover map in conjunction with mapped case data was used to determine disease-land type associations using the Fisher’s exact test deploying simulated p-values . Crude and adjusted odds ratios indicate that melioidosis in this region is related to gender ( males ) , race , occupation ( farming ) and co-occurring chronic diseases , particularly diabetes . Spatial analyses of disease incidence , however , showed that disease risk and geographic clustering of cases are related strongly to land cover types , with risk of disease increasing non-linearly with the degree of human modification of the natural ecosystem . These findings indicate that melioidosis represents a complex socio-ecological public health problem in Kedah , and that its control requires an understanding and modification of the coupled human and natural variables that govern disease transmission in endemic communities .
Melioidosis , once thought to be restricted to Southeast Asia and northern Australia [1] , is now increasingly diagnosed in other tropical regions , including across Africa , the Caribbean , and other parts of Asia [2 , 3] . Thriving in soil and surface water , the causative Burkholderia pseudomallei saprophytic bacterium can bear extreme environments , including enduring starvation for long durations , a likely biological factor underlying its survival and potential transmission across broad geographical regions [4] . The main mode of transmission involves physical contact of lesions with contaminated reservoirs; however , inhalation or ingestion of particles in the air can also serve as dissemination mechanisms . B . pseudomallei has been found to be the causal agent for approximately 20% of community acquired bacteremias in north- eastern Thailand , [4] but the pathogen can also induce a wide spectrum of clinical manifestations ranging from pneumonia , internal organ abscesses to septicemia . The disease is notably lethal , with overall mortality rates found to range anywhere between 19 and 54 percent in different communities [2 , 5] . Previous work has shown that peninsular Malaysia may be at high risk for the disease , with hospital cases recorded from practically all regions of the country [6–11] , and calculated annual incidences ranging from 4 . 3 per 100 , 000 in the eastern state of Pahang [6] to as high as 16 . 35 per 100 , 000 recently reported by us in the northwestern agricultural state of Kedah [7] . These studies have also highlighted the high fatality rate , as well as provided data suggestive of the myriad social and ecological factors that may govern disease transmission in this region [6–11] . Despite the extent and potentially high public health significance of the disease in the country , definitive information on risk factors are still constrained by the fact that previous studies have invariably focused on analyses of patient case records . Although such case series-based studies are useful for estimating relative disease incidences and for quantifying the prevalence of exposure or risk factors associated with the disease [12] , these analyses are limited by the quality of patient selection , observation period , and time-invariant or fixed confounder effects [12 , 13] . Here , we extend our previous case series study by employing a hospital-based matched case-control investigation to carry out a more powerful and definitive examination of the demographic , socioeconomic , and landscape risk factors that may govern melioidosis incidence in the Alor Setar region of Kedah state . Although it is well-known that melioidosis in Southeast Asia is associated with rice farming [2 , 14] , the relative risk for disease across the major landscapes occurring in this region , including in the case of Kedah , has never been quantified . We therefore analyzed the spatial distribution of patient-cases in this study to both identify for the first time the areas of high risk in the state [15] , and to assess the landscape features that are likely to be associated with this soil and water-mediated infectious disease in this region .
All case-patients were extracted from the Melioidosis Registry , established in 2005 and containing all confirmed melioidosis cases referred to HSB . Confirmation of melioidosis at HSB is rigorous and is done by culture , serology , or a combination of both tests [7] . Microbial detection of B . pseudomallei in blood cultures is achieved using the BACTEC9420 Instrumented Blood Culture System ( Fluorescent Series , Becton Dickinson ) . Cultures of other bodily fluids were performed using blood agar and MacConkey’s medium , and the API 20 NE biochemical identification system ( BioMérieux ) for B . pseudomallei . Serology tests are based on detection of B . pseudomallei using the Indirect Fluorescent Antibody ( IFA ) method . Since culture tests are the gold standard for diagnosis , only culture positive cases were included in this study , resulting in an initial selection of 254 out of the original 488 cases from the Melioidosis Registry for the above study period . By contrast , 384 controls were initially collected from patients admitted to the orthopedics department in HSB at the same time as the selected case-patients . These controls were matched with case-patients for age , gender , race , home address , and admission date to HSB ( +/- one week ) . Addresses were matched to the nearest village or street . Following the matching effort , we obtained 242 matched pairs of cases and controls . Demographics , risk factors , including smoking , alcohol usage and occupation , and underlying illnesses , were compared between controls and case-patients . Occupation was categorized into three classes representing low to high occupational risk for contracting melioidosis based on likely exposure . Those considered to have low occupational risk were in sales , executive positions , academia , and other job roles with minimal outdoor exposure; the medium group consisted of children , housewives , and those with service jobs; while those in the high occupational risk group were solely farmers . For continuous variables , either the Student’s t test or the Mann-Whitney U test was used , and for categorical variables either the Pearson χ2 test or Fisher’s exact tests was used as required . A conditional logistic regression was used to calculate the adjusted odds ratios ( OR ) for associations of melioidosis with different underlying illnesses . All analyses were performed using SPSS 20 and R 3 . 0 . 1 [16 , 17] . Cases were mapped on Google Earth based on patient address to obtain latitude and longitudinal coordinates . Overall , locations were identified for 175 case-patients; controls were not mapped because of locational matching with case-patients . The original latitude and longitudinal coordinates were then projected to Kertau UTM Zone 47 for Malaysia to create a case distribution map using ArcGIS 10 . 1 [18] . The presence of significant clusters in the spatial case data was assessed using the scan statistics feature in SaTScan 9 . 0 [19] . A continuous Poisson model , which tests the hypothesis that cases follow a homogeneous spatial Poisson process with constant density throughout the study area [20] , was applied with the following parameters: scanning for areas with high rates in a circular manner , a maximum spatial cluster size set to 50% of the population at risk , and no geographical overlap [17] . In order to determine the prevalence of melioidosis within different landscapes occurring in the study area , we first overlaid the case point data onto a 5 arc minutes land cover map of East Asia and the Pacific [21] , clipped to cover the Kedah state boundary . This allowed the count of cases falling in each land cover type . A gridded population map of Kedah was then extracted from the Gridded Population of the World [22] and overlaid on the land cover map in order to derive the population count within each land cover type . The melioidosis prevalence in each land cover classification was estimated using the following equation: ( number of cases/population count ) * 100 , 000 . Spatial variation in the prevalence of the disease between the different land cover types was assessed by applying the Fisher’s exact test with simulated p-values based on 2000 replicates [14] . The proportions of each land cover type within and outside the identified disease cluster were calculated using ArcGIS [18] , and statistical differences in the composition of proportionate landcover types within and outside the cluster were evaluated using a Dirichlet regression model for testing variations in compositional data between categorical variables [23] . The project was approved by the National Institute of Health ( NIH ) and Malaysian Research and Ethics Committee ( MREC ) . All patient data used in this study was anonymized .
Demographic characteristics , and underlying risk factors and illnesses , are displayed in Table 1 for the matched cases and controls . Overall , there was a preponderance of males ( 76 . 4% ) , the predominant race was Malay ( 93 . 0% ) , and the mean age was 46 . 6 , with males slightly older than females ( Table 1 ) . Subject occupation appeared to play a significant role in disease acquisition , with case-patients found to be engaged proportionately more in the high risk occupation of farming ( 16 . 9% in case-patients versus 10 . 3% in controls: χ2 = 4 . 49 , p < 0 . 034 ) , while they were significantly underrepresented in the low risk occupations of sales , executive positions and academics ( 10 . 7% versus 22 . 3% in controls: χ2 = 11 . 74 , p < 0 . 001 ) . While no significant associations between case-patients and either alcohol use or smoking were found for this population , varying degrees of statistically significant relationships , by contrast , were found between case-patients and diabetes ( χ2 = 43 . 269 , p < 0 . 001 ) , chronic renal failure ( χ2 = 8 . 593 , p = 0 . 003 ) , and other immunocompromised states ( χ2 = 4 . 03 , p = 0 . 045 ) . Crude ORs ( Table 2 ) from running ordinary univariate logistic regressions supported these findings for diabetes , ( OR = 3 . 46 , 95% CL:2 . 38–5 . 04 ) and chronic renal failure ( OR = 4 . 04 , 95% CL:1 . 30–2 . 05 ) , but not other diseases ( chronic lung failure , HIV and other immunocompromised states ) ( OR = 1 . 20 , 95% CL:0 . 83–1 . 75 ) . By contrast , adjusted ORs from the conditional logistic model showed that all three ( diabetes: OR = 4 . 13 , 95% CL:2 . 62–6 . 51; chronic renal failure: OR = 1 . 95 , 95%CI 1 . 19–3 . 19; and other diseases: OR = 10 . 0 , 95% CL:1 . 28–78 . 12 ) comprised significant risk factors for the disease . They also show that the odds of acquiring melioidosis is likely to be 4 times higher among diabetics and about 2 times higher among patients with chronic renal failure . The case-fatality ratio among culture-confirmed melioidosis cases was 41 . 8% ( 95% CL: 35 . 5–48 . 4 ) while the case-fatality ratio among melioidosis-suspected cases was 19 . 7% ( 95% CL: 14 . 7–25 . 9 ) ( Table 3 ) . Risk factors which significantly affected the relative risk of death among the confirmed cases include age group ( with risk of death significantly low amongst the youngest section of the present cases ) , having diabetes mellitus , having an unknown-risk for occupation , or having an unknown smoking status . Among melioidosis-suspected cases , the risk factors significantly affecting the risk of dying include having an underlying infection of diabetes mellitus or chronic renal failure . The spatial distribution of melioidosis cases superimposed on the Kedah land cover map is shown in Fig 1 , and indicates a concentration of cases in the urban and semi-urban settings surrounding Alor Setar . The continuous Poisson spatial scan statistic detected one likely primary cluster ( log likelihood ratio = 505 . 18 , radius = 25 . 57 km , centroid coordinates = 100 . 467 , 6 . 15 , p<0 . 001 ) that included HSB ( Fig 1 ) . No significant secondary cluster was detected . Table 4 shows the prevalence of the disease across the five major land types or landscapes expected to influence melioidosis transmission within the present study location . These land cover types were derived by combining the finer scale environments shown in Fig 1 ( see details of the land cover combinations carried out as shown in Table 4 ) , and might be considered to represent a gradient in the underlying environmental exposure risk to acquiring melioidosis . They also represent a graded increase in the degree of human modification of the natural ecosystem from forested areas to the urban setting . Table 4 shows that , as expected , melioidosis prevalence varied markedly between these land cover types , with results from the Fisher’s exact test showing this variation to be highly statistically significant ( χ2 = 45 . 019 , p< 0 . 0005 ) . We estimated the risk of living in these environments by calculating an exposure OR based on the joint distributions of cases and populations at risk in each land type . This showed that while living in areas with large-scale irrigation-based agriculture represented the most risk for acquiring melioidosis ( OR = 2 . 24 , 95% CL:1 . 57–3 . 17 ) , followed by mixed agriculture/pastoral environments ( OR: 1 . 55 , 95% CL:1 . 02–2 . 31 ) ; living in forest-associated areas , either the mixed forested ( OR: 0 . 55 , 95% CL:0 . 34–0 . 86 ) or forestry areas ( OR: 0 . 28 , 95% CL:0 . 08–0 . 73 ) , protected against the disease . Living in urban Alor Setar also provided some protection , but this protection did not attain statistical significance ( OR: 0 . 83 , 95% CL: 0 . 59–1 . 13 ) . Fig 2 plots the ORs in relation to the five land cover types , and illustrates the non-linearity in the impact of the investigated land cover types on melioidosis risk . Fig 3 shows the proportions of the five landscapes observed within the primary high infection cluster ( Fig 1 ) in comparison to the proportions recorded outside the cluster . The proportions outside the cluster were obtained by calculating the respective areas of each land cover type occurring within a convex hull placed around the locations of cases outside the primary cluster ( Fig 1 ) . The results show that the proportions of each land cover type varied markedly within and outside the primary cluster , with high risk land cover types ( mixed agriculture and in particular , agricultural areas ) occurring proportionately more within the primary cluster ( Dirichlet Regression with intercept only versus one including cluster effects: deviance = 220 . 5142 , df = 5 , p<0 . 001 ) . A multivariate logistic regression analysis , controlling for the effects of age , gender and occupational differences between cases located within and outside the primary cluster , confirmed that the observed spatial clustering of melioidosis cases found in this study was a direct function of living in such “pathogenic” environments ( Table 5 ) .
Although Malaysia is increasingly recognized as an important endemic focus for melioidosis [7 , 24] , and previous investigations have highlighted the basic epidemiological features of the disease among cases , this is the first case-control study from that country conducted specifically to elucidate the socio-epidemiological and landscape risk factors , as well as magnitudes of their association , for bacteremic disease . In order to more effectively evaluate the impacts of these diverse etiological factors , we have also used an efficient matched case control study design to undertake the present analysis . Our study populations were representative of the ethnic make-up of Kedah state , with 93 . 0 percent of study subjects being Malay . The Census of Malaysia reports an equal proportion of males and females within the state of Kedah [25] . However , 76 . 4 percent of the case-patients in our study population were males while 23 . 6 percent were females , supporting suggestions that there is a strong tendency for infections to occur in males in this region [24 , 26] . The mean age of male subjects was slightly higher than that of females ( Table 1 ) , but since our previous study indicated that age may have a non-linear rise and fall relationship with cases [7] , this age difference is unlikely to underlie the higher occurrence of the disease in males . On the other hand , as males in the present community engage more in activities related to greater contact with soil ( e . g . farming , small-scale foodstuff or recreational gardening ) , this result corroborates previous conclusions [27 , 28] that the elevated disease prevalence/incidence observed for males is due to their higher occupational exposure to the disease [27–29] . The strong occupationally-related hazard for acquiring melioidosis was further corroborated by our analysis of socio-economic risk factors for the disease ( Table 1 ) . Thus , while individuals engaged in activities with a generally low outdoor exposure appeared to be protected from the disease , those in occupations with higher outdoor exposure were at elevated risk of being infected and sick . In particular , the results indicate that due to a potentially prolonged exposure to soil sources and thus higher probability of inhalation , ingestion or inoculation of B . pseudomallei , rice farmers , a large proportion of who were also males , in this region form the most vulnerable group for infection . This finding is in line with results from other studies conducted in Southeast Asia with a comparable extensive rice farming agricultural profile as Kedah state [30 , 31] . However , as our spatial analysis of landscape risk indicates , this observed association with occupation may be a complicated one . The key result here is that after controlling for occupation , individuals living in high risk environments , viz . agriculture/mixed agriculture regions , are still at higher risk for acquiring infection , suggesting that rather than occupation per se it is the place in which these activities take place that is more important . Excess alcohol consumption has been identified as a predisposing factor for melioidosis in tropical Australian populations [26] . However , the religion most commonly practiced in Kedah is Islam , which prohibits practicing Muslims from the consumption of alcohol . Thus , only a diminutive one percent of our study group consumed alcohol , showing its insignificance as a risk factor in this particular population , as was also found to be the case in previous studies from Malaysia [10] and Singapore [32] . By contrast , our results reinforce the results from different communities in tropical Australia and across the Southeast Asia region , which indicate that patients with undermined immune systems , specifically those with diabetes and chronic renal diseases , are significantly at higher risk for developing melioidosis [4 , 24–26] . The finding that diabetes is a risk factor for melioidosis is of particular concern in Malaysia due to its high diabetes prevalence and the observed trend of increasing prevalence [33] , which has been attributed to the adoption of more “Westernized” diets and population shifts in exercise [34] . However , while patients with chronic lung disease were also at a significantly higher risk for the disease in the northern tropical Australian study setting , this subgroup was not a major constituent in our study ( Table 1 ) , and was also omitted in the Thailand investigation due to difficulties in risk estimation using retrospective records [25 , 26] . These disparate results , and the finding from a concluded randomized control trial [35] , has cast doubt on the involvement of the complex polymorphonuclear leukocyte ( PMNL ) network in increased risk to melioidosis observed amongst those with the above underlying diseases [29 , 36–38] , suggesting that research to uncover more complex mechanisms in diverse endemic populations must continue to be a priority if more effective therapeutic or other interventions are to be developed for controlling such deadly co-morbidities . The overall case-fatality rates among culture-confirmed and suspected cases were 41 . 8% and 19 . 7% , respectively ( Table 3 ) , illustrating the lethal nature of the disease in Kedah communities despite the use of recommended standard antibacterial agents for therapy . For both confirmed and suspected cases , case-fatality rates did not differ significantly by sex or ethnicity but were higher for patients with diabetes and for suspected cases , patients with chronic renal failure ( Table 3 ) . These findings are similar to results from other studies carried out in Southeast Asia and Australia [26 , 27] . Our analysis of the spatial distribution of melioidosis cases , indicated , firstly , the presence of a significant primary cluster of elevated cases occurring in close proximity to the main city of Alor Setar , where HSB is also located ( Fig 1 ) . While this may reflect the confounding effect of patient proximity to the hospital , the fact that HSB serves as the primary reference hospital for the disease in the state of Kedah suggests that this influence is unlikely to be the major cause of the identified disease cluster . This conclusion is further supported by the results of our analysis of the spatial risks associated with acquiring melioidosis for populations residing in different landscapes . The major finding here is that while human-unaltered forested areas are protective , the highest risks occurred with agricultural-related modifications of the environment , particularly in the case of large-scale irrigation-based agriculture , which in the Kedah context is primarily related to vast rice cultivation . Intriguingly , the results show that the risk of acquiring the disease declined , and indeed was neutral in the urban setting of Alor Setar , which may be considered to represent the most human modified of the land types studied here . These findings support the growing evidence from landscape-based epidemiological studies , which suggest that forest modification and fragmentation may be a major driver for the emergence of many vector- and water-borne diseases [39 , 40] . We suggest that for melioidosis , the specific mechanisms , apart from variations in human population density , which could underlie this association may be primarily related to an increase in the habitat and hence abundance of B . pseudomallei as forest ecosystems are altered to produce agricultural land types . For example , the protection afforded by intact forests may be due to their ecohydrological functions that may regulate water- or soil-borne pathogen emergence by filtering pathogen-laden runoff and modulation of the amplitude of flows during seasonal rain [35] . However , we did not measure the presence or abundance of B . pseudomallei in the different land use types , so this additional analysis would need to be conducted to test our proposed hypothesis . However , our finding that the risk of acquiring melioidosis may be reduced in the urban setting indicates that the association between infectious disease emergence and human modification of the natural ecosystem is complex and will depend on the specific pathogen . For melioidosis , we suggest that urban areas may represent a neutral risk setting because although on the one hand , they represent centers of high population density , this is countered by the presence of lower pathogen habitats , higher socio-economic development of the population , and better provision of health and other services . This study has several limitations . Firstly , melioidosis may remain latent for years before symptoms develop; thus , the place of residence at the time of infection may differ from the place of current residence . Secondly , we selected controls from the orthopedics department , which may introduce a selection bias as individuals in the orthopedics department may not be representative of the entire melioidosis-negative population . Individuals in the orthopedics department were not clinically tested for the presence of B . pseudomallei; however , it is unlikely the controls are infected as melioidosis is a notifiable disease although they may be latently infected . An additional source of selection bias may have been introduced by the decision to recruit individuals from a single hospital; distance to hospital may play an important role in determining which individuals choose to seek treatment . However , melioidosis is a notifiable disease and clinically positive individuals are referred to HSB from all over Kedah state , so this may diminish the importance of the distance to hospital variable . Overall , our results indicate that complex and interacting socio-ecological factors may underlie the transmission of melioidosis in Kedah . This conclusion also indicates that if we are to gain a fuller understanding of the components and their modification required to prevent the future emergence as well as reduction of the disease amongst at-risk populations in Malaysia , and elsewhere in the tropics , it is essential that we deploy an investigatory approach that examines melioidosis as an outcome of coupled human and natural variables acting at various spatial scales from the individual host to the environment . | Although the public health significance of melioidosis as a particularly highly fatal emerging infectious disease threat in the tropics is being recognized , data on the risk factors and the geographic distribution of the disease is still limited . Previous studies have also largely been based on the investigation of case series data . Here , we extend these studies by employing a hospital-based matched case-control study to carry out a more robust examination of the demographic , socioeconomic , and landscape risk factors that may govern melioidosis incidence in the Alor Setar area of Kedah , State , a key disease endemic region of Malaysia . Our results indicate that melioidosis in northern Malaysia is significantly associated with gender , race , occupation , co-occurring chronic disease as well as living in risky landscapes , with disease risk increasing and declining non-linearly with the degree of ecosystem modification . These findings denote that the disease represents a complex socio-ecological public health problem in Kedah , and that its control requires an understanding and modification of both the human and natural variables that underlie disease transmission in this setting . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"medicine",
"and",
"health",
"sciences",
"melioidosis",
"geographical",
"locations",
"diabetes",
"mellitus",
"bacterial",
"diseases",
"endocrine",
"disorders",
"medical",
"risk",
"factors",
"infectious",
"disease",
"control",
"infectious",
"diseases",
"epidemiology",
"endocrinology",
"metabolic",
"disorders",
"agriculture",
"people",
"and",
"places",
"asia",
"biology",
"and",
"life",
"sciences",
"malaysia",
"renal",
"failure",
"nephrology"
] | 2019 | Socio-epidemiological and land cover risk factors for melioidosis in Kedah, Northern Malaysia |
Extensive apoptosis is often seen in patterning mutants , suggesting that tissues can detect and eliminate potentially harmful mis-specified cells . Here , we show that the pattern of apoptosis in the embryonic epidermis of Drosophila is not a response to fate mis-specification but can instead be explained by the limiting availability of prosurvival signaling molecules released from locations determined by patterning information . In wild-type embryos , the segmentation cascade elicits the segmental production of several epidermal growth factor receptor ( EGFR ) ligands , including the transforming growth factor Spitz ( TGFα ) , and the neuregulin , Vein . This leads to an undulating pattern of signaling activity , which prevents expression of the proapoptotic gene head involution defective ( hid ) throughout the epidermis . In segmentation mutants , where specific peaks of EGFR ligands fail to form , gaps in signaling activity appear , leading to coincident hid up-regulation and subsequent cell death . These data provide a mechanistic understanding of how cell survival , and thus appropriate tissue size , is made contingent on correct patterning .
Defective cells are often eliminated by apoptosis during development and tissue homeostasis [1–4] . This has been particularly well studied during the process of cell competition , whereby unfit cells are eliminated when confronted with normal cells within a growing tissue [5] . Excess apoptosis is also seen in mutants that lack essential developmental determinants , a phenomenon that has been observed in a variety of model organisms , including zebrafish embryos lacking the signaling molecule Sonic Hedgehog [6] , mice lacking the negative Wnt signaling regulator Adenomatous polyposis coli ( APC ) in the developing neural crest [7] , and Drosophila segmentation mutants [8–12] . These observations have suggested the existence of a quality control system that detects conflicting or nonsense patterning inputs and , as a result , initiates apoptosis in response . Although apoptosis was first observed in Drosophila segmentation mutants over 30 years ago [8–10] , relatively little is known about the molecular basis of cell elimination in this context . Previous studies have revealed that apoptosis occurs within and around the areas where the missing developmental regulators are normally required . Thus , in the pair-rule mutant odd-skipped ( odd ) , apoptosis is seen in alternate stripes that encompass the areas where odd is normally expressed [13 , 14] . Likewise , in embryos lacking the anterior determinant Bicoid , ectopic cell death occurs primarily at the anterior of the embryonic epidermis [13] . In Drosophila , apoptosis is initiated by a double inhibition mechanism: the activity of ubiquitously expressed inhibitor of apoptosis proteins ( IAPs ) , which prevent caspase activation , is inhibited by the so-called IAP antagonists reaper ( rpr ) , head involution defective ( hid ) , grim , and sickle ( skl ) [15] . Most apoptotic events are initiated by the transcriptional up-regulation of one or more IAP antagonists . Indeed , hid is up-regulated in a pattern that prefigures apoptosis in segmentation mutants , i . e . , at the anterior of bicoid mutants and in alternate stripes in odd mutants [13] . Therefore , it appears that cells within the segmental pattern “know” that they are missing key fate determinants and activate hid expression in response . Here , we take advantage of the genetic tools in Drosophila to investigate how apoptosis is triggered in mispatterned epidermal cells .
Consistent with an earlier report [13] , extensive apoptosis was detected in Drosophila embryos lacking genes acting at various steps of the segmentation cascade , including mutants of the terminal gene tailless ( tll ) , the gap gene krüppel ( kr ) , the pair-rule gene fushi-tarazu ( ftz ) , or the segment polarity gene hedgehog ( hh ) ( Fig 1A–1E ) . In each instance , cleaved Death caspase-1 ( Dcp1 ) , a marker of apoptosis , was strongly enriched in the areas where the mutated patterning gene is known to be required during normal development . For example , in a null ftz mutant generated for this study ( ftzΔ . attP ) , seven bands of apoptotic cells appeared in the segments where Ftz is normally required in the wild type ( Fig 1E , see also [12] ) . Thus , in ftzΔ . attP mutants , alternate segments undergo massive apoptosis , while almost no cell death is detected in the remaining , normally patterned segments , which therefore serve as a useful control . For this reason , ftz mutants were selected as a prototypical condition to investigate mispatterning induced apoptosis . Among the four known proapoptotic genes of Drosophila , hid is the key mediator of apoptosis in bicoid mutant embryos [13] . To assess the involvement of hid in ftz mutants , we created a clean hid allele ( hidΔ . attP ) by replacing the first coding exon , which encodes the IAP-binding motif , with an attP integrase site [16–18] . This mutation was recombined with ftzΔ . attP , and the resulting double mutant embryos were stained with anti-cleaved Dcp1 . Although some signal remained in the head region , immunoreactivity was much lower than in single ftzΔ . attP mutants ( Fig 1F and 1I ) . In contrast , abundant apoptosis was detected in ftzΔ . attP mutants lacking the closely related IAP antagonists rpr and skl ( S1 Fig ) . We conclude that , as in bicoid mutants , hid is the main mediator of apoptosis in ftz mutant embryos . Thus , in the absence of Hid , cells that would normally be eliminated are forced to contribute to the final pattern . To assess the fate of these “undead cells” , hidΔ . attP ftzΔ . attP double mutants were allowed to undergo terminal differentiation into first instar larvae , and the pattern of denticles was examined . Alternate denticle belts were missing while numerous nondescript ectopic denticles appeared at the posterior of each remaining belt ( S2 Fig ) . In addition , the cuticle appeared particularly convoluted , suggesting that tissue size is at least partially restored compared to the single mutant . Therefore , as previously suggested for bicoid mutants [13] , it appears that undead cells can contribute to the epidermis but are unable to adopt their normal fate . To visualize the expression of hid in mispatterned embryos , we generated an authentic transcriptional reporter by integrating a cDNA encoding green fluorescent protein ( GFP ) into the attP site of hidΔ . attP [16] . These animals will be referred to as hidΔ . GFP to indicate that this genetic modification creates a null allele ( Δ ) as well as a fluorescent reporter ( GFP ) . To increase signal intensity , the hidΔ . GFP sensor was examined in homozygous conditions throughout , which has the dual benefit of doubling the copy number of the GFP reporter in the genome whilst simultaneously preventing apoptosis , which could eliminate cells before the GFP signal has time to accumulate . In homozygous hidΔ . GFP embryos , a weak GFP signal could be detected throughout the epidermis , suggesting that low-level hid expression occurs at subapoptotic levels during normal development ( Fig 1G ) . In hidΔ . GFP ftzΔ . attP double homozygotes , increased levels of GFP fluorescence were detected in broad stripes , resembling the bands of caspase immunoreactivity seen in ftzΔ . attP single mutants ( Fig 1H and 1J ) . Time-lapse imaging showed that striped GFP fluorescence arises around stage 11 of embryogenesis ( S1 Movie ) , approximately 4 hours after the time when pair-rule genes contribute to segmental patterning [19 , 20] . Likewise , cleaved Dcp1 immunoreactivity became only detectable at stage 11 in homozygous ftzΔ . attP embryos ( S3A–S3C Fig ) . Therefore , loss of ftz appears to trigger hid expression and apoptosis with a delay , well after Ftz has fulfilled its patterning role in wild-type embryos . The same delay was noted in bicoid mutants [13] , suggesting that it could be a general feature of segmentation mutants . Mismatched or nonsense cell fates could conceivably be recognized through local cell interactions . With the aim of identifying the relevant mediator , we used the ubiquitous actin5C-Gal4 driver to overexpress modulators of major signaling pathways in a ftzΔ . attP mutant background and stained the resulting embryos with anti-cleaved Dcp1 ( summarized in S1 Table ) . One manipulation , overexpression of an activated form of the Drosophila epidermal growth factor receptor ( EGFRact , [21] ) , significantly reduced cleaved Dcp1 immunoreactivity ( Fig 2A–2D ) . Since EGFR signaling is required for survival of the cells that secrete naked cuticle [22] and since EGFR signaling has been previously implicated in suppressing hid activity [23 , 24] , we set out to determine whether the excess apoptosis and up-regulation of hid in ftz mutants could be accounted for by a loss of EGFR activity . If EGFR signaling is needed to keep hid expression off , we would expect hid expression to be up-regulated in EGFR mutants . Indeed , strong ubiquitous GFP fluorescence was observed in embryos homozygous for both EGFRF3 , an amorphic allele , and hidΔ . GFP ( S4A and S4B Fig ) . This suggests that most , if not all , epidermal cells require EGFR signaling to keep hid expression low . Conversely , the prosurvival activity of EGFR signaling appears to be largely mediated by repression of hid activity since almost no apoptosis was detected in EGFRF3 hidΔ . attP double mutants ( S4D Fig ) . In EGFRF3 single mutant embryos , cell death was widespread ( S4C Fig ) but did not occur until embryonic stage 11 ( S3D–S3F Fig ) , the same time at which apoptosis becomes detectable in ftz mutants . We conclude that , in the absence of EGFR signaling , most cells of the epidermis up-regulate hid expression and undergo apoptosis around mid-embryogenesis . To visualize EGFR signaling activity in mispatterned ftz embryos , we stained embryos with an antibody that recognizes phosphorylated extracellular signal–regulated kinase ( dpERK ) [25] . As validation , we first stained EGFRF3 homozygotes . No immunoreactivity was detectable above background levels ( S5 Fig ) , a strong indication that EGFR is the principal contributor of ERK phosphorylation in the embryonic epidermis . Anti-dpERK was next used to compare the patterns of EGFR signaling activity in ftzΔ . attP mutant and control embryos . The hidΔ . GFP allele ( homozygous ) was included to allow simultaneous assessment of hid expression and to avoid possible confounding effects of apoptosis on signaling activity . In control hidΔ . GFP embryos , dpERK immunoreactivity was observed in a wide range of tissues , including the epidermis but also the peripheral nervous system , oenocytes , and the tracheal precursors ( Fig 2E ) [26–29] . In the epidermis , dpERK immunoreactivity was detectable in nearly all cells , though not uniformly so . Signal intensity followed a broadly undulating pattern along the anterior/posterior ( A/P ) axis , with smooth peaks near segment boundaries and intervening shallow troughs ( Fig 2G ) . Little GFP fluorescence was detectable in the epidermis ( Fig 2E′ and 2G ) , confirming that hid transcription ( as reported by hidΔ . GFP ) remains low in normally patterned embryos ( see also Figs 1G and S3A ) . In the absence of ftz activity , dpERK immunoreactivity dropped significantly in seven broad bands ( Fig 2F and 2H ) , leaving only half the number of signaling peaks intact . At the same time , a complementary pattern of GFP fluorescence ( from the hidΔ . GFP reporter ) appeared ( Fig 2F and 2H ) . We conclude that , in ftz mutant embryos , the transcriptional activation of hid increases in regions where EGFR signaling activity fails to reach the threshold level required for epidermal cell survival . Other segmentation mutants , including tll , kr , and hh , also displayed anticorrelated hidΔ . GFP reporter activity and dpERK immunoreactivity ( Fig 3A–3C ) . Importantly , the pattern of hid expression in these mutants corresponded to that of apoptosis ( as shown in Fig 1A–1D ) . Embryos lacking another segmentation gene , patched , were particularly informative , as they had increased dpERK immunoreactivity and did not up-regulate hidΔ . GFP ( Fig 3D ) . Crucially , they also showed no excess apoptosis , as assayed with anti-cleaved Dcp1 ( S6 Fig ) . This observation confirms the tight correlation between EGFR signaling and the absence of hid expression . It also shows that not all instances of patterning error are associated with apoptotic cell death . EGFR signaling has previously been widely implicated in cell survival in the embryo and other tissues [22 , 30 , 31] . Known Drosophila EGFR ligands include the TGF-α homologues Spitz , Keren , and Gurken , which require activation by Rhomboid proteases [32–35] , as well as the Rhomboid-independent neuregulin-like ligand Vein [36] . To assess the nature of the ligands involved in the survival of embryonic epidermal cells , we first examined embryos homozygous for a deficiency ( Df rho-1 , 3 ) spanning the rhomboid ( rho-1 ) and roughoid ( rho-3 ) loci . In this background , a relatively mild apoptotic phenotype was observed ( S7B Fig ) . Likewise , homozygous vein ( vn ) mutants were also characterized by a relatively minor increase in the number of apoptotic cells compared to wild-type embryos ( S7C Fig ) . In contrast , simultaneous removal of both ligand systems , in Df rho-1 , 3 vnL6 double homozygotes , led to widespread apoptosis ( S7D Fig ) , consistent with the severe cuticle phenotype previously reported [22] . We conclude that multiple ligands act redundantly to elicit the EGFR signaling landscape that , at mid-embryogenesis , prevents apoptosis . In agreement with previously described expression patterns [36 , 37] , vn and rho-1 were found to be expressed in a generally segmental manner in wild-type embryos ( Fig 4A and 4C ) . Thus , peaks of ligand production correspond roughly to the areas of high-dpERK immunoreactivity seen around segmental borders ( Fig 2G ) . We therefore considered whether changes in ligand production could account for the observed changes in EGFR signaling in patterning mutants . The pattern of rho-1 expression is expected to depend on the segmentation network since it is determined by segment polarity genes [37 , 38] , which are themselves regulated by upstream pair-rule and gap genes . It is likely that the segmentation network similarly controls vn expression . Therefore , as expected , peaks of vn and rho-1 expression failed to appear in alternate segments of ftzΔ . attP mutants ( Fig 4B and 4D ) , foreshadowing the loss of dpERK immunoreactivity ( Fig 2F ) , hid up-regulation ( Fig 1H ) , and apoptosis ( Fig 1E ) . Importantly , the same logic can explain the pattern of apoptosis in other segmentation mutants , which are all expected to affect the pattern of EGFR ligand production within their realm of action . It also neatly explains the absence of apoptosis in patched ( ptc ) mutants , in which ectopic peaks of rho-1 expression arise [37] . During normal development , EGFR signaling is lowest in cells located around the middle of each segment , which are likely to be exposed to the lowest amounts of ligand ( Fig 2G ) . We asked whether these cells are particularly susceptible to apoptosis . To map the pattern of apoptosis along the A/P axis , we performed in toto multiview single plane illumination microscopy ( mSPIM ) on wild-type embryos stained with anti-cleaved Dcp1 and anti-Engrailed , which marks the anterior edge of each segment boundary . This approach allowed us to visualize , at a given time point , all epidermal apoptotic events , which are relatively rare during normal development ( see Fig 1A and Fig 2A ) . Raw image files were generated ( Fig 4E ) , segmented ( Fig 4F ) , and the position of each caspase-positive cell was then mapped and normalized relative to the nearest anterior and posterior segment boundary . These data were then tabulated in a histogram displaying the frequency of apoptosis at various positions along the A/P axis within each segment ( Fig 4G ) . Cell death was relatively infrequent near segment boundaries , where vn and rho-1 are most highly expressed ( Fig 4A–4D ) , while apoptosis was more abundant in the middle of the intervening areas . We conclude that , in the cells that are most distant from a source of ligand , EGFR signaling activity falls near , and occasionally below , the threshold level needed to protect against apoptosis . Cells in these regions are therefore more likely to commit to an apoptotic cell fate . Despite the observed correlation between the distance to segment boundaries and the rate of apoptosis in wild-type embryos , it is clear that cells located near the sources of EGFR ligand production can still undergo apoptosis . In this context , it is worth pointing out that only a subset of embryonic apoptotic events are regulated by the EGFR–Hid axis , since residual apoptosis is still detected in hid mutants [39] , and Hid-independent apoptotic processes have been characterized in the Drosophila embryo [40] . Therefore , it is likely that additional factors , besides EGFR signaling , contribute to the control of apoptosis in the embryonic epidermis .
Excess apoptosis in patterning mutants gives the impression that mispatterned cells are recognized as deleterious and eliminated during development . We have shown that excess apoptosis in embryonic patterning mutants can be explained by the requirement of EGFR signaling for cell survival and the fact that the segmentation cascade culminates in the segmental production of EGF ligands ( model summarized in Fig 4H ) . Thus , in segmentation mutants , sources of EGF ligand production are lost , leaving remaining sources unable to reach all intervening cells . This results in tissue loss until all cells are brought back within signaling range , restoring a “normal” distance between segment boundaries . EGFR signaling has previously been linked to compartment size regulation in the embryo [41] . Specifically , it was proposed that epidermal cells in the posterior regions of each segment compete for a finite amount of Spitz emanating from an anterior source . Our results are consistent with an extended version of this model , whereby multiple EGFR ligands ( including Spitz and Vein ) produced primarily at either side of the segment boundary control the size of entire segments , not just posterior compartments . The pattern of apoptosis in wild-type embryos ( Fig 4G ) suggests that segments are slightly oversized upon completion of patterning and that the extra cells are eliminated as a result of subthreshold EGFR signaling and hid activation . Accordingly , there is no need to invoke the existence of a mechanism that recognizes and eliminates mis-specified cells , and mispatterning-induced cell death could be the byproduct of a size control system . As we have shown , the lack of EGFR signaling triggers hid expression only at stage 11 , when proliferation is largely complete [42] . This would serve as a particularly suitable time for a size control checkpoint and pruning of excess cells would be an effective means of ensuring reproducible segment dimensions in a fast-growing embryo with no compensatory proliferation . Similar prosurvival signaling by limited ligand diffusion from specific locations could form the basis of tissue size sensing and explain mispatterning-induced apoptosis in a variety of developmental contexts . It is intriguing that , while EGFR signaling over a relatively low threshold is required for cell survival , higher signaling activity ( as occurs near the segment boundary ) contributes to the specification of pattern elements ( i . e . , denticles ) [22 , 37 , 43–45] . Thus , the same ligands contribute to both patterning and cell survival , perhaps ensuring a tight coordination of these two essential developmental activities .
Flies were maintained at 25°C on standard fly food containing yeast , agar , and cornmeal . w1118 was used as a wild-type control throughout . Fly stocks used in this study are listed in Table 1 . To identify regulators of apoptosis in mispatterned cells , a collection of UAS lines was assembled . Each expressed a gene that modulates the activity of a signaling pathway in the presence of Gal4 ( S1 Table ) . These so-called UAS lines were crossed to the null ftz allele ftzΔ . attP , and male progeny containing both the ftz allele and the UAS transgene were crossed to females bearing the ubiquitous actin5C-Gal4 driver recombined to ftzΔ . attP . The embryos produced from this cross were then stained for cleaved Dcp1 . Accelerated “ends out” homologous recombination was used to generate the hidΔ . attP and ftzΔ . attP alleles ( as described in [18] ) . Homology arms of approximately 2 kb were PCR amplified from genomic DNA . These fragments were cloned into a targeting vector ( pTVCherry ) containing a GMR > white eye marker and an attP landing site flanked with FRT sites . UAS-rpr was included downstream of the 3′ homology arm . Targeting vectors were integrated into the genome via P element–mediated transformation to generate so-called donor lines . To release the homology cassette from its genomic location , donor flies were crossed to hs-FLP and heat shocked at 24-hour intervals throughout larval development . Adult female progeny with mottled eye color ( indicating the presence of both the targeting vector and hs-FLP constructs ) were crossed to the strong ubi-Gal4 driver to eliminate individuals that had either failed to excise the homology cassette or had undergone an unsuccessful recombination event . Resultant white+ candidates were then isolated and verified with PCR . To generate the hidΔ . GFP reporter , a GFP cDNA was cloned into the RIVCherry vector [18] , which contains an attB sequence and a 3xPax3-cherry selection marker . This construct was injected into hidΔ . attP embryos along with a source of ΦC31 integrase , and progeny were screened for Cherry expression . Rainbow Transgenic Flies , Inc . ( Camarillo , California , United States ) and Bestgene , Inc . ( Chino Hills , California , United States ) were used for embryo injection services . Embryos were collected overnight , transferred to baskets , and washed with water . Embryos were then dechorionated using 75% bleach for 3 minutes and washed again . After drying on a paper towel , embryos were transferred with a brush to microcentrifuge tubes containing equal volumes of heptane and 6% formaldehyde in phosphate buffered saline ( PBS ) . Fixation was performed for 25 minutes on a rotating platform . The fixative was then removed and replaced with 1 ml methanol . Embryos were shaken vigorously for 45 seconds to remove the vitelline membrane , and the heptane was removed . Embryos were washed twice more in methanol and moved to −20°C for long-term storage . Fixed embryos were rehydrated in 0 . 3% triton in PBS ( PBT ) and blocked for 30 minutes at room temperature in 4% bovine serum albumin in PBT . Embryos were incubated with primary antibody at 4°C overnight , washed for 1 hour in PBT at room temperature , and then incubated in a solution of Alexa Fluor 488 , 568 , or 633 conjugated species appropriate secondary antibodies ( Thermo Fisher Scientific , 1:1 , 000 ) for 2 hours . Further washes ( 1 hour ) were conducted before samples were mounted in Vectashield containing DAPI ( Vector Laboratories ) . Primary antibodies used in this study are listed in Table 2 . Confocal microscopy was conducted with a Leica SP5 confocal microscope using a 20x ( Leica , NA 0 . 7 ) or 40x ( Leica , NA 1 . 25 ) oil immersion objective . Typically , 8 confocal planes were imaged at 1 . 8 μm intervals and processed using Fiji ( NIH ) and Photoshop ( Adobe ) . Samples were prepared using the hanging drop method described in [47] . A small drop of oxygen permeable Voltalef 10S oil was spotted onto a cover slip . A dechorionated embryo was then placed inside the oil and oriented with a 27-G hypodermic needle . The sample was then inverted so that the oil containing the sample hung below the cover slip . Images were acquired from above at 10 minute intervals using a Leica SP5 confocal microscope and processed using imageJ ( NIH ) and Photoshop ( Adobe ) as described above . Fluorescently labeled samples were imaged on a multiview light sheet microscope [48] . The optics consisted of two detection and illumination arms . Here , each detection arm formed an epifluorescence microscope using in sequence: a water-dipping lens ( Apo LWD 25x , NA 1 . 1 , Nikon Instruments , Inc . ) , a filter wheel ( HS-1032 , Finger Lakes Instrumentation LLC ) equipped with emission filters ( BLP01-488R-25 , BLP02-561R-25 , Semrock , Inc . ) , a tube lens ( 200 mm , Nikon Instruments , Inc . ) , and sCMOS camera ( Hamamatsu Flash 4 . 0 v2 . 0 ) . The imaging produced an effective pixel size of 0 . 26 μm . The illumination arms each consisted in sequence: a water-dipping objective ( CPI Plan Fluor 10x , NA 0 . 3 , Nikon Instruments , Inc . ) , a tube lens ( 200 mm , Nikon Instruments , Inc . ) a scan lens ( S4LFT0061/065 , Sill optics GmbH & Co . KG ) , and a galvanometric scanner ( 6215h , Cambridge Technology , Inc . ) . The illumination arms were fed by a combination of lasers ( 06-MLD 488 nm , Cobolt AB , and 561 LS OBIS 561 nm , Coherent , Inc . ) . Samples were translated through the resulting light sheet using a linear piezo stage ( P-629 . 1CD together with E-753 controller ) . Multiple rotation views were achieved using a rotational piezo stage ( U-628 . 03 with C-867 controller ) in combination with a linear actuator ( M-231 . 17 with C-863 controller , all Physik Instrumente GmbH and Co . KG ) . Microscope operation was done using Micro Manager [49] , running on a Super Micro 7047GR-TF Server , with 12 Core Intel Xeon 2 . 5 GHz , 64-GB PC3 RAM , and hardware Raid 0 with 7 2 . 0 TB SATA drives . Four views with a rotational offset of 90° were recorded at 1 μm sectioning . Fusion of individual views is based on a diagnostic specimen used to determine an initial guess for an affine transformation that maps each view into a common frame of reference . Together with the data , this initial guess was used with a rigid image registration algorithm [50] to fuse individual views and resulted in isotropic resolution of 0 . 26 μm in the final registered image . ImSAnE [51] was used to obtain tissue cartographic projections showing global maps of the embryo surface of interest . Digoxigenin labeled antisense RNA probes were synthesized from Gold Collection ( Berkeley Drosophila Genome Project ) clones containing full length versions of the vn or rho-1 cDNAs ( clone number LP21849 and LD06131 , respectively ) . Vectors were linearized with the EcoRI restriction enzyme and used as a template for in vitro transcription as per manufacturer instructions ( Roche , SP6/T7 DIG RNA-labeling kit ) . Fixed embryos were bathed in 3% hydrogen peroxide in methanol for 15 minutes to remove endogenous peroxidase activity and rinsed extensively in PBT . Embryos were then incubated for 3 minutes in 10 μg/ml proteinase K in PBT and washed in 2 mg/ml Glycine in PBT . Samples were next washed in a hybridization buffer before the relevant probe was added , and the samples incubated overnight at 55°C . The next day , the embryos were washed once more in PBT and incubated overnight at 4°C with a sheep anti-DIG antibody diluted at 1:2 , 000 . On the third day , embryos were washed and incubated with a biotinylated anti-sheep secondary antibody , and the final signal was developed using a Tyramide-FITC signal amplification kit ( Thermo Fisher Scientific ) . Standard immunostaining was then performed on these samples , using the protocol outlined above . To measure fluorescence intensity along the A/P axis , rectangular regions of interest with a fixed height of 75 μm were drawn over four contiguous segments . In each instance , efforts were made to ensure the same embryonic regions were analyzed between samples . GFP ( from hidΔ . GFP ) , dpERK , and Engrailed intensity profiles were generated across the regions of interest in Fiji [52] , and these data were imported into MATLAB ( Mathworks ) for further analysis . GFP and dpERK intensity profiles were fitted to the peaks of engrailed expression ( i . e . , segment boundaries ) along the A/P axis to normalize data to segment width , which can vary between samples . This was achieved by creating 50 evenly spaced intervals along the A/P axis of each segment before extracting the raw data values and averaging the resulting numbers across all samples analyzed . standard error of the mean was calculated for each point , and data was plotted using the boundedline . m function developed by Kelly Kearney and downloaded from the MATLAB stack exchange . To measure the frequency of apoptosis along the segmental A/P axis , the Engrailed , and cleaved Dcp1 channels were separated from the raw mSPIM image files and then segmented using the open source iLastik software ( University of Heidelberg ) . Segmented images were imported into ImageJ and reduced to single pixel lines and points using the software’s inbuilt skeletonization algorithms . From these skeletons , individual segment boundaries were manually classified and masks were generated that covered each of the segments to be analyzed . Processed files were then transferred to MATLAB for further analysis . Distances were calculated from each Dcp-1 positive pixel to the nearest anterior and posterior Engrailed positive pixel using a k-nearest neighbors algorithm , and these measurements were used to determine the position of each apoptotic cell along the segmental A/P axis as a percentage of segment width . Cells located in the immediate vicinity of the ventral midline ( defined as 20% of linear Dorsal to Ventral distance ) were excluded from the analysis to avoid apoptotic figures in the developing nervous system . The retained data were collated and tabulated as a histogram . | In many tissues , defective cells are eliminated by a process called apoptosis . This process prevents the emergence of rogue cells , which could be detrimental to normal physiology . Apoptosis is particularly apparent in developing embryos that lack appropriate positional information , and it has been suggested that in the absence of clear positional instructions , cells are unable to acquire a defined fate and commit suicide as a result . Here , we have used mutant fruit fly embryos lacking essential segmental determinants to identify the molecular signals that trigger apoptosis in response to mispatterning . We found that cells do not trigger apoptosis in response to conflicting fate determinants . Instead , mispatterning disrupts a tissue size control system that removes excess cells in oversized segments . Specifically , correct patterning information leads to the segmentally repeated production of survival signals , which activate the epidermal growth factor receptor , and this system is disrupted in patterning mutants leading to reproducible patterns of apoptosis . We propose that a similar , though less obvious , process also occurs in normal embryos . In such embryos , each segment would initially comprise a slight excess of cells and would then be trimmed down to a size specified by the pattern of survival signal production and the range of these signals . We suggest that a similar regulatory logic could ensure the coordination of tissue patterning and size in a variety of developing tissues . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"skin",
"cell",
"death",
"invertebrates",
"medicine",
"and",
"health",
"sciences",
"integumentary",
"system",
"morphogenic",
"segmentation",
"cell",
"processes",
"egfr",
"signaling",
"green",
"fluorescent",
"protein",
"animals",
"endocrine",
"physiology",
"animal",
"models",
"developmental",
"biology",
"luminescent",
"proteins",
"model",
"organisms",
"drosophila",
"melanogaster",
"experimental",
"organism",
"systems",
"growth",
"factors",
"epidermis",
"embryos",
"morphogenesis",
"drosophila",
"epidermal",
"growth",
"factor",
"research",
"and",
"analysis",
"methods",
"embryology",
"animal",
"studies",
"proteins",
"endocrinology",
"short",
"reports",
"insects",
"arthropoda",
"biochemistry",
"signal",
"transduction",
"eukaryota",
"cell",
"biology",
"anatomy",
"apoptosis",
"physiology",
"biology",
"and",
"life",
"sciences",
"cell",
"signaling",
"organisms"
] | 2018 | EGFR signaling coordinates patterning with cell survival during Drosophila epidermal development |
French Guiana is a territory that has a decades-long history of dengue outbreaks and more recently , in 2014 , a chikungunya outbreak . Zika virus ( ZIKV ) emerged in late 2015 and subsequently led to an important outbreak . A cross-sectional phone survey was conducted among the general population during the outbreak in June 2016 with a total of 1 , 129 individuals interviewed to assess perceptions , knowledge and behaviors regarding zika infection . The population seemed aware of zika , and perceived the infection as a more serious health threat than other common mosquito-borne diseases . Furthermore , both the perceptions and behaviors related to zika and its prevention were found to vary considerably among different social groups , geographic areas and gender; less educated female participants were found to perceive the disease as more worrisome and were less likely to adopt protective behaviors . Moreover , female population has been particularly responsive to awareness campaigns and rapidly understood the extent of risks associated with ZIKV infection . These results revealed that ZIKV appeared at the time of the survey as a new health threat that concerns the public more than chikungunya and dengue fever with differences observed among subgroups of population . These results have implications for the development of multifaceted infection control programs , including strategies for prevention and awareness , helping the population to develop an accurate perception of the threat they are facing and encouraging behavior changes .
French Guiana , an overseas region and department of France located in the Amazonian forest complex , is composed of two main inhabited geographical regions: a central , urbanized area including a coastal strip along the Atlantic Ocean , where a large part of the population lives , and a more remote area along the Surinamese and Brazilian frontiers called the “interior area” ( Fig 1 ) . In French Guiana , Ae . aegypti has been responsible for several major dengue fever outbreaks over the past few decades and a recent outbreak of chikungunya [17 , 18] . The emergence of ZIKV has been considered of particular concern because the territory has the highest fertility rate in the Americas ( 3 . 5 children per woman ) , with an infant mortality rate ( 1 . 2% ) that is three times higher than in metropolitan France ( 0 . 4% ) [19] . On the 22nd of January in 2016 , local health authorities launched an official epidemic alert following the rapid spread of ZIKV in the most inhabited part of the territory [20] . The public health response to the outbreak of ZIKV disease included distribution of information about the importance of mosquito bite prevention and the use of physician services for pregnant women with potential symptoms , as well as recommendations by health authorities to delay pregnancy . A massive media campaign was conducted during the outbreak including television and radio spots , prospectus , posters ( displayed on the roads and in medical centers ) and also newspapers . Messages were about how to prevent mosquito bites or with all zika symptoms listed and there was also specific prospectus for pregnant women . Moreover , all pregnant women were invited to be carefully monitored with a blood sample collected at each trimester of pregnancy then analyzed at the Arbovirus Reference Center at Pasteur Institute of French Guiana [21] .
A cross-sectional phone survey about “beliefs , attitudes and practices” among the general population of French Guiana was conducted from June 15–30 in 2016 . Eligibility criteria included ( i ) having a landline or mobile phone , ( ii ) being at least 18 years old and ( iii ) indicating consent for participation . The sample design was based on a random 2-stage selection procedure , stratified according to the type of phone ( 50% mobile and 50% land line ) , municipalities and age . Households were randomly selected at the first stage then one participant was randomly selected among adults living in the selected household . The sample calculation included investigation of 1 , 100 individuals . The survey was conducted by Ipsos , a French consulting firm that used the CATI system ( Computer Assisted Telephone Interviews ) and CONVERSO software . To reach the largest possible portion of the population , interviews were carried out from Monday to Friday between 10 am and 8 pm and on Saturday between 10 am and 5 pm . According to the French legislation , the survey protocol and processing of data collection was subject to a declaration to CNIL , the French National Agency responsible for ethical issues and protection of individual data collection under no . 2043940 . All the information was collected anonymously , and the participants were informed of their rights to access and rectify personal information . Data were recorded using a standardized questionnaire administered by phone . Aside from socio-demographic variables , collected data were grouped into four general categories: 1 ) environmental variables and exposure to mosquito , 2 ) perceptions of the illness and risk of contracting ZIKV infection , 3 ) perceptions and practices of protective behaviors promoted by the public health authorities to control the spread of the disease and 4 ) self-reported frequency of protective behavior in response to the zika epidemic . The questionnaire covered of a wide range of topics such as the type of housing , the presence or absence of potential breeding sites , and potential factors associated with breeding sites . Respondents were also asked how frequently they were bitten by mosquitoes ( response options: ‘Never’ , ‘Seldom’ , ‘Sometimes’ or ‘Often’ ) . Participants were also asked whether they were educated about “Aedes mosquitoes” , how frequently they practiced outdoor activities , and during which time of the day mosquito bites occurred . Participants were asked to report the occurrence of an acute febrile illness consistent with presumptive ZIKV infection during the outbreak , and if they had dengue and/or chikungunya virus infection . If the answer was “YES” , we asked them if they had consulted a doctor . A broad range of personal beliefs was investigated , especially regarding perceptions of the health threat , i . e . , qualitative judgments ( based on closed-end questions using unordered response options ) , and quantitative judgments ( based on questions using a Likert response scale with a numerical value ranging from 0 to 10 ) that individuals expressed when asked to evaluate a specific illness and the risk of contracting it [22] . To characterize these perceptions within the population , questions were drawn from the existing methodological literature using the Brief Illness Perception Questionnaire ( B-IPQ ) [23] . This questionnaire measures six components: the identity- the symptoms that patients associate with the illness; the cause- the personal ideas about etiology; the timeline- the perceived duration of the illness; the consequences- the expected effects and outcome; and the treatment control- the effectiveness of treatment methods for recovery from the illness; and the perceived coherence- whether people think the threat is easy to understand . Complementary questions were adapted from the risk perception literature devoted to transmissible infectious diseases . Worry , perceived severity , perceived exposure , and perceived susceptibility to the disease were also assessed to explore the multiple components of risk perception [24] . Except for the perceived cause and identity of the disease , respondents were asked to use an 11-point Likert response scale ranging from 0 to 10 to characterize their mental and emotional representations of the health threat associated with a variety of vector-borne diseases . In line with the health beliefs model [25] , participants were asked how often they practiced health-protective behaviors as they relate to various vector control methods ( e . g . , wearing long-sleeved clothing or using repellent ) and vector control measures ( e . g . , covering water receptacles ) . Participants were then asked whether those behavioral recommendations were effective in preventing mosquito bites ( response options: ‘Very ineffective’ , ‘Somewhat ineffective’ , ‘Somewhat effective’ , ‘Very effective’ and ‘Unsure’ ) . To obtain a mean protection score for each respondent , a level of protection score was created as the sum of each mean used . Statistical analyses were performed using STATA12 software ( Stata Corp . , College Station , TX , USA ) [26] and SPAD8 [27] . Primary sampling units and strata were considered for calculating estimations according to the design effect . Post-stratification weights were used to correct potential biases due to misrepresentation of demographic characteristics , including gender and education level . All estimations were obtained using STATA12 “svy” commands . Bivariate analyses were conducted using Chi-square tests to compare proportions , and linear combinations tests and regression combined with Wald tests were used to compare mean scores of risk perceptions . The level of statistical significance was set to ( p = 0 . 05 ) . Data were adjusted according to gender and level of education . In addition , a multiple correspondence analysis and a hierarchical cluster analysis were performed to determine the natural groupings of observations regarding the level of knowledge about zika disease and its issues in order to cluster the population in different groups according to their literacy . South America and French Guiana layers were drawn using data from OpenStreetMaps ( http://www . openstreetmap . org ) and mapping operations have been done using QGIS 2 . 18 software [28] .
Over half of the participants reported having knowledge of Aedes mosquitoes ( 59 . 1% , 95% Confidence Interval ( CI ) : 54 . 8–63 . 3 ) , and 87 . 9% ( 95% CI: 84 . 7–90 . 5 ) properly identified zika as a vector-borne disease . The most commonly mentioned symptoms associated with zika disease were fever and myalgia ( 91% , 95% CI: 87 . 9–93 . 4 and 60% , 95% CI: 55 . 7–64 . 2 , respectively ) , followed by headache and arthralgia ( 49 . 2% , 95% CI: 44 . 9–53 . 4 and 44 . 8% , 95% CI: 40 . 6–49 , respectively ) . Symptoms mentioned the least frequently were conjunctivitis ( 8 . 5% , 95% CI: 6 . 4–11 ) and neurologic disorders ( 1 . 6% , 95% CI: 0 . 9–2 . 8 ) . Most participants were aware of zika transmission issues: 79 . 3% ( 95% CI: 75 . 6–82 . 6 ) knew that ZIKV can be transmitted from mother to child , 55 . 7% ( 95% CI: 51 . 5–59 . 9 ) knew that ZIKV can be sexually transmitted . Finally , 54 . 4% ( 95% CI: 50 . 1–58 . 6 ) declared to be well-informed about zika disease . However , 17 . 3% ( 95% CI: 14 . -20 . 9 ) reported that there was a vaccine against zika . The most popular sources of information about the disease were television , radio and posters ( 85% , 95% CI: 81 . 4–87 . 9 , 66% , 95% CI: 61 . 9–70 and 63 . 1% , 95% CI: 58 . 8–67 . 2 , respectively ) . The proportions of respondents claiming to have been previously infected with zika , chikungunya or dengue were of 8 . 8% ( 95% CI: 6 . 8–11 . 5 ) , 14 . 3% ( 95% CI: 11 . 4–17 ) and 20 . 7% ( 95% CI: 17 . 7–24 . 1 ) , respectively . Among those who reported sudden onset of high fever during the zika outbreak ( 10 . 5% , 95% CI: 88–17 ) , 83 . 8% claimed to have seen a doctor . The WHO recommended a number of IVM actions for individuals to prevent mosquito bites and thus the spread of the virus , thereby reducing their own personal risk . Among respondents who declared being bitten by mosquitoes either often or every day , 40 . 5% ( 95% CI: 36 . 4–44 . 6 ) and 66 . 5% ( 95% CI: 60 . 1–72 . 3 ) , respectively reported taking preventive measures . When asked about the effectiveness of several preventative measures , the most commonly reported were emptying of water from receptacles , use of bed nets , and covering storage containers , whereas the less effective measures were extensive insecticide spraying and the use of a vaporizer for outdoor insecticides . The most frequently reported preventive measures were closing the door and reducing outdoor activities ( 47 . 7% , 95% CI: 43 . 4–52 ) and using window nets ( 32 . 43% , 95% CI: 28 . 7–36 . 4 ) . Overall , almost 20% ( 95% CI: 15 . 7–22 . 1 ) of the population took at least 5 actions to prevent mosquito bites . As shown in Fig 2 , zika received a significantly higher mean score than did dengue and chikungunya as a response to questions about worry , perceived severity , perceived consequences and perceived exposure ( p<0 . 01 ) . Regarding zika treatment , its perceived efficiency was significantly lower than dengue and chikungunya treatments ( p<0 . 01 ) . Moreover , zika appeared to be less understood than chikungunya disease ( p<0 . 01 ) . Nonetheless , ZIKV infection was perceived as more easily avoidable than dengue ( p = <0 . 01 ) . The distribution of risk perceptions and behaviors among several subpopulations are presented in Table 1 . Analysis revealed an association between age , perceived worry , level of understanding , and behaviors , both with a gradient in responses . Respondents between 18 and 25 years old were more worried about zika ( p = 0 . 016 ) , more likely to adopt protective measures ( p<0 . 01 ) , and they claimed to understand the disease less compared to claims of older age groups ( p = 0 . 047 ) . Individuals living in the interior were less worried about zika than those living in the coastal area ( p<0 . 01 ) and were more likely to adopt protective measures ( p = 0 . 019 ) . Our results show that previous infection with dengue , chikungunya or zika virus was associated with different perceptions of zika disease . Specifically , individuals reporting previous infection judged the disease less severely ( p = 0 . 05 ) , worried less ( p = 0 . 01 ) , felt more exposed ( p = 0 . 017 ) , had lower estimates for control ( p<0 . 01 ) and treatment efficiency ( p<0 . 01 ) , and felt more vulnerable in the context of outbreak ( p = 0 . 049 ) . Finally , respondents with a high level of education characterized the disease as being less severe ( p<0 . 001 ) , as affecting patients to a lesser extent ( p<0 . 001 ) , and as being less avoidable ( p = 0 . 026 ) ; highly educated participants also better understood the disease ( p<0 . 01 ) and judged the treatment as less efficient than respondents with a lower level of education ( p<0 . 001 ) . As shown in Table 1 , women were more afraid of zika than were men . Women were significantly more worried about zika ( p<0 . 001 ) , felt more exposed ( p<0 . 001 ) , and characterized the disease as more severe ( p<0 . 001 ) and as affecting the patient more than did men ( p<0 . 001 ) . Although scores were higher for women in terms of understanding of the disease and control and treatment efficacy , these differences were not statistically significant . Interestingly , women were worried about disease independently of the level of knowledge , whereas among men , there was an association between knowledge and risk perception ( p = 0 . 03 ) ; 65% of men with strong knowledge of the disease were worried vs . 50% among those who were not knowledgeable . When analyzed according to specific subgroups , women presented nuanced responses with regard to the level of education ( Table 2 ) . The impact of education on perceptions was higher among women than men . Low-educated women worried more about the disease ( p = 0 . 02 ) and were less knowledgeable ( p<0 . 001 ) . Moreover , the level of education was also associated with the adoption of protective behaviors ( p<0 . 01 ) . Even if the most popular sources of information ( television , radio and posters ) were reported by the major part of respondents , some variations were observed . Participants aged between 18 and 25 years were more likely to access to social network ( p = 0 . 01 ) and highly educated participants were more likely to access poster ( p = 0 . 03 ) . Participants who had heard about zika from television ( p = 0 . 010 ) , radio ( p<0 . 001 ) , poster campaign ( p<0 . 001 ) , social network ( p = 0 . 035 ) and from leaflets ( p<0 . 001 ) reported to be significantly more informed about zika disease . However , only participants having been aware through leaflets ( p = 0 . 023 ) were more knowledgeable about zika disease and its issues than those who did not . Men and women were equally split on the issue of leaving French Guiana in case of pregnancy during a zika outbreak , with 52% and 49% claiming they should have left the territory . Both men and women seemed to be compliant with the recommendations of the WHO , with 85% claiming that they would use a condom during sexual activity . However , approximately 65% of women claimed they would postpone a pregnancy in case of zika outbreak , whereas only 51% of men claimed they would postpone pregnancy . Women indicated that they would worry about undergoing pregnancy during a zika outbreak ( 75% ) .
The results of our survey indicated that the public perceived zika disease as a more serious potential health threat than other common mosquito-borne diseases , even though a range of elements within cognitive representations of zika were statistically and graphically found to be anchored on those of chikungunya . The survey helped to identify a subgroup of population shaped by specific risk perceptions and behaviors that deserves further attention given the importance of the public understanding and mental representation of illnesses in the adoption of effective protective behaviors . If this assessment seems difficult to conduct in an epidemic setting , high-risk groups identified may be targeted as a priority in case of a new emergence . | Although dengue fever has been a focus of many awareness campaigns in Latin America , very little information is available about beliefs , attitudes and behaviors regarding vector-borne diseases among the population of French Guiana . Following the end of the first chikungunya outbreak and at the initial onset of the first zika outbreak , a quantitative survey was conducted among 1129 individuals aiming to study the emotional , cognitive and behavioral response to the risk of zika infection and assess variations among different groups of population . People from French Guiana were found to perceive zika substantially differently from other Aedes mosquito-borne diseases . Overall , ZIKV appeared at the time of the survey as a new health threat that makes the population more scared than chikungunya and dengue fever . Furthermore , both the beliefs and behaviors related to zika and its prevention were found to vary considerably among different social groups , gender and geographic areas . Education had an impact on perceptions and behaviors among women . Female population has been particularly responsive to awareness campaigns and rapidly understood the extent of risks associated with ZIKV infection . Overall , findings emphasize the importance of developing appropriate and relevant strategies helping population to engage in protective behaviors adapted to the health threat they are facing . Given the importance of the public response and precautionary actions to control the spread of an emergent threat , additional research on risk perceptions and other behavioral determinants is warranted . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"invertebrates",
"medicine",
"and",
"health",
"sciences",
"zika",
"fever",
"pathology",
"and",
"laboratory",
"medicine",
"chikungunya",
"infection",
"infectious",
"disease",
"epidemiology",
"pathogens",
"tropical",
"diseases",
"microbiology",
"animals",
"viruses",
"rna",
"viruses",
"neglected",
"tropical",
"diseases",
"infectious",
"disease",
"control",
"insect",
"vectors",
"infectious",
"diseases",
"medical",
"microbiology",
"behavior",
"epidemiology",
"microbial",
"pathogens",
"disease",
"vectors",
"insects",
"arthropoda",
"mosquitoes",
"eukaryota",
"flaviviruses",
"viral",
"pathogens",
"biology",
"and",
"life",
"sciences",
"viral",
"diseases",
"species",
"interactions",
"organisms",
"zika",
"virus"
] | 2018 | Emerging trends of Zika apprehension in an epidemic setting |
The rate at which a cytotoxic T lymphocyte ( CTL ) can survey for infected cells is a key ingredient of models of vertebrate immune responses to intracellular pathogens . Estimates have been obtained using in vivo cytotoxicity assays in which peptide-pulsed splenocytes are killed by CTL in the spleens of immunised mice . However the spleen is a heterogeneous environment and splenocytes comprise multiple cell types . Are some cell types intrinsically more susceptible to lysis than others ? Quantitatively , what impacts are made by the spatial distribution of targets and effectors , and the level of peptide-MHC on the target cell surface ? To address these questions we revisited the splenocyte killing assay , using CTL specific for an epitope of influenza virus . We found that at the cell population level T cell targets were killed more rapidly than B cells . Using modeling , quantitative imaging and in vitro killing assays we conclude that this difference in vivo likely reflects different migratory patterns of targets within the spleen and a heterogeneous distribution of CTL , with no detectable difference in the intrinsic susceptibilities of the two populations to lysis . Modeling of the stages involved in the detection and killing of peptide-pulsed targets in vitro revealed that peptide dose influenced the ability of CTL to form conjugates with targets but had no detectable effect on the probability that conjugation resulted in lysis , and that T cell targets took longer to lyse than B cells . We also infer that incomplete killing in vivo of cells pulsed with low doses of peptide may be due to a combination of heterogeneity in peptide uptake and the dissociation , but not internalisation , of peptide-MHC complexes . Our analyses demonstrate how population-averaged parameters in models of immune responses can be dissected to account for both spatial and cellular heterogeneity .
Cytotoxic T lymphocytes ( CTL ) prevent the spread of intracellular pathogens through T cell receptor ( TCR ) recognition of pathogen-derived peptides presented on MHC class I molecules on the surface of infected cells . CTL may have several modes of action but their canonically understood role is to kill cells recognised as infected , either through delivery of lytic mediators through the target cell membrane or engaging ligands on the cell surface that induce apoptosis . Quantifying the kinetics of CTL killing has been of interest for many years [1]–[19] ( see ref . [20] for a review ) and is important for at least two reasons . First , knowledge of the rate at which individual CTL can survey and kill cells allows us to derive estimates of the numbers or tissue densities of CTL required to contain an infection . Second , developing tools to measure the kinetics of the different processes involved in lytic activity ( locating cells , forming stable conjugates , lysing the infected cell and dissociating from it ) may help us to understand how ineffective or exhausted CTL are functionally impaired or to identify bottlenecks in the lytic process that may be potential targets for augmenting CTL responses . Early studies of CTL-target dynamics were performed almost exclusively in vitro but more recently there has been some focus on data from splenic killing assays , using variants and generalizations of the experimental and modeling approach taken by Barchet et al . [21] and Regoes et al . [11] . There , mice are challenged with a pathogen and following the clearance of infection , a mixture of isogenic splenocytes either pulsed with pathogen-derived peptides or left as unpulsed controls is injected intravenously . Proportions of both populations accumulate in the spleen , where the peptide-pulsed cells can be killed by resident epitope-specific CTL . Models of the kinetics of the transferred cell populations in the spleens in the hours following transfer have yielded estimates of the rate at which single spleen-resident CTL are able to survey and kill . These models assume that CTL and targets are interacting in a well-mixed environment and have provided reasonable descriptions of the data with the assumption that peptide-pulsed targets are lost with first-order kinetics . When CTL are present in excess – that is , at high effector:target ( E:T ) ratios – one can assume that the total rate at which targets are killed is not limited by the time each CTL takes to lyse its target [22] . The simplest model of CTL activity then assumes that the per-capita rate of loss of targets is , where is a measure of CTL density or numbers in the spleen . The units and magnitude of dictate the interpretation of the constant , but if is measured as a proportion of all surveyable cells in the spleen , is the rate at which one CTL can move between cells of any type , multiplied by the probability of lysis upon engagement with a peptide-pulsed or infected cell ( Text S1 , section A ) . We term the ‘effective surveillance rate’ . If killing is assumed to occur with 100% efficiency , is simply the rate of CTL surveillance , and has been estimated to be in the range 1–35 cells per minute in a variety of experimental infection systems [11] , [12] , [16] , [18] . In simple models of well-mixed CTL and targets , knowledge of this parameter , the time taken for CTL to lyse infected cells , and the uncontrolled pathogen growth rate are sufficient to calculate the critical CTL density required to control an infection [22] . These studies took a rather coarse-grained view of CTL-target dynamics that may mask several potential sources of heterogeneity and CTL biology . First , the assays are performed with mixed splenocyte populations but these models assumed all cells are detected and killed with equal efficiencies . Are some cell populations intrinsically easier to kill than others ? Second , although CTL have been shown to be able to respond in a dose-dependent manner to very low levels of peptide-MHC ( pMHC ) ligands on a cell surface [23] , [24] , it is not clear whether this susceptibility varies across cell types , or at what stage in the killing process any effect of pMHC availability is manifest . Third , the models assume the spleen to be a single compartment , but it is a heterogeneous environment with areas enriched for T cells , B cells and red blood cells , raising the possibility that cell populations are not well-mixed and cells of different types may be exposed to different CTL densities . To explore these issues we used a combination of in vivo and in vitro killing assays to examine the influence of target cell heterogeneity , peptide dose and spatial heterogeneity on the kinetics of CTL activity in vivo .
We revisited the splenic killing assay in the setting of influenza infection ( Figure 1 ) . The experimental system is detailed in Materials and Methods , but briefly , TCR transgenic F5 T cells specific for the NP68 influenza epitope were transferred to congenic mice and challenged with systemic administration of live influenza virus . Our calculations depended on being able to enumerate the CTL capable of participating in lysis of pulsed targets . Tetramer staining one week after challenge revealed that transferred F5 cells outnumbered endogenous CTL specific for the NP68 influenza epitope roughly 500-fold in the spleen ( Text S1 , section B ) . We therefore assumed that F5 numbers were a reasonable approximation to the number of antigen-specific effector cells in the spleen . To assess the cytotoxic activity of these CTL , total splenocytes from donors were pulsed with four different doses of peptide , with each dose associated with a different level of cell dye , then transferred intravenously to influenza-immunised hosts . T and B cell targets were identified by the expression of either TCR or B220 , yielding 8 different target cell populations . The original model of the kinetics of transferred cells in blood and spleen [11] assumed pulsed and unpulsed cells flow from blood to spleen in the same ratio as they are present in the inoculum , and that transferred , unpulsed cells are not lost from the spleen after entry over the course of the assay . We found direct evidence challenging both of these assumptions ( Text S1 , section C ) , so as a starting point we extended the model to allow for ( 1 ) a rate of enrichment of unpulsed cells relative to pulsed cells in the blood over time , and ( 2 ) a rate of loss of the transferred cell populations from the spleen through egress and/or death due to non-CTL-related mechanisms . The basic model is then represented with the following: ( 1 ) ( 2 ) ( 3 ) ( 4 ) where in equations 3 and 4 we have inserted the solutions for the time-dependent densities of transferred cells in blood ( equations 1 and 2 ) . All populations are assumed to enter the spleen from the blood at per-capita rate . Unpulsed and pulsed cells die in the blood or migrate into locations other than the spleen at rates and , respectively . is the per-capita rate of loss of pulsed targets in the spleen due to lysis by CTL , and and ( 0 ) are the initial concentrations of pulsed and unpulsed cells in the blood , respectively . The quantity of interest is the ‘fractional killing’ in the spleen or the extent of loss of pulsed relative to unpulsed cells , for each transferred cell population , in which the quantity corrects for departures from a 1∶1 ratio of pulsed to unpulsed cells in the inoculum ( see Materials and Methods ) . The fractional killing therefore lies in the range [0 , 1] . Using this quantity controls to some extent for variation across animals in the number of cells recovered from the spleen . Because it is dimensionless , units of measurement of cell numbers in blood and spleen do not need to be specified in this calculation . The constant in equations 3 and 4 relates the units of cell numbers in the blood to those in the spleen , and appears linearly in the solutions to equations 1 – 4 . In combination with the spleen influx rate it therefore disappears from the ratio ( Text S1 , section D ) . Before using equations 3 and 4 to estimate the killing rate from the fractional killing , the total rate that unpulsed cells leave the blood ( ) and the splenic loss/egress rate ( ) were estimated by fitting equation 3 to the timecourse of unpulsed cells in the spleen ( Text S1 , section C ) . We chose to measure cell population sizes as proportions of total splenocytes . This measure exhibited a much smaller coefficient of variation than total numbers in the spleen ( Data S1 ) , possibly due to variation in spleen size across animals and associated differences in the total rate of ingress of lymphocytes . The excess rate of loss of pulsed over unpulsed targets from the blood , , was estimated independently from observations of the ratio of pulsed to unpulsed transferred cells in the blood up to 18 hours post-transfer ( Text S1 , section C ) . Using these estimates ( shown in Table 1 ) left only the target cell death rate due to CTL , , to be estimated from the timecourse of the fractional killing , . The further assumption of mass-action at the whole spleen level corresponds to where is the effective surveillance rate and is the measured number of F5 CTL in the spleen as a proportion of total splenocytes . Equations 1-4 predict that the fractional killing should asymptotically approach 1 , simply because the per-capita rates of loss of all populations are constant over time and the loss rate is higher for pulsed than unpulsed cells . However the loss of pulsed targets stopped short of 100% , an effect more pronounced for T than B cells and increasingly apparent at lower peptide doses . To account for this saturation , we explored three extensions of the basic model , illustrated in Figure 2 and detailed in Text S1 , section D . We combine egress and non-specific cell death in the spleen in a single loss rate . Similarly the rate combines both death in the blood and the net rate of extra-splenic sequestration of unpulsed targets . An integrated model of spleen-blood kinetics would allow for a proportion of targets lost from the spleen to re-enter the circulation , a process that would contribute to the enrichment for unpulsed cells in the blood over time . Instead we represent the population dynamics of cells in the blood semi-empirically with equations 1 and 2 , which recapitulate the kinetics of the pulsed:unpulsed ratio in blood and of unpulsed cells in the spleen ( net influx slowing to zero by 6h , followed by a net loss ) . Killing also slows to near zero by 6h ( Figure 3A ) and our estimates of parameters related to CTL activity are insensitive to the splenic loss rate , suggesting it is reasonable to use our description of blood kinetics . This assumption also eliminates the need to estimate the potentially different non-specific death rates in blood and spleen . We fitted these three models to the data , estimating the CTL-mediated killing rate and the additional parameter for each model ( , or ) separately for the two target cell populations . To complete the analysis we found that the addition of a third parameter , a time lag , was required to account for the fact that the fractional killing curves did not extrapolate cleanly to zero at the beginning of the assay ( Figure 3A ) . represents a delay between transfer and the first evidence of death of pulsed cells . It might comprise the mean time taken to migrate from blood into the spleen and across the marginal zone into white pulp where we would expect specific CTL to be resident in the greatest numbers . It may also comprise the handling time , the time taken for a target to be killed following conjugation with a CTL . Because splenic effector:target ratios were high in these assays ( Text S1 , section E ) , few CTL are expected to kill more than once and so the handling time is not expected to limit the total rate of killing , but it may act to delay the appearance of the first apoptotic cells . If the times to complete any or all of these process are non-exponentially distributed , or if more than one process is operating , there may be a discernable delay before killing is observed . We model the effect of migration and/or handling time by setting the killing rate for times and for , where is time-dependent for the decay model , , and a constant for the hidden-target and hybrid models . We found that our estimates of the time lag did not vary significantly across peptide doses and so assumed that it took values specific to the T or B cell populations only . Due to the non-nested nature of the models we compared their abilities to describe the data using the Akaike Information Criterion ( AIC ) [25] . Since the three models contained equal numbers of parameters , the AIC could be calculated simply as n log ( RSS ) where RSS is the residual sum-of-squares and is the number of datapoints . This quantity is the negative of twice the ( maximum ) log-likelihood up to an additive constant in which disappears when comparing models . We found strongest support for the hybrid model and the weakest for the decay model , for both T cell and B cell targets ( Hidden-target vs . hybrid model , for T cells , for B cells; Decay vs hybrid model , for T cells , 17 for B cells ) . The interpretation of these differences is that the relative probability of the hidden-target and not the hybrid model minimising the information lost in describing the data is for T cells and for B cells . However the fits were visually indistinguishable on both the absolute scale and the logit-transformed scale on which fitting was performed , and parameter estimates were comparable . Graphs of fits and parameter estimates for the Hybrid model are shown in Figure 3 and parameter estimates for all three models are shown in Table 1 . Reduced models with no pre-killing lag time yielded substantially inferior fits ( for the hybrid model , for B and T cells respectively ) . Our key observations are that every model indicated that within the susceptible populations B cells were killed significantly more slowly per-capita than T cells at all doses except the lowest , and for both populations the rate of killing of susceptible targets fell with peptide dose ( Figure 3B and Table 1 ) . While we will argue that spleen-averaged estimates may be misleading , to allow comparison with estimates of the effective surveillance rate from other studies and pathogens , we quote here the spleen-averaged where was the average number of F5 cells as a proportion of splenocytes across all animals and all timepoints . The effective surveillance rates of F5 effectors for T cell targets ( with 95% CIs ) ranged from 0 . 14 ( 0 . 12–0 . 16 ) /min at the lowest peptide dose to 0 . 44 ( 0 . 40–0 . 50 ) /min at the highest dose , and from 0 . 15 ( 0 . 13–0 . 17 ) /min to 0 . 27 ( 0 . 24–0 . 30 ) /min at low/high doses for B cells . Interpreting these figures in terms of numbers of targets surveyed per unit time is complicated by the fact that combines both the base cell-cell surveillance rate and the probability that a CTL kills a pulsed target on encounter , , which may be ; so the estimates of at higher peptide doses where one might expect killing efficiency to be greatest [26] are likely closest to the base surveillance rate but still a lower bound . Mempel et al . [27] used intravital imaging to study the lysis of peptide-pulsed B cell targets by specific CTL in a popliteal lymph node , and estimated that CTL take minutes in vivo to survey cells not expressing cognate antigen . Our figures are consistent with this if one assumes that in their system the time for a CTL to migrate between surveyable cells was short . Estimating from the target cell death rate also requires the assumption that all target cell populations are exposed to the same spleen-averaged density . This assumption is questionable , as we see below . Another caveat is that is the rate of killing of susceptible targets only; if any targets are refractory or inaccessible to CTL , then models that assume all targets are susceptible such as in [11] will underestimate . Estimates of derived from these models may then also be lower bounds . Nevertheless the effective surveillance rates in the setting of influenza infection and F5 transgenic CTL are slower than those estimated using LCMV or polyomavirus infections [11] , [12] , [16] , [18] , an issue we return to in the Discussion . Modeling splenic egress and enrichment of unpulsed targets in the blood was required to explain the kinetics of unpulsed targets in the spleen and of the pulsed:unpulsed ratio in the blood , respectively ( Text S1 , section C ) . Including these processes in the models of killing improved their qualities of fit slightly ( for the hybrid model , neglecting splenic egress and blood enrichment increased the AIC by 1 . 6 for B cells and 1 . 9 for T cells ) but had no substantial impact on parameter estimates or the relative support for the three models . This is likely due in part to a disparity in timescales . Our estimates of the time for the pulsed:unpulsed ratio to halve in the blood were 4 . 3 h for T cells and 11 . 4 h for B cells ( , Table 1 ) . In contrast , the total rate of influx of unpulsed cells into the spleen is proportional to in these models and was estimated to halve in 1 . 3 h for T cells and 1 . 5 h for B cells ( , Table 1 ) . Thus , the differential influx of the pulsed and unpulsed cells into the spleen became apparent only after the majority of targets had accumulated . Further , we assumed that the splenic egress or non-specific loss rate applied to pulsed and unpulsed populations equally , and so including this process only weakly influences the timecourse of the pulsed:unpulsed ratio in the spleen . We note that other estimates of times to transit the spleen are lower , and that T cells may egress more rapidly than B cells [28] , [29] , suggesting that killing in the spleen may indeed contribute to the enrichment for unpulsed cells in the blood . The delay before killing , , was comparable for T and B cells at approximately 75 minutes . T and B cells take 20–30 minutes following intravenous transfer to cross the marginal zone of the spleen and enter the white pulp [30] . A handling time between encounter and breakup of the target cell has been observed in many studies . A study of killing of B cells in vivo found CTL took between 9–25 minutes to lyse targets after conjugation [27] . Note that the models predicted an increase in the fractional killing in the spleen during the lag period , most strongly for T cells ( Figure 3A and Table 1 ) . However by neglecting recirculation we predict this is due to the slow but significant enrichent for unpulsed over pulsed T cell targets from the blood into other anatomical locations , and not killing in the spleen itself . To investigate whether we could detect evidence of mass-action killing operating , following ref . [12] we allowed for the base killing rate to be specific to each mouse , , and assumed it was linearly proportional to the density of effector CTL in the spleen of that animal , . Here is the measured number of F5 CTL as a proportion of all splenocytes , and is the effective surveillance rate . We assumed that was constant across animals but took values specific to each target cell type ( T or B cell ) and peptide dose . Effector:target ratios were greater than 3 in these assays ( Text S1 , section E ) , so if CTL and targets were well-mixed in the spleen and moving randomly , mass-action might be expected to hold . However , we found that across all peptide doses and target cell types the models using a simple population-average killing rate described the data significantly better than those with mouse-specific , mass-action killing rates . Indeed we saw no significant positive correlation between fractional killing and F5 CTL , either in total numbers or as a fraction of splenocytes , at any peptide dose or timepoint , after correcting for multiple comparisons ( Text S1 , section F ) . We saw roughly 40% variability across animals in the total number of F5 CTL recovered from the spleens , . However the dependence of the rate of loss of targets on the size of the CTL population might be expected to be an increasing function of their density , rather than of total numbers alone; the rate of encounter of targets with a given number of randomly dispersed CTL would be expected to vary with spleen volume if the populations are well-mixed . F5 CTL numbers as a proportion of all splenocytes were highly consistent across animals and timepoints ( , or variation ) . We conclude that our data do not provide sufficient power to support or refute the mass-action hypothesis at the whole-spleen level , nor do they allow us to quantify any functional dependence of killing rates on CTL densities . Ganusov et al . [19] performed these assays in the context of LCMV infection , using adoptive transfer of specific CTL and varying their number over three orders of magnitude , and did indeed find evidence for a linear dependence of splenocyte death rates on CTL numbers or frequencies , and so in the analyses below we retain the possibility that mass-action operates . Another key observation derived from the hybrid model is that the average rate at which pulsed target cells lost susceptibility increased by an order of magnitude as peptide dose fell from 10−6 M to 10−9 M , most rapidly for T cells . This effect was reflected in the hidden-target model by the susceptible fraction declining with dose , and generally being smaller for T cells than B cells ( Figure 3B and Table 1 ) . These dose-dependencies suggest that heterogeneity in susceptibility derives from properties intrinsic to target cells rather than global effects such as cells migrating into regions in the spleen inaccessible to CTL . In the light of these observations , the following is a mechanism of loss of susceptibility compatible with the hybrid model: Peptide-pulsed targets would be expected to exhibit a unimodal distribution of pMHC densities at each dose , and if peptide is lost this distribution will shift towards lower pMHC densities with time . The lower the dose of peptide , the closer to any threshold of detection the initial distribution will be , and so the average rate of loss of susceptibility across the entire target population ( the rate in the hybrid model ) will increase , as observed . To explore the peptide-loss hypothesis , we quantified the expression and turnover of MHC class I on T and B cells in vitro by blocking transport of MHC class I to the cell surface and observing the kinetics of its internalisation ( see Materials and Methods ) . We found T cells expressed MHC class I on the cell surface at 2–3 fold lower levels than B cells , and MHC was lost approximately twice as rapidly ( Figure 4; half-lives of approximately 11 and 21 h for T and B cells respectively ) . If these peptide-MHC turnover rates in vitro reflect the rate of turnover in vivo , then in our splenic cytotoxicity assays pMHC densities on the T cell targets fall only approximately 2-fold before the bulk of surviving pulsed cells are below a threshold of detection and killing has stopped . This means that for loss of MHC to underlie arrest in the hybrid model , peptide doses covering four orders of magnitude result in four target cell populations clustered closely in pMHC expression just above the limit of detection . Parsimony then suggests that it is unlikely that MHC internalisation alone explains the arrest of killing , as suggested previously [13] . MHC turnover imposes only a lower limit on the rate of loss of visibility of targets , as peptide may dissociate from the MHC class I to which it is bound . While we do not have an estimate of the lifetime of the NP68/H-2Db complex used in our assays , peptide-MHC class I dissociation half-lives of 1–4 h have been reported in other systems [31]–[33] and so loss of peptide from targets remains a potential explanation for loss of susceptibility . However we cannot exclude a contribution from heterogeneity in peptide uptake by targets such that at decreasing peptide doses , an increasing proportion of pulsed targets are already below a threshold of detectability by CTL at the beginning of the assay . The hidden-target model , which describes the data with fidelity comparable to the hybrid model , at least visually if not statistically , is an expression of this heterogeneity in initial conditions . Finally , the dependence of the B cell killing rate on peptide dose was weaker than that of T cells ( Figure 3B and Table 1 ) . One explanation for this is that the higher level of MHC expression on B cells means at higher peptide doses the population lies more completely within a saturating region of the curve relating dose to susceptibility to CTL . All three models indicated that the per-capita rate of killing by CTL was lower for B cell targets than for T cells . This may be because B cells are intrinsically less susceptible to lysis at a given peptide dose , that B and T cells encountered different local densities of CTL , or that the CTL were less motile in B cell areas than in T cell areas . To begin to discriminate between these ( non-exclusive ) possibilities , we used microscopy to quantify the distribution of effector CTL within the spleen ( Figure 5A ) . F5 CTL were indeed distributed heterogeneously , with the majority ( 60% ) in T cell areas , roughly 4-fold fewer ( 13% ) in B cell follicles , and more than a quarter ( 27% ) in red pulp ( Figure 5B ) . Since at least a proportion of the intravenously injected T and B cell targets may migrate to their respective areas in the spleen , it then seems likely that the different target cell populations are exposed to different local densities of CTL . To further explore the relative contributions of this spatial heterogeneity and potential differences in susceptibility to lysis , we used this spatial data together with the T and B cell death rates to estimate the relative ability of F5 CTL to kill peptide-pulsed B and T cells . To separate the effects of CTL numbers and susceptibility we began by assuming that mass-action held within the T and B cell areas and killing of each population was restricted to its relevant area . Targets and CTL might reasonably be assumed to be randomly distributed within each . In the hybrid and hidden-target models the per-capita rate of killing of the susceptible population of peptide-pulsed T cells in the spleen is . If these T cells are restricted to T cell areas and are exposed to specific CTL at local density , this killing rate must be equal to where is the effective surveillance rate in T cell areas . So , and similarly for B cells . Note that this calculation does not depend on the relative volumes of B and T cell regions in the spleen , which we estimated to be ( mean s . e . m . ) by raising the ratio of their areas in the imaged sections to the power . Different degrees of crowding of T and B cell targets in the spleen are represented by different values of the parameter , which relates density units in blood and spleen but disappears from the estimation of for each population . Then ( 5 ) Here and are decomposed into the probabilities of lysis following encounter with a CTL , and , and the base CTL surveillance rates in B and T cell areas , and . In this calculation we have replaced the local densities of CTL in each of the T and B cell areas ( as fractions of total splenocytes within each region ) with and , the measured local CTL densities in units of cells per of spleen section . Because these sections were of the order one cell width deep , the CTL densities measured per unit volume are then approximately these cell numbers per unit area divided by the section depth . We assume the densities of total surveyable cells are the same in T and B cell areas , and so . Using the target cell death rates derived from the hybrid model , we estimate that on a per-CTL basis B cells are killed 3-5 times more rapidly than T cell targets , with the lower boundary of the 95% confidence interval lying above 1 , at all peptide doses ( Figure 5C , solid circles ) . The hidden target model yielded very similar estimates ( Figure 5C , open circles ) . In summary , under the assumptions of mass-action and restriction of the populations to their respective areas in the spleen , the difference in local densities of CTL was too large to explain the difference in killing rates of T and B cells pulsed with the same dose of peptide , and so the relative paucity of CTL in B cell areas is compensated to a degree by a higher effective surveillance rate ( ) . This difference in might stem from susceptibility to lysis; for instance , the 2–3 fold difference in MHC expression by B cells might contribute to a higher probability of detection by CTL upon encounter , , at a given peptide dose . It might also arise from differences in the motility of CTL within T and B cell areas , and . To narrow down the possibilities even further , we wanted to estimate the intrinsic susceptibilities of T and B cells to lysis by CTL and assess their dependence on peptide dose , while minimising any effects of spatial heterogeneity or CTL motility . Lysis is a multi-stage process . The CTL must encounter and survey the cell , detect that it bears the relevant peptide , form a stable conjugate , initiate lysis and eventually disengage from the apoptotic cell . We wanted to identify at which stage ( s ) of the killing process any differences between T and B cells or across peptide doses were manifest most strongly . To do this we performed an in vitro cytotoxicity assay using T and B cell targets pulsed as before with different doses of the peptide , and co-localised with F5 CTL activated in vitro ( see Materials and Methods ) . These populations were co-cultured for between 0 and 120 minutes , allowing us to follow the kinetics of free targets ( S ) , the number of CTL-target conjugates in which the CTL had not degranulated ( stained negative for LAMP1a at the cell surface , ) , and the number of conjugates in which the CTL had degranulated ( LAMP1a detected at the cell surface , ) , assumed to indicate lysis . The following generalisation of the hybrid or decay models described the kinetics of these populations well ( Figure 6 ) : ( 6 ) ( 7 ) ( 8 ) The term proportional to is the total rate of formation of conjugates between targets and LAMP1a− CTL . At the beginning of the assay the expected time for a given target to become conjugated is . is the initial rate at which a conjugate dissociates without lysis; is the initial rate at which a CTL in a conjugate becomes LAMP1a+ through degranulation; and is the rate at which a LAMP1a+ conjugate dissociates . CTL were in excess in this assay and so we assumed LAMP1a+ cells did not kill again . With this assumption , LAMP1a+ conjugates arose directly from LAMP1a− conjugates only . Inspection of the data revealed that the formation of conjugates and killing slowed considerably over the course of the assay , appearing to stop completely after roughly an hour ( Figure 7 ) . This was much earlier than the timescale of arrest of killing in vivo ( Figure 3A ) , and seems unlikely to be the result of loss of peptide-MHC from the target cells . We observed a roughly 50% loss of effector CTL numbers over the 2h timecourse , accounting for both free CTL and those in conjugates . We speculate that as well as dying , the CTL became increasingly functionally impaired , perhaps related to the release of cytotoxic factors into the culture medium . To capture this behaviour we allowed rate constants to change with time such intially they reflect the interactions between targets and fully-functional CTL , but by , conjugate formation had stopped , and the efficiency of progression to degranulation was zero; ( 9 ) Parameter estimates are shown in Figure 8 and rate constants are quoted as their inverses ( timescales , in minutes ) . CTL and targets formed conjugates at similar rates ( ) for T and B cells at each peptide dose , but conjugates were slower to form at lower doses ( Figure 8A ) . CTL-B cell conjugates progressed to degranulation ( LAMP1a+ , which we assumed led to lysis ) after roughly 15 minutes , independent of dose , while the mean time to degranulation for CTL-T cell conjugates was roughly 30 minutes slowing to 45 minutes at the lowest peptide dose ( Figure 8B ) . We saw considerable uncertainty in the rate of dissociation without lysis , the failure rate ( Figure 8C ) , but this process was relatively slow and the mean lifetime of LAMP1a− conjugates was determined largely by the time to degranulation ( Figure 8D ) . This led to high ( 50–90% ) efficiencies of lysis , , at the beginning of the assay , but uncertainty in obscured any potential variation in efficiency with peptide dose ( Figure 8E ) . Similarly we detected no significant differences in the rate of change of parameters , indicating that a progressive loss of CTL functionality affected the killing of all cell populations equally ( Figure 8F ) . Lastly , we estimated that degranulated ( LAMP1a+ ) conjugates took between 100 and 200 minutes either to break up or for the target cell to disintegrate ( Figure 8G ) , again with no significant T-B differences . Multiple CTL bound to single targets may shorten the time taken to kill [4] . We saw evidence for formation of conjugates comprising more than two cells at approximately 10% of doublet numbers after one hour , and stable over time ( Text S1 , section G ) . The data were not sufficient to parameterise the dynamics of theses multiples . However , one could reasonably assume that triplets form by a second CTL joining a CTL-target doublet , particularly because CTL were in excess ( Text S1 , section G ) . The estimated pre-lytic doublet loss rate will then comprise both breakup into singlets and formation of LAMP1a− triplets , then quadruplets , etc . , within which targets may or may not have an increased probability of being killed . Therefore , by neglecting multiples we may overestimate the true conjugate breakup rate , and so our efficiency of killing is a lower bound , with an error of 10% or less . Again , CTL are in excess in these assays and so estimates of the doublet conjugate formation rate and the doublet lysis rate will be unchanged . In summary , we found that while the rate of conjugate formation fell with peptide dose , there were no detectable differences in the ability of CTL to conjugate with T or B cells; and while T cells subsequently progressed to lysis more slowly than B cells , there were no detectable differences in the efficiency of lysis across cell type or peptide dose . A similar conclusion regarding the effect of peptide dose on conjugation was reached by Jenkins et al . [34] , who measured the impact of the avidity of TCR-pMHC interations on lysis using transgenic OT-I CTL specific for the OVA257–264 peptide . There , the avidity of TCR interactions , assumed positively correlated with peptide dose , impacted the rate of formation of conjugates but had no significant effect on the proportion of conjugates exhibiting clustering of tyrosine kinases at the contact site , an early indicator of TCR signaling and progression to lysis . In contrast , lytic efficiency was found to vary with dose in an in vitro tissue model of killing of HIV-derived peptide-pulsed targets [26] . Measurements of rates of conjugate formation and lytic efficiency are somewhat definition-dependent and correlated , however . We may be overestimating lytic efficiency and underestimating the rate of conjugate formation since what we define as a conjugate has remained stable for long enough to be detected by flow cytometry .
The in vivo killing assay and the imaging indicated that if killing of T and B cells was restricted to their respective areas in the spleen and rates were locally linear in CTL numbers , differences in local CTL densities were too great to explain the differences in killing rates of the two populations . This suggested that the effect of excess CTL in T cell areas may be partly compensated by more efficient CTL surveillance in B cell areas – either by an increased rate of encounter with cells of all types , or by B cells being killed with a higher probability than T cells following conjugation ( Figure 5C ) . However the rate of formation of conjugates with CTL , and the probability of progression to lysis , were indistinguishable for T and B cells in vitro , and so together these assumptions and observations prompt the conclusion that the rate at which CTL survey cells of any type is higher in B cell follicles than in T cell areas . While that remains to be tested , the assumption of killing of each population being entirely restricted to their respective areas in the white pulp is questionable . Both target cell populations enter the spleen through the circulation and enter the white pulp via the marginal zone . At least a proportion of B cells then migrate through CTL-rich T cell areas en route to B cell follicles [30] . Lymphocytes egressing from the spleen do so by transiting the red pulp [29] , where more than a quarter of splenic F5 CTL resided in our assay ( Figure 5B ) . Bajénoff et al . [30] observed that between 3 h and 8 h after intravenous transfer of isolated splenocytes , B cells were continuing to accumulate in the white pulp from the marginal zone that separates the red and white pulp , roughly a third were resident in B cell follicles , and the remainder were co-localised with T cells . If similar migration patterns and kinetics apply in our assays , the difference in the average CTL densities encountered by the two target cell populations over the assay may be smaller than that inferred simply from the CTL densities in T cell areas and B cell follicles alone . It is possible that this effect alone may account for the differences in T-B killing rates . It is also not critically dependent on the mass-action assumption , requiring only that death rates are increasing functions of CTL density over the conditions found here . While we saw no significant differences in the ability of CTL to conjugate with T or B cells in vitro , or in the probability of conjugation resulting in lysis , CTL-T cell pairs took more than twice as long to either break up or progress to lysis . This difference in handling time will not affect the ability of CTL to control an infection when they are in excess , but will become important at lower E:T ratios when an increasing proportion of CTL will be sequestered in conjugates at any time , and so may become limiting [22] . This difference may make growing populations of infected T cells intrinsically more difficult to control than B cells , in the absence of spatial or peptide-dose effects . Note here we are referring to the rate of progression to lysis following encounter , or single-cell behaviour . This is distinct from the population-level killing rate which depends only the rate at which CTL can encounter and identify targets , and not on the handling time , when CTL are in excess or the mean time to locate the next pulsed target us much longer than the handling time . We predicted a delay of more than an hour before killing of targets within the spleen was evident . However in an LCMV infection model , Barber et al . [35] observed substantial loss of peptide-pulsed cells in the spleen within 15 minutes , relative to target numbers in uninfected control animals . A similarly rapid decline of pulsed relative to unpulsed targets was observed in a polyoma virus infection model [16] , [36] . It is possible that this faster loss derives in part from the systemic nature of those infections , which might lead to greater extra-splenic sequestration or killing of pulsed targets in rapidly perfused organs such as liver and lung . Indeed in an LCMV model , Graw et al . [18] saw a roughly four-fold enrichment for unpulsed transferred cells in the blood by 4 hours , compared to the two-fold enrichment in our assay ( Text S1 , section C ) . However Barber et al . [35] saw that the early loss of pulsed targets in the spleen was attenuated in mice with CTL lacking Perforin , a membrane pore-forming protein involved in the delivery of cytolytic molecules to the target cell . We might expect filtering of pulsed targets from the blood by 15 minutes to be similar in these and WT mice , since it is initially TCR- and not Perforin-dependent . This strongly suggests that lysis was indeed occurring in the spleen within 15 minutes of cell transfer . Estimates of the time CTL take to kill targets have varied widely across systems , from minutes [26] , [27] to hours [37] , and so a discrepancy of this magnitude is perhaps not surprising . F5 CTL may simply take longer to kill; we found mean handling times of 30 minutes or longer for T cell targets in vitro , although handling times with B cell targets were shorter ( Figure 8B ) . The longer delay before killing is apparent may also derive from the time taken for CTL and targets to encounter each other . By day 7 the influenza infection is well controlled and so levels of inflammation are likely lower than in the LCMV system , which we speculate may result in reduced CTL motility; and the spatial distribution of specific CTL that we found in our system may differ from those in LCMV infection models [38] , which might result in differences in the mean time for ingressing targets to enter CTL-rich areas of the spleen . Our results recapitulate previous findings that peptide dose influences susceptibility to lysis by CTL ( see , for example , [16] , [24] , [34] ) . Threshold effects have also been observed . Purbhoo et al . [24] demonstrated a sigmoid relation between peptide dose and the extent of lysis at one timepoint in an in vitro cytotoxicity assay , with a location and steepness that varied with the particular TCR and peptide but over ranges of peptide doses comparable to ours . They showed that as few as two pMHC within the interface between the T cell and its target were sufficient to induce lysis at least in a proportion of contacts , an effect saturating at between 4-200 pMHC , consistent with other studies [39] , [40] . Along similar lines , Henrickson et al . [33] showed in an LCMV model that a sharp threshold of peptide dose given to dendritic cells ( DC ) exists for activation of specific CD8 T cells , corresponding to between 30 and 60 pMHC complexes per DC . If dissociation of peptide from MHC generates the refractory or ‘invisible’ targets in the in vivo assay , these results suggest that these targets have reached very low surface densities of specific pMHC . It is then possible that the greater proportion of refractory cells among the T cell targets derives in part from their 2–3 fold lower levels of MHC class I expression ( Figure 4 ) . Further , the fact that we and others [24] observe incomplete killing even in the populations receiving high doses of peptide suggests heterogeneity in peptide uptake can be substantial . Our work builds on other studies that used the splenic killing assay and exposes different sources of heterogeneity that need to be considered when estimating rates of CTL surveillance . However , the issues that we raise highlight the need for measurements of CTL efficacy performed with live replicating pathogens in relevant tissues , for several reasons . First , there appears to be considerable variation across in vivo cytotoxicity assays in the parameters defining CTL activity , likely deriving from differences in microenvironment , TCR specificity , the mode of CD8 T cell priming and hence effector quality , and target cell susceptibility . Second , both the in vitro and in vivo analyses confirmed earlier findings that peptide dose influences the ability of CTL to detect pulsed targets , but it is not known what peptide doses yield physiologically relevant levels of cognate pMHC on target cells . Third , the influenza infection is well-controlled by the time of the assay 7 days post-challenge; inflammation in the spleen even while killing of peptide-pulsed targets is occurring is presumably low , and so CTL motility and any ability to home to targets may differ between this scenario and one in which an infection is ongoing . Finally , while we have focused on the lytic mode of CTL action , they may also control the spread of intracellular pathogens by non-lytic mechanisms [41] , [42] that will presumably not be manifest in assays using peptide-pulsed targets . The parameters defining how CTL survey and kill infected cells are key elements of models of the within-host dynamics of intracellular pathogens . Deterministic models assuming homogeneous mixing of components of the immune system and infected cells have been used widely and have provided many mechanistic insights into the progression and control of viral infections ( for a review , see for example Ref . [43] ) . While the functional forms of the terms in these models may be appropriate for describing the dynamics of an infection , the parameters they contain are usually compound quantities and may implicitly average over spatial and cellular heterogeneity . Characterising this heterogeneity is important when attempting to make more detailed quantitative statements regarding host-pathogen interactions . For example , a substantial number of CTL in our in vivo assay resided in the red pulp , and would only have been encountered by the proportion of splenocytes that egress from the white pulp over the course of the assay . Depending on the time take to transit the red pulp , it may be that these CTL contribute very little to killing of targets . Estimates of per-CTL killing rates will then be too low if these CTL ( enumerated following the homogenisation of whole spleens ) are assumed to be co-localised with splenocytes only . With increasing availability of in-vivo imaging data , quantitative immunologists will be able to characterise the within-host ecology of infections in more detail , and specifically the critical sizes of effector cell populations needed for immunity .
The UK Home Office Project Licence 80/2506 ( Development and function of innate and adaptive immune responses ) covers all animal experiments conducted at the NIMR . Ly5 . 1 C57BL/6J , Ly5 . 2 C57BL/6J , and F5 . Rag1-/- mice were bred and maintained in a conventional pathogen-free colony at the National Institute for Medical Research , London , UK . All lines were of H-2b haplotype . Animal experiments were performed in accordance with UK Home Office regulations . The following monoclonal antibodies and cell dyes were used: CD45 . 2 PE-Cy7 , CD45 . 1 FITC , TCRβ APC , B220 PE-TexasRed ( all eBioscience ) , H-2Db PE ( BioLegend ) , LiveDead nearIR and CellTrace Violet ( both Invitrogen ) , and H-2Db-ASNENMDAM dextramer-PE ( Immudex ) . Samples were acquired on CyAn ADP ( Dako Cytomation ) , Canto-II ( BD ) or Fortessa X20 ( BD ) flow cytometers , and analysis was performed with FlowJo software ( Treestar ) . Cell culture medium was RPMI supplemented with 10% FCS , 2mM glutamine , 1% penicillin/streptomycin and 50 µM β-mercaptoethanol ( all Sigma ) . Splenocytes from naive Ly5 . 1 or Ly5 . 2 C57BL/6J mice were cultured with NP68 peptide ( influenza NP366-374 , strain A/NT/60/68 , ASNENMDAM , Mimotopes ) at 10−6 M , 10−7 M , 10−8 M , 10−9 M , or in culture medium alone for unpulsed cells , for 2 hours at 37°C . These cells were then labelled with CellTrace Violet ( CTV ) at either 10 µM , 2 . 5 µM , 625 nM , 156 nM or 40 nM , respectively . Following peptide pulse and CTV labelling , target cells were mixed together in equal ratios . Ly5 . 1 C57BL/6J mice were injected IV with 2 million lymph node cells from Ly5 . 2 F5 . Rag1-/- mice and A/NT/60-68 influenza virus , to generate a spleen-resident population of NP68 specific CTL . Seven days later , 10 million Ly5 . 2+ target cells were injected per recipient mouse . At indicated timepoints from 0 . 5–24 hours after injection of targets , mice were sacrificed and spleen and blood were harvested for analysis by flow cytometry . Care was taken in the timing of both injection and sacrifice for individual mice , and organs were harvested directly into ice cold media , to ensure an error of no more than 5 minutes in the reported timepoints . Target and effector cells were distinguished from host cells by expression of Ly5 . 2; CTV fluorescence was used to identify target cells that had been pulsed with different doses of peptide , while effector cells were CTV-unlabelled . Staining for TCRβ and B220 was used to identify T and B cell targets . To generate effector CTL , lymph node cells from Ly5 . 2 F5 . Rag1-/- mice were activated in vitro for three days in the presence of NP68 peptide ( 10−8 M ) . Activated blasts were purified by Ficoll ( GE Healthcare ) density-gradient centrifugation and expanded for a further four days in the presence of 10 nM IL-2 ( Peprotech ) . Ly5 . 1+ target cells were prepared as described above . CTL and target cells were added to wells at an E:T ratio of at least 5∶1 , and briefly centrifuged to initiate cell contact . Cells were co-cultured at 37°C for the indicated period of time ( 10 minutes – 2 hours ) in the presence of anti-LAMP1a ( eBioscience ) to detect degranulation of CTL during the culture period . At the end of the culture period , cells were immediately fixed with IC fixation buffer ( eBioscience ) to preserve E:T conjugates . Samples were then stained and analysed by flow cytometry , with the addition of a known number of AccuCount fluorescent particles ( Spherotech ) to determine cell counts . Target and effector cells were identified by expression of Ly5 . 1 or Ly5 . 2 respectively , and E:T conjugates by dual fluorescence for these markers along with forward scatter area and width characteristics to identify doublets . Staining for TCRβ and B220 was used to identify T and B cell targets , and CTV fluorescence to identify cells that had been pulsed with different doses of peptide . Ly5 . 1 C57BL/6J mice were injected IV with 2 million lymph node cells from Ly5 . 2 F5 . Rag1-/- mice and A/NT/60-68 influenza virus , to generate a spleen-resident population of NP68 specific CTL . Seven days later , at the time when in vivo cytotoxicity assays were performed , mice were sacrificed and spleens harvested for analysis . Each spleen was cut in two , and the weight of each segment recorded . One segment was processed for cell counting and analysis by flow cytometry; the other segment was immediately frozen in liquid nitrogen . Frozen spleen segments were then embedded in OCT compound ( VWR International ) . At least three non-consecutive sections 7 µM thick were cut from each spleen , and stained with antibodies to Ly5 . 2 , IgD and CD4 ( all eBioscience ) . Separate images for each fluorescence channel were collected at 20x magnification on a Leica SP5 confocal microscope , and analysed using ImageJ software ( NIH ) . IgD and CD4 fluorescence was used to manually identify regions of interest corresponding to B cell zones , T cell zones and red pulp , and Ly5 . 2 fluorescence was subsequently used to enumerate CTL within each of these regions . Single cell suspensions were prepared from the spleen of C57BL/6J mice and incubated at 37°C in culture medium for the indicated periods of time in the presence of 5 µg/mL Brefeldin A ( Sigma ) or vehicle control ( DMSO , Sigma ) . Cells were then washed with PBS and stained for TCRβ , B220 and H-2Db for analysis by flow cytometry . Ordinary differential equation models , described in Results , were used to simulate the flux of peptide-pulsed and unpulsed splenocytes from blood into the spleen and the killing of pulsed targets within the spleen . Parameters were estimated separately for T and B cell target populations at each peptide dose by fitting these models to the logit-transformed fractional killing , where and are the numbers of pulsed and unpulsed transferred cell populations recovered from the spleen . The correction factors were close to unity and were the ratio of each peptide-pulsed population to the unpulsed population in the inoculum . Estimates of were obtained from the transfer of targets taken from the prepared splenocyte population into naive animals , and observing the proportions of the different target cell populations as they flowed into the spleen . Closed-form solutions to the models were obtained using Mathematica [44] and fitted to the data using the nls function in [45] . Data were logit-transformed to ensure the normality and heteroscedasticity of the distribution of residuals . Arcsin square root , complementary log-log and probit transforms yielded similar parameter estimates and qualities of fit . All data used in this manuscript are provided as Supporting Information ( Data S1 ) . | Measurements of the rates at which a single cytotoxic T lymphocyte ( CTL ) can survey for infected cells , and kill them upon encounter , are important for constructing predictive models of vertebrate immune responses to intracellular pathogens . The surveillance rate has been estimated previously using combinations of modeling and experiment , making the assumption that CTL and target cells are well-mixed and that all cell types are killed with equal efficiency . In this study we take an iterative approach with theory and experiment to go beyond such models and detail the effects of cellular heterogeneity , the spatial organisation of the tissue within which killing is taking place , and the influence of the level of expression of peptides on the target cell surface . We demonstrate that determining the degree of co-localisation of effector and target cells , and the level of peptide expression on targets , are most important for improving estimates of CTL killing rates . Further , while the probabilities of killing upon conjugation of CTL with T and B cell targets are similar , T cells take substantially longer to kill than B cells , an effect that may be important when CTL numbers are limiting . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"infectious",
"disease",
"immunology",
"theoretical",
"biology",
"clinical",
"immunology",
"population",
"modeling",
"biology",
"and",
"life",
"sciences",
"immunology",
"computational",
"biology",
"immune",
"response"
] | 2014 | Spatial Heterogeneity and Peptide Availability Determine CTL Killing Efficiency In Vivo |
Chromosome segregation errors in human oocytes are the leading cause of birth defects , and the risk of aneuploid pregnancy increases dramatically as women age . Accurate segregation demands that sister chromatid cohesion remain intact for decades in human oocytes , and gradual loss of the original cohesive linkages established in fetal oocytes is proposed to be a major cause of age-dependent segregation errors . Here we demonstrate that maintenance of meiotic cohesion in Drosophila oocytes during prophase I requires an active rejuvenation program , and provide mechanistic insight into the molecular events that underlie rejuvenation . Gal4/UAS inducible knockdown of the cohesion establishment factor Eco after meiotic S phase , but before oocyte maturation , causes premature loss of meiotic cohesion , resulting in destabilization of chiasmata and subsequent missegregation of recombinant homologs . Reduction of individual cohesin subunits or the cohesin loader Nipped B during prophase I leads to similar defects . These data indicate that loading of newly synthesized replacement cohesin rings by Nipped B and establishment of new cohesive linkages by the acetyltransferase Eco must occur during prophase I to maintain cohesion in oocytes . Moreover , we show that rejuvenation of meiotic cohesion does not depend on the programmed induction of meiotic double strand breaks that occurs during early prophase I , and is therefore mechanistically distinct from the DNA damage cohesion re-establishment pathway identified in G2 vegetative yeast cells . Our work provides the first evidence that new cohesive linkages are established in Drosophila oocytes after meiotic S phase , and that these are required for accurate chromosome segregation . If such a pathway also operates in human oocytes , meiotic cohesion defects may become pronounced in a woman's thirties , not because the original cohesive linkages finally give out , but because the rejuvenation program can no longer supply new cohesive linkages at the same rate at which they are lost .
In both mitotic and meiotic cells , sister chromatid cohesion is required for accurate chromosome segregation , and the cohesive linkages that hold sister chromatids together depend on the cohesin complex which forms a DNA-entrapping ring [1] , [2] . In addition to holding sister chromatids together , cohesion plays several additional essential roles during meiosis . The integrity of the synaptonemal complex , a meiosis-specific structure that holds homologs in close proximity during recombination , depends on cohesion proteins and crossovers between homologs are reduced in cells in which cohesion is compromised [3] . In addition , cohesion along the arms of sister chromatids provides an evolutionarily conserved mechanism that keeps recombinant homologs physically associated until anaphase I [4]–[6] . By maintaining chiasmata , arm cohesion promotes proper orientation and microtubule attachments of homologous chromosomes on the metaphase I spindle and is therefore crucial for accurate segregation of homologs during the first meiotic division . The timeline of human oogenesis presents a daunting challenge for the maintenance of meiotic cohesion [7] . Oocytes undergo meiotic DNA replication , establish sister chromatid cohesive linkages and complete meiotic recombination during fetal development . Before birth , oocytes enter a prolonged diplotene arrest ( known as dictyate ) , and resumption of meiosis occurs only as individual oocytes are recruited for ovulation . Because the majority of human oocytes remain arrested for decades , the continued physical association of recombinant homologs and their accurate segregation during meiosis I demands that cohesion along the arms of sister chromatids remain intact during this extended timeframe . Chromosome segregation errors during female meiosis are the leading cause of miscarriages and birth defects in humans [8] . Furthermore , the risk of producing aneuploid gametes increases exponentially as women age . A correlation between advanced maternal age and increased incidence of single chromatids prior to the second meiotic division has been reported for human oocytes obtained from cancer patients and from women undergoing in vitro fertilization [9] , [10] . While the mechanisms underlying the maternal age effect are likely to be complex , work in Drosophila and mice also indicates that meiotic cohesion weakens with age and supports the hypothesis that deterioration of meiotic cohesion plays an important role in age-related segregation errors in human oocytes [6] , [11]–[14] . Based on work in budding yeast , it is widely accepted that under normal conditions , cohesive linkages are only established during S phase [1] , [2] . However , if this were the case for human oocytes , the same cohesin complexes used for cohesion establishment in the human fetal ovary still would be present in adult oocytes years later . On first reflection , this provides a satisfying explanation for why cohesion defects would be more prevalent in the oocytes of older women . However , is it really possible that the same cohesin rings remain intact on meiotic chromosomes for even five years , much less 25 ? An alternate possibility is that maintenance of meiotic cohesion is an active process that utilizes a specialized rejuvenation program to establish new cohesive linkages throughout the extended timeframe of prophase I . Precedence for cohesion establishment outside of S phase comes from work in budding yeast , which has demonstrated that under certain conditions vegetative cells can establish functional cohesive linkages during G2 [15]–[19] . Moreover , in both Drosophila and mouse oocytes , localization of the cohesin loader Nipped-B along the arms of meiotic chromosome during pachytene has been observed [20]–[22] , consistent with the possibility that cohesin complexes are loaded and converted to functional linkages during meiotic prophase . Here we utilize Drosophila to test the hypothesis that cohesion rejuvenation occurs during meiotic prophase . The Drosophila oocyte provides an excellent system to study the maintenance of meiotic cohesion because prophase I lasts approximately six days [23] , and the linear array of oocytes within each of the ovarioles comprising the ovary permits one to monitor chromosome morphology at progressive stages during meiotic prophase . In addition , a simple genetic assay allows us to measure the fidelity of meiotic chromosome segregation . We have used a Gal4/UAS inducible RNAi strategy to ask whether cohesion defects occur if we reduce cohesion regulators or cohesin complex subunits after meiotic cohesion is established normally during meiotic S phase . We find that a rejuvenation program operating during prophase I is necessary to sustain a level of meiotic cohesion that is sufficient for chiasma maintenance and accurate chromosome segregation . Our data support the model that rejuvenation of meiotic cohesion requires Nipped-B-dependent loading of newly synthesized cohesin complexes and Eco-dependent establishment of new cohesive linkages . Furthermore , Eco-mediated cohesion rejuvenation does not depend on induction of double-strand breaks and therefore differs from the damage-induced cohesion re-establishment pathway that operates in yeast cells . We raise the possibility that the rejuvenation pathway we have uncovered in Drosophila oocytes may represent an evolutionarily conserved mechanism to ensure that an adequate number of cohesive linkages remain present during the extended period that metazoan oocytes stay arrested in prophase I .
We reasoned that if it were to occur , rejuvenation of cohesion during the extended prophase I period might utilize factors that are normally required for cohesion establishment . Therefore , we began our analysis by focusing on the cohesion establishment factor Eco , ( also known as Deco , Drosophila Eco1 [24] ) . In yeast , Eco1 acetyltransferase activity is required to establish cohesion during S phase [1] , [2] . In order to ask whether Eco activity is required to keep meiotic cohesion intact after its original establishment , we employed a Gal4/UAS inducible RNAi strategy that allowed us to leave Eco levels and activity unaffected during meiotic S phase , when cohesion is established , and to induce RNAi-mediated decay of Eco transcripts only after cohesive linkages are generated . To accomplish this , we used the mat-α-tubulin-Gal4-VP16 driver ( hereafter abbreviated matα driver ) which has previously been shown to begin expression during mid-prophase [25] . Using a UASp-Actin-GFP reporter , we verified that the matα driver is not active until after meiotic DNA replication and would therefore allow us to induce Eco knockdown after normal cohesion establishment ( Figure S1 ) . We used the matα driver to induce expression of a UAS-Eco RNAi hairpin transgene ( VDRC . Eco . 35982 , Table S1 , hereafter referred to as Eco RNAiGD ) and performed single molecule FISH [26] to quantify the number of Eco transcripts in control and Eco knockdown ( KD ) germline cysts at different stages during oogenesis ( Figure S2 ) . In germarial region 3 , the number of Eco transcripts per area for control and Eco KD was the same ( p = 0 . 95 , Figure S2B ) , confirming that Eco knockdown does not commence until after meiotic S phase . We are confident that our assay is sensitive enough to detect a change in the number of Eco transcripts in the germarium because we observed a significant reduction ( ∼14% ) in region 3 for the viable allelic combination eco1/eco2 compared to the control ( p = 0 . 018 , Fig S2B ) . At later stages , when matα driver expression is more robust , Eco transcripts were significantly reduced in Eco RNAiGD egg chambers , confirming that matα driving Eco RNAiGD reduces Eco transcripts in germline cells , but only after cohesive linkages are established during meiotic S phase . To begin to explore whether Eco activity is required to maintain meiotic cohesion , we investigated whether reduction of Eco during meiotic prophase impacts the integrity of the synaptonemal complex ( SC ) , a tripartite proteinacious structure that holds homologous chromosomes in close proximity during the process of meiotic recombination [27] . Mutations in cohesion proteins have been shown to disrupt the formation and/or maintenance of the SC [28]–[35] . In wild-type Drosophila oocytes , full-length SC forms in region 2A of the germarium and remains intact until stage 6 [36] , [37] . To monitor SC stability in Eco KD oocytes , we stained for the SC transverse filament protein C ( 3 ) G [36] , [38] . In control ovarioles ( Eco RNAiGD transgene , no driver ) , we observed long continuous C ( 3 ) G threads until stage 6 ( Figure 1 ) when normal SC disassembly occurs at the end of pachytene . However , when Eco is knocked down in mid-prophase I ( matα driving Eco RNAiGD transgene ) , we observed a number of SC defects . Figure 1 shows representative images of the SC in control and Eco KD oocytes at different stages , as well as quantification of the defects we observed . In order to quantify the severity of the defects , we utilized three categories ( broken threads , short threads , or spots ) to describe oocytes that lacked normal , continuous C ( 3 ) G threads . These categories represent the range of defects we observed , with spots corresponding to the most severe disruption of the SC . When Eco is knocked down during mid-prophase I , the SC appears normal in all germarial stages ( regions 2A , 2B and 3 ) , but premature disassembly is visible beginning at stage 2 ( Figure 1 , Eco KD ) , after expression of the matα driver begins . The majority of stage 2 oocytes contain continuous SC , but minor defects ( broken threads ) are apparent in approximately 35% of the Eco KD oocytes . By stage 3 , the majority of oocytes exhibit SC defects; the percentage of oocytes with broken threads increases to 40% , while 25% display a more severe phenotype ( short threads ) . As oocytes progress through prophase I , the severity of the defects increases . By stage 4 , premature disassembly of the SC is visible in all Eco KD oocytes examined , the majority of which contain only short threads of SC . At stage 5 , 35% of the Eco KD oocytes contain only spots of SC signal , and the majority of the stage 6 oocytes belong to this category ( Figure 1 ) . These results demonstrate that when Eco is knocked down after S phase , progressive deterioration of the SC occurs . Because the Drosophila germline is somewhat refractory to RNAi [39] and because the Eco RNAiGD vector ( modified pUAST vector pMF3 ) is not efficiently expressed in the germline [40] , [41] , we overexpressed Dicer-2 [41] to increase the efficacy of the Eco RNAiGD hairpin . Dicer-2 is a component of siRNA-dependent RISC ( RNA induced silencing complex ) and is required for siRNA-mediated silencing [42] . As shown in Figure 1 , SC defects are enhanced both in the number of oocytes affected and the severity of the defects when the matα driver induces overexpression of a UAS-Dicer-2 transgene simultaneously with Eco RNAiGD . Overall , these data argue that Eco activity during meiotic prophase is essential to stabilize the synaptonemal complex and are consistent with a role for Eco in maintaining meiotic cohesion after meiotic S phase . If Eco were required to rejuvenate meiotic cohesion after S phase , we would expect RNAi-mediated reduction of Eco activity during prophase to result in chromosome segregation errors . Therefore , we used our standard genetic assay ( see Materials and Methods ) to measure the fidelity of chromosome segregation in Eco KD and control oocytes . Induction of the Eco RNAiGD transgene with the matα driver caused a significant increase in meiotic nondisjunction ( p = 0 . 004 , Figure 2A ) , and chromosome segregation errors increased even more substantially in Eco KD oocytes when Dicer-2 was over-expressed ( p<0 . 0001 , Figure 2A ) . Our findings that chromosome segregation errors in Drosophila oocytes increase when Eco is reduced during prophase I indicate that Eco activity is required after S phase to ensure accurate chromosome segregation during meiosis . We carried out a number of controls to rule out nonspecific effects and confirm that the SC and segregation defects we observed in Eco KD oocytes were indeed due to reduction of Eco during meiotic prophase . Expression of two additional Eco hairpin transgenes ( Eco RNAiV1 and Eco RNAiV22 , see Table S1 ) with the matα driver caused premature disassembly of the SC ( Figure S3 ) , and induction of the Eco RNAiV1 hairpin also resulted in meiotic NDJ ( Figure S3 ) . The low fertility of Eco RNAiV22 KD females ( even in the absence of UAS-Dcr-2 ) prevented us from measuring meiotic NDJ in this genotype . However , together these data verify that the effects of Eco RNAiGD transgene expression were not due to nonspecific targets . In addition , in oocytes in which matα-induced Dicer-2 overexpression occurred in the absence of any RNAi hairpin transgene , we observed normal SC ( Figure S4A ) and no significant increase in NDJ ( Figure S4B ) confirming that the enhanced SC and NDJ defects we observed when Dicer-2 was overexpressed in a Eco KD background were a result of the increased RNAi efficiency . Finally , we also validated that the onset of SC defects at stage 2 in Eco KD oocytes ( Figure 1 ) occurred as a natural consequence of the temporal expression of the matα driver and not because the germarium is refractory to RNAi . When we used the nanos-Gal4-VP16 driver [43] to induce the expression of Eco RNAiGD in the germarium , we observed premature disassembly of the SC beginning in germarial region 2B ( Figure S5A & B ) , providing evidence that Eco knockdown in the germarium can be achieved with the nanos driver . Moreover , knockdown of Eco using the nanos driver resulted in a significant increase in segregation errors ( p<0 . 0001 , Figure S5C ) . Most likely , nanos-driven knockdown of Eco impacts the establishment of cohesion during meiotic S phase as well as any prophase I functions of Eco protein . One prerequisite for accurate segregation during meiosis I is that homologous chromosomes must remain physically associated until anaphase I when they segregate to opposite poles . After crossovers are formed , it is cohesion along the arms of sister chromatids that keeps recombinant homologs tethered to each other and chiasmata stabilized [4]–[6] . If knockdown of Eco after meiotic S phase causes loss of cohesion during prophase , we would expect chiasmata to become destabilized and recombinant chromosomes to missegregate during anaphase I . Therefore , we utilized a genetic assay [11] that allowed us to monitor for loss of chiasma maintenance in Eco KD oocytes . We first verified that crossover formation was not severely disrupted when Eco RNAiGD was induced with the matα driver in combination with Dicer-2 overexpression . We monitored crossover frequency in four intervals along the X chromosome in Eco KD and control oocytes and found no significant difference between the two genotypes ( Figure S6 ) , indicating that crossovers form normally when Eco is reduced during mid/late pachytene . To obtain direct evidence that Eco KD during prophase I causes loss of cohesion , we assayed the recombinational history of the missegregating chromosomes in Eco RNAiGD oocytes . To obtain these data , we performed a standard NDJ test using Eco KD and control females heterozygous for an X chromosome with several visible markers , including one located proximal to the centromere ( Figure 2B ) . By performing an additional cross with the Diplo-X progeny arising from the NDJ test , we were able to determine what fraction of missegregating chromosomes had undergone one or more crossovers and , based on the centromere-linked y+ marker , whether segregation errors occurred primarily in meiosis I or in meiosis II . The results from two independent experiments shown in Figure 2C demonstrate that the majority of bivalents that exhibit segregation defects in Eco KD oocytes are recombinant and that segregation errors occur primarily during meiosis I . In the first experiment , 18 of the 27 Diplo-X females arising from chromosome NDJ in Eco KD oocytes harbored at least one recombinant chromosome . In the second experiment , 13 of the 16 Diplo-X females contained at least one recombinant chromosome . In addition , because 27 of the 31 Diplo-X females harboring recombinant chromosomes were heterozygous for the centromere proximal y+ marker ( Experiments 1 and 2 combined , see Figure S7 ) , we conclude that the majority of segregation errors arose from meiosis I NDJ ( Figures 2C and S7 ) . Missegregation of recombinant chromosomes during meiosis I supports the hypothesis that when Eco is knocked down after meiotic S phase , crossovers are formed but chiasmata are not stabilized due to loss of arm cohesion . Moreover , it is important to note that our assay underrepresents the percentage of recombinant bivalents that missegregate , because it is possible for a Diplo-X female to inherit two non-recombinant chromatids from a recombinant bivalent ( Figure 2B ) . Together , these results demonstrate that Eco activity is required to maintain meiotic cohesion after cohesive linkages are formed during S phase and that Eco-mediated rejuvenation of cohesion during meiotic prophase is necessary for chiasma maintenance and accurate chromosome segregation . Our analysis of Eco KD oocytes indicates that Eco activity is required to maintain cohesive linkages during meiotic prophase I . One possibility is that re-acetylation of SMC3 molecules within existing cohesive rings is required to stabilize meiotic cohesion during the prolonged period of prophase I . Alternatively , Eco-mediated rejuvenation of cohesion during prophase I could involve establishment of new linkages . Matα-driven knockdown of cohesin subunits should only impact meiotic cohesion if the latter were true . Therefore , we used the matα driver to induce expression of SMC1 RNAiV22 , SMC3 RNAiV20 , or Stromalin ( SA ) RNAiV20 hairpins ( see Table S1 ) to reduce synthesis of cohesin subunits after establishment of meiotic cohesion . These transgenic constructs utilized the Valium 20 or Valium 22 vectors optimized for expression in the Drosophila germline [39] . Like reduction of Eco during meiotic prophase , knockdown of SMC1 , SMC3 or SA using the matα driver results in premature disassembly of the SC starting at stage 2 ( Figure 3 ) . A comparison of each KD genotype with its corresponding control genotype ( UAS RNAi , no driver ) is shown in Figure 3 . Similar to what we observed for Eco KD oocytes , both the number of oocytes affected and the severity of the defects escalate as cohesin knockdown oocytes progress through prophase I . In addition , the phenotype at each stage of oogenesis is very similar for each of the cohesin knockdowns . These data indicate that synthesis of new cohesin subunits is required during meiotic prophase I to keep the SC intact . Given the similarity of SC defects in Eco KD and cohesin KD oocytes , we asked whether chromatin localization of SMC1 was perturbed when Eco or SMC1 proteins were reduced after meiotic S phase . One possibility is that premature disassembly of the SC occurs in these genotypes because of disruption of the cohesin-enriched chromosome cores that form a scaffold for the axial elements of the SC [29] , [35] , [44] , [45] . We performed C ( 3 ) G and SMC1 co-immunolocalization experiments with ovarioles of Eco KD and SMC1 KD females and their respective controls ( no driver ) and compared region 3 oocytes ( before SC defects occur ) with those at stage 4 ( when SC defects are pronounced ) . In both Eco KD and SMC1 KD oocytes , long continuous SMC1 threads were visible in region 3 oocytes , coincident with intact SC ( Figure 4 ) . However , in both genotypes , the SMC1 signal was restricted to short threads and spots in stage 4 oocytes , similar to that of the C ( 3 ) G signal . This pattern contrasts sharply with the extensive SMC1 threads visible in stage 4 oocytes for both control genotypes ( Figure 4 ) . Our findings indicate that SMC1 protein synthesis and Eco activity are required after meiotic S phase to maintain cohesin-enriched chromosome cores during pachytene . These data support the model that chromatin association of newly synthesized cohesin subunits occurs during prophase I and depends on Eco . We also found that X chromosome NDJ increases when cohesin subunits are knocked down after meiotic S phase ( Figure 5 ) . Matα driver induced expression of an SMC1 RNAiV22 or SMC3 RNAiV20 hairpin during mid-prophase increased chromosome missegregation significantly ( p = 0 . 011 and p<0 . 0001 , respectively ) . Of particular note , SMC3 KD oocytes were extremely subfertile , and progeny were obtained in only one of three NDJ tests performed . SA KD oocytes were sterile in all NDJ tests attempted . The sterility observed when cohesin subunits are knocked down during prophase most likely stems from embryonic lethality following effective maternal depletion of these essential proteins using the germ-line optimized Valium 20 and Valium 22 expression cassettes . Regardless , the significant increase in NDJ observed when SMC1 or SMC3 is reduced indicates that new synthesis of cohesin subunits during meiotic prophase is required for accurate chromosome segregation . To verify that meiotic cohesion is lost when SMC1 is knocked down after meiotic S phase , we used our recombinational history assay to determine the genotype of missegregating chromosomes in SMC1 KD oocytes . The number and distribution of crossovers along the X chromosome are normal when SMC1 KD is induced with the matα driver ( Figure S8 ) ; however , although chiasmata form , they are not maintained . In two independent experiments , the majority of Diplo-X females arising from segregation errors harbored at least one recombinant chromosome , and segregation errors occurred primarily during meiosis I ( Figures 5 and S8 ) . These data indicate that maintenance of meiotic cohesion and chiasmata require incorporation of newly synthesized cohesin subunits into functional cohesive linkages during prophase I . We have shown that the synthesis of new cohesin subunits during prophase I is required for maintenance of cohesion and SC integrity . These data suggest that rejuvenation requires either replacement of individual cohesin subunits within pre-existing rings or the loading of new intact cohesin complexes . To further investigate the mechanism of rejuvenation , we asked whether Nipped-B , the Drosophila Scc2 ortholog [46] , is required after meiotic S phase to maintain cohesion . In S . cerevisiae , cohesin rings form normally in scc2 mutants but do not associate with chromosomes [47] . In Drosophila oocytes , Nipped-B co-localizes with SMC1 and SMC3 along the arms of pachytene chromosomes [20] , supporting the model that loading of cohesin complexes continues to occur after meiotic S phase . We reasoned that if loading of new cohesin rings during prophase I is required for cohesion maintenance , knockdown of Nipped-B using the matα driver would cause meiotic defects . We performed experiments using two different hairpins ( Nipped-B RNAiV20 and Nipped-B RNAiV22 , see Table S1 ) so that concerns of off-target effects could be eliminated . Figure 6 shows that when the matα driver induces expression of either of the Nipped-B hairpins , premature disassembly of the SC begins at stage 2 . The observed defects are very similar for the two Nipped-B constructs , and both the percentage of oocytes affected and the severity of the defects increases during prophase I progression . If we simultaneously overexpress Dicer-2 in Nipped-B knockdown oocytes , SC defects are modestly enhanced . Importantly , the phenotypes that are manifest in Nipped-B KD oocytes closely resemble those observed when Eco or cohesin subunits are reduced during prophase I . Unfortunately , we were unable to monitor the fidelity of chromosome segregation because Nipped-B KD flies were sterile , even in the absence of Dicer-2 overexpression . However , our finding that SC defects arise when Nipped-B is reduced after S phase supports the model that cohesion rejuvenation involves the loading of new cohesin complexes , not substitution of new subunits into preexisting chromatin bound cohesin rings . Under normal conditions , establishment of cohesive linkages occurs only during S phase . However , a notable exception has been described in yeast vegetative cells exposed to DNA damage during G2 . In response to double-strand-breaks ( DSBs ) , Eco1-mediated re-establishment of cohesion occurs throughout the genome during G2 [15] , [17] . One possibility is that Eco-mediated rejuvenation of cohesion in Drosophila oocytes is a programmed response to the DSBs that initiate crossovers during early meiotic prophase . We set out to test this hypothesis by determining whether Eco is still required to maintain arm cohesion in the absence of meiotic DSBs . If rejuvenation of cohesion only occurs in response to DSBs , then SC defects should be absent in Eco KD oocytes that lack DSBs . In order to genetically eliminate meiotic DSBs , we utilized a null allele of the mei-W68 gene , which encodes the evolutionarily conserved Spo11 endonuclease required for formation of meiotic DSBs [48] . In mei-W681 mutant oocytes , meiotic DSBs are eliminated and crossovers do not occur , but the temporal program of SC assembly and disassembly is normal [49] . We compared the morphology of the SC in mei-W681 Eco KD oocytes ( Eco RNAiGD and matα driver ) and mei-W681 oocytes in which the Eco RNAiGD was not expressed ( Eco RNAiGD , no driver ) . Two independent experiments are shown in Figure 7 . We observed long , continuous SC in mei-W681 oocytes ( no driver ) with normal disassembly commencing at stage 6 . In contrast , when Eco was knocked down in mei-W681 oocytes , premature disassembly of the SC was evident in stage 2 and became progressively more pronounced . Therefore , even in the absence of Spo-11 induced DSBs , Eco is required after meiotic S phase to maintain the integrity of the SC . These data support the model that cohesion rejuvenation during meiosis occurs though a novel mechanism that is distinct from DNA damage induced cohesion re-establishment during G2 in vegetative yeast cells .
Here we describe the first evidence that maintenance of meiotic cohesion during prophase I is an active process and provide mechanistic insight into this rejuvenation pathway in Drosophila oocytes . The defects that we observe when SMC1 , SMC3 or SA mRNAs are knocked down after meiotic S phase indicate that newly synthesized cohesin proteins are required during prophase I for sustained cohesion until the meiotic divisions . Moreover , our finding that SC stability depends on the Scc2 ortholog Nipped-B during prophase I suggests that loading of new cohesin complexes , and not replacement of individual subunits within existing cohesin rings , occurs during prophase I . Finally , our observation that the cohesion establishment factor Eco is required after meiotic S phase argues that cohesion maintenance and chiasma stabilization require new cohesive linkages to be formed during meiotic prophase I . Together our findings indicate that the cohesive linkages established in Drosophila oocytes during meiotic S phase are insufficient for cohesion to remain intact throughout prophase I . Accurate chromosome segregation requires more than passive endurance of the original cohesive linkages established during meiotic DNA replication . Our data support the model that cohesive linkages turn over during the protracted timeframe of meiotic prophase and that in order for oocytes to ensure levels of meiotic cohesion sufficient for accurate chromosome segregation , replacement cohesin complexes must be loaded onto the meiotic chromosomes by Nipped-B and made cohesive by the action of the acetyltransferase Eco . We use the term “rejuvenation” to describe this active process of loading cohesin complexes and generating replacement cohesive linkages during meiotic prophase I . Although it is possible that intact linkages are targeted for replacement by the rejuvenation program , we favor the model that rejuvenation acts to replace linkages that are lost due to normal turn over . At this time , we cannot rule out the possibility that Nipped-B performs a function other than cohesin loading during meiotic prophase . However , the striking similarities between the SC defects in cohesin KD and Nipped-B KD oocytes ( Figures 3 and 6 ) support the model that continuous loading of cohesin by Nipped-B during meiotic prophase is required for sustained cohesion and chiasma stabilization ( Figure 4 ) . Furthermore , we have previously reported that Nipped-B localizes along the arms of meiotic chromosomes during pachytene [20] and these results agree with recent reports that Nipbl , the mammalian Scc2 ortholog , localizes along the chromosome axes of meiotic chromosomes in mouse spermatocytes and oocytes [21] , [22] . Interestingly , although Nipbl largely dissociates from spermatocyte chromosomes by late pachytene , it remains associated with diplotene chromosomes in mouse oocytes [21] , consistent with the hypothesis that cohesin loading is required throughout the extended arrest period that mammalian oocytes undergo . Our Nipped-B and cohesin knockdown results also argue that loading of cohesin complexes onto existing lateral elements is able to occur within the context of a fully formed SC . These results nicely complement the evidence in budding yeast that the transverse filament protein Zip1 is continuously incorporated into the existing complex during pachytene [50] . Together , these results highlight the dynamic nature of the synaptonemal complex , a complex structure found almost universally in meiotic cells undergoing recombination . How do oocytes generate new cohesive linkages outside the context of meiotic S phase ? In budding yeast , cohesion can be established during G2 in response to DSBs [15] , [17] . Given that induction of DSBs and initiation of meiotic recombination occur early in the meiotic program , we reasoned that cohesion rejuvenation during meiosis might be mechanistically similar to the DSB-induced re-establishment pathway in vegetative yeast cells . However , when we abolish meiotic DSBs using a null allele of mei-W68 ( Drosophila Spo-11 ) , we find that reduction of Eco after meiotic S phase still results in premature disassembly of the SC . These results indicate that even in the absence of Spo-11 induced DSBs , an Eco-mediated rejuvenation pathway is required to stabilize the SC . Our experiments also address the possibility that meiotic DSBs initiate a signaling cascade in Drosophila oocytes that promotes conditions permissive for Eco activity outside of S phase . If this were the case , we might expect that turnover of cohesive linkages during prophase in mei-W68 oocytes would result in premature disassembly of the SC because Eco is not active . However , we and others ( Figure 7 and [49] ) observe normal timing of SC disassembly in mei-W68 oocytes that contain wild-type levels of Eco . Although we cannot rule out the possibility that a low number of Spo11-independent DSBs occur in Drosophila oocytes during prophase I , our results are consistent with the model that cohesion rejuvenation in oocytes is not a programmed response to the induction of Spo-11 triggered DSBs or their repair , and represents a novel mechanism that is distinct from that described in G2 vegetative yeast cells . Another noteworthy example of cohesion establishment outside of S phase has been reported in budding yeast vegetative cells [18] . Under normal conditions , phosphorylation of Eco1 beginning in late S phase creates a phospho-degron recognized by the SCF ubiquitin ligase , and destruction of Eco1 prevents additional cohesive linkages from forming after S phase [19] . Although three kinases participate in this pathway [19] , the initiating phosphorylation event is catalyzed by CDK1 , and yeast expressing an Eco1 mutant protein that cannot be phosphorylated by CDK1 are able to establish new cohesive linkages during G2 [18] . Significantly , during prophase I in metazoan oocytes , which lasts for days ( fruit flies ) to decades ( humans ) , CDK1 activity is silenced , primarily through translational inhibition of cyclins [51] . If CDK1-induced destruction of Eco1 orthologs is conserved in metazoans , inhibition of CDK1 during meiotic prophase I arrest may provide a key regulatory mechanism that allows Eco1 orthologs to remain active in metazoan oocytes beyond meiotic S phase . This would allow rejuvenation of cohesion to occur during the extended period in which oocytes must sustain a number of cohesive linkages that will be adequate to support accurate chromosome segregation . In yeast and human cells , it is well established that the acetyltransferase activity of Eco1/Esco is required for formation of cohesive linkages [52] , [53] . While Eco1-mediated acetylation of SMC3 is essential for cohesion establishment during S phase , DSB-induced cohesive linkages that are formed in G2 require acetylation of the α-kleisin cohesin subunit [54] . Although we know that Drosophila Eco is necessary for prophase I rejuvenation of meiotic cohesion , future experiments will be needed to determine whether the acetyltransferase activity of Eco is required for this process and to identify the substrate ( s ) of Eco1 during meiotic prophase . Is arm cohesion more dependent on rejuvenation than centromeric cohesion in Drosophila oocytes ? One interpretation of our recombinational history analyses would support this notion . However , caution is required because our assay relies solely on genetic markers and therefore only allows us to capture information about the final segregation outcome . This makes it difficult to compare our data directly with recent cytological studies of mouse oocytes that observed age-dependent weakening of centromeric cohesion prior to anaphase I [12] , [14] . The clustering of Drosophila oocyte chromosomes within a compact karyosome structure during prophase I [23] precludes our ability to perform a cytological analysis similar to those for mouse oocytes . Still , from both Eco and SMC1 KD oocytes , we recovered Diplo-X progeny that were homozygous for the centromere-linked marker ( Figures 2C , 4C , S7 and S8 ) , indicating that at least in some cases , centromeric cohesion is definitely impacted when the rejuvenation pathway is compromised . However , most Diplo-X oocytes arising from knockdown of Eco or SMC1 during prophase I were heterozygous for a centromere-linked marker , consistent with disruption of arm cohesion causing loss of chiasmata and missegregation of homologous chromosomes during the first meiotic division . These results fit nicely with observations in both human and fly oocytes that bivalents with a distal crossover are more vulnerable to segregation defects [55]–[58] , presumably because the closer a crossover is to the end of the chromosome , the shorter the region of arm cohesion is that holds recombinant homologs together . Interestingly , in ord null mutants that lack both arm and centromeric meiotic cohesion , random segregation of sister chromatids results in reductional segregation errors ( homologs ) that outnumber equational errors ( sisters ) by a factor of at least 3 to 1 [59] , [60] . Therefore , our results are not inconsistent with loss of both arm and centromeric cohesion yielding a random segregation outcome . Finally , it is important to note that defects solely in centromeric cohesion prior to the first meiotic division could theoretically lead to missegregation events that yield a gamete heterozygous for a centromere-linked marker ( for examples see [8] , [61] ) . So , although our data support the conclusion that both arm and centromeric cohesion defects arise from knockdown of Eco or SMC1 in Drosophila oocytes during meiotic prophase , we cannot assign their relative contributions to the segregation errors we observe . Our data support the model that cohesive linkages turn over in Drosophila oocytes during the normal timeframe of meiotic prophase ( ∼6 days ) and that replacement linkages are required to ensure cohesion . However , recent studies in mouse oocytes have led to the opposite conclusion – namely , that turnover of cohesin does not occur during meiotic prophase [62] , [63] . What is the basis for this apparent contradiction ? One possibility is that differences in meiotic progression in fly and mouse oocytes have led to divergent mechanisms for the regulation of meiotic cohesion during prophase I . For instance , once mouse oocytes exit from dictyate arrest and mature , they complete meiosis I and remain arrested in metaphase II until fertilization . Unlike mammalian oocytes , Drosophila oocytes arrest at metaphase I and passage through the oviduct triggers resumption and completion of meiosis , even in the absence of fertilization . Perhaps the requirement to stabilize chiasmata that are under tension during metaphase I arrest requires new supplementary linkages to be formed in Drosophila oocytes but not mouse oocytes . This seems unlikely , however , given that under normal conditions Drosophila females lay fertilized eggs continuously and the metaphase I arrest usually lasts less than two hours [23] . Another possibility is that differences in the experimental tools and approaches used in the fly and mouse studies account for these contradictory results . The matα driver that we use to induce knockdown of cohesin subunits or cohesin regulators becomes active during mid-pachytene , although robust expression does not occur until late pachytene ( Stage 4 , Figure S1B ) . As such , we are manipulating Eco , Nipped-B and cohesin levels earlier during meiotic prophase than the mouse experiments that utilized the GDF-9 promoter to drive Cre recombinase and inactivate the SMC1β gene in developing oocytes [62] or those that utilized the ZP3 promoter to drive expression of TEV-resistant Rec8 during the growing phase that precedes ovulation [63] . In addition , one potential problem with ectopic Rec8 expression is that an imbalance in the normal stoichiometry of cohesin subunits may have prevented TEV-resistant Rec8 from entering the nucleus [64] . Whether rejuvenation of meiotic cohesion is a conserved feature of metazoan meiosis remains to be demonstrated . However , it is hard to comprehend why fruit flies possess a mechanism to actively keep cohesion intact during a six-day time frame if no similar program exists in mammalian oocytes during their much longer prophase I arrest . Under conditions of normal meiotic progression in Drosophila oocytes , rejuvenation ensures that the number of cohesive linkages is sufficient to promote accurate chromosome segregation . However , when Drosophila oocytes are forced to “age , ” and spend approximately 20 times longer in diplotene [11] , cohesion is lost prematurely and chromosomes missegregate . The observation that under “aging” conditions , the normal rejuvenation pathway is incapable of sustaining cohesion in Drosophila oocytes raises the intriguing possibility that rejuvenation becomes less efficient with age . If a meiotic cohesion rejuvenation pathway also operates in human oocytes , and its effectiveness declines with age , cohesion defects may become pronounced in older women not because the original cohesive linkages finally give out , but because the rejuvenation program can no longer supply new cohesive linkages at the same rate at which they are lost .
Flies were reared at 25°C on standard cornmeal molasses medium . Please see Table S1 for the complete genotypes and origin of stocks used in this study . Please see Text S1 for detailed descriptions of the cross schemes utilized to generate flies for genetic and/or cytological experiments . For X chromosome NDJ assays , B+ experimental females were crossed to males containing an attached X∧Y , v f B chromosome . In this scheme , progeny arising from normal as well as Diplo-X and Nullo-X gametes can be recovered and distinguished based on eye shape and sex . Total %NDJ and P values were calculated according to Zeng et al . [65] . sc cv v f-y+/y females were used to measure crossover frequency along the X chromosome as well as to perform NDJ tests with subsequent recombinational history analysis and male progeny were scored for each of the visible markers . Six sets of ovaries from newly eclosed females fattened overnight with extra yeast and males were dissected in 1X PBS , splayed using a tungsten needle , and fixed for 20 minutes in a mixture of 600 ul heptane and 200 ul of 2% unbuffered formaldehyde ( EM grade , Ted Pella ) containing 0 . 5% Nonidet P-40 ( Surfact-Amps NP-40 , Pierce ) . All incubations and washes were done on a rotating platform at room temperature unless otherwise noted . Ovaries were rinsed three times with 1X PBST ( 1X PBS with 0 . 2% Tween-20 ( Surfact-Amps 20 , Pierce ) ) and blocked for one hour in 1X PBS with 1% BSA . Ovaries stained only with C ( 3 ) G antibody , were incubated overnight at 4°C with primary antibody diluted in antibody buffer ( 1X PBS with 0 . 01% Tween-20 and 0 . 5% BSA ) . For SMC1 immunolocalization , ovaries were incubated in primary antibody for 2 hours at room temperature . SMC1 primary and secondary antibody incubations were completed before C ( 3 ) G primary and secondary antibody incubations were performed . Following primary antibody incubation , ovaries were rinsed three times , washed 3×20 min in 1X PBST and incubated with the appropriate secondary antibodies diluted in antibody buffer for one hour . Subsequently , ovaries were rinsed three times and washed for 20 minutes each in 1X PBST , 1X PBST containing 0 . 1 ug/ml DAPI , and 1X PBS containing 0 . 01% Tween 20 . After the final wash , ovaries were separated into individual ovarioles with a tungsten needle , transferred to #1 . 5 18-mm poly-L-lysine-coated coverslips , and mounted in 20 µl of Prolong Gold Antifade reagent . C ( 3 ) G mouse monoclonal antibody , clone 1A8-1G2 [38] , was diluted 1∶1000 and detected using Cy3-conjugated anti-mouse secondary antibody . For simultaneous immunolocalization of C ( 3 ) G and SMC1 , guinea pig polyclonal SMC1 antibody [29] was diluted 1∶2000 and detected using Cy3-conjugated anti-guinea pig secondary antibody , and C ( 3 ) G was detected using either Cy5-conjugated or Alexa Fluor-488 conjugated anti-mouse secondary . All secondary antibodies were used at a final dilution of 1∶400 . Secondary antibodies conjugated to Cy3 and Cy5 were obtained from Jackson Immunoresearch Laboratories and the Alexa-488 conjugated secondary antibodies were obtained from Molecular Probes . To characterize the onset of expression for the matα driver , ovaries from young females containing a UASp-Actin-GFP reporter ( B-071 ) driven by matα-Gal4-VP16 were dissected in 1X PBS , and the anterior region of each ovary splayed slightly . Ovaries were fixed for 5 minutes at room temperature in 1X PBS containing 4% formaldehyde ( EM grade , Ted Pella ) and rinsed three times in 1X PBS . Nuclei were stained by incubating fixed ovaries in 1 µg/ml Hoechst 33342 for 15 min , followed by a brief rinse and a 15 min wash in 1X PBS . Individual ovarioles were separated using a tungsten needle , transferred to #1 . 5 18-mm poly-L-lysine-coated coverslips , and mounted in 15 µl of Vectashield . Coverslips were sealed with nail polish , and slides stored at 4°C until imaging . Images were acquired using a Nikon A1RSi laser scanning confocal controlled by NIS Elements ( version 4 . 13 ) . All images were collected using a 40X oil Plan Fluor DIC ( NA 1 . 3 ) objective and sequential scanning mode . Single slices were captured using unidirectional scanning with a 407 nm laser ( for DAPI ) and 488 nm laser ( for GFP ) . The single molecule FISH probes were designed using the Stellaris Probe Designer and ordered from Biosearch Technologies ( http://www . singlemoleculefish . com ) . The probe set consisted of a mixture of 48 DNA oligonucleotides ( 20 mers ) complementary to the Eco open reading frame . In designing the probes , the zinc finger and acetyltransferase domains of Eco were excluded , as well as any regions with homology to non-Eco sequences within the Drosophila genome . Probes were conjugated to Quasar 570 dye . Ovaries from 8 young females held with yeast and males for one day were dissected and slightly splayed in 1X PBS and then transferred to a 1 . 5 ml tube for fixation in 4% formaldehyde in 1X PBS for 15 minutes at room temperature on a nutator . After rinsing 3 times and washing twice for 5 min with 1X PBS to remove the fixative , ovaries were stored in 1 ml of 70% ethanol at 4°C for at least twelve hours . After removing the 70% ethanol , ovaries were incubated in 2X SSC containing 10% formamide for 10 min and then incubated in 100 ul of hybridization buffer ( 2X SSC , 10% formamide , 100 µg/ml dextran sulfate , 2 mM vanadyl ribonucleoside complex , 20 µg/ml BSA and 1 mg/ml E . coli tRNA ) containing 50 nM probe overnight at 37°C with gentle rotation in a dark chamber . Following hybridization , all washes and incubations were performed at room temperature with rocking . Ovaries were rinsed once in 400 µl of 2X SSC containing 10% formamide , and then washed for 10 min in an additional 400 µl of the same buffer . For visual identification of germline cells , ovaries were incubated for 2 hours in a mixture of ORB mouse monoclonal antibodies , clones 4H8 and 6H4 [66] each at 1∶30 dilution in 2X SSCT ( 2X SCC containing 0 . 2% Tween 20 ) . Ovaries were rinsed 3X and washed 3×10 min in 2X SSCT and then incubated for one hour in Alexa Fluor-488 conjugated anti-mouse secondary diluted 1∶400 in 2X SSCT . Ovaries were rinsed 3X and washed 2×10 min in 2X SSCT followed by a 20 minute incubation in 2X SSCT containing 0 . 1 µg/ml DAPI and an additional 10 minute wash in 2X SSCT . A tungsten needle was used to separate ovarioles before mounting on a #1 18-mm poly-L-lysine-coated coverslip with 20 µl of Prolong Gold Antifade reagent . Imaging was performed using a Nikon A1RSi laser scanning confocal system controlled by NIS Elements ( version 3 . 22 ) . All images were collected using a 100X oil CFI Apochromat TIRF objective ( NA 1 . 49 ) and sequential scanning mode . Single focal planes in which the oocyte nucleus was visible were captured using unidirectional scanning with a 407 nm laser ( for DAPI ) , 488 nm laser ( for ORB ) and 561 nm laser ( for smFISH ) . Control , knockdown and eco mutant images were acquired using the same settings and processed identically . Captured images were imported into Volocity 5 . 5 for quantification of mRNAs in individual germ-line cysts . ORB staining allowed each egg chamber to be cropped in order to remove the surrounding layer of follicle cells so that mRNA quantification was limited to the oocyte and nurse cells . Quantification of the number of mRNA signals was carried out using the following protocol in Volocity 5 . 5: First , a “find objects by % intensity” task was applied to the smFISH channel to set a threshold to best identify the bright mRNA spots . Second , a “remove noise from object , medium filter” task was applied . Then an “exclude objects by size” was added to remove the background signal from the measurements . This measurement sequence allowed determination of the total number of mRNA spots within the germ-line ROI for a single optical section . Because there is some variability in the size of egg chambers even at the same stage , the # of mRNA spots/area was used for comparison of different genotypes . | Meiosis is a specialized type of cell division that gives rise to sperm and eggs . In a woman's thirties , errors in meiotic chromosome segregation rise exponentially , significantly increasing the probability that she will conceive a fetus with Down Syndrome ( Trisomy 21 ) . Accurate chromosome segregation during meiosis depends on protein linkages ( cohesion ) that hold sister chromatids together . The widely held view is that under normal conditions , cohesion can only be established during DNA replication , and the original cohesive linkages formed in fetal oocytes are gradually lost as a woman ages . However , it seems unlikely that the same cohesion proteins could survive for even five years , much less 25 years . Here we show that Drosophila oocytes possess an active rejuvenation program that is required to load newly synthesized cohesion proteins and to establish new cohesive linkages after meiotic DNA replication . When we reduce the proteins responsible for rejuvenation after meiotic S phase , cohesion is lost and meiotic chromosomes missegregate . If such a rejuvenation pathway also exists in human oocytes and becomes less efficient with age , oocytes of older women may no longer be able to replace cohesive linkages at the same rate that they are lost . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"developmental",
"biology",
"medicine",
"and",
"health",
"sciences",
"model",
"organisms",
"women's",
"health",
"cell",
"biology",
"genetics",
"biology",
"and",
"life",
"sciences",
"molecular",
"cell",
"biology",
"research",
"and",
"analysis",
"methods"
] | 2014 | Rejuvenation of Meiotic Cohesion in Oocytes during Prophase I Is Required for Chiasma Maintenance and Accurate Chromosome Segregation |
A large fraction of the proteins that are being identified as key tumor dependencies represent poor pharmacological targets or lack clinically-relevant small-molecule inhibitors . Availability of fully generalizable approaches for the systematic and efficient prioritization of tumor-context specific protein activity inhibitors would thus have significant translational value . Unfortunately , inhibitor effects on protein activity cannot be directly measured in systematic and proteome-wide fashion by conventional biochemical assays . We introduce OncoLead , a novel network based approach for the systematic prioritization of candidate inhibitors for arbitrary targets of therapeutic interest . In vitro and in vivo validation confirmed that OncoLead analysis can recapitulate known inhibitors as well as prioritize novel , context-specific inhibitors of difficult targets , such as MYC and STAT3 . We used OncoLead to generate the first unbiased drug/regulator interaction map , representing compounds modulating the activity of cancer-relevant transcription factors , with potential in precision medicine .
While the number of high-value , candidate therapeutic target proteins has increased dramatically over the past five years , most of them lack a corresponding FDA-approved or late-stage investigational ( i . e . , clinically relevant ) small-molecule inhibitor . Furthermore , a large number of these are considered undruggable and may thus benefit from small molecules inducing potent , albeit indirect inhibition , within a specific tumor context . For instance , ibrutinib , a Bruton’s Tyrosine Kinase ( BTK ) inhibitor , can effectively abrogate aberrant NF-kB activity in human B cells , with clinically relevant application to treatment of the ABC subtype of diffuse large B cell lymphoma [1] . A key problem in addressing this challenge is the lack of generalizable methodologies for the efficient and systematic prioritization of small molecule compounds as direct or indirect inhibitors of an arbitrary protein of interest . Throughout this manuscript , we will use the word ‘compound’ for short to refer to small molecule compounds . Consistently , by compound targets and compound activity we refer to the proteins targeted by the small molecule compound and its pharmacological activity , respectively . Indeed , high-throughput screens ( HTS ) mostly rely on ad hoc , experimental gene reporter assays , whose design , testing , optimization , and miniaturization is laborious and inefficient . In addition , most of these assays are limited to reporting on the activity of a single target protein or of a specific protein class ( e . g . , protein kinases [2] ) . Computational HTS approaches , such as quantitative structure activity relation ( QSAR ) analysis [3] and virtual screening [4] , rely on availability of structural models for both the ligands and the target protein and thus on prior knowledge from related compound’s binding assays or from X-ray/NMR target structure elucidation [3] . For instance , the similarity ensemble approach ( SEA ) , which predicts new target-ligand relationships based on their similarity to established target-ligand sets , is widely adopted [5] . However , results completely depend on the availability of ligand analogs , whose structure has been previously elucidated . Critically , these methods lack cell-context specificity and are limited to assessing only direct , high-affinity binding compounds , thus missing small-molecule compounds that may indirectly modulate the activity of a target protein , as is the case for ibrutinib . These compounds cannot be assessed by QSAR , because they do not represent high-affinity ligands of the target protein of interest but rather of one of its major context-specific up-stream regulators . In addition , these methods are not effective for protein families that lack specific binding pockets , such as transcription factors ( TFs ) [6] , even though these comprise many of the best established tumor dependencies . Indeed , TFs such as ESR1 , NOTCH1 , MYC , GATA3 , and ERG , among many others , are frequently aberrantly activated in cancer [7] . In addition , many TFs have been recently elucidated as Master Regulators of tumor cell state , which are organized in highly interconnected modules or tumor checkpoints [8] , including key synthetic lethal combinations , such as STAT3 , CEBPB , and CEBPD in mesenchymal glioblastoma [9] or CENPF and FOXM1 in malignant prostate carcinoma [10] . Recently , several perturbational strategies have been proposed to measure differential gene expression following systematic chemical perturbations of specific cell lines , such as the connectivity map ( CMAP ) [11] and the Library of Integrated Network-based Cellular Signatures ( LINCS ) [12] . However , since most small molecule compounds affect the activity rather than the expression of target proteins , these data cannot elucidate targets but rather their ability to modulate the entire gene expression signature of a cell . We recently introduced DeMAND , a method for the interrogation of cell context specific networks , to infer drug mechanism of action ( MoA ) [13] . While being very efficient to capture direct as well as indirect context-specific targets [13] , DeMAND requires at least six gene expression profiles per compound . As a result , while it is very effective for elucidating the MoA of individual compounds of interest , it is not optimally suited to the reverse problem , i . e . , prioritizing candidate protein inhibitors from large-scale perturbational profiles , especially when fewer than six perturbational profiles per compound are available . We thus developed OncoLead , a novel and highly generalizable methodology for the efficient and systematic identification of small molecules that directly or indirectly inhibit a target protein of interest . OncoLead leverages the Virtual Inference of Protein activity by Enriched Regulon analysis ( VIPER ) algorithm [10 , 14]—a network-based algorithm for the assessment of protein activity from gene expression data—to assess the effect of a panel of drugs on protein activity from individual expression profiles . We limit our analysis to ~7 , 000 regulatory proteins ( RPs ) , including ~2 , 000 transcription factors ( TFs ) and ~5 , 000 signaling proteins ( SIGs ) , whose regulatory ‘activity’ may be modulated by a small-molecule compound . While these represent only ~30% of the human genome , they capture an important component of relevant tumor dependencies that may benefit from targeted inhibitor availability . Briefly , given two cellular states ( e . g . , baseline and compound-perturbed ) , OncoLead uses the differential expression of a protein’s transcriptional targets ( i . e . , its regulon ) as an accurate and highly reproducible multiplexed endogenous reporter assay for its activity [15 , 16] . For a given RP , the regulon comprises its context-specific direct or indirect transcriptional targets [17] . This approach is especially well suited to the screening of large libraries of compounds for two reasons: first , it can accurately infer compound-mediated protein activity modulation from a single perturbational profile ( e . g . , RNASeq following perturbation ) ; second , its performance is essentially unaffected when RNASeq depth is reduced from 30M to 0 . 5M reads [18] , thus allowing highly-multiplexed characterization of the activity of compounds at low cost . We first show that OncoLead can effectively assess differential activity for established targets of the compound , even when these are not differentially expressed following compound’s perturbation . To accomplish this goal , we leveraged two public databases including the Connectivity MAP ( CMAP ) [11] and the Library of Integrated Network-based Cellular Signatures ( LINCS ) ( http://lincs . hms . harvard . edu/ ) , as well as one in vivo dataset , containing gene expression profiles ( GEPs ) obtained post-treatment from patients’ tumor tissue . For each cell line or tissue represented in the datasets , the analysis was performed using networks representing the transcriptional targets of the candidate compound-targeted proteins in tissue lineage-matched contexts . We used the algorithm to assemble the first comprehensive , cell-context-specific map of inhibitors targeting RPs . The associated resource , which includes a comprehensive map of RP-compound’s interactions , is available as a supplementary file linked to this publication . We then show that the algorithm is effective in elucidating novel tumor-specific inhibitors of undruggable targets . Specifically , OncoLead was highly effective in inferring novel breast-cancer-specific inhibitors of MYC and STAT3 , which were experimentally validated .
OncoLead assesses whether a compound is an effective inhibitor/activator of a given regulatory protein , based on its effect on the transcriptional level of the protein’s regulon—i . e . its set of direct and indirect transcriptional targets—to infer the regulatory protein’s differential activity; see Methods and [18] . For simplicity , we call compound’s mode-of-action ( CMoA ) to the full repertoire of proteins , whose activity is significantly affected following perturbation with the compound . These include both direct targets as well as context-specific downstream effectors of compound’s activity , and thus effectively representing the context-specific compound’s MoA . Clearly , the accuracy of our inferences of protein activity depends on the quality of the protein regulons . Due to lineage specific chromatin remodeling and co-factor availability , protein regulons are highly cell context specific [19 , 20] . In this work , we used the ARACNE algorithm [21] for context-specific inference of the regulatory network . We have previously shown that regulons inferred by ARACNE are particularly suited for VIPER analysis [18] . As shown in Table 1 , ARACNE-based regulon inference was performed using tumor-context matched gene expression profiles ( GEP ) from The Cancer Genome Atlas consortium ( TCGA ) [22] , and relevant tumor context matched GEO datasets [23] , when available . We further complemented the ARACNE networks by incorporating evidences from other resources , including direct TF to target genes interaction evidences from chip-seq or chip-chip data ( ChEA database ) [25] , direct or indirect protein-protein interactions from the STRING database [24] , and indirect functional associations inferred upon RNAi-mediated gene silencing experiments collected from the GEO database [26] ( see Methods ) . Integration of these different evidences was performed at the inferred protein activity level ( see Methods and Fig 1A ) . To quantitatively assess interactome quality , we computed the Interactome Reliability Scores ( IRS ) as the area under the curve ( AUC ) representing the number of statistically significant OncoLead-inferred CMoA proteins as a function of the p-value threshold ( see Methods ) . The rationale , as previously discussed [10] , is that less accurate interactome models yield fewer statistically significant proteins , and thus lower IRS than the more accurate ones . Indeed , IRS scores decreased monotonically when protein interactions were increasingly randomized ( 0%–100% ) using a degree-preserving randomization algorithm [27] ( Fig 1B ) . Furthermore , confirming our hypothesis , tissue-matched interactomes systematically achieved the best IRS performance against the corresponding cell line specific signatures ( Fig 1C ) . Gene expression signatures ( GES ) representing each cell line following compound’s perturbations were then analyzed using OncoLead on multiple networks to generate integrated results . This produced a sparse 3-dimensional matrix of protein activity signatures [ΔAP , L , C] , representing the relative differential activity ( treatment with compound vs . DMSO control ) of each target protein , P , expressed as Normalized Enrichment Score ( NES ) , in cell line L , with compound C . This matrix thus provides a quantitative representation of the CMoA of all tested compounds , across all profiled cell lines ( S1 and S2 Tables ) . We then expanded this analysis by leveraging an extensive collection of gene expression profiles , representing treatment of multiple cell lines with various compounds and shRNAs targeting different genes , available from the LINCS repository ( http://lincs . hms . harvard . edu/ ) . These datasets provide a limited representation , restricted to only 978 reporter genes ( L1000 ) measured by a multiplexed Luminex assay . Within the constraints of such reduced representation , we used this dataset to build an experimental gold standard dataset ( GSD ) of compounds affecting the activity of specific target genes , by matching the signature of compound’s perturbations to those of shRNA mediated silencing . We limited our analysis to 1 , 365 compounds yielding statistically reproducible transcriptional responses ( see Methods ) , and 92 shRNA-mediated silencing assays for which ( a ) target gene silencing could be confirmed by L1000 measurements at >3 standard deviations from the controls mean and ( b ) the gene was represented in the interactome as regulator ( i . e . RP ) . This resulted in distinct gene silencing assays for each cell line , from a minimum of 16 in VCAP prostate cancer cells to a maximum of 151 in A375 melanoma cells . To assemble a suitable experimental GSD , compound’s perturbations were matched to gene-silencing assays by Pearson correlation analysis of the corresponding , cell-matched L1000 signatures . Thus , each of the 1 , 365 most reproducible compound’s perturbations were associated to a list of shRNA-mediated gene silencing assays , ranked from the one with the most correlated to the one with the most anti-correlated L1000 signature . The rationale is that the gene-silencing assays with signatures most correlated to a compound’s perturbation signature represent proteins whose activity is inhibited by the compound . We then computed the DTPA of each target protein for each of the perturbed cell lines , using the full gene expression profile , from regression analysis of the L1000 signature , see Methods . Thus , for each target protein , we rank-sorted all compounds by DTPA score , from its strongest predicted candidate inhibitor ( i . e . , that with the largest negative DTPA ) to its strongest activator ( i . e . , that with the largest positive DTPA ) . Finally , we assessed these predictions by reciprocal gene set enrichment analysis ( GSEA ) [30] of the OncoLead-predictions against the experimentally-prioritized target modulators in the GSD . Specifically , for each target protein , we computed the NES representing enrichment of DTPA ranked inhibitors in statistically significant GSD inhibitors ( p = 0 . 05 ) . Enrichment was statistically significant for most proteins targeted by small molecule compounds ( NES > 1 . 96; p < 0 . 05 , shown in green for RPs , Fig 2C and S2 Fig ) . This includes 112/151 proteins in A375 cells ( 74% ) , 68/106 in A549 cells ( 64% ) , 33/45 ( 73% ) in HA1E cells , 19/25 ( 72% ) in HCC515 cells , 75/145 ( 52% ) in HEPG2 cells , 105/137 ( 52% ) in HT29 cells , 97/120 ( 81% ) in MCF7 cells , 85/97 ( 88% ) in PC3 cells , and 16/16 ( 100% ) in VCAP cells . Overall 609/842 testable proteins ( 72% ) yielded OncoLead-inferred candidate inhibitors that were strongly enriched in experimentally assessed ones , based on the GSD . This is especially remarkable considering that LINCS L1000 assays directly measure expression of only 978 genes . As a result , on average , only 1/20th of regulon targets is directly measured by these assays while other targets are imputed . In addition , shRNA-mediated silencing may have significant off-target effects . Taken together , these data suggest that the method represents an effective strategy to prioritize candidate inhibitors for arbitrary proteins of interest . To test whether OncoLead may be effective in elucidating the targets of specific compounds in vivo , we used gene expression data obtained from patient-derived tumor biopsies before and after therapeutic intervention . Specifically , we leveraged a dataset generated by Miller et . al ( GSE20181 ) [31] , consisting of primary breast tumor samples profiled after Letrozole treatment , including at 30-days ( short term: ST ) and 90-days ( long-term LT ) , compared to pre-treatment profiles ( PT ) . Letrozole blocks estrogen synthesis in postmenopausal patients by inhibiting the aromatase enzyme . This abrogates estrogen receptor activation in breast cancer cells . Individuals profiled in this dataset include 36 estrogen deprivation responsive and 14 non-responsive patients . Response was assessed based on whether significant tumor size reduction was observed at 90-days post-treatment . Four differential expression signatures were analyzed , including ST:PT ( 30-days VS . pre-treatment ) and LT:PT ( 90-days VS . pre-treatment ) , across both responsive and non-responsive patients . DTPA vectors were obtained by OncoLead analysis of these signatures , using the TCGA patient-derived Breast Carcinoma interactome ( Table 1 ) . As expected , ESR1 activity was significantly reduced following Letrozole treatment in the responsive group ( Fig 2D ) , with longer treatment inducing stronger ESR1 activity reduction ( pLT:PT = 0 . 01; pST:PT = 0 . 046 ) . Strikingly , however , OncoLead-inferred ESR1 activity was not significantly affected by Letrozole in non-responsive patients ( pLT:PT = 0 . 51; pST:PT = 0 . 095 ) . Furthermore , differential ESR1 expression was not statistically significant following Letrozole treatment ( p-value > 0 . 05 , by Student’s t-test ) at either time point and for either responsive or non-responsive patients , suggesting that CMoA analysis correctly captured ESR1 inhibition even though its expression levels were not affected ( Fig 2D ) . Since the IRS of each perturbation summarizes the effect of such perturbation on the inferred activity of regulatory proteins , we decided to use the IRS as a metric for the bioactivity of the small molecule compound . Specifically , we evaluated the IRS score across all cell lines and compound’s perturbations in the Connectivity Map ( CMAP ) dataset . As expected , progressive degradation of the gene expression signatures , by randomly permuting increasingly larger subsets of gene expression values , was associated with a proportional decrease in the IRS ( Fig 1D ) . We selected Irinotecan for this test because it showed one of the highest IRS values in CMAP . Among the 1 , 294 CMAP compounds , HDAC , topoisomerase , CDK , and estrogen receptor antagonists presented the largest overall IRS in the MCF7 luminal breast cancer cell line . These compounds represent well-known cancer drug classes , currently under investigation in breast cancer clinical trials , and thus likely to be highly bioactive in these cells . The same analysis performed on the other two CMAP cell lines ( i . e . , PC3 and HL60 ) , consistently identified HDAC , HSP90 , NF-KB , topoisomerase , proteasome and protein synthesis inhibitors among the compounds with highest IRS . Interestingly , we observed a low IRS for a large proportion ( ~55% ) of the compounds profiled in MCF7 cells in CMAP , suggesting poor bioactivity of those compounds at the profiled concentrations . The specific highest IRS in MCF7 ( breast cancer , BRCA ) , HL60 ( acute promyelocytic leukemia , APL ) , and PC3 ( prostate cancer , PRAD ) cells was achieved by fulvestrant ( ESR1 antagonist ) , tretinoin ( all-trans retinoic acid ) , and pioglitazone , respectively . Estrogen antagonists represent the standard of care in BRCA adjuvant therapy[32] . Indeed , based on NCI60 data[33] , fulvestrant achieves 50% growth inhibition ( GI50 ) at a substantially lower concentration in MCF7 compared to HL60 ( -LogGI50 ( M ) = 8 vs . 5 . 2 ) . Similarly , all-trans retinoic acid represents the standard of care in APL[34] . Indeed , based on drug sensitivity profile in COSMIC data , HL60 cells were more sensitive to tretinoin than MCF7 and PC3 ( LogIC50 ( uM ) = 0 . 83 , 4 . 0 and 5 . 8 for HL60 , MCF7 , and PC3 , respectively ) . Finally , pioglitazone , a PPAR-γ agonist approved by the FDA as anti-diabetes drug , showed higher IRS in PC3 compared to the other two cell lines . Interestingly , PPAR-γ agonists are currently in phase 2 clinical trials for AR-independent prostate cancer[35] . Our analysis provides a comprehensive map of transcription regulator ( TR ) , whose activity is modulated by a large repertoire of compounds across the CMAP and LINCS perturbational databases . This represents a large set of cellular contexts , including: MCF7 ( breast ) , PC3 ( prostate ) , HL60 ( leukemia ) in CMAP , and A375 ( skin ) , A549 ( lung ) , HA1E ( kidney ) , HCC515 ( lung ) , HEPG2 ( liver ) , HT29 ( colon ) , MCF7 ( breast ) , PC3 ( prostate ) , and VCAP ( prostate ) in LINCS ( S1 and S2 Tables ) . Specifically , a compound-TR activity matrix was generated for each cell line , with rows representing TR proteins and columns representing compound’s perturbations . This matrix can be used to perform a variety of analyses , including identifying optimal context-specific compounds to inhibit or activate an arbitrary TR protein , as well as to compute the similarity between compounds .
Traditional drug discovery has been focused on one-drug / one-target strategy . More recently , the approach has been expanded to poly-pharmacological effects of the drugs , meaning that one drug can interact with multiple on-target and off-targets resulting in combination of wanted and unwanted effects [51] . However , MoA discovery is still limited to compound’s direct binders [4 , 5] . In this work , we expand the concept of drug MoA by including direct and indirect mediators of compound’s effect . We limit our approach however , to proteins having a transcriptional regulatory role in the cell ( TR ) . Given their central role as regulators of cell state [10 , 17] , limiting CMoA definition to TRs dramatically reduce the dimensionality of the problem with no reduction of sensitivity to detect compound’s biological activity . Due to signaling and transcriptional network rewiring , inclusion of non-direct targets in CMoA makes the approach exquisitely cell context specific . While small molecule compounds affect mostly the activity of the directly or indirectly targeted proteins , we currently have no methods in our biochemistry arsenal to directly measure protein activity . While important advances in the field of proteomics have made possible the quantification of proteins and protein isoforms in a close-to proteome-wide fashion , protein activity is not solely determined by protein or protein isoform abundance , but also depends on proper cellular localization and interaction with co-factors . Gene expression , on the other hand , can be actually profiled with high accuracy and at relative low cost , making it one of the phenotypic read-outs of choice for current drug-screen efforts . mRNA abundance however , is not directly associated with coded protein activity , especially after compound-mediated short-term perturbations , but can be interpreted , using context-specific models of transcriptional regulation , to infer changes in the regulatory proteins activity [10] . At the core of our approach is the VIPER algorithm [10 , 18] , which makes possible several fundamental characteristics: ( 1 ) the analysis can be performed at the single-sample level , allowing its application to FDA-approved drug repositioning in cancer precision medicine protocols ( see below ) , ( 2 ) results are very robust to regulatory model accuracy and expression profile quality , and ( 3 ) results are almost insensitive to partial transcriptome coverage , making it particularly suited for the analysis of low-depth RNAseq expression profiles following drug perturbation [18] . In fact , we have shown that VIPER-activity signatures obtained from 1 million reads per sample are virtually identical to the ones obtained from 30 million reads , while GES were dramatically different . This remarkable quality enables us to infer drug MoA from ultra-low cost high-multiplex expression profile analysis following drug perturbation . VIPER can accurately identify regulators whose activity is modulated by the compound . However , in ~10% of the cases , it may switch the effect directionality ( e . g . , infer a protein as activated when is in fact inhibited ) [20] . This occurs because , at steady state , auto regulatory loops may exist that induce inverse correlation between protein activity and mRNA expression . As a result , targets may be correctly inferred but their relationship ( activated vs . repressed targets ) may be inverted . We recently introduced DeMAND , a method for the elucidation of compound MoAs , based on compound’s perturbational gene expression profiles[13] . Although OncoLead and DeMAND are both network-based , they are fundamentally different and complementary , both in their formulation and , more importantly , in their practical applications . More specifically , while DeMAND focuses on compound-mediated dysregulation of protein interactions , thus requiring at least six distinct perturbations for reliable predictions , OncoLead directly infers changes in protein activity based on their regulon differential expression , thus requiring a single sample . Using the same benchmarks that were used to evaluate performance of the DeMAND algorithm , the perturbational profiles of fourteen compounds in LY3 , we tested the algorithm complementarity of the OncoLead and the DeMAND . OncoLead outperformed DeMAND almost by the same margin by which DeMAND outperformed a naïve t-test analysis , as reported in[13] ( S4A Fig ) . When integrating the predictions from the two methods , each protein was assigned the best of the OncoLead or DeMAND scores . The integrated results outperformed both individual method predictions , confirming the significant complementarity of these methods ( S4A Fig ) . For example , while DeMAND missed the direct doxorubicin target ( TOP2A ) , this was effectively captured by OncoLead ( ranked 25 out of 6 , 819 targets , p = 4 . 7x10-22 ) . Conversely , while OncoLead missed geldanamycin’s target HSP90AA1 , it was effectively discovered by DeMAND ( ranked 6 out of 7499 targets , p = 1x10-25 ) . In general , OncoLead is better suited to analyzing proteins that have a direct regulatory role , while DeMAND is more effective on proteins involved in non-regulatory interactions , such as those involved in heat stress responses . As a result , the integration of both methods is optimally suited for the elucidation of compound’s MoA . A clear limitation of our approach however , is that the definition of drug MoA is limited to the proteins represented as regulators in the network models . This leaves out non-protein mediators of drug effect as well as proteins not considered as regulators ( TRs in this manuscript ) . However , because any drug-induced phenotypic change reflected in transcriptome changes will be partially mediated by transcriptional regulators , genome-wide coverage of TRs by the regulatory models should be enough to capture , at least partially , the MoA of any bioactive compound . This study presents a new framework for defining compound’s MoA . We have shown that OncoLead-MoA captures the direct targets for known drugs while GES does not . Moreover , by interpreting drug-induced GES with models of transcriptional regulation , OncoLead-MoA is insensitive to signature noise while providing information about the bioactivity of the compounds , even when no replicated profiles are available . This is because non-informative signatures , generated from non-bioactive compounds exposure , will be poorly explained by the regulatory models and hence , their IRS scores will tend to be small , compared to informative , bioactive compound-derived expression signatures ( S4B Fig ) . We envision a strong impact of our approach for drug repositioning in precision medicine . Specifically , our method can be applied to infer drugs targeting patient-specific and currently ‘undruggable’ targets . For instance , we have shown that tumor subtype-specific MRs constitute tumor dependencies[8–10] , which are usually non-tractable from the current repertoire of FDA-approved drugs . However , inhibitors for such MRs can be inferred from gene expression data of appropriate models after drug perturbation , as we have shown here for MYC and STAT3 . We can envision a framework for cancer precision medicine in which we first infer the MRs for a single-patient in an unbias genome-wide fashion . Then by matching the patient’s MR profile with the full FDA-approved drug OncoLead-MoA , in the spirit of the Connectivity Map [11] , we select drugs or drugs combinations that not only target the top patient MR , but that comprehensively target a significant proportion of them . The elevated noise in single-tumor gene expression profiles makes this approach to be unfeasible if based only on gene expression . Conversely , interpretation of the signatures based on regulatory models , which are based on tenth to hundreds of genes per protein regulon , makes OncoLead results highly reproducible [20] , and the single-patient MR signatures , as well as the single-sample based drug MoA obtained from drug-screen experiments , extremely robust ( S4C Fig ) . If MRs are enriched in tumor dependencies [9 , 10] , then this approach should prioritize drugs being specifically toxic for the patient’s tumor .
GSEA uses the Kolmogorov-Smirnov statistic and tests the enrichment of a gene set on a gene signature generated by t-test , fold-change or other methods[30 , 53] . The distance between drugs based on gene expression or CMoA profiles was computed using signature distance algorithm . It takes the 5% most up-regulated and down-regulated genes ( gene expression value or OncoLead activities ) of one sample 1 ( 2 ) and computes its enrichment on the other sample 2 ( 1 ) to get value DIS1-2 ( DIS2-1 ) . Then take the average of DIS1-2 and DIS2-1 to get the distance between sample 1 and 2 . The distance between drugs based on GI50 sensitivity profiles from NCI60 was computed using spearman correlation of the sensitivity scores across different cell lines . MCF7 and SNB19 cells were cultured in RPMI ( Invitrogen ) and supplemented with 10% FBS ( Invitrogen ) and 1% penicillin-streptomycin ( Cellgro ) . Cell viability was measured using CellTiter-Fluor Cell Viability Assay ( Promega , G6080 ) . Cignal STAT3 reporter ( luc ) kit was purchased from Quiagen ( CCS-9028L ) . MYC reporter was bought from Qiagen/SABiosciences ( CCS-012L ) . Reporter activity was determined by Dual-Glo Luciferase Assay System ( Promega , E2920 ) . Compounds were purchased from Sigma , Prestwick and Spectrum . MCF7 cells in RPMI 10% FBS with antibiotics were plated at the density of 2x104 cell/well onto 96-well flat-bottom plate one day before transfection . Cells were transfected with MYC reporter using the jetPrime ( Polyplus ) delivery system . Reporter mixture contains an inducible MYC responsive firefly luciferase construct and constitutively express Renilla construct ( ratio 40:1 ) . A mixture of non-inducible firefly luciferase reporter constitutively expressing Renilla construct was used ( ratio 40:1 ) as a negative control . Constructs constitutively expressing GFP , firefly luciferase and Renilla luciferase constructs ( 40:1:1 ) were used as positive control . 24 hours after cell seeding , culture medium was replaced with fresh medium ( 100 ul ) and drugs were added to the cells in duplicates . 24 hours after drug treatment , cell viability and MYC reporter activity was measured . Serial dilutions of drugs were prepared in DMSO to keep the same final concentration of DMSO at 0 . 8% ( S4 Table ) . DMSO only was used as a negative control for drugs . Gene reporter activity following drug treatment was normalized to cell viability ( Firefly/CellTiter-Fluor ) and compared to the negative controls ( DMSO ) . MCF7 cells were first transfected with STAT3 and stimulated using 40ng/ml IL6 . 24 hrs later cells were treated with drugs at 10uM . For the control , corresponding amount of DMSO was added to the cells . 24hrs after the drug treatment , cell viability was determined and reporter activity was measured . Each experiment was performed in triplicate . The experimental procedure for SNB19 was the same for MCF7 except stimulation using IL6 . Galliela Lactone and Static were included in the experiment as positive controls and DMSO was used as negative control . | Most transcription factors are considered “undruggable” in conventional drug discovery . However , a large number of them are discovered to be key tumor dependencies . Thus , targeting these difficult targets has been a challenge for cancer drug discovery . Here , we introduce a novel method , OncoLead , that applies biological networks to identify candidate inhibitors that either directly or in-directly block the activities of these targets . This approach is confirmed by known target-inhibitor interactions in public databases . Furthermore , we predicted new inhibitors for MYC and STAT3 , which are validated by in vitro assays . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"biotechnology",
"luciferase",
"assay",
"chemical",
"compounds",
"small",
"molecules",
"gene",
"regulation",
"applied",
"mathematics",
"regulatory",
"proteins",
"dna-binding",
"proteins",
"organic",
"compounds",
"biochemical",
"analysis",
"simulation",
"and",
"modeling",
"algorithms",
"regulator",
"genes",
"enzyme",
"assays",
"mathematics",
"transcription",
"factors",
"gene",
"types",
"bioassays",
"and",
"physiological",
"analysis",
"research",
"and",
"analysis",
"methods",
"transcriptional",
"control",
"proteins",
"gene",
"expression",
"chemistry",
"regulons",
"biochemistry",
"organic",
"chemistry",
"genetics",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences"
] | 2017 | Systematic, network-based characterization of therapeutic target inhibitors |
Community Directed Treatment with ivermectin is the cornerstone of current efforts to eliminate onchocerciasis . However recent studies suggest there are foci where long-term annual distribution of the drug alone has failed to ensure elimination thresholds are reached . It is important to achieve high levels of compliance in order to obtain elimination targets . An epidemiological and entomological evaluation conducted in the western region of Cameroon in 2011 revealed that two health districts remained with a high prevalence of infection , despite long-term distribution of ivermectin since 1996 . This paper explores potential factors that may have contributed to the non-interruption of transmission , focusing on ivermectin treatment compliance and the importance of systematic non-compliance within the population . A mixed methods approach was used , including a population-based survey to assess treatment compliance and factors associated and qualitative assessments including focus group discussions and in-depth interviews with key programme stakeholders and drug distributors . Compliance was reported at 71 . 2% ( 95%CI: 61 . 7–79 . 2%;n = 853/1198 ) . The key factors related to compliance in the most recent round related to either programmatic and delivery issues , primarily absenteeism at the time of the campaign or alternatively individual determinants . An individual’s experience of side effects in the past was strongly associated with non-compliance to ivermectin . Other factors included ethnicity , how long lived in the village and age . There was a high percentage of reported systematic non-compliance at 7 . 4% ( 95% CI: 4 . 3–12 . 3%; n = 86/1165 ) , higher amongst females . This group may be important in facilitating the sustainment of on-going transmission . Efforts to reduce the number of systematic non-compliers and non-compliance in certain groups may be important in ensuring the interruption of transmission in the study area . However , in areas with high pre-control force of transmission , as in these districts , annual distribution with ivermectin , even if sustaining high levels of compliance , may still be inadequate to achieve elimination . Further studies are required to better understand the transmission dynamics and focus of on-going transmission in the study districts .
Onchocerciasis is caused by the filarial parasite , Onchocerca volvulus , which is transmitted by Simulium blackflies . In humans , the adult stage ( macrofilariae ) are found as worm bundles , within sub-cutaneous nodules or more deeply in the body , living for an average of ten years [1] . The female adult worm produce microfilariae , which are found circulating in the skin and can migrate to the eye [2] . Host immunological responses to infection are responsible for the cutaneous and ocular morbidities , associated with the infection [3] . Ivermectin has been shown to be a safe and potent O . volvulus microfilaricide [4 , 5] and has been licensed for the human treatment of onchocerciasis since 1987 . Additionally , ivermectin also has an embryostatic effect on the female adult worm , temporarily reducing microfilarial production [4] . Within three to four months following treatment , microfilarial production slowly resumes [5] , however , repeated exposure to ivermectin over time , potentially has a cumulative effect on female worm fertility , with the recovery of microfilarial production likely never reaching pre-treatment levels [6 , 7] . In respect to the longevity of the adult worm , the effect of ivermectin is believed to be limited [7] and therefore any control efforts require long-term , regular distribution of the drug for the entire lifespan of the adult worm . The community-directed treatment with ivermectin ( CDTi ) approach was therefore developed by the African Programme for Onchocerciasis Control ( APOC ) , to ensure a sustainable model for the routine distribution of ivermectin in the community [8] . APOC was set up in 1995 , with the initial aim of control of infection and elimination of onchocerciasis as a public health problem . However , there is growing evidence , especially in meso and lower hyper-endemic settings , that the distribution of ivermectin alone , can lead to the interruption of O . volvulus transmission in Africa [9 , 10] , with proof of principle in Senegal , Mali [11] , Uganda [12] , Nigeria [13] and in Sudan [14] . This has led to the recent shift in focus beyond just control of onchocerciasis to the interruption of transmission and the elimination of infection [15] . However , there are also examples of foci where long-term annual ivermectin distribution alone seems to have been inadequate in interrupting transmission [16–19] . A number of models developed to represent O . volvulus transmission dynamics [20–22] show that various factors may influence the successful elimination; these range from the pre-control endemicity and magnitude of inter-treatment transmission [20] , the duration , frequency , timing and coverage of ivermectin [20–22] and individual dynamics and heterogeneous interactions between the vector , parasite and host [23] . One of the key parameters essential for the success of elimination programmes built on preventative chemotherapy ( PCT ) , is the sustained high level compliance amongst the population at risk [24] . Drug coverage is an often reported metric in PCT programmes , however , there is often a disparity between ‘coverage’ which refers to the proportion of eligible people who received drugs , as compared to ‘compliance’ referring to the proportion of eligible people who actually ingested the drugs [25] . Evidence suggests the importance of compliance on residual infection rates [26] . If a significant proportion of the population systematically fail to comply with treatment , then potentially a proportion of the parasite reservoir remains untreated [27] . This may help facilitate transmission and the potential of recrudescence or re-infection amongst those treated , reducing the chances of successful elimination in the foci [28] . The role of systematic non-compliers and the importance of this group has been highlighted in onchocerciasis transmission models [20] but further studies are required in order to determine the infection status of this group and to further explore the impact of systematic non-compliers on progress towards elimination . Risk factors shown to be associated with treatment coverage and compliance , range from programme and delivery issues to individual recipient characteristics . Programme level factors include issues with the drug delivery or mode of distribution , failure of the community volunteers to distribute the drugs , absenteeism at the time of the campaign or issues of trust in the drug distributors . Individual level factors range from awareness of the campaign or disease , perceived risk from infection and risk or benefit of taking the drug , including the fear of side effects [25 , 28–33] . Serious adverse events as a result of ivermectin treatment , are a significant risk in some patients co-infected with Loa Loa filariasis [34] . Quantifying and understanding drug compliance is extremely important , especially with the recent paradigm shift from control to elimination in onchocerciasis programmes . Therefore , further studies are needed to determine the impact of non-compliance on progress towards elimination . The study reported here aimed to further explore factors related to non-compliance in two health districts in the west region of Cameroon . An area where ivermectin distribution has been in place since 1996 but the interruption of transmission , as shown in the 2011 epidemiological and entomological evaluation , has not been achieved [18] .
This study focused specifically on Foumbot and Massangam health districts , in the west region of Cameroon . Both districts are rural , with the total population of Foumbot at 93 , 071 and Massangam at 39 , 776 ( Community-Directed Drug Distributor census 2014 ) . The predominant ethnic group is Bamoun and the two dominant religions are Islam and Christianity . The dominant vegetation cover in the area is that of degraded forest . There are two rainy seasons ( March to May and September to October ) , facilitating black fly reproduction and disease transmission . There are a number of rivers in the area , including the Mbam River , River Nja ( in Massangam ) and the River Noun ( in Foumbot district ) . The Simulium species found in the area is Simulium damnosum complex , specifically Simulium squamosum A [35] . The west region of Cameroon has implemented annual ivermectin distribution since 1996 . Initial treatment coverage in the region was low ( 30–35% ) but it increased to 60–65% upon the introduction of the CDTi approach in 1999 [National Onchocerciasis Control Programme , personal communication , 11 September 2015] . More recent data from Massangam district , showed a high geographic coverage ( number of villages in the district that received ivermectin/total number of villages ) , whilst the therapeutic coverage ( number of persons that received ivermectin/ total population ) was more variable , ranging from 67 to 85% ( Table 1 ) . In Foumbot district , the geographic coverage was also generally high , remaining at 100% after 2006 . Between 2004–2014 , the therapeutic coverage , fluctuated between 76% up to 84% . In Massangam health district , where lymphatic filariasis is supposedly endemic [36] , co-administration with albendazole was introduced in 2011 . Both districts have a low risk of severe adverse events ( SAE ) due to loiasis [37] and no SAEs have been reported from these districts since the onset of the mass administration of ivermectin in 1996 ( National Onchocerciasis Control Programme , personal communication , 11 September 2015 ) . An epidemiological and entomological evaluation was conducted in 2011 and found that although the prevalence of microfilariae and nodules were significantly reduced as compared to the 1996 baseline , transmission of O . volvulus had not been interrupted . In particular , prevalence of infection remained particularly high ( microfilaria prevalence over 40% ) in select communities in Foumbot and Massangam . Two out of three fly collection sites had infective rates of 0 . 19% and 0 . 18% and an annual transmission potential of 70 ( Foumbot ) and 300 ( Massangam ) respectively [18] . The data collection was carried out in December 2014 , three to four months following the last round of community ivermectin distribution , conducted in August and September 2014 . The study used a mixed methods approach . The survey aimed to verify the reported therapeutic and geographical coverage of ivermectin at the level of the two health districts and understand reasons for non-compliance . The survey followed a two-stage cluster sampling methodology , with the primary cluster ( primary sampling unit ) the village selected using probability proportional to size . The secondary cluster , the household , was selected using the household listing approach or where household census information was not available , a modified random walk methodology . A questionnaire was administered to everyone normally resident in the household i . e . resident in the last six months , recording key demographic information , if they received and swallowed ivermectin and albendazole where relevant , the reason if they did not , historical ivermectin compliance and information on side effects from taking the drugs . If anyone was absent at the time of the household visit , the survey team made one return visit later that day . If still not available and if possible , a household member answered on their behalf and this was recorded on the questionnaire . Households that refused to participate were not replaced . The sample was calculated to estimate a coverage of 80% with 5% precision , at a 95% confidence level . Taking into account a design effect of 4 ( based on recommendations from other treatment coverage surveys [38 , 39] ) and non-response of 20% , a total of 1 , 180 individuals were to be sampled from 236 households across 20 villages . Additional operational and programmatic factors , as well as population and social dynamics which may have contributed to sub-optimal compliance , were explored through qualitative methodologies . In five purposefully selected villages focus group discussions ( FGDs ) were held with the community , two ( one male and one female ) per village , conducted in French or Pidgin English . These villages were selected based on their availability of historical data on onchocerciasis and to ensure a wide geographical distribution across the two districts . Individuals that took part in the FGDs were included based on having lived in the village for a minimum of 10 years but ideally at least the last 20 years . Key informant interviews were held with both district onchocerciasis focal persons and Community-directed Drug Distributors ( CDDs ) in the five selected villages . The qualitative assessment was conducted at the same time as the quantitative survey , and was led by social science researchers with experience in conducting qualitative research . The study was approved by the national ethical review committee “Le Comité National d’Ethique de la Recherche pour la Santé Humaine ( CNERSH ) ” in Cameroon . The study objectives and procedures were explained to all participants in their local languages and written informed consent was obtained from all participants before they were included in the study . Informed assent was provided by minors ( under 21 years of age ) and caregivers provided consent for their participation . All survey questionnaires were double-entered into a pre-designed EpiData 3 . 0 database which had in built consistency and validation checks . Further consistency , data range and validation checks were also performed in STATA 12 . 0 ( StataCorp . 2011 . Stata Statistical Software: Release 12 . College Station , TX: StataCorp LP ) , the software also used for the analysis . Descriptive statistics were employed to present simple frequencies of the dependent variables ( ivermectin compliance and systematic non-compliance ) and its distribution by sex , age , education , occupation , ethnicity , years lived in the village , history of taking the drug , perceived risk from onchocerciasis and self-reported side effects experienced in the past . Chi-squared tests and multivariate logistic regression models were used to assess for the association between various explanatory variables and the dependent variables , participation in the last treatment ( ivermectin ) round and systematic non-compliance . Age and sex and variables shown to be associated in the univariate analysis with the dependent variable , were included in the multivariate logistic regression model . The likelihood ratio test was used to determine the model of best fit . The dataset was presumed to be self-weighted ( based on the probability proportional to size sampling ) but the analysis was adjusted for the cluster sampling methodology using robust standard errors based on observed between cluster variability . For the qualitative component , all interviews and FGDs were audio-recorded and transcribed verbatim . For the analysis , a three stage thematic coding approach was undertaken , using the interview topic guide to help structure the analysis . This was complemented with a more iterative approach which drew on aspects of grounded theory and allowed for new themes and ideas to develop from the interviews and FGDs . NVivo 10 ( NVivo qualitative data analysis software , QSR International Pty Ltd . Version 10 , 2012 ) was used to aid the coding and analysis of the transcripts . Attempts were made to triangulate the data from both the quantitative and qualitative methodologies and to determine where the findings from both results were similar or disparate .
The median age of the survey participants was 27 years old ( mean 31 years ) , with 54 . 0% of them being female ( 95% CI: 50 . 8–57 . 1% ) . Of those aged over 15 years , most had attended formal education , while around a third had not completed primary level ( 38 . 1%; 95%CI: 30 . 6–46 . 2% ) . The majority of respondents identified themselves as Muslim ( 73 . 6%; 95%CI: 60 . 5–83 . 5% ) and of Bamoun ethnicity ( 80 . 1%; 95%CI: 66 . 2–89 . 2% ) . The primary occupation of the respondents was farming ( 49 . 8%; 95%CI: 38 . 1–61 . 4% ) . Most had lived in the village all of their life ( 69 . 0%; 95%CI: 61 . 1–75 . 9% ) and about a quarter ( 23 . 5%; 95%CI: 15 . 4–34 . 2% ) travelled outside the village for periods of two weeks or more . Overall , 71 . 2% ( 95% CI: 61 . 7–79 . 2% ) of participants ( all ages ) , stated they had taken ( swallowed ) ivermectin during the last distribution . There was no evidence of a difference in compliance between the two districts ( p = 0 . 27 ) . In Massangam district , where ivermectin was co-administered with albendazole , 59 . 2% ( 95% CI: 29 . 1–83 . 7% ) received both drugs . A number of reasons for non-compliance were explored in the study , these related to either programmatic/delivery issues ( shown in light grey ) or individual/community factors ( shown in dark grey Table 2 ) . Being absent at the time of the MDA delivery was a key factor impacting on non-compliance , reported by a third of those not taking the drug in the last round . This was sometimes related to seasonal migration , with some migrating to the area only for a few months for transhumance or for farming in the fertile plains before travelling to the towns and cities to sell their produce . Two other key programmatic/delivery issues were the failure of the drug distributor to deliver ivermectin to the household and the lack of awareness by the individual as to the MDA campaign . The survey data was corroborated by the qualitative findings , which suggested that the inability of participants to access ivermectin was an important barrier , although it was not always clear whether CDDs failed to deliver the drugs to the household or the individuals were just absent at the time of the drug distribution campaign . The socio-demographic characteristics independently associated with drug uptake were age , ethnicity and years of residency in the village ( Table 3 ) . The lowest compliance was in young adults ( aged 20–34 years ) at 61 . 3% ( 95% CI: 50 . 4–71 . 1% ) . Individuals that identified themselves as Bamileke were nearly three times more likely to take up treatment as compared to the largest ethnic group , the Bamoun ( OR = 2 . 9; 1 . 8–4 . 5 p = 0 . 01 ) . Those who had lived in the village for less than five years were less likely to have taken drugs that those that had lived there for a longer period . Other characteristics , including sex , occupation , education , religion , or perceived personal risk of onchocerciasis showed either weak or no association with ivermectin compliance . Among individual/community level factors , a fear of side effects was the main reason for not taking ivermectin in the last round ( Table 2 ) . A total of 166 ( 14 . 3%; 11 . 7–17 . 4% ) individuals ( aged five years or over ) self-reported they had ever experienced a side effect after taking ivermectin and there was a strong association between reporting side effects in the past and the likelihood of non-compliance in the last round ( odds ratio = 0 . 3; 0 . 1–0 . 5 , Table 3 ) . With regards to side effects , the qualitative data corroborated the survey findings indicating that side effects were a major influencing factor on whether or not to take the drugs . The interviews and FGDs highlighted a variety of side effects from ivermectin mentioned by the community , from minor ailments such as swellings and itches to more serious concerns about visual impairment , potential sterility and death . Additionally some did not take the drugs at the time of the distribution as they were reportedly saved for their alternative benefits , typically to kill hair lice . Qualitative data suggests that the majority of the local community were aware of onchocerciasis and generally accepted ivermectin to be effective in treating the disease and related morbidities . Only 6% of participants believed the drug was ineffective ( Table 2 ) . Onchocerciasis was particularly important in Makouopsap village where a number of individuals were known to be blind as a result of infection . The black fly ( local name “moute moute” ) was also known to be an issue in the studied communities , especially for farmers , who worked in the fertile areas by the side of the river . One issue highlighted in the qualitative part of the study was the role of CDDs in ensuring high quality campaigns and high treatment coverage . More specifically , two aspects of the CDD role were discussed ( i ) CDDs’ attrition and motivation and ( ii ) CDDs’ relationship with the community . For example , the district health officials interviewed in the study pointed out that CDDs’ motivations and subsequently attrition rates varied greatly and were undoubtedly reflected in the duration and quality of the MDA campaigns . The CDDs worked primarily as volunteers and their willingness to distribute drugs was driven largely by their attachment to the local community , their religious beliefs and their commitment to good health . Some CDDs felt that their work was not appreciated by the community , many experienced problems ranging from the community apathy and mistrust around CDDs’ motivations to negative reactions and insults in response to adverse side effects . Supervision and support from the health facility workers did not appear to be a strong motivating factor . In fact , a number of CDDs expressed concern over a lack of support they received in handling potential adverse events and they often had to spend their own money to care for the patient with complications . The only positive incentive mentioned by CDDs was training . However , attending training often resulted in financial losses for CDDs , which also discouraged them from work , as one CDD explained: The majority ( 67 . 1% ( 95% CI: 13 . 6–28 . 7% ) ) of survey respondents stated they always took the drugs , while 20 . 1% ( 95% CI: 13 . 6–28 . 7% ) took the drug at least once in the last 5 years . There was however a substantial number of individuals aged at least 5 years and above ( 7 . 4% ( n = 86; 95% CI: 4 . 3–12 . 3% ) ) who stated they had never taken ivermectin , herein referred to as systematic non-compliers . Due to the potential of recall bias , individuals were not asked as to the reasons they did not take ivermectin , beyond the most recent round . However , when analysing the reasons given by systematic non-compliers as to why they did not take ivermectin in the last MDA , the most commonly reported reasons were they were absent during the campaign , 36 . 6% ( 95% CI:18 . 7–59 . 2% ) and fear of side effects , 21 . 8% ( 95% CI:14 . 1–32 . 1% ) . The majority ( 63 . 1% ( 95% CI: 53 . 7–71 . 6% ) ) of systematic non-compliers were female ( OR = 1 . 7; 1 . 1–2 . 6 , p = 0 . 02 ) . Reasons provided by the female non-compliers included fear of sterility and issues related to irregular menstruation , which they thought may be a result of taking ivermectin . Being from the Mbororo tribe ( nomadics ) was strongly associated with systematic non-compliance ( p<0 . 001 ) . Individuals from this group were ten times more likely to state they had never taken the drug as compared to the largest tribe , the Bamoun ( OR = 10 . 9; 6 . 5–18 . 0 ) .
Prior to the introduction of the CDTi approach in the study area , there were some issues in attaining consistently high ivermectin coverage , although more recently reported coverage has been higher . The study , however , showed compliance for the last round of ivermectin distribution was lower ( 71 . 2%; 95%CI: 61 . 7–79 . 2% ) than that reported through the health system reports ( 80% in Foumbot and 81% in Massangam ) and the proportion of systematic non-compliers was high , at 7 . 4% . Poor compliance and particularly high levels of systematic non-compliance have been highlighted in some earlier research [26 , 40] as likely contributors to the potential non-interruption of transmission in this area . The study suggests that the main reasons for individuals not taking the drugs can be broken down into two key areas , programmatic/delivery issues and individual/community factors . Programmatic delivery issues related to low coverage included the absence of individuals on the day of the distribution , often the result of seasonal migration , a common occurrence in the study area . Additionally , there was a noted failure of CDDs to deliver the drug , although it was unclear whether this was a failure on the part of the CDD or was related to individual’s absence at the time of the campaign . Finally , there was an unawareness of the MDA campaign , this could be related to poor sensitization or again linked to the migratory patterns of the population in this area , it is likely interventions to address both issues will need to factored in future campaigns . With regard to individual/community level factors , the fear of or experience of side effects associated with ivermectin was the main reason for non-compliance despite the fact that the area is at low risk of loiasis related SAEs [37] and no SAEs had been previously reported . The finding is consistent with earlier research [25 , 41] . Other individual characteristics associated with non-compliance were ethnicity , younger age and shorter residency in the area . With regards to ethnicity , the Bamileke , who had the highest level of compliance in the last round are not indigenous to the study area and it was not clear as to the exact reason for their higher compliance rate , an issue that needs to be further explored in future research . The lowest levels of compliance in those aged 20–35 years is probably related to increased work and mobility amongst this group , while lower compliance among those who had moved into the village in the last five years , may be to do with a lack of awareness of the campaign and/or the risks of onchocerciasis . Interestingly , there was reference to the topical use of ivermectin added to the hair as a means to kill lice , suggesting that the drug was in some cases valued for its alternative benefits . This highlights the importance of CDDs directly observing the individuals swallowing the drugs to ensure ivermectin is used as intended . Issues of CDDs’ motivation and their relationship with the community came out as a potential mediating factor of compliance in the qualitative part of the study . The findings suggest that poor CDD motivation and mistrust between the community and the CDDs can be related to both a poor quality campaign and poor drug compliance . These data however , comes from the qualitative interviews only and should therefore be treated as a hypothesis , which needs to be tested in future research using quantitative methods . A high proportion of the population reported they never took ivermectin , with strong evidence that women were more likely to be systematic non-compliers , likely related to the fear of side effects related to infertility . The Mbororo tribe were also strongly associated with non-compliance , a group of nomadic pastoralists that migrate and therefore are likely to miss the MDA campaign , unless specific measures are taken to adequately reach this group . The study has a number of limitations . First , the original intention of this research had been to investigate prevalence and intensity of infection amongst those individuals that had not taken ivermectin in the most recent round of MDA , to determine if they are a reservoir of infection facilitating on-going transmission . Unfortunately , due to a high number of individuals refusing to have a skin snip biopsy and due to a loss of follow-up of participants that had participated in the coverage survey , not enough data was generated to be able to report reliable findings . Future studies should try and look at this question in more detail , perhaps using a different methodology . A further limitation of the study was the lack of verification of the reported compliance in the study with the CDD records , with only 59 . 1% of records available at the time of the study . With regards to recall bias , which can be a limitation in coverage surveys , we tried to minimise it by the short period between the treatment round and the survey and by taking a conservative definition of systematic non-compliance , defined here as having never taken ivermectin , which is more likely to be accurately remembered over time . Despite these limitations , the study has important programmatic implications . Future MDA campaigns need to make efforts to increase ivermectin compliance and specifically address the large percentage of systematic non-compliers . Potential interventions include intensifying awareness of the benefits of ivermectin , improving CDDs’ support and motivation and better coordination and timing of MDA . Consideration needs to be given to seasonal migration patterns and to higher levels of buy-in from tribal leaders , particularly amongst the largest ( Bamoun ) ethnic group and the Mbororo nomadic pastoralists . The delivery of MDA through the better use of the kinship system , applied for example in Uganda [42] , may also help improve compliance amongst various ethnic groups . Future success of MDA campaigns must also address the community perception and fear of side effects , especially amongst females , potentially ensuring greater involvement of women in the CDTi approach . It is also important to ensure better support to CDDs in educating communities and managing adverse events . It is important that MDA programmes in the study areas intensify mop up campaigns or potentially move to semi-annual distribution . Finally , epidemiological models suggest that in areas where pre-control force of transmission is high , like in the study area [18] , annual ivermectin distribution even with high compliance , may be inadequate to interrupt transmission and achieve elimination by 2025 [43] . It is therefore important that further studies are carried out to better understand the transmission zones in the two health districts and delineate the potential focus of high transmission . With this knowledge , tailored activities can be developed to ensure the maximum chance of interrupting transmission of O . volvulus in the study area and achieving elimination . | Community Directed Treatment with ivermectin is the cornerstone of current efforts to eliminate onchocerciasis . Ivermectin distribution alone has been shown to be able to interrupt transmission but there are foci where long-term distribution of the drug alone has failed to ensure elimination thresholds are reached . Two health districts in the western region of Cameroon remain with high prevalence of infection despite annual distribution of ivermectin since 1996 . The study aims to explore factors related to non-compliance in two health districts in the west region of Cameroon . Nearly 30% of the population did not take ivermectin during the most recent round of mass drug administration and there was a significant proportion of the population that had reportedly never taken the drug . The key factors related to drug compliance in the most recent round , related to either programmatic and delivery issues , primarily absenteeism at the time of the campaign , or alternatively individual determinants , such as side effects associated with the drug , ethnicity , age and years lived in the village . Efforts to reduce the number of systematic non-compliers and non-compliance in certain groups are likely to be important in ensuring the interruption of transmission in the study area . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"invertebrates",
"medicine",
"and",
"health",
"sciences",
"onchocerca",
"volvulus",
"tropical",
"diseases",
"geographical",
"locations",
"parasitic",
"diseases",
"animals",
"onchocerca",
"ethnicities",
"pharmaceutics",
"neglected",
"tropical",
"diseases",
"pharmacology",
"onchocerciasis",
"africa",
"cameroon",
"adverse",
"reactions",
"drug",
"distribution",
"drug",
"delivery",
"pharmacokinetics",
"people",
"and",
"places",
"helminth",
"infections",
"nematoda",
"biology",
"and",
"life",
"sciences",
"population",
"groupings",
"drug",
"interactions",
"organisms"
] | 2016 | Factors Associated with Ivermectin Non-Compliance and Its Potential Role in Sustaining Onchocerca volvulus Transmission in the West Region of Cameroon |
Neural mass models ( NMMs ) are increasingly used to uncover the large-scale mechanisms of brain rhythms in health and disease . The dynamics of these models is dependent upon the choice of parameters , and therefore it is crucial to be able to understand how dynamics change when parameters are varied . Despite being considered low dimensional in comparison to micro-scale , neuronal network models , with regards to understanding the relationship between parameters and dynamics , NMMs are still prohibitively high dimensional for classical approaches such as numerical continuation . Therefore , we need alternative methods to characterise dynamics of NMMs in high dimensional parameter spaces . Here , we introduce a statistical framework that enables the efficient exploration of the relationship between model parameters and selected features of the simulated , emergent model dynamics of NMMs . We combine the classical machine learning approaches of trees and random forests to enable studying the effect that varying multiple parameters has on the dynamics of a model . The method proceeds by using simulations to transform the mathematical model into a database . This database is then used to partition parameter space with respect to dynamic features of interest , using random forests . This allows us to rapidly explore dynamics in high dimensional parameter space , capture the approximate location of qualitative transitions in dynamics and assess the relative importance of all parameters in the model in all dimensions simultaneously . We apply this method to a commonly used NMM in the context of transitions to seizure dynamics . We find that the inhibitory sub-system is most crucial for the generation of seizure dynamics , confirm and expand previous findings regarding the ratio of excitation and inhibition , and demonstrate that previously overlooked parameters can have a significant impact on model dynamics . We advocate the use of this method in future to constrain high dimensional parameter spaces enabling more efficient , person-specific , model calibration .
Neural mass models ( NMM ) approximate the average behaviour of large populations of neurons and therefore provide an efficient way to simulate electrographic data in order to understand the mechanisms of brain ( dys- ) function . They have been used to understand a wide variety of physiological and pathophysiological activities of the brain , including the alpha rhythm [1 , 2] , sleep rhythms [3–5] , brain resonance [6] or dynamics resulting from conditions such as epilepsy [7–11] , schizophrenia [12] and dementia [13] . In particular , mechanisms underlying these conditions can be uncovered by inverting NMMs given dynamic data and studying the meaning of model parameters [14–17] . However , maintaining a sense of biological realism in NMMs results in a high dimensional parameter space . The presence of many parameters renders the estimation of parameters from data , or model inversion , a challenging task because it is difficult to systematically and exhaustively explore large hypervolumes in order to identify subvolumes that are plausible . In order to reduce dimensionality , subsets of parameters can be fixed based on a priori assumptions . Both the choice of initial values for parameters and the boundaries of the parameter space that are searched are often constrained [18] . Unfortunately , these constraints are often based on previously used values that have sometimes arisen arbitrarily in the literature . For example , the majority of parameters used in the study of [11] are taken directly from a previous study [19] . This study itself used previous parameter values [20 , 21] . Ultimately these values were derived from studies made in the 70s [1 , 22–26] ( see Fig 1 for a summarised history of typically cited parameter values for the NMM ) . In these early derivations of NMM , parameters that could be experimentally determined were estimated but their uncertainties were not always measured [1] . Such parameters at the macroscopic level of NMM are often presumed to relate directly to properties of individual neurons but aggregated , for example , to mean values [28] . However , large variability has been shown to exist in parameters measured directly from neurons and even parameters that are considered to be quasi-certain in the modelling community , such as membrane time constants , have been shown to vary significantly in experiments [29] . Furthermore it remains unclear exactly how parameters of NMMs relate to microscopic properties of nervous tissue . Under standard values of NMM parameters , important insight has been gained regarding the generation of spontaneous or evoked electrographic recordings . For example , epileptiform rhythms have been shown to be induced by alterations to the excitatory/inhibitory balance in models [30] . However , fixing default values a priori in order to study the generation of particular dynamics does not allow to understand the behaviour of the system at unexplored , potentially plausible parameter values . Thus we cannot discover whether other regions of parameter space permit the same or different conclusions . When specifying prior distributions for model inversion ( for example using the Kalman filter or Dynamic Causal Modelling frameworks [18 , 31] ) we usually , therefore , do not know to what extent any resulting inference is dependent upon the particular choice of priors or whether unexplored regions of parameter space could also provide reasonable solutions . High dimensionality of parameter space is a particular problem in such settings since inversion algorithms become computationally demanding . It is therefore often prohibitive to explore a large parameter space or conduct inference under alternative choices of priors . The same can also be said for the use of global non-deterministic searches , for example based on evolutionary algorithms [32] . Typically , the parameter space of NMMs contains nonlinear manifolds which delineate parameter sets that give rise to qualitatively the same dynamics . These manifolds can therefore be studied to understand the emergence of different dynamic regimes . Traditional approaches to mapping dynamics over changes in parameters include bifurcation analyses and simulation studies . Some models have been extensively studied via these methods [33 , 34] , which typically only examine two parameters simultaneously . Clearly , in high dimensional systems such as NMM , we expect that changing a third parameter could affect the distribution of dynamics obtained . As such , the Jansen model [19] have been studied comprehensively by simultaneously altering 3 parameters [30] . A potential downside to such analyses is that results can be cumbersome and difficult to summarise , thus moving beyond 3 parameters with these techniques would prohibit a succinct evaluation of the role of each of parameter . Another approach is to extend multiple bifurcation analyses in a single parameter across further dimensions , whilst classifying different bifurcations and their prevalences [35] . Although this is a valid approach to understanding some elements of the complexity over large dimensional parameter spaces , it does not give a comprehensive overview of the role that each parameter plays . Even if very enlightening about the role and codependence of a few parameters , bifurcation theory cannot be use to simultaneously approach all parameters . In high dimensions ( e . g . D ≥ 3 ) these methods soon become computationally intractable and are not able to characterise the effect on dynamics of changing all parameters simultaneously . On the other hand , studying a restricted number of parameters is unsatisfactory . It would therefore be highly beneficial to develop approaches to understand the repertoire of NMM dynamics over all parameters that cannot be sufficiently constrained . Such an approach would facilitate choosing appropriate priors and initial parameter settings in model inversion algorithms . It would also facilitate a deeper understanding of complex , high dimensional models . Approaches such as global sensitivity analyses ( variance-based methods [36] , screening [37] or generalised models [38] ) have previously been used to identify the existence of relationships between dynamics and parameters . However , these methods do not allow to quantify the impact that changes in multiple ( e . g . all ) parameters have on dynamics , or to identify specific regions of parameter space in which changes in dynamics occur . In this study we therefore introduce a new methodology for the characterisation of NMM dynamics simultaneously over all parameters . The NMM is simulated a large number of times with different combination of parameters . The simulations are then classified according to their dynamics using pre-specified features . The relationships between parameter space and dynamics are then studied using the classical machine learning method of decision trees and random forests [39] . The decision trees can map the parameter space and the random forest can be used to rapidly characterise the dynamics of the NMM under previously unexplored parameter combinations . Importantly , the resulting statistical model yields natural means by which to quantify the relative importance that each parameter plays in the generation of dynamic characteristics of interest , without restricting analyses to low dimensional subspaces . We use this method to demonstrate the significance of previously overlooked NMM parameters for both physiological and pathophysiological rhythms .
In this section we give a brief descriptive overview of the approach and provide further mathematical details of each component in subsequent sections . To do so , we transform the NMM into a statistical model , which is a function that maps parameters onto a quantification of these important features . This statistical model can then be analysed to understand the relationship between the NMM parameters and the dynamics ( see Fig 2 for a general overview ) . The first step consists of choosing a NMM and defining a plausible parameter space , i . e . some constraints on the extreme values that each parameter can take . In this study we use a variation of the Jansen and Rit model introduced in the context of epilepsy [11] . This model , which we refer to as the Wendling model , has 11 parameters . The second step in the methodology consists of transforming the mathematical model into a database . To do this the NMM is simulated a large number of times using different parameters , which are chosen using a latin hypercube design . This is a space filling design which allows to efficiently explore the whole parameter space given a fixed number of simulations [40] . Each simulation is then classified in terms of some chosen characteristics . Here , we choose to focus on characteristics that are often used to define healthy and epileptiform rhythms , i . e . amplitude , frequency and number of peaks per period . The amplitude was defined as the maximum minus the minimum of the simulation . In cases for which the amplitude was greater than zero , i . e . the simulation was not constant , the frequency of the cycle and the number of peaks per period were calculated . The number of peaks can be used , for example , to characterise pathological dynamics . One of our aims is to characterise qualitative changes in model dynamics over the features above , since such an approach would enable us to find boundaries in parameter space over which dynamics change . We therefore seek to “classify” dynamics , rather than , for example , estimate quantitative features . Studying the database with classical statistics such as the joint distribution of the likelihood of seizure dynamics gives new insights into the model , but does not yield a comprehensive analysis . The final step is to fit the data with a statistical model . Here , we choose to use a tree approach , which cuts the parameter space into rectangular regions of different sizes and is amenable to high dimensional analyses . These regions are created with the aim that each one contains similar dynamics and so trees approximate the parameter space in a simple and interpretable way . Of course , we do not expect that the parameter space can be completely mapped to a set of rectangular regions , each containing homogeneous dynamical features . Some regions therefore contain dynamics with different features and the proportion of space in the region filled by particular dynamics is useful information . For example , one can ask whether certain regions contain a high density of seizure dynamics or exclude regions with certain features from further analyses . The statistical model captures defined characteristics of the mathematical model and summarises them in an efficient way , therefore facilitating the estimation of sensitivity of the dynamics to variations in a particular parameter . Thus critical , or important parameters for a given dynamics can be found . The extension of the Jansen-Rit model [19] introduced by Wendling et al . [41] considered in this paper has classically been used to study transitions to seizure dynamics . It is a neurophysiological model , i . e . one that has been built to understand interactions in nervous tissue at the macro- or meso-scopic level . It has previously been shown to display a repertoire of important dynamics which occur at ictal and inter-ictal states , for example in temporal lobe epilepsy [11 , 41] . The model is based on the assumption of the existence of four populations of neurons: pyramidal cells; excitatory interneurons; slow and fast inhibitory interneurons . The activity of each population is governed by the interactions between them . Each population is characterized by: The total potential of the pyramidal cell population is given by the aggregated contributions of the three feedback loops of inter-neurons connected to it . This is the output of the model , in analogy with recordings of electroencephalography ( EEG ) [42] . These interactions can be summarise in the following set of ordinary differential equations: z ˙ 1 ( t ) = z 6 ( t ) ( 1 ) z ˙ 6 ( t ) = A a S { z 2 ( t ) - z 3 ( t ) - z 4 ( t ) } - 2 a z 6 ( t ) - a 2 z 1 ( t ) ( 2 ) z ˙ 2 ( t ) = z 7 ( t ) ( 3 ) z ˙ 7 ( t ) = A a ( p + C 2 S { C 1 z 1 ( t ) } ) - 2 a z 7 ( t ) - a 2 z 2 ( t ) ( 4 ) z ˙ 3 ( t ) = z 8 ( t ) ( 5 ) z ˙ 8 ( t ) = B b C 4 S { C 3 z 1 ( t ) } - 2 b z 8 ( t ) - b 2 z 3 ( t ) ( 6 ) z ˙ 4 ( t ) = z 9 ( t ) ( 7 ) z ˙ 9 ( t ) = G g C 7 S { C 5 z 1 ( t ) - z 5 ( t ) } - 2 g z 9 ( t ) - g 2 z 4 ( t ) ( 8 ) z ˙ 5 ( t ) = z 10 ( t ) ( 9 ) z ˙ 10 ( t ) = B b C 6 S { C 3 z 1 ( t ) } - 2 b z 10 ( t ) - b 2 z 5 ( t ) ( 10 ) The biological meaning of the NMM parameters is given in Table 1 . As highlighted in the introduction the values of these parameters or their possible ranges are often based on previously used values that have sometimes arisen arbitrarily in the literature . As further experiments are conducted over time , it is possible to gain an improved insight into the range that NMM parameters could take . Examination of the experimental literature reveals that neuronal level mechanisms , which are often assumed to map to NMM parameters , can vary significantly from one species to another , as well as within species [29] ( neuroelectro . org ) . Therefore the plausible range of NMM parameters can be large . The parameters A , B , G , C and P have traditionally been considered to be highly uncertain and dynamics have therefore been studied over substantial ranges of these parameters [11 , 19 , 43] . In contrast , the membrane time constants a , b and g have often been considered as relatively certain [11 , 19] . However , experimental studies point towards the contrary . For example , there is a large uncertainty of dendritic time constants of the somatic response due to synaptic input for single neurons [44 , 45] . Ranges for these values have been shown to be large , from 25 s−1 [46] to 140 s−1 [47] for pyramidal neurons . Similarly , the membrane time constant of inhibitory neurons ( related to b ) could also be considered uncertain , with values ranging from 6 . 5 s−1 to 110 s−1 [48] . We use these experimentally determined ranges for values of a and b in our study ( see Table 2 ) . It is more difficult to find a plausible range for g; values used can be traced back to 1993 [49] , in which the authors indicated a large uncertainty . We therefore implement a large range for this parameter ( 350 to 650−1 ) . C was previously fixed at 135 [19] based on interesting dynamics occurring near this value . Here , we chose to use the initial range of uncertainty in [19] from 0 to 1350 . v0 was considered uncertain in previous studies and has also therefore been examined across a range of values , for example 2 to 6 mV [19] . Here , we extend the study from 2 to 10 . e0 is often fixed at 2 . 5 s−1 but a range from 0 . 5 to 7 . 5 s−1 has been recorded [24] , and therefore we use this range . Finally there is very little information about r , the value was found experimentally , but without information regarding uncertainty [23] . We therefore studied the range of this parameter from 0 . 3 to 0 . 8 mV . A summary of ranges of parameter values implemented in our study is given in Table 2 . The NMM was simulated 2 , 000 , 000 times varying 11 parameters A , B , G , P , a , b , g , C , v0 , e0 and r using a latin hypercube design to explore the parameter space . The simulations were computed using ODE45 in MATLAB ( Runge–Kutta method ) . Each time , 20 seconds of EEG activity were simulated , the first 10 seconds were removed to eliminate transients . Simulations were performed in parallel over 4 CPUs each running at 3 . 5 GHz . It took approximately 4 days to simulate the whole data base ( i . e . 2 , 000 , 000 simulations ) . We are interested in understanding the relationship between parameters of the NMM and its dynamics . This understanding can be achieved through an explicit mapping between regions of parameter space and qualitatively different dynamics ( e . g . steady states and oscillations ) . Previous studies have analysed the dynamics of NMMs by characterising features of simulations . Different properties of dynamics have been used for characterisation , such as the power spectrum [11] , amplitude or variance [51 , 52] and more nuanced features such as the number of spikes within a period of a specific rhythm [32 , 33] . These studies have demonstrated that NMMs can recreate key types of epileptiform dynamics such as slow spike-wave rhythms and theta spikes , which are important rhythms for generalised and focal epilepsies , respectively . Based on these previous studies , we consider three key features of simulations that are relevant for delineating different types of dynamics within the NMM: amplitude , frequency and number of peaks per cycle . We use these features to classify regions of parameter space according to the nature of the emergent dynamics . For example , alpha activity corresponds to low-amplitude oscillations with a frequency of around 10Hz . Alternatively , seizure dynamics in this model correspond to low-frequency oscillations ( 2-8Hz to take into account focal and generalized seizure activity ) with additional peaks that correspond to “spikes” or “poly-spikes” in EEG ( c . f . Fig 3 ) . We formalise this idea by denoting P ( Y|X ∈ R ) as the probability of the dynamics Y given that the parameter set X belongs to region R , which is a hypercube subset of the full parameter space . For example we could have: Y = { 0 , if the dynamic state of interest is seizure dynamics ; 1 otherwise . and a region defined , for example , by X ∈ R = X ∈ ( ( A < 5 ) ∩ ( 15 < B < 30 ) ) . In this case P ( Y = 1|X ∈ R ) represents the likelihood of observing seizure dynamics when the parameter A is inferior to 5mV , B is between 15 and 30 mV and the other parameters are not constrained . The value of P ( Y = 1|X ∈ R ) is given by P ( Y = 1 | X ∈ R ) ) = ∫ x ∈ R P ( Y = 1 | x ) d x ( 11 ) Since the function mapping X onto Y is unknown , we take a sampling approach and use the database created by the simulations defined in the section Wendling model . We can therefore estimate P ( Y = 1|X ∈ R ) for the given region R by P ^ ( Y = 1 | X ∈ R ) = 1 | χ | ∑ χ y . ( 12 ) where χ = {x|x ∈ R} and |χ| denotes the cardinality of the set . P ( Y = 1|X ∈ R ) can be further used to determine which parameters are important to find certain dynamic regimes . An obvious question that arises is how to choose R . A first approach consists of fixing the regions Ri∈[1:m] such that each region has the same size , i . e . the parameter space is cut into pre-defined regions . Another approach consists of partitioning the parameter space by selecting M “optimal” regions , R1 , … , RM . By optimal , we mean the number of regions M is as small as possible such that in each region the discrepancy of the event Y is low . In principle , this results in a more efficient mapping of the dynamics of the model onto its parameter space . Furthermore the boundaries between regions are useful , as they indicate which parameters have an important role in the emergence of dynamics of interest . Effectively they describe the transitions between different dynamic types that can correspond to bifurcations or other types of phase transition in the underlying dynamic model . To define optimal regions , we use an approach called decision tree learning algorithms [53] . Here , the parameter space is partitioned recursively into rectangular disjoint subspaces . The size of each region is determined by ensuring that it consists , as far as possible , of only a single type of dynamics . Tree-based methods are a conceptually simple , yet very powerful tool to study highly nonlinear functions for the purpose of regression or classification . These methods are inherently non-parametric; no assumptions are made regarding the underlying distribution of parameter values . They can be trained quickly and also provide a vehicle to efficiently predict the output of new simulations . We focus on classification and regression tree ( CART ) algorithms [53] . These produce binary splits recursively from the root ( the complete parameter space ) , to its leaves ( the regions corresponding to dynamics of a single type ) . In general , finding the optimal partitioning of parameter space is a NP-complete problem [54] . Therefore , decision tree learning algorithms are based on heuristics whereby locally optimal decisions are made within each region of the tree . Whilst such an approach is not guaranteed to give the globally-optimal decision tree , CART methods have been shown to give good results in practice [53] . Here , we summarise the approach , which is described in detail in [39] . Formally we have a data set consisting of n points in R p , xij where i ∈ [1: n] and j ∈ [1: p] . The set of outputs consists of the class of observed dynamics Yi of each simulation . Suppose that we have a partition into M regions , R1 , … , RM . For a given region Rm the splitting stage is chosen by finding an optimal split point in terms of the impurity criterion described Eq ( 14 ) . We seek the jth split parameter and split point , s , such that the cost function: argmin { s , j } I R L ( m , j , s ) + I R R ( m , j , s ) ( 13 ) is minimised . Here RL ( m , j , s ) = {x|x ∈ Rm , x . j ≤ s} and RR ( m , j , s ) = {x|x ∈ Rm , x . j > s} are respectively the potential left and right split of the region of interest . The measure I R m of region impurity represents the quality of classification in a region . By this we mean how well a region of parameter space maps onto model dynamics of a single type . It is defined by the Gini index I R m = 1 N m ∑ k = 1 K p ^ m k ( 1 - p ^ m k ) ( 14 ) where p ^ m k = 1 N m ∑ x i . ∈ R m I ( y i = k ) ( 15 ) is the proportion of class k observations in a given region Rm . When IR = 0 the region is pure , and there is only a single class of dynamics . By contrast a large Gini index indicates a region with large impurity , and thus contains parameters that map onto different types of dynamics in the model . For each region , the determination of the split points can be done very quickly ( o ( p × n ) operations ) and hence by scanning through all of the inputs , determination of the best pair ( j , s ) is feasible in finite time . An example of a tree and its construction can be found in Fig 4 . To estimate a new set of parameters x , the class with the largest frequency k ( m ) = argmax k p ^ m k is attributed to x . Furthermore P ^ ( y ∈ k ) = p ^ m k .
2 , 000 , 000 simulations were computed on the whole parameter space as described in section Wendling model , using the ranges in Table 2 . Analysing these simulations , we found that the dynamics of the Wendling model can predominantly be categorised as steady state ( 62 . 3% of parameter space ) . The remaining simulations were classified by frequency and number of peaks ( see Fig 3 for a description of dynamics ) . The dynamics ‘spike and wave’ or ‘poly-spike and wave’ , which are characteristic of seizure dynamics , represent 5 . 8% of the parameter space . This number can be considered as the likelihood to find seizure dynamics when random parameters are used . Fig 6 provides a 2-dimensional representation of the distribution of steady state and seizure dynamics throughout the whole parameter space . It allows us to detect the combination of parameters that are particularly prone to producing seizure dynamics or steady states in the model . It can be seen that the parameter subspace in which seizure dynamics can be found is large and is not concentrated in small sub-regions . The top right of Fig 6 demonstrates that seizure dynamics can be observed across most parameter values; there are few combinations of two parameters for which , regardless of other parameter values , seizure dynamics cannot exist . Examples are the inverse mean time in the excitatory and slow inhibitory loop ( a and b ) , which give rise to dark blue regions in Fig 6 ( low likelihood of seizure dynamics ) . Specific combinations of parameters A , B , C , e0 or r can also preclude seizure dynamics . In contrast , the subfigures for parameters of the fast inhibitory loop ( G and g ) appear quite homogeneous , and therefore do not change the likelihood of seizure dynamics . However , varying the value of some parameters does reduce the likelihood of observing seizure dynamics: reducing parameters of the excitatory loop ( A and a ) ; the connectivity coefficient ( C ) ; the maximum firing ( e0 ) ; and the inflexion point ( v0 ) and the slope ( r ) of the sigmoidal nonlinearity . On the other hand , increasing the inverse mean time in the slow inhibitory loop ( b ) also reduces the probability of observing seizure dynamics . Intermediate values of the input ( P ) and the average slow inhibitory gain ( B ) increase the chance of observing seizure dynamics . Particular combinations of pairs of parameters such as the average synaptic gains ( A and B ) or the inverse time scales ( a and b ) can significantly alter the chance of observing seizure dynamics . For example , there is a linear combination of a and b for which the proportion of dynamics in the seizure class is greater than 30% . The lower triangle of Fig 6 indicates that steady state dynamics can be observed in a very large proportion of the parameter space . It can be seen that small values of A or C force the system to be at steady state . Explorations such as undertaken in Fig 6 are informative and give a good preliminary indication of the role that each parameter plays in constraining the model dynamics . Nevertheless , in more than two dimensions , visualisation becomes difficult . For example , extending Fig 6 to 3 dimensions would require 1 , 000 2D plots . Therefore , we used tree statistics ( see section Building a tree ) to efficiently summarise how a change in a parameter can impact the dynamics of the model . Fig 7 presents one such tree that describes the segmentation of parameter space according to the density of seizure dynamics . Recall , that for each branch the tree algorithm scans through the sub-parameter space to identify the optimal separation between the maximum and minimum likelihood of observing the feature of interest ( seizure dynamics in this case ) . Fig 7 is a relatively small tree used to illustrate the method . At the root of this tree , the first parameter used to partition parameter space is the inverse time scale of the slow inhibitory loop , b . b ≥ 60 reduces the probability of observing seizure dynamics and produces a region that represents 49% of the parameter space . This region , which represents nearly half of the parameter space , contains only 10% of all parameter combinations that lead to seizure dynamics . Taking b < 60 again yields approximately half of the total parameter space ( 51% ) , but this region contains 90% of all parameter combinations that lead to seizure dynamics . Since this region is large , and the probability of observing seizures in the whole space is low ( 5 . 8% ) , the density of seizures in this region is low at 10% . The next branch cuts through the average slow inhibitory gain at B = 32 . Above this value , 19% of the parameter space remains and this contains 12% of all parameter combinations that yield seizure dynamics . The remaining 32% of parameter space accounts for 78% of seizure dynamics . Choosing A ≥ 2 . 1 further increases the density of seizure dynamics to 17% , incorporating 73% of all parameter sets that lead to seizure dynamics . Further adding the criterion that v0 ≥ 4 . 8 leads to a region with highest density of seizure dynamics ( bottom right region in Fig 7 ) . This region represents 15% of the total parameter space and the proportion of seizure dynamics in this region is 22%; thus it accounts for 57% of all parameter combinations that result in seizure dynamics . However it is possible to create larger trees with more regions giving a finer resolution . There is of course a trade-off as larger trees segment the parameter space into more ( smaller ) hypercubes , making them more cumbersome to analyse ( see supplementary materials S1 and S2 Figs for more examples ) . The main conclusion to be drawn from the large tree presented in the supplementary material is that the dependency of dynamics on parameter space is complex: transitions between dynamics can vary between regions . For example , an increase of B or P can either increase or decrease the likelihood of seizure dynamics . However , other parameters exhibit robust transitions; a split at r around 0 . 52 appears consistently , and e0 and v0 tend to slightly increase the seizure likelihood when their values increase . Fig 6 seems to show different results from [11] . To recall; in [11] , the presence of seizure dynamics would appear only for B superior to 20 and A superior to 5 ( other parameters at standard value as in Table 2 ) . In contrast , in our Fig 6 the likelihood of seizure when B is superior to 20 is low and higher for small values of B . These results illustrate that although a projection of the parameter space in 2-dimensions is helpful to gain a quick understanding of the parameter space , it does not capture all of its aspects . In Fig 7 , with the help of the tree algorithm , the manifold is well approximated . Indeed one can see that even for large B ( B > 32 ) seizure dynamic can appear with the standard values of parameters [11] ( fourth leaf from the left ) . Furthermore the tree shows that this change appears around A = 5 . 4 . There are other , larger manifolds in Fig 7 at small values of B . These manifold are the ones which influence Fig 6 the most and “mask” the results of [11] . To generalise the example of Fig 7 , we computed the variable importance of model parameters over a random forest of 100 trees . Clearly , the importance of a parameter depends on the characteristics we are interested in . Results regarding the presence of steady states , oscillations with different amplitudes and frequencies , as well as seizure dynamics are provided in Table 3 . We find that A , B , C and v0 are important parameters for transitioning between steady state dynamics and the different types of oscillations . Interestingly , the amplitude of oscillations was less dependent on A and instead strongly dependent on C and e0 . This might seem surprising given the importance of A in observing oscillations in the first place . This contrast demonstrates how the relative importance of a parameter is strongly dependent on the observed feature of interest ( e . g . frequency vs the amplitude ) . The input from other regions of the cortex ( P ) can affect the emergence of oscillations but has a marginal role in tuning the amplitude and the frequency of these oscillations . The connectivity constant ( C ) is important for governing the amplitude but not the frequency of an oscillation . In fact , few parameters ( A , B , a , b and G ) are important for determining the frequency of oscillations . All parameters except g were found to play a role in the generation of transitions between dynamics , but with varying importance . The frequency of an oscillation was found to be predominantly dependant on the inhibitory slow loop parameters ( B and b ) . These parameters were also found to be crucial for producing seizure dynamics . This observation confirms the finding in Fig 7 that when these parameters split the space they reduce impurity . Overall the excitatory pair of pyramidal and excitatory interneurons and the slow inhibitory loop are important to create oscillations in the Wendling model . The output of the Wendling model is sensitive to a change of any of these parameters as indicated by the NVI measurements ( Table 3 ) . Fig 6 demonstrated a potentially important relationship between the parameters A and B and the parameters a and b . We further investigated this by incorporating two artificial parameters rA/B and ra/b which are respectively the ratio of A over B and the ratio a over b . Fig 8 shows that smaller values of ra/b lead to a lower likelihood of observing seizure dynamics . A ratio less than 1 . 6 gives a likelihood of observing seizure dynamics of 0 . 56% in a very large sub-region that contains 57% of the parameter space . At the opposite extreme , the region on the right of the figure contains 40% of all seizure dynamics in only 5% of the whole parameter space . In this region the proportion of seizures is nearly 50% . It is interesting to note that low values of ra/b reduce the likelihood of seizure dynamics , whereas for rA/B , small values ( <0 . 19 ) or large values ( >1 . 5 ) reduce the likelihood of seizure dynamics . A more highly resolved version of this tree can be found in supplementary materials S2 Fig . We recomputed the NVI , incorporating these two new parameters over a random forest of 100 trees . The results are in Table 4 . It is clear that for steady state transitions or frequency of oscillations rA/B is the most important , whereas ra/b is most important for transitions to seizure dynamics . Aside from the amplitude of oscillations , the normalised variable importance of the ratios rA/B and ra/b are larger than for the parameters taken individually .
In this study , we introduced a new approach to explore the parameter space of high dimensional NMMs . In contrast to classical studies that considered parameters individually , or in pairs , we used a random forest approach in order to study the entire parameter space simultaneously . Our approach relies on the creation of a database of dynamic features derived from forward simulations . Other statistical approaches could be used to study the database , but they all suffer from particular deficiencies . For example , artificial neural network models have a vast number of hyperparameters that cannot be interpreted [66] . Support vector machines [67] result in boundaries between regions of parameter space that are not split according to single parameters , and therefore one has to integrate over all parameters to understand the importance of each . Kernel methods such as Gaussian processes [68] rest upon the assumption of “smoothness” of data , i . e . , proximal parameter sets are assumed to yield similar simulations , which is clearly not the case close to bifurcations . Another approach combines trees and Gaussian process [69 , 70] , but that approach requires prior assumptions on parameters , limiting its use when this information is not to hand . In contrast , the approach we employed provides an efficient way to study the influence of model parameters on their dynamics: trees are computationally fast , make no a priori assumptions on either the type of model or parameter values , and can handle data that are represented on different measurement scales [71] . We thereby demonstrated that random forests are a useful tool to study the dynamics of NMMs . The implementation of the random forest approach [56] , overcomes the issue that each implementation of CART produces a single tree that is locally optimal . A drawback is that the random forest approach introduces some loss of interpretability , but the final solution is more representative of the global optimum . This is particularly important for the NVI which measures the relative contribution of parameters to an observed dynamic feature of interest of the model ( e . g . a steady-state , oscillation or spike-wave ) . By this we mean that effectively , the NVI indicates which parameters are critical for segmenting the total parameter space into regions in which a feature of interest is more or less likely to be observed . Further , the NVI provides a principled approach for determining whether or not parameters can be fixed , hence reducing the number of parameters to be calibrated from observable data . A consistently low NVI across all features of interest means that the considered parameter plays little role in any dynamic change and can therefore be fixed to an arbitrary value within a given physiological range . For example , in our study of the Wendling model , g has little effect on determining transitions from steady-state to oscillations , or in determining the amplitude and frequency of those oscillations . It can therefore be fixed , meaning that the parameter space explored in subsequent calibration is smaller . On the other hand , some parameters have a high NVI for specific features of interest , and are therefore important for observing that specific feature without playing an important role in altering other aspects of the dynamics . For example , e0 is critical in determining the amplitude of oscillations , but plays a marginal role in the appearance of other features . Therefore if amplitude is not a particular feature of interest , e0 could be fixed . When considering networks of dynamical systems , the number of parameters can rapidly become very large , so NVI is an important tool for managing this increase in complexity . For example , one could use the framework presented herein to examine whether there are certain network structures in which certain edges can be given fixed weights , thereby reducing the dimension of an optimisation or calibration problem . The notion of importance is defined using NVI due to its robustness and ability to measure the influence of parameters on dynamics [72] . However , the notion behind importance is somewhat nebulous , and it is difficult to directly attribute a small difference in NVI to the relative importance of a specific parameter . The pragmatic approach we have adopted , is to consider parameters with values of NVI > 0 . 1 as playing a role in governing the feature of interest . In contrast , parameters with a NVI close to 0 can be disregarded . In the present study we defined importance specifically in the context of changes in parameters causing changes in asymptotic dynamics . This is relevant for the case in which bifurcations give rise to epileptiform activity . However , there are other possible model scenarios in which changes in dynamics could occur , such as for example , intermittency , bistability and excitability [73] . In these cases , we would seek to characterise importance with respect to changes in unstable invariant sets of the system , for example boundaries of basins of attraction . Furthermore , importance as we have defined it in the context of the NMM does not imply that a parameter is crucial for changes in dynamics at the individual level . For example , it might be necessary to model some seizures using transitions between dynamics that occur only in small regions of parameter space . It is important to highlight that in the random forest approach , other definitions of importance exist , such as the permutation importance or the conditional permutation importance [61] . However , these approaches suffer from lack of robustness [74] , hence our focus on NVI . Our analyses of the full parameter space of the Wendling model show that parameters of the slow inhibitory loop ( b and B ) play the most important role ( in term of NVI ) in the emergence of seizures . The time scale of the slow inhibitory loop ( b ) is the most important parameter; a small change in its value can transform steady state dynamics into seizure dynamics robustly , i . e . for the majority of combinations of other parameters in the model . We found the excitatory loop , governed by a and A , together with the offset of the sigmoid function ( v0 ) to be the second most important components of the system for the emergence of seizure dynamics . These are followed by the other parameters of the sigmoid function ( v0 and r ) and the parameter that scales connectivity between the different populations of neurons ( C ) . Interestingly , changes in the fast inhibitory loop ( parameters g and G ) do not play an important role in the generation of seizure dynamics . We note that a low value of NVI in the context of our study does not mean that a parameter is irrelevant to the emergence of other brain dynamics not captured by the choice of features . Furthermore , parameters with low NVI may play a role in determining transitions between dynamics in specific subsets of parameter space; NVI is purely a global measure . The parameters governing the magnitude of input from other areas ( P ) or the scaling of intrinsic connectivity ( C ) , for example were shown herein to have little ( global ) effect on the emergence of seizure dynamics , but in a priori constrained sub-regions have been shown capable of governing transitions in NMMs [7 , 8] . Table 4 shows a comparison of parameter importance when different features are considered . Parameters of the slow inhibitory loop , b and B , as well as the ratio of time scales ra/b , showed relatively high importance across all features . It is therefore possible that these parameters are important for transitions between dynamics in general . Verifying this will require exploration of additional features in model dynamics . We found that the ratio of parameters of the excitatory and inhibitory loops play an important role in the generation of all the features we considered , with the exception of amplitude of oscillations . The ratio of time scales ( ra/b ) is the most important factor governing emergence of seizures , whereas the ratio of gains ( rA/B ) is most important for the onset of cycles and the frequency of these cycles . Reducing ra/b robustly reduces the likelihood of seizures regardless of other parameter values ( see e . g . Fig 8 and supplementary materials , S2 Fig ) . rA/B on the other hand , presents an intermediate range of values that have highest likelihood of seizure dynamics . Our finding of the importance of rA/B for the emergence of seizure dynamics is in line with previous experimental observations . For example , [75] found that the ratio of Glutamine to GABA Levels is larger in people with idiopathic generalized epilepsies compared to healthy controls . This also aligns with the action of some antiepileptic drugs , for example those acting via modulation of neurotransmitters such as GABA [76] , the potentiation of which would be reflected in our model by an increase in B , and hence a decrease in rA/B . Furthermore , Our finding of the importance of rA/B for the emergence of seizure dynamics confirms previous modelling results [30] Interestingly , since the highest likelihood of emergent seizure dynamics was found to be for intermediate values of rA/B , this would suggest that , depending on the choice of other parameters , decreasing the ratio of excitation to inhibition could also produce a route into seizure dynamics , in line with evidence of the possibility of heightened inhibition at seizure onset [77] . Our finding that the slow inhibitory and excitatory synaptic gains have more influence than the fast inhibitory loop is in line with previous modelling results [33 , 34] , as are our findings that the parameter r and the ratio ra/b are important for dynamics of the NMM [35 , 78 , 79] . Few experimental studies have investigated the role that different time constants might play . However , it has been shown that chloride ion homeostasis is perturbed in patients with mesial temporal lobe epilepsy [80] , and intracellular chloride ion concentrations have been shown to play a role in the time constants of postsynaptic potentials [81] . This therefore presents a possible biophysical interpretation for the importance of ra/b . Interestingly , a recent study utilising dynamic causal modelling applied to a zebrafish model of seizures also demonstrated the potential importance of excitatory and inhibitory synaptic time constants [82] . In our study , we obtained these results using a method in which the influence of all parameters was analysed simultaneously and a complete characterisation of the relative importance of all parameters was possible . In fact , this analysis revealed new combinations of parameters that can potentially govern the emergence of seizure dynamics in the Wendling model , for example v0 . In addition , given our finding that the ratio ra/b is most important for seizure generation it would be interesting to explore the known effect of drugs that could target the inverse mean time ratio ra/b . [11] presented detailed , two-dimensional analyses of the effects that changing system parameters have on emergent dynamics . One of the findings of [11] was that seizure dynamics predominantly occur when B > 20 . However , our results ( Fig 6 ) show that the likelihood of seizures when B > 20 appears rather low ( but not zero ) and is in fact higher for small values of B . These results indicate that although a projection of the parameter space in 2-dimensions is helpful to gain a quick understanding of the system , it does not capture the global picture . In our Fig 7 , with the help of the tree algorithm , we did indeed find that for large B ( >32 ) seizure dynamics occur for the range of parameters used by [11] ( fourth leaf from the left in Fig 7 ) . Furthermore the tree shows that this change appears around A = 5 . 4 . However , our analysis in Fig 7 demonstrates that there are other regions of parameter space , for lower values of b that contain seizure dynamics . The approach presented herein relies on the construction of a statistical model of dynamics based on simulations . This means that we cannot uncover the dynamic mechanisms that govern the emergence of the features studied , for example the presence of unstable invariant sets or changes in stability . However , our approach could be combined with traditional methods such as numerical continuation [83]; we would first constrain parameter space by using NVI to identify the most important parameters , together with transition boundaries and then perform more detailed analyses therein . Studies including [73] and [84] describe four alternative mathematical mechanisms underlying the emergence of seizures: bifurcation ( a parameter is slowly varied so that the system crosses a bifurcation point ) , bistability ( backround and seizure attractors co-exist , with perturbations allowing transitions between the two ) , transient excitability ( the seizure dynamics occurs due to a complex trajectory elicited by a perturbation ) and intermittency ( background and seizure dynamics are part of the same attractor ) . In this study we have focussed on a detailed explanation of the bifurcation mechanism ( e . g . how small changes in system parameters can lead to abrupt changes in emergent dynamics ) . Specifically , we find for the chosen Wendling model that under the bifurcation assumption changes in the slow inhibitory loop or the excitatory loop are most likely to underpin the emergence of seizures . It is important to highlight that this finding is specific to the chosen model and further , that it does not exclude the other three possibilities . To explore the possibility of transient excitability and bistability , we would need to extend our statistical model to include system variables ( e . g . initial conditions ) and properties of perturbations as parameters . We investigated the impact of initial conditions by considering them as parameters and found their NVI to be close to zero , indicating that regions of bistability are small in the context of global changes in parameters . Another possible extension to the results presented herein would be to consider different dynamic models or different characteristic features of their dynamics . For example [85] or [86] focused their attention on the power spectrum of the model in comparison with clinically recorded data . Future work could focus on power spectra as a feature of interest , enabling an appropriate characterisation of the importance of parameters for generating alpha activity in NMMs . In summary , we presented a framework for the global characterisation of the dynamics of NMMs . Our methods have the potential to advance patient-specific model representations , for example by first determining the relative importance of parameters , and then reducing the parameter space to a subset in which model calibration from data becomes tractable . Such an approach will become increasingly important as the emphasis on networked dynamical systems of the brain increases . Here the number of model parameters grows rapidly , beyond the point for which established approaches such as Kalman filtering [15] or genetic algorithms [32] , that work directly with the dynamical system of interest , can be effective . | Understanding the workings of the healthy brain and the disruptions that lead to disease remains a grand challenge for neuroscience . Given the complexity of the brain , mathematical models are becoming increasingly important to elucidate these fundamental mechanisms . However , as our fundamental understanding evolves , so models grow in complexity . If the model has only one or two parameters , formal analysis is possible , however understanding changes in system behaviour becomes increasingly difficult as the number of model parameters increases . In this article we introduce a method to overcome this challenge and use it to better elucidate the contribution of different mechanisms to the emergence of brain rhythms . Our method uses machine learning approaches to classify the dynamics of the model under different parameters and to calculate their variability . This allows us to determine which parameters are critically important for the emergence of specific dynamics . Applying this method to a classical model of epilepsy , we find new explanations for the generation of seizures . This method can readily be used in other application areas of computational biology . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"medicine",
"and",
"health",
"sciences",
"engineering",
"and",
"technology",
"trees",
"neuroscience",
"simulation",
"and",
"modeling",
"decision",
"analysis",
"decision",
"tree",
"learning",
"systems",
"science",
"mathematics",
"statistics",
"(mathematics)",
"management",
"engineering",
"artificial",
"intelligence",
"plants",
"research",
"and",
"analysis",
"methods",
"epilepsy",
"computer",
"and",
"information",
"sciences",
"decision",
"trees",
"animal",
"cells",
"dynamical",
"systems",
"machine",
"learning",
"cellular",
"neuroscience",
"statistical",
"models",
"eukaryota",
"cell",
"biology",
"neurology",
"neurons",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"physical",
"sciences",
"organisms"
] | 2018 | Classifying dynamic transitions in high dimensional neural mass models: A random forest approach |
Genome reduction has been observed in many bacterial lineages that have adapted to specialized environments . The extreme genome degradation seen for obligate pathogens and symbionts appears to be dominated by genetic drift . In contrast , for free-living organisms with reduced genomes , the dominant force is proposed to be direct selection for smaller , streamlined genomes . Most variation in gene content for these free-living species is of “accessory” genes , which are commonly gained as large chromosomal islands that are adaptive for specialized traits such as pathogenicity . It is generally unclear , however , whether the process of accessory gene loss is largely driven by drift or selection . Here we demonstrate that selection for gene loss , and not a shortened genome , per se , drove massive , rapid reduction of accessory genes . In just 1 , 500 generations of experimental evolution , 80% of populations of Methylobacterium extorquens AM1 experienced nearly parallel deletions removing up to 10% of the genome from a megaplasmid present in this strain . The absence of these deletion events in a mutation accumulation experiment suggested that selection , rather than drift , has dominated the process . Reconstructing these deletions confirmed that they were beneficial in their selective regimes , but led to decreased performance in alternative environments . These results indicate that selection can be crucial in eliminating unnecessary genes during the early stages of adaptation to a specialized environment .
Bacterial genomes have the potential to rapidly change their size and content as a result of various mechanisms such as deletion , duplication and horizontal gene transfer . The net expansion or contraction at the genome scale is thus a function of both the rate at which these events occur and the subsequent filters imposed by natural selection and/or genetic drift [1] , [2] . Although most bacterial genomes have remained relatively constant in size due to an apparent overall balance of these forces [3] , distinct strains within a species can differ remarkably in gene content [4] . This finding has led to categorizing the genome into the core and accessory ( or auxiliary ) components , the former being present in nearly all members , and the latter being present in only a subset of strains [5] . The population biology and selective environment of microbes each contribute to the tempo and mode of genomic change . Of primary importance is the effective population size ( Ne ) of a species , as this influences the efficacy of selection versus drift . Repeated bottlenecks , such as those experienced by intracellular endosymbionts ( which also participate in little , if any horizontal gene transfer ) , result in tremendous rates of sequence change and ineffective selection to maintain functions required for host-independent lifestyle . This often leads to loss of many genes that are essential for the free-living microbes and massive genome shrinkage ( ex: 77% in the intracellular symbiont of aphids , Buchnera aphidicola and genomes as small as Hodgkinia cicadicola ( 144 kb ) ) [2] , [6]–[9] . On the other hand , simply living on a restricted set of resources in a relatively constant environment can also result in reduced genomes despite very large Ne , such as observed for the plankton Prochlorococcus and Pelagibacter [10] , [11] . For these it has been suggested that the major force driving genome reduction is streamlining , defined as when “selection acts to reduce genome size because of the metabolic burden of replicating DNA with no adaptive value” [11] . In addition to DNA synthesis , deletions also eliminate producing the RNA molecules and proteins encoded by that region . Beyond external factors , the genomic structure of microbes and mechanisms of gene gain and loss make it possible for large regions to come and go in single events . Accessory genes are disproportionately found on extrachromosomal replicons that are subject to potential loss . Alternatively , even when present on the main chromosome accessory genes are often found as discrete genomic islands disrupting an otherwise syntenic chromosome between strains in a species . This can result in gains or losses via various mechanisms such as homologous or site-specific recombination and phage integration/excision [12] , resulting in punctuated large-scale gene content changes . Large-scale reductions of accessory genomes via these events may be a critical mechanism in early stages of genome shrinkage . Although either drift or selection could contribute to genome reduction observed in nature , we lack direct evidence to distinguish between the lack of purifying selection to maintain the genes lost versus positive selection for their loss . Genomic analyses of chronic infections , such as Pseudomonas aeruginosa in cystic fibrosis patients , have repeatedly observed large deletions [13]–[15] . This rapid loss of genomic islands could simply be due to high rates of recombination and drift ( or hitchhiking ) . Alternatively , the instability of the accessory genome could be due to selection , either for reduced genome length ( i . e . , streamlining ) or beneficial gene loss , such as has been shown for Shigella flexneri , a facultative intracellular pathogen of primates [16] . Laboratory-evolved populations of bacteria present the unique opportunity to address the forces involved in genome reduction under selective regimes that tilt the relative efficacy of selection versus drift . ‘Mutation accumulation’ experiments purposefully use single-colony bottlenecks at each transfer to maximize drift [17] . In contrast , the more typical experimental evolution regimes maintain an Ne often in the millions , allowing selection to dominate [18] , [19] . To date , genome reductions found in the above experiments have tended to be modest ( up to 4% of the genome in mutation accumulation experiments , and 1% in larger populations ) . The DNA loss rates observed have been low ( ∼2 bp per generation ) , and the regions lost have largely been inconsistent across lineages . With the exception of small ( 1 . 6–7 kb ) deletions in the ribose gene cluster of Escherichia coli [20] , none of these genome reductions have been tested for their fitness effects . As such , it remains unclear whether these observed genome reductions imparted an advantage in the selective environment , whether fitness effects scale with the length of DNA removed , and/or whether such events generate tradeoffs across other environments . Here , we used experimental evolution to investigate the role of large-scale deletions in adaptation and specialization . We evolved populations of the α-proteobacterium Methylobacterium extorquens AM1 , a member of the dominant genera found on leaf surfaces [21] , [22] . Like other bacteria that utilize single-carbon ( C1 ) compounds ( e . g . , methanol ) as growth substrates , M . extorquens AM1 has also specialized to grow on a very limited array of multi-C compounds ( e . g . , succinate ) , and has been a model for exploring rapid metabolic specialization during adaptation [22] . Across 32 populations evolved for 1500 generations in one of four different nutrient regimes we found 80% of these deleted the same genomic region that encompasses up to 10% of the genome . By reconstructing these deletions under the ancestral genetic background we have demonstrated that they rose in frequency due to selection; however , the advantage gained was not a generic effect of shortening genome length , but was specific to the region lost and imparted an advantage ( or disadvantage ) that depended upon the environment .
In order to examine the potential role of large-scale deletions in adaptation of M . extorquens AM1 we analyzed genome content from replicate evolved populations . Eight parallel populations were grown at a large Ne ( ∼2 . 5×108 ) in each of four different nutrient regimes ( 32 populations in total ) : methanol , succinate , mixture of methanol and succinate , or alternating between methanol and succinate . After 1500 generations , the evolved populations increased fitness in their selective environments by 15 to 37% compared to their wild-type ancestor ( Figure S1 ) . As reported previously [22] , a couple of these strains were actually less fit than the ancestor , which likely represent genotypes that exist due to frequency-dependent interactions such as cross-feeding . For comparison , we also maintained 10 lineages on solid medium for 1500 generations that we transferred through single-cell bottlenecks to maximize the strength of drift . To determine the extent to which large-scale deletions contributed to adaptation , we used comparative genomic hybridization ( CGH ) to uncover chromosomal changes in 44 isolates from the 32 evolved populations ( Table S1 ) . Like many bacteria , the 6 . 9 Mb genome of M . extorquens AM1 has multiple replicons of varying sizes ( 5 . 5 Mb chromosome , 1 . 3 Mb megaplasmid present at one copy per chromosome , and 3 plasmids between 25–44 kb present between 1–3 copies per chromosome ) [23] , a total of 23 distinct deletions were identified , some of which in more than one lineage ( Table S2 ) . Over 91% of the deletion events were due to homologous recombination between matching sequence regions , and of these , 86% were between co-directional pairs of one of the 142 insertion sequences ( ISs ) present in the genome of M . extorquens AM1 [23] . Most notable were the extensive , repeated changes to the megaplasmid: 36 of the 44 isolates screened by CGH contained deletions spanning a single region that ranged from 23 kb to 641 kb ( Figure 1A ) . The largest of these deletions removed 24 . 7% of the accessory genes ( unique to M . extorquens AM1 versus strain DM4 ) [23] and 2 . 7% of shared , core genes . This represents the largest parallel losses observed during laboratory adaptation thus far . Previous experiments either observed an occasional large deletion ( 200 kb ) [17] or repeated loss of small regions ( <7 kb ) [20] . Applying a PCR-based screen to 56 additional isolates revealed 51 more with deletions in this region ( Table S1 ) . Despite this overall parallelism , the precise borders of these deletions were somewhat different . We broadly classified these into three classes of deletion types ( DT1 , 2 , and 3; Figure 1A ) . A DT1 event with borders precisely at a co-directional pair of ISs had been independently identified by genome re-sequencing of a methanol-evolved isolate from a population initiated with a different starting genotype [24] . These deletion types were present at significantly different proportions across the four nutrient regimes ( Figure 1B ) . Moreover , since distinct subtypes of deletions coexisted in some populations at changing frequencies ( Figure 1B and Text S1 ) , the larger deletions may have occurred stepwise , as proposed for similar events in the genomes of B . aphidicola strains [25] . The observed parallelism across replicates could be due to either an unusually high rate of occurrence and/or a selective advantage conferred by the events themselves . For example , the parallel deletions of regions of the ribose operon of glucose-evolved E . coli were shown to partly depend upon a high rate of transposition and subsequent recombination [20] . To address this possibility , we examined the 10 populations transferred through single-colony bottlenecks for 1500 generations . None of the defined deletion types were detected by PCR , which is significantly unlikely to be observed given the rate they appeared in the large Ne populations ( P<0 . 0001 ) ( Text S1 ) . In order to directly test for a possible selective advantage of these deletions , we reconstructed deletions in the wild-type ancestor and tested whether these were individually beneficial in their selective environments . We created four deletions that represent the largest class found ( engineered type 1 , ET1 ) , the half of ET1 that was commonly lost ( ET2 ) , a small region , itself only observed once , at the intersection of all identified deletions ( ET3 ) , and a fourth region ( ET4 ) that , although never observed to be lost in the evolved populations , removed the alternative half of DT1 and was equivalent in length to ET2 ( both ∼300 kb ) ( Figure 2A ) . With the exception of ET3 in the methanol/succinate switching environment , all deletion types were individually beneficial in the selective environments that they were observed in , with up to a 15% selective advantage for ET1 in succinate medium ( Figure 2B ) . The nearly neutral fitness effect of ET3 indicated that the beneficial effect was not due to removing this shared region . Interestingly , the fitness effect of ET1 was approximately the same as expected from the two half deletions ( ET2 and ET4 ) , suggesting that there is little epistasis between these two regions ( Figure S2 and Text S1 ) . Two lines of evidence refuted the hypothesis that the physiological basis of the fitness advantage of the large-scale deletions was simply due to a shorter genome , and rather suggested that loss of specific gene ( s ) was the primary benefit . The streamlining hypothesis that genome reduction is driven by metabolic efficiency of a shorter genome would predict that: 1 ) the magnitude of benefit would scale with size of the deletion and 2 ) the benefit would be reasonably similar across multiple environments . First , we found that selective advantage did not correlate with deletion size . This is most clearly demonstrated by comparing ET2 and ET4 , which have equivalent lengths . These two ∼300 kb deletions exert quite different effects , whereas ET1 ( which is twice as large ) and ET4 behaved quite similarly . Second , we found that , although the marginal benefits of ET2 and ET3 were relatively constant across different growth substrates , but the phenotype of ET1 and ET4 varied markedly . This included being a disadvantage during growth on methanol when transferred from succinate , which appears to be due to a longer transition time between nutrients and decreased fitness during stationary phase on succinate ( Figure 2C , 2D and Figure S3 ) . This result is in consistent with a recent report where no correlation between genome size and selection intensity was found across a variety of natural isolated bacteria [26] . The high prevalence of observing DT1 in populations evolved in succinate and the methanol-succinate mixture is in accord with the above phenotypes , but other factors such as epistatic interactions with previous mutations may account for the surprisingly high frequencies of DT1 in methanol and low frequency in methanol/succinate switching environments . Given that the various deletions appeared ( above the limit of detection ) in the second half of this 1500 generation experiment , other mutations would have already been present that may alter the selective effect of these losses ( Text S1 ) . Although the large-scale deletions from the megaplasmid of M . extorquens AM1 were beneficial in the laboratory environment , further tradeoffs suggest that loss of this region would have consequences in natural environments . Unlike regular plasmids , megaplasmids and minichromosomes ( or ‘chromids’ ) are long-term replicons residing in more than 10% of bacterial genomes across markedly different life styles [27] . The GC% , coding density and the percentage of repeat region of the megaplasmid in AM1 are compatible with the main chromosome but very different from the other 3 small plasmids present in this genome , indicating its long-term existence in this strain . We first examined the potential functions of the megaplasmid by COG analysis , which showed an overrepresentation of genes related to metabolic functions in deleted regions ( X2 test , P<0 . 0001 , Figure S4 ) . This is consistent with the observation of other reduced genomes [7] . Furthermore , the predicted functions of the genes encoded within the deleted regions were consistent with a lack of essentiality due to duplicate copies of potentially essential genes on the main chromosome . Indeed , with a fairly liberal definition of homology ( minLrap≥0 . 8; maxLrap≥0; identity≥60% ) there are 159 genes on the megaplasmid that have the homologs on the main chromosome . As 113 of these are found in DT1 , a significant overrepresentation relative to the rest of the megaplasmid ( X2 test , P<0 . 0001 ) , and all regions with synteny with the chromosome for ten or more genes are located in the deleted region , some of the benefit may have come from removing these possible redundancies . On the other hand , many genes putatively involved in stress responses would have been lost ( Table S5 ) . We therefore tested the deletion types across a panel of stresses , revealing that ET1 and ET4 had decreased resistance to ampicillin and arsenate ( Figure 2E and 2F ) , and increased growth at the upper end of the temperature range of M . extorquens AM1 ( Figure 2G ) . The loss of two sigma factors ( Figure S5 ) and genes shown to be involved in leaf surface colonization [28] by these deletions lends further support that some of these genes contributed to the ecology of this strain . These data have provided a rare opportunity to demonstrate that selection for gene loss contributed to the repeated , large-scale removal of accessory functions from adapting genomes . The selection regime we applied was a seasonal environment of growth and starvation , but since only one or two supplied resources and all other environmental factors were held constant , this rendered many functions unnecessary . It is quite common for plasmids bearing antibiotic resistance genes or toxins to be lost when these functions go unrewarded; however , this process differs substantially what is described here in terms of the scale of genome change , the presence of genes that would be essential if it were not for a duplicate copy , as well as the mechanism of loss ( unfaithful segregation vs . homologous recombination ) . Given that so many genes were lost in these deletion events , future work will be required to pin down whether few or many genes contribute to the observed phenotypes and by how much . Numerous stress response genes were lost in these events , and it has been in multiple cases that there can be tradeoffs between growth capacity and stress response in environments ranging from chemostats [29] to long-term stationary phase [30] . Similarly , the deletions that sped growth on most substrates led to an impaired capacity to deal with nutrient switches , starvation , and the toxic effects of an antibiotic and a toxic metal . Although these large-scale losses were successful due to the benefit they conferred in the flask , it is quite likely that they would impart tradeoffs in components of the natural environment inhabited by Methylobacterium . Selection-driven loss of accessory genes can rapidly limit the niche of a given lineage , resulting in restricted lifestyle and lowering both Ne and access to horizontal gene transfer with other members of the species . Indeed , aspects of our laboratory conditions and starting strain - a sudden restriction in niche breadth and now unnecessary accessory functions present in contiguous islands - commonly occurs in natural environments , such as the establishment of chronic infections by opportunistic pathogens where analogous deletion events have been identified [13]–[15] . Smaller , more isolated populations in which purifying selection for previously useful functions is absent can lead to further genome reductions as drift becomes increasingly relevant . Thus , although in other scenarios bottlenecks leading to loss of functions via drift could initiate specialization , our results emphasize the potential for selection-driven , large-scale deletions of unnecessary genes as a route towards a limited niche and the beginning of a path leading to further genome changes .
This paper examines isolates from 32 populations that were founded from two nearly isogenic strains of wild-type Methylobacterium extorquens AM1 , CM501 and CM502 , which have pink and white colony color , respectively [31] . These populations evolved in four different environments each with 8 replicates ( odd numbers founded by CM501; even by CM502 ) : methanol ( M , 15 mM , ‘A’ populations ) , succinate ( S , 3 . 5 mM , ‘B’ populations ) , a mixture ( MS ) of methanol ( 7 . 5 mM ) and succinate ( 1 . 75 mM ) ( ‘C’ populations ) , and alternating ( M/S ) between methanol ( 15 mM ) and succinate ( 3 . 5 mM ) ( ‘D’ populations ) . The general selective regime , minimal medium and culturing conditions utilized were described previously along with the initial examination of the dynamics of adaptation and specialization of the A and B populations [22] . The C and D populations were evolved in the same conditions except for the mixed or alternating substrate conditions . Briefly , populations were grown in 9 . 6 mL of medium and cultured at 30°C in 50 mL flasks with 225 rpm shaking . Serial transfers were performed every 48 hours using 1/64 dilutions ( i . e . , 6 generations ) with a population size at the end of each cycle of ∼2×109 . Three or four evolved isolates were obtained from generation 1500 of each population with preference for different colony morphologies , where apparent . From each population , one or two isolate ( s ) were chosen to test in genomic microarray analysis , and the remaining colonies were screened for deletions via PCR ( Table S1 ) . Ten mutation accumulation lines were founded by CM501 and prorogated at 30°C on solid media comprised of half nutrient agar and half ‘hypho’ agar containing succinate ( 7 . 5 mM final concentration ) [31] to allow rapid colony formation . For each lineage , every 3 . 5 days the last colony on the streak line was picked as a random sample and streaked on a new plate . The population expanded from one cell to approximately 106 cells in a colony each passage , representing ∼20 generations , and was repeated 75 times ( ∼1 , 500 generations ) . DNA isolation was performed using the Wizard Genomic DNA Purification Kit ( Promega , Madison , WI ) following the manufacturer's protocol . Deletions in evolved strains were identified using comparative genomic hybridization arrays performed by MOgene Inc . ( St . Louis , MO ) , a certified Agilent service provider . The custom arrays spotted with the ancestor genome were designed , printed , and probed as described [32] . Without the necessity to detect quantitative signals , each sample was labeled with either Cy3 or Cy5 and hybridized once with a sample labeled with the other dye . In total , 25 hybridizations were done for 45 samples ( 44 evolved strains and the ancestor ) , including three control experiments ( Table S2 ) . To confirm each deletion , one primer outside ( p1 & p4 ) and inside ( p2 & p3 ) that region was designed for each side . The deletion was confirmed if fragment was amplified by p1 & p4 but no product was amplified by p1 & p2 or p3 & p4 . For fragments shorter than 1 kb , exact junctions were verified via sequencing . Products longer than 1 kb were analyzed via restriction digests to compare with the predicted patterns from the genome sequence . All confirmed deletions were consistent with array results . The precise junctions of all deletions on the main chromosome were identified except two of the deletions in CM1055 and CM1820 due to the presence of multiple repeat elements around their flanking regions . For the deletions on the megaplasmid , we only focused on the parallel pattern of the deletions and did not confirm each various subtype with their slightly different endpoints . For detecting deletions in isolates not screened via CGH , we designed 4 pairs of primers to amplify regions across DT1 , each with upstream and downstream pairs ( Table S3 ) . We classified isolates into 4 major types based on the PCR results ( DT1 with 2 subtypes ) : DT1a ( negative results from all ) , DT1b ( negative result from R1 , R2 , and R3 and positive result from R4 ) , DT2 ( negative result from R1&R2 and positive result from R3&R4 ) , DT4 ( positive result from R1&R2 and negative result from R3&R4; not found in any population ) , and no deletion ( positive results from all ) . The deletion in CM1194 was categorized as DT3a based on the array data; the deletion in CM1182 was categorized as DT3b based on the negative result from R3 but positive results from the other 3 . Allelic exchange plasmids for generating deletion mutants were constructed based on pCM433 , a sacB-based suicide vector [31] . PCR products of regions upstream and downstream of each deletion were amplified and consecutively cloned into pCM433 to generate pML4 , pML5 , pML7 and pML9 ( Table S4 ) . In order to reduce false-positives , a second selection marker , kan , with loxP excision sites amplified from pCM184 [33] was introduced into each of the plasmids between the upstream and downstream regions to generate pML10 , pML11 , pML12 and pML13 , respectively ( Table S4 ) . Each of these donor plasmids was then introduced into the wide type strain ( CM501 ) via triparental conjugations as previously described [31] . Single-crossover recombinants were selected with tetracycline ( Tet , 10 µg mL−1 ) and then double-crossover recombinants were selected with kanamycin ( Kan , 50 µg mL−1 ) and sucrose ( 5% wt/vol . ) . For each deletion type , we saved three independent clones through the cloning steps . These were each confirmed by PCR to contain the correct deletion and all three were tested for a consistent phenotype . The kan marker was then excised by cre recombinase as before [33] to generate the desired unmarked deletion mutants ( Table S1 ) . We performed the growth and fitness assays following a previously described procedure [22] with a few modifications . Briefly , three replicate cultures of each strain were inoculated and acclimated in minimal medium supplemented with carbon sources in 48-well plates ( Corning , Lowell , MA ) at 30°C , 650 rpm , 1 mm orbit and a total volume of 640 µL in each well . Growth curves were then obtained by following the change in OD600 ( Victor2 plate reader , Perkin Elmer , Waltham , MA ) . The transition time between growth phases observed during growth on MS ( M∶S = 7∶1 ) was estimated as before [34] . Growth rates and regression lines for each phase were calculated ( Phase I: y = a1+b1x; Phase II: y = a2+b2x ) , the OD600 at the time of transition ( ODt ) was determined as the average of two OD600 values with the minimum change during the transition phase , and the effective transition time was obtained as the difference between the two time values ( x1 , x2 ) where the estimated regression lines were equal to ODt . Fitness of each strain was measured as before [22] by competing each evolved or constructed strain against a fluorescently labeled ancestor ( CM1179 ) strain in 48-well plates with initial volumetric ratio of 1∶1 . Due to the small fitness changes for certain strains , competition assays were run for 4 cycles of growth ( i . e . , 8 days ) . The ratio of non-fluorescent cells in mixed populations was measured by passing population samples before ( R0 ) and after 4 cycles of competition growth ( R4 ) through a BD LSR II flow cytometer ( BD Biosciences , San Jose , CA ) . Fitness values ( W ) were calculated by following equation:To estimate the fitness cost on succinate during stationary phase , cultures were also sampled at hour 28 ( early stationary phase ) , such that the fitness cost was estimated as the difference in fitness values calculated between hours 0 to 28 vs . 0 to 48 ( using 9 replicates per strain ) . Disc diffusion assays were done to test for sensitivity on formaldehyde , SDS , peroxide , a trace metal mix , salt , arsenate and amplicillin . Bacteria were grown to stationary phase ( OD600∼1 . 5 ) in regular hypho medium supplied with 3× succinate ( 10 . 5 mM ) . Five mL of this culture was mixed with 60 mL of 42°C pre-warmed soft agar ( 0 . 75% , with 15 mM succinate ) , and 5 mL of this mixture was poured onto hypho agar plates with 15 mM succinate . Disks were placed at the center of the plates and aliquots ( 5 µL ) of formaldehyde ( 37% ) , SDS ( 10% ) , peroxide ( 30% ) , a trace metal mix ( 1000× ) ( Delaney et al . unpublished ) , NaCl ( 1 M ) , sodium arsenate ( 10% w/v ) or amplicillin ( 100 mg/mL ) were added on the filter discs . Diameters of growth inhibition were measured after 4 days . Exponential-phase cells growing on succinate ( OD600∼0 . 5 ) were used in heat shock and UV resistance assays . Cells were transferred to 55°C for 15 min for heat shock or exposed to 312 nm UV light for 15 min for UV resistance assays . Suspensions were then diluted and plated onto hypho agar containing 15 mM succinate , and colonies were counted after 4 days of 30°C incubation . Additionally , succinate-grown cultures were tested in liquid medium with the following treatments: formaldehyde ( 1–5 mM ) , SDS ( 10−1–10−5‰ ) , peroxide ( 10−1–10−5‰ ) , trace metal mix ( 2–20× ) , salt ( 5–500 mM ) , ampicillin ( 12 . 5–50 µg/mL ) , sodium arsenate ( 20–50 mM ) , UV exposure prior to growth ( 1–20 min ) , or heat stress during growth ( 32–36°C ) . Final OD600 or relative growth rates were calculated as the ratio of treatment to control . | Many free-living bacteria are known to commonly lose large portions of their genomes , but it is unclear what evolutionary forces drive these changes . Is this due to random loss in small populations , as is thought to be the case for the extreme genome degradation of intracellular symbionts , or due to selection ? And if it is beneficial , is it directly caused by replicating a shorter genome or advantageous loss of the genes themselves ? We uncovered that most replicate populations of Methylobacterium extorquens AM1 evolved in the laboratory for 1 , 500 generations lost nearly 10% of their genome . Through reconstructing these deletions , we demonstrated that these losses were indeed beneficial , but the advantage did not scale with length of genome lost , and were even deleterious in alternative environments . These findings suggest that the initial stages of genome shrinkage may be driven by selection , ultimately leading to a more streamlined , specialized organism . | [
"Abstract",
"Introduction",
"Results/Discussion",
"Materials",
"and",
"Methods"
] | [
"bacteriology",
"genetic",
"mutation",
"genome",
"evolution",
"population",
"genetics",
"microbiology",
"mutation",
"mutation",
"types",
"microbial",
"evolution",
"microbial",
"physiology",
"biology",
"evolutionary",
"genetics",
"genetic",
"drift",
"adaptation",
"natural",
"selection",
"genetics",
"genomics",
"evolutionary",
"biology",
"genomic",
"evolution",
"bacterial",
"evolution",
"evolutionary",
"processes",
"genetics",
"and",
"genomics"
] | 2012 | Repeated, Selection-Driven Genome Reduction of Accessory Genes in Experimental Populations |
Cohesin regulates sister chromatid cohesion during the mitotic cell cycle with Nipped-B-Like ( NIPBL ) facilitating its loading and unloading . In addition to this canonical role , cohesin has also been demonstrated to play a critical role in regulation of gene expression in nondividing cells . Heterozygous mutations in the cohesin regulator NIPBL or cohesin structural components SMC1A and SMC3 result in the multisystem developmental disorder Cornelia de Lange Syndrome ( CdLS ) . Genome-wide assessment of transcription in 16 mutant cell lines from severely affected CdLS probands has identified a unique profile of dysregulated gene expression that was validated in an additional 101 samples and correlates with phenotypic severity . This profile could serve as a diagnostic and classification tool . Cohesin binding analysis demonstrates a preference for intergenic regions suggesting a cis-regulatory function mimicking that of a boundary/insulator interacting protein . However , the binding sites are enriched within the promoter regions of the dysregulated genes and are significantly decreased in CdLS proband , indicating an alternative role of cohesin as a transcription factor .
Cohesin is an evolutionally conserved multisubunit protein complex consisting of an SMC1A and SMC3 heterodimer , and at least two non-SMC proteins SCC1 ( also known as RAD21 or MCD1 ) and SCC3 ( also known as SA or STAG ) . Cohesin controls sister chromatid cohesion during S phase with Nipped-B-Like ( NIPBL ) facilitating its loading and unloading [1] . ESCO2 possesses acetyltransferase activity and is involved in the establishment of cohesion [2] . Cohesin is loaded onto chromatin during G1/S phase in budding yeast and during telophase of the preceding cell division in vertebrates . Loading of cohesin also happens during G2/M phase when double strand DNA breaks are generated [3] . Removal of cohesin from chromosome arms begins during prophase and completes by separase-mediated dissolving of the remaining cohesin from centromeres during anaphase [3] . Although no consensus DNA sequence for cohesin binding has been demonstrated , cohesin is enriched at heterochromatin [4] and DNA double-strand breaks ( DSBs ) [5] . A large amount of intact and free cohesin is associated with chromosomes for most of the cell division cycle because of a separase independent mechanism [6]–[8] . A noncanonical role for cohesin as a key regulator of gene expression has been proposed [9] . The Drosophila NIPBL homolog , Nipped-B , alleviates the gypsy insulator function by assisting in long distance enhancer–promoter interactions to activate cut and Ultrabithorax expression . Nipped-B and cohesin colocalize and bind preferentially to active transcription units [9] . Recently , CTCF was reported to colocalize with cohesin and required for cohesin binding to chromatin [10] , [11] . CTCF is the only protein known to bind to all vertebrate chromatin insulators and was initially identified as a transcription factor that binds to mammalian c-MYC promoters [12] . In addition , CTCF is well studied for its role in regulating genomic imprinting , and is proposed to regulate higher order chromatin structures such as intra- and interchromosomal association [13] , [14] . CTCF is required for Tsix transactivation and involved in maintaining both X inactivation and escape domains , it stabilizes the repetitive sequences in several genetic disorders , and has been suggested to act as a tumor suppressor gene [15] . CTCF binding sites have been mapped in the human genome [16] . Cornelia de Lange syndrome ( CdLS , Online Mendelian Inheritance in Man [OMIM] 122470 , 300590 , and 610759 ) is a dominant disorder with multisystem abnormalities including characteristic facial features , hirsutism , upper extremity defects , gastroesophageal dysfunction , growth , and neurodevelopmental delays . The incidence is about one in 10 , 000 , with most cases being sporadic . There is equal gender distribution with a high degree of phenotypic variability . About 60% of the probands with CdLS have heterozygous mutations in the NIPBL gene , whereas 5% have mutations in the SMC1A gene , and one patient was found to have a mutation in the SMC3 gene [17] , [18] . Other multisystem developmental disorders have been found to be caused by mutations in cohesin-related genes , such as Roberts-SC phocomelia ( RBS , OMIM 268300 ) an autosomal recessive disorder caused by either homozygous or compound heterozygous mutations in the ESCO2 gene [19] . These disorders have collectively been termed “cohesinopathies . ” Given the paucity of sister chromatid cohesion defects observed in individuals with CdLS [20] , we hypothesize that it is the newly established role of cohesin in gene regulation that results in the multisystem phenotype when disrupted . To study the effects of disruption of cohesin on gene expression in human cells we have utilized lymphoblastoid cell lines ( LCLs ) from individuals with CdLS that harbor known heterozygous mutations in the cohesin regulator NIPBL and cohesin structural component SMC1A and applied a genome-wide approach to analyze gene transcription and cohesin binding .
LCLs from 16 severely affected CdLS probands with NIPBL protein-truncating mutations as well as 17 age , gender , and race matched healthy controls were used as training samples for assays on the Affymetrix HG-U133 plus 2 . 0 expression arrays , six additional individuals including one CdLS proband , one healthy control , two RBS probands ( a related cohesinopathy ) , and two Alagille syndrome ( AGS ) probands ( an unrelated multisystem dominant developmental disorder caused by disruption in the Notch signaling pathway ) served as testing samples ( Table S1A and S1B ) . 27 , 995 probe sets ( 12 , 740 nonredundant genes ) were considered to be expressed in human LCLs . Unsupervised sample clustering by principle component analysis ( PCA ) of all the expressed probe sets was able to separate probands from controls in the training set indicating these two groups have different gene expression patterns ( Figure 1A ) . Differential expression of these 27 , 995 probe sets was ranked by F = ( between group variance ) / ( within group variance ) . Permutation analysis was performed 100 times and false discovery rate ( FDR ) was calculated for each F score , whereas redundancy was collapsed by keeping the ones with the highest F scores . We have identified a group of 1 , 915 probe sets ( 1 , 501 nonredundant genes ) with FDR<0 . 05 and 420 probe sets ( 339 nonredundant genes ) with FDR<0 . 01 ( Tables S2 and S3 ) that are differentially expressed in CdLS . Heatmap representation of the expression levels of these genes clearly demonstrates that the expression of the 420 probe sets is remarkably different between CdLS probands and controls ( Figure 1B ) . NIPBL itself had the highest ranking , with FDR = 0 and a fold change of −1 . 34 . In order to examine whether CdLS probands could be differentiated from controls through expression profiling , Leave-One-Out cross validation was performed on the training set . The top 400 probe sets were selected on each of the 33 rounds that corresponded to an FDR≈0 . 01 . The left-out samples were successfully classified into two distinct groups using nearest centroid method with the exception of two controls and one proband that were misclassified ( Figure 1C ) . The area under the receiver operating characteristic ( ROC ) curve is 0 . 985 with test accuracy of 91% ( 95% confidence interval = 76%–98% ) . Nearest centroid classification method was further performed on the six testing samples based on the identified 420 probe sets ( FDR<0 . 01 ) . The one healthy control and two individuals with AGS were classified as controls; one CdLS and two RBS probands were classified as probands ( Figure 1C and Table S4 ) . RBS is due to the mutations in the ESCO2 gene that also regulates cohesin , whereas AGS is an independent genetic disorder due primarily to mutations in the JAG1 gene , a member of the Notch signaling pathway . Thus , although limited to only two samples , it appears that RBS shares a similar transcription profile with CdLS , consistent with these two disorders being caused by disruption of the cohesin pathway . It is of interest that whereas the two RBS probands were classified as CdLS , their discriminant scores ( DS ) were actually midway between the scores of CdLS probands and the controls suggesting RBS might have an intermediate transcription profile to CdLS and controls ( Figure 1C ) . Clustering-based feature selection was carried out on the 339 nonredundant genes ( 420 probe sets , FDR<0 . 01 ) to identify independent pathways or functional groups . Five clusters were discovered ( Table S5 ) and 32 genes ( Table S6 ) were chosen for further custom array validation according to smaller FDR , bigger fold change , and less redundancy . A cohort of 101 samples including individuals with different phenotypes ( healthy , CdLS , or other disorders ) and various gene mutations ( Table S7 ) were measured on custom arrays carrying 56 probes mapped to the 32 selected genes ( Tables S6 and S8 ) . We have followed a step-wise procedure to identify classifiers according to different CdLS subgroups , and applied nearest centroid classifications on all 101 samples ( Table S9A and S9B ) . Detailed analysis is described in Text S1 . A 23-gene classifier can be used to categorize CdLS probands with NIPBL mutations from the rest of the samples including non-CdLS , CdLS probands with SMC1A mutations , and CdLS probands without an identifiable gene mutation . This indicates that the expression of these 23 genes is tightly correlated to NIPBL function . To improve the generality of the classifier , we randomly selected 15 mild probands with NIPBL mutations as a new training group . Expression of ten of the 23 genes was significantly different between this group and the original 17 controls and was also capable of subclassifying all CdLS probands from non-CdLS controls , regardless of the gene mutations or clinical presentations of the probands . This suggests that expression of these ten genes is affected by a CdLS specific disease process . For both classifications , the expression levels of the classifier genes are tightly correlated to disease severity . A clear progression of increasing discriminant scores ( DS ) can be seen from healthy controls through mild , moderate , and severe CdLS probands ( Figure 2A and 2B ) . In addition , we have identified three genes NFATC2 , PAPSS2 , and ZNF608 that could be used as biomarkers for CdLS ( Figure S1A and S1B ) . The phenotype associated gene expression profiles strongly suggest either direct or indirect roles for the identified genes . Cohesin is a multisubunit complex constructed from SMC1A , SMC3 , RAD21 , and SCC3 subunits . Mutations in SMC1A or SMC3 and the cohesin regulator NIPBL lead to the human developmental disorder CdLS . To test the hypothesis that cohesin regulates transcription through its chromatin binding activity and that this association is regulated by NIPBL activity we undertook whole genome mapping of cohesin binding sites in LCLs from two healthy controls and one severely affected CdLS proband with an NIPBL mutation . Because of our inability to identify an effective antibody with high specificity against NIPBL or SMC1A , we chose an antibody against RAD21 ( one of the other key components of the cohesin complex ) to map genome-wide cohesin binding sites . Chromatin immunoprecipitation ( ChIP ) using a polyclonal antibody against human RAD21 was performed and products were hybridized on Affymatrix 2 . 0 tiling arrays . The score of model-based analysis of tiling-arrays algorithm ( MAT ) was calculated and probes were mapped to genomic positions . Peaks representing genomic regions bound by hRAD21 were identified with a p<10−6 and FDR<0 . 01 . The 54 , 675 probe sets on Affymetrix HG-U133 plus 2 . 0 expression array can be unambiguously mapped to 15 , 162 unique RefSeq mRNAs including 10 , 378 transcribed and 4 , 784 nontranscribed genes in LCLs . 78% of the 15 , 162 mapped genes do not harbor intragenic cohesin sites ( Here , “intragenic” means genomic region from the transcription start site [TSS] of a gene to the transcription termination site [TTS] of the same gene ) , and cohesin binds at variable distances outside those genes . 22% of the 15 , 162 mapped genes harbor intragenic cohesin sites , this number is reduced to 19 . 0% in the silent nontranscribed genes ( p≤7 . 2e−6 ) and no change in the disease neutral genes ( 22 . 9% , p≤0 . 0864 ) ; on the contrary , more of the differentially expressed genes harbor cohesin sites ( 27 . 0% , p≤7 . 44e−5 ) ( http://145 . 18 . 230 . 98/Service/Statistics/Binomial_proportions . html ) ( Table S10 ) suggesting a correlation between intragenic cohesin binding and gene expression . For the 22% of genes with intragenic cohesin sites , cohesin preferentially binds to a narrowed region surrounding the TSSs or the TTSs with frequency at the TTSs only half of that at the TSSs . The 100-kb regions spanning upstream and downstream of the genes have only background levels of cohesin binding ( Figure 3A and 3B ) . Among controls , the degree of cohesin binding within +/− 1 kb of the TSSs is greatest for those genes that are actively transcribed and especially in those genes that are differentially expressed in the NIPBL mutant cell lines , whereas the silent nontranscribed genes have the same , or lower level , of enrichment as the background level ( Figure 4A ) . In addition , cohesin binding is enriched at the 5′-UTRs only for actively transcribed genes , with no binding difference at exons , introns , or 3′-UTRs between the actively transcribed and silent genes ( Figure 4B ) . Identification of overrepresented cohesin binding near promoters suggests that cohesin may regulate gene expression as a transcription factor . In spite of this , the majority of the expressed genes ( 78% ) do not harbor any cohesin binding sites in their intragenic regions , indicating most of the genes in the human genome may be regulated by cohesin independent pathways or cohesin is involved in their expression regulation through an alterative mechanism . We further evaluated 13 genomic loci based on their gene expression alterations to validate their cohesin binding status by the more sensitive method of ChIP-quantitative PCR ( qPCR ) . Out of these 13 loci , two regions are equally bound by cohesin in both healthy and CdLS cells , two regions are not bound by cohesin in either healthy or CdLS cells , and the remaining nine loci demonstrated significant loss of cohesin binding in CdLS cells as compared to control cells . The ChIP-qPCR results are consistent with cohesin binding alterations detected by ChIP array studies ( Figure S2A and S2B , Table S11 ) . The total number of cohesin binding sites is reduced by 29 . 7% ( 9 , 530 versus 13 , 560 ) in the examined CdLS proband , but the total number of binding sites at TSSs is reduced by 43 . 4% ( 448 versus 792 ) in the same proband , suggesting that cohesin is more likely to be removed from TSSs ( 43 . 4% versus 29 . 7% , p = 5 . 9e−8 ) ( Table 1 ) . The 10 , 378 genes expressed in LCLs have been statistically ranked for their misexpression in CdLS probands as described above . In the controls , there exist 666 LCL expressed genes that have cohesin binding sites mapped to the +/− 1-kb vicinity of TSSs , and 107 of them are identified as differentially expressed in CdLS ( FDR<0 . 05 ) . In CdLS , only 376 such genes have cohesin sites around their TSSs ( 376 versus 666 , reduced by 43 . 5% ) , only 53 of the 107 differentially expressed genes still maintain their TSS/cohesin association , whereas the rest have all lost their TSS cohesin binding sites ( 53 versus 107 , reduced by 50 . 5% ) in the proband ( Table 2 ) . At the TSSs , the number of cohesin sites on differentially expressed genes is significantly reduced in CdLS , whereas the reduction is moderate for the nondifferentially expressed genes , and only minimal for those silent nontranscribed genes ( Figure 4A ) . The binding between cohesin and TSSs of expressed genes is highly correlated to the CdLS phenotype . In our identified panel of differentially expressed genes in CdLS , Fisher testing on the ChIP data shows that these genes tend to attract more cohesin to their TSSs in control cells under the healthy condition ( p = 10e−4 ) than the neutral genes do , whereas under the diseased condition in CdLS cells , those genes tend to lose their capability to recruit cohesin and associate with cohesin at a similar level to the neutral genes and have lost their statistically significant difference ( p = 0 . 1 ) ( Table 2 ) . This 2-kb region ( +/− 1 kb surrounding the TSS ) was further analyzed for the entire group of 10 , 378 genes expressed in LCLs that have been ranked for their differential expression in CdLS probands as described above . Cohesin enrichment was clearly identified in control cells for the top ranked genes and a dramatic decrease in binding is demonstrated in the CdLS cells , suggesting that the genes that harbor more cohesin sites around the promoter regions are more likely to be misexpressed in CdLS ( Figure 4C ) . Moreover , this difference was even more remarkable if we narrowed the analyzed region to the +/− 100-bp central area surrounding TSSs ( Figure S3 ) . To summarize , in control LCLs cohesin preferentially binds to transcribed genes at the TSSs as compared to the silent nontranscribed genes . The binding sites are even more enriched for the differentially expressed genes . In CdLS , cells tend to lose cohesin binding globally , however the cohesin sites at TSSs are more likely to be lost , most notably for the differentially expressed genes where loss of cohesin binding at the TSSs results in a binding frequency approaching the background level . The preferential binding to promoter regions suggests cohesin may play a role as a transcription factor . Recently cohesin has been functionally linked to CTCF , an insulator capable of blocking enhancers or preventing the spread of epigenetic signals [15] . In our study , the ion transporter protein ATP11A is significantly downregulated in CdLS ( FDR = 0 . 027 ) , although the fold change is small ( −1 . 24 ) . ATP11A locates within ENCODE region ENr132 on Chromosome 13 with four other genes . Therefore , the ENCODE datasets obtained from GM6990 , a similar EBV-transformed human B cell line ( http://genome . ucsc . edu/ , http://www . sanger . ac . uk/PostGenomics/encode/data-access . shtml ) , were able to be adapted for our analysis [21] . There are six CTCF and two cohesin binding sites in this area , both cohesin sites overlap with CTCF sites . In controls , this area can be split into three chromatin regions according to multiple histone modification makers ( Figure 5A and 5B ) [22]–[24] , and cohesin and CTCF colocalize at the border . Region 1 harbors only one gene C13orf35 , which is not expressed in LCLs . Region 3 harbors three genes , MCF2L , F7 , and F10 , which are all expressed comparably in LCLs from both controls and probands . The ENCODE study has shown that chromatin-silencing marker H3K27me3 is enriched in region 3 , but open chromatin markers H3K4 me1/me2/me3 , H3K36me3 , and H3K79me3 , and DNaseI sensitive sequences are underrepresented , indicating chromatin in this region is condensed and transcription repressed [22]–[24] . In region 2 , on the other hand , H3K4 is highly methylated , H3 tails are vastly acetylated , and multiple DNaseI sensitive sites appeared; meanwhile , H3K27 methylation level is quite low indicating region 2 is an active open chromatin domain . ATP11A is the only gene located in region 2 and differentially expressed in CdLS . Of note , the cohesin binding site between regions 2 and 3 at Chromosome 13: 112 , 645 , 000–112 , 645 , 600 is lost in CdLS ( Figure 5A and 5B ) . ChIP-qPCR was then performed using specific primers to amplify this binding locus in an expanded sample set including three healthy controls and three CdLS probands ( Figure 5C ) . Two of the three probands , PT2 and PT12 , have NIPBL truncating mutations with severely affected clinical features and have been included in the whole genome expression array studies as described above; the third proband CDL-017 has a mutation in the SMC1A gene and manifests a much milder phenotype ( Tables S1 and S7 ) . Cohesin binding site 1 ( Chromosome 3: 79653256–79653385 ) , which was not lost in CdLS according to our qualitative array analysis was therefore used as a positive binding control . By quantitative PCR assays , the enrichment of cohesin bound to site 1 was not found to be changed between controls and the probands , which is consistent with the array findings . However , cohesin binding was dramatically reduced , within Chromosome 13: 112 , 645 , 000–112 , 645 , 600 among CdLS probands including the individual with the SMC1A mutation ( Figure 5C ) . Although cohesin binding was not completely lost in CdLS by ChIP-qPCR , the result is consistent with the missing binding peak seen in the qualitative ChIP array analysis . Although this dataset is limited , it suggests that cohesin possesses a function as an insulator/boundary protein , in addition , functional NIPBL is required for this process . With disruption in the NIPBL mutated or cohesin subunit SMC1A mutated human cells , the silent chromatin signals from region 3 appear to be able to cross the boundary and migrate into region 2 to inhibit ATP11A transcription . Cohesin and CTCF may function cooperatively at this locus owing to their colocalization . In addition , both CTCF binding sites remained intact in CdLS , which may explain why the downregulation of ATP11A was not dramatic ( −1 . 24 ) . Ingenuity Pathways Analysis ( IPA ) ( Ingenuity Systems , Inc . , http://www . ingenuity . com ) was used to analyze the identified differentially expressed genes . Out of the 339 genes with FDR<0 . 01 , 150 genes are documented in cancer , neurological , hematological , skeletal and muscular , and dermatological diseases; 150 genes are identified as major players in embryonic and tissue development , hematological and immune system development and functions; in addition , 153 genes have well established cellular and molecular functions in cell death , cell proliferation , and cell cycle regulation . We have further analyzed the biological functions and canonical pathways mediated by the 23- and 10-gene classifiers and the three biomarkers as validated by target array ( Table S12 ) . Interestingly , more than 60% ( 15 out of 23 ) of the identifier genes harbor intragenic cohesin binding sites , which is much higher than the average genome level ( 22% ) . Moreover , some of these genes have completely lost their cohesin association in CdLS ( Table S12 ) . Both groups of classifier genes are tightly related to pathways of cell death , cellular development , and tissue morphology . 12 out of these 23 genes are involved in 47 known biological functions or disease conditions . Five of these 12 genes are also part of the 10-gene classifier , including two genes , NFATC2 and PAPSS2 , which are the identified biomarkers for CdLS . The 23-gene classifier could differentiate CdLS probands with NIPBL mutations suggesting the expression of these 23 genes are tightly controlled by NIPBL; whereas the 10-gene classifier is less powerful and only able to identify CdLS probands without the ability to differentiate subgroups of probands with different gene mutations , suggesting that these ten genes are related to terminal events during CdLS pathogenesis . Therefore , the common five genes , PAPSS2 , NFATC2 , MAP3K5 , LTB , and PHF16 , which are involved in multiple reported events by IPA and are shared by the two classifiers , might be involved in cellular functions that are universally affected in CdLS . Presumably mutations in NIPBL , SMC1A , or SMC3 , or as of yet unidentified CdLS causing gene mutations , will all result in alterations of the related biological functions controlled by these five genes . On the other hand , the seven unique genes with functional roles that were excluded from the 10-gene classifier , KIFAP3 , AIM1 , BBS9 ( PTHB1 ) , TSPAN12 , TRERF , ARHGAP24 , and ID3 , probably represent cellular functions affected more specifically by NIPBL mutations . Four genes , PAPSS2 , NFATC2 , MAP3K5 , ADCY1 , were identified to be involved in 32 canonical pathways by IPA; they also regulate multiple biological functions as mentioned above . ADCY1 is the single gene out of the above four genes that exists in the 23-gene classifier but is excluded from the 10-gene classifier; thus the specific canonical pathways regulated by ADCY1 ( i . e . B cell receptor signaling , RAR activation , sulfur metabolism , and endoplasmic reticulum stress pathways ) , could largely depend on normal functions of NIPBL . Two out of the three biomarkers , NFATC2 and PAPSS2 , are reported to be involved in multiple biological functions and canonical pathways by IPA analysis . The third biomarker ZNF608 is a novel protein with very minimal known functions . However , the zinc finger protein family members are known to be the major players in many molecular and cellular pathways . One of the biomarkers , NFATC2 , is involved in multiple signaling pathways during development and affecting skeletal myogenesis , chondrogenesis , axon growth , and guidance [25] , [26] . Two NFATC negative regulators both locate to the Down syndrome critical region of human Chromosome 21 , Nfatc2−/− and Nfatc4−/− double-knockout mice have physical and cognitive features resembling human Down syndrome [27] . Dysregulation of NFATC2 in the postnatal nervous system may contribute to mental deficiency in CdLS . Another biomarker , PAPSS2 , plays a pivotal role in the biosynthesis of sulfate donors for sulfotransferase reactions . Its activity is important for normal skeletal development; recessive mutations on PASS2 cause the genetic disorder spondyloepimetaphyseal dysplasia ( SEMD ) , Pakistani type and degenerative osteoarthritis [28] . Papss2−/− knockout mice have shortened limbs , reduced axial skeletal length , and complex facial features . Its transcripts were also present in the heart and brain in mouse embryos [29] .
Cohesin consists of four major proteins SMC1A , SMC3 , SCC1 , and SCC3 . NIPBL plays a role in shuttling cohesin onto and off the chromatin , although the exact mechanism of its action is poorly understood . All proteins in this pathway are evolutionally conserved from yeast to human [30] . Cell-cycle related sister chromatid cohesion , DNA repair , and homologous rearrangement are well established roles for the cohesin apparatus . A role for cohesin in regulating gene expression has also been proposed and appears to be more sensitive to subtle dosage alterations of the cohesin apparatus and its regulators than its canonical function in sister chromatid cohesion [31] . In both yeast and Xenopus , the loading of cohesin onto chromatin in G1 phase is functionally separable from the establishment of sister chromatid cohesion at S phase [32] , [33] . In Drosophila , Nipped-B mediates interactions between the promoter and remote enhancers for cut and Ultrabithorax; heterozygous Nipped-B mutants diminish cut expression in the emergent wing margin without showing cohesion defects indicating sister chromatid cohesion is independent from cohesin regulated gene expression [34] . In mice , Pds5b mutants have multiple CdLS-like defects without flawed sister chromatid cohesion [35] . In humans , CdLS is caused by heterozygous loss-of-function mutations in the NIPBL ortholog of Nipped-B and , in a smaller percent of cases , by mutations in the SMC1A or SMC3 cohesin subunit genes [17] , [18] , [36] , [37] . Given the constellation of developmental abnormalities present in individuals with CdLS , with only a subset showing minor cohesion defects [20] , [38] , it is likely that the alterations of cohesin regulation and structure seen in these individuals result in gene expression dysregulation . We chose to use an easily accessible but a seemingly developmentally irrelevant tissue , LCLs , for these studies . We hypothesized that congenital genetic disorders might arise , in part , through dysregulation of expression of specific genes and that expression differences between affected and unaffected individuals might be present in tissues other than disease presenting tissues . These cells also provide an invaluable resource of naturally occurring mutant cohesin proteins ( both structural and regulatory components of cohesin ) that can be used to study the cellular processes regulated by this complex and specifically the impact on regulation of gene expression . Surprisingly these cells may also provide valuable insight into human developmental processes as well . We have identified specific gene expression profiles for CdLS that are capable of classifying probands and tightly correlate with disease severity . Cohesin preferentially binds to promoter regions of the actively expressed genes suggesting a role as a general transcription factor . These binding sites are significantly reduced in NIPBL mutant CdLS samples . This result is likely due to NIPBL's direct role in cohesin loading on chromatin , which in turn affects transcriptional regulation at specific loci and would contribute to the CdLS pathogenesis . Out of the 339 dysregulated genes with FDR<0 . 01 , 202 were upregulated ( 59 . 6% ) and 137 were downregulated ( 40 . 4% ) , more genes were reactivated than inhibited with mutations in NIPBL ( 59 . 6% versus 40 . 4% , p = 3 . 44e−17 ) suggesting that NIPBL and cohesin can result in both negative ( as transcriptional repression ) and positive ( as transcriptional activation ) effects on expression . A similar percentage of upregulated and downregulated genes was also observed among the 1 , 501 dysregulated genes with FDR<0 . 05 . Moreover , 71 of the above 339 genes ( 20 . 9% ) and 207 of the 1501 genes ( 13 . 8% ) have fold changes larger than 1 . 5 , whereas the highest fold changes are −4 . 2 and +4 . 6 , respectively . Although the majority of expression levels seemed only mildly perturbed , developmental deficits in CdLS are likely due to a cumulative change in multiple genes . Another reason for less remarkable expression differences could be the LCL tissue type used for this study , with bigger fold changes in more genes possibly present in more directly affected tissues of , e . g . , brain or limb , and at specific times during embryonic development . However , it is also possible that the transcriptional dysregulation may be directly mediated by NIPBL through a yet uncharacterized mechanism and the reduced cohesin binding may be a secondary effect . In our study , a 30% reduction in NIPBL message was able to trigger a 29 . 7% ( 9 , 530 versus 13 , 560 ) reduction in cohesin binding sites in CdLS probands and further affects the transcription of specific genes . The central components for sister chromatid cohesion , RAD21 ( SCC1 ) , SMC1A , SMC3 , STAG2 , ESCO1 , ESCO2 , and PDS5A ( also known as SCC-112 ) , are all expressed similarly between controls and CdLS probands with NIPBL mutations . However , STAG1 , PDS5B ( also known as APRIN ) , MAU2L ( KIA0892 ) , as well as several other genes with functions related to sister chromatid cohesion were significantly dysregulated in NIPBL mutant CdLS probands ( FDR<0 . 05 ) ( Table S13 ) , suggesting that the cohesin pathway itself is affected by mutant NIPBL . MAU2 ( KIAA0892 ) is the putative human homolog of scc4 in Caenorhabditis elegans [31] , [39] . It forms an essential loading complex with NIPBL that regulates cohesin-chromatin association , sister-chromatid pairing , and mitotic checkpoints in HeLa cells . Physical association between NIPBL and MAU2 is indispensable for their stability , as depletion of either of the two proteins would subsequently diminish the cellular level of the other one [39] . In our study , decreased NIBPL transcription ( −1 . 33 , FDR = 0 ) was able to upregulate the transcription of MAU-2 ( +1 . 11 , FDR = 0 . 026 ) , suggesting a functional compensation may exist for cohesin loading in CdLS . A cohesin-independent mechanism has also been suggested to exist . Condensin complexes [40] , origin recognition complexes ( ORCs ) [41] , centromere complexes [42] , and DNA catenation [43] have each been reported to play a role in mediating cohesin-independent sister chromatid cohesion . Genes involved in these functions are also found to have dysregulated expression in NIPBL mutant individuals ( Table S13 ) . This finding indicates a subset of genes regulated by NIPBL are tightly involved in sister chromosome segregation events , but expression alteration may be required to pass a certain threshold in order to induce visible cohesion defects . This observation could explain why cell lines derived from CdLS probands did not demonstrate significant sister chromatid pairing problems . In contrast to CdLS , cohesion defects have been reported in three human developmental disorders: RBS ( OMIM 268300 ) [19] , Rothmund–Thomson syndrome ( RTS , OMIM 268400 ) [44] , and α-Thalassemia/mental retardation syndrome , X-linked ( ATRX , OMIM 301040 ) [45] . Interestingly , although the expression of the RBS disease causative gene ESCO2 was not dysregulated in CdLS cell lines , the other two disease genes , ATRX and RECQL4 , both demonstrated dysregulation in NIPBL mutant cell lines ( Table S13 ) . Several cohesin targets have been identified . Steroid hormone ecdysone receptor ( EcR ) , which is the Drosophila homolog of human NR1H3 , was suggested to be regulated by Smc1 , and Runx3 was identified as a direct target of Rad21 in zebrafish [46] , [47] . The fact that both of these genes were also significantly dysregulated ( FDR<0 . 05 ) in CdLS probands with NIBPL mutations indicates that NIPBL may first affect cohesin proteins and subsequently dysregulate cohesin targets . Surprisingly , we did not find that cohesin directly binds to these two genes in the cell line studied , which raises the possibility that cohesin may regulate their expression over long distances . When comparing ChIP-on-chip results for Nipped-B and/or SMC1A binding sites in three different Drosophila cell types [48] , homologs of 20 differentially expressed human genes in CdLS probands ( FDR<0 . 05 ) were also found to be bound by NIPBL and cohesin ( unpublished data ) . Eight of these 20 genes are also bound by cohesin in humans suggesting they may be cohesin targets in both Drosophila and humans . It also suggests that cohesin mediated transcription is a conserved biological event . Moreover , most of the binding sites were lost in CdLS cells indicating dysregulated gene expression correlates with loss of cohesin binding . Among the eight genes , KMO , ELL2 , and ARHGAP17 have cohesin binding at TSSs; ROBO1 , UBE2H , MED13L , RASA3 , and PDPK1 had cohesin binding within intronic regions . One of these genes , ROBO1 ( homolog of lea in Drosophila ) , is of particular interest as it was found to have a fold change of 4 . 6 , which is the largest among all the genes on the array . ROBO1 has been associated with dyslexia , a neurocognitive disorder of language and graphic processing that could be due to the abnormal migration and maturation of neurons during early development . We have identified groups of 23 , 10 , and 3 genes as CdLS classifiers or biomarkers that are capable of differentiating CdLS from non-CdLS samples . The expression levels of these genes also correlate to the phenotypic severity of this disorder , although it is not clear at this time how the dysregulation of these particular genes might contribute to the phenotypes . More than 60% of the identifier genes harbor intragenic cohesin binding sites with some of them lost in CdLS proband . The obvious overrepresentation of genes carrying intragenic cohesin binding sites among the CdLS classifier genes further suggests that expression of the dysregulated genes is tightly related to the availability of cohesin binding . Overall , the majority of genes do not carry known cohesin binding sites , indicating that cohesin may play an upstream role in regulating human genes , or cohesin may enact regulation on some of the genes through distal cis- or trans-regulatory elements . The potential role for cohesin independent NIPBL regulation can not be excluded . Cohesin has recently been found to be physically and functionally associated with the vertebrate insulator protein CTCF . In our study cohesin binds to only ∼20% of genes intragenically . This distribution does not change much between expressed genes and silent genes , and between differentially expressed genes in CdLS and disease neutral genes . Cohesin could be involved in gene regulation , like CTCF , by either binding to promoter elements and having a direct influence on the transcriptional machinery or by binding to intergenic cis-elements such as insulators to control gene expression from remote distances [49] . In our study , we have detected a potential boundary effect of cohesin at the ATP11A gene locus that suggests , for the first time in humans , that cohesin may bind to insulators and regulate transcription . Reduced cohesin binding at this locus was further validated in three additional CdLS probands by the more sensitive ChIP-qPCR including probands with either NIPBL mutations or cohesin subunit SMC1A mutation . However , cohesin does not exactly mimic the function of CTCF , at least in LCLs . Some CTCF target genes , such as PIM-1 [50] and APP [51] , although expressed in LCLs , are neither dysregulated in CdLS nor do they lose cohesin binding at their regulatory regions . On the other hand , the CTCF target gene , BRCA1 [52] , was downregulated in CdLS ( −1 . 2 , FDR = 0 . 017 ) but without losing cohesin binding sites . Additional quantitative analysis or ChIP-qPCR to study more genomic loci will delineate a clearer picture of cohesin and CTCF effects on transcriptional regulation . The role cohesin plays in imprinting and X inactivation remains unclear [53] . In summary , we have undertaken a genome-wide approach to study gene expression and cohesin binding in NIPBL mutant human samples . On the basis of our data and previously reported studies , we propose that NIPBL may be involved in modulating cohesin function through various mechanisms . Besides its canonical role in regulating sister chromatid segregation proposed by Haering et al . [54] ( Figure 6A ) , cohesin may also regulate transcription ( 1 ) as an insulator protein by acting alone or with CTCF , or ( 2 ) as a transcription factor by binding to promoter elements . While regulating transcription , NIPBL may also serve as a cohesin shuttle to chromatin that leads to decreased cohesin binding when NIPBL is mutated . Data from this study are quite consistent with this role . Whether this loading mechanism either partially overlaps with , or is completely independent from NIPBL-mediated sister chromatid cohesion remains unknown ( Figure 6B ) . NIPBL and cohesin may very well form one protein complex binding to regulatory elements of target genes , with NIPBL mutations affecting the regulatory capacity of this complex ( Figure 6C ) . The colocalization of NIPBL and cohesin seen in Drosophila studies could be consistent with this model [9] . A third possibility is that NIPBL is able to maintain an accessible chromatin structure for cohesin binding whereas defective NIPBL leads to reduced accessibility for cohesin at specific chromosomal loci ( Figure 6D ) .
This study was conducted according to the principles expressed in the Declaration of Helsinki . The study was approved by the Institutional Review Board of The Children's Hospital of Philadelphia and Misakaenosono Mutsumi Developmental , Medical , and Welfare Center . All patients provided written informed consent for the collection of samples and subsequent analysis . All participants were evaluated by one or more experienced clinicians . Gene mutations were confirmed by sequencing , and most of the cases have been previously reported by our laboratories [17] , [55] , [56] . LCLs were grown uniformly in RPMI 1640 with 20% fetal bovine serum ( FBS ) , 100 U penicillin/ml , 100 µg streptomycin/ml sulfate , and 1% L-glutamine . To identify differentially expressed genes between CdLS probands and controls , age and gender matched samples from 16 normal controls of European descent and 17 clinically severely affected probands of European descent with NIPBL protein-truncating mutations ( nonsense or frameshift ) were chosen as the training set for the discriminate analysis . To validate the expression pattern obtained from the training set , six samples including one healthy control , one Egyptian CdLS proband , two Roberts syndrome probands , and two Alagille probands were used as the testing set . All 39 cell lines were grown anonymously and the processing of these 39 cell lines were randomized by genotypes to eliminate batch effects that may contribute to genotype-specific gene expression . Samples are listed in Table S1A and S1B with detailed description . For custom array analysis , detailed information of these samples is listed in Table S7 . Out of these 101 samples of European descent , the training set included 17 healthy controls and 14 severely affected CdLS probands with NIPBL protein-truncating mutations . All 31 samples were also used for the training in Affymatrix array analysis . For the testing set , all new samples were selected , which included four healthy controls , six severely affected probands , 13 moderately affected probands ( nine have NIPBL mutations and four do not have an identifiable mutation ) , and 34 mildly affected probands ( 26 have NIPBL mutations and eight do not have an identifiable mutation ) . We have also included nine CdLS probands with SMC1A mutations , as well as four samples with different genetic diagnoses ( two AGS , one Roberts syndrome , and one unknown multisystem genetic disorder ) . As above , samples were processed anonymously and randomly . 5×106 exponentially growing cells were seeded in 15 ml media in a 75-ml Falcon flask , and fed exactly after 24 h . After an additional 24 h on day 3 , 8 ml of the media was removed and cells were pelleted by centrifuge and RNA extraction was performed immediately . Total RNA from each sample was extracted with the RNeasy Mini-kit ( Qiagen ) , synthesis of double-stranded cDNA was performed using SuperScript Double-Stranded cDNA Synthesis kit ( Invitrogen ) , and cleaned up with GeneChip Sample Cleanup module ( Affymetrix ) . The resulting products were then used to synthesize biotin-labeled cRNA with Enzo Bioarray High Yield RNA Transcript Labeling kit ( Enzo Life Sciences ) and further fragmented to 35–200-bp oligos . All procedures were done according to manufacturer's instructions . 30 µl fragmented cRNA at the concentration of 500 ng/µl was sent for hybridization in the microarray facility at The Children's Hospital of Philadelphia . Microarray hybridizations were performed by using HG-U133 plus 2 . 0 GeneChips ( Affymetrix ) . The HG-U133 plus 2 . 0 contains ∼54 , 000 25-mer probe sets that covers approximately 47 , 000 transcripts and variants out of which 38 , 500 are well-characterized human genes . After hybridization and washes , arrays were scanned and analyzed both for genes that were present and for expression level using Microarray Analysis suite ( MAS ) 5 . 0 using default settings according to manufacturer's instructions . The same RNA isolation process was performed as above . 32 genes were selected by clustering-based feature selection and 59 probes were designed ( Table S8 ) . Probe designing , RNA labeling , and hybridization were conducted using the Ziplex workstation ( Xceed Molecular , http://www . xceedmolecular . com/ ) . In brief , concentrations of the isolated RNA were determined by measuring the absorbance at 260 nm . All total RNA samples were of acceptable purity ( ratio of the absorbance at 260 nm to 280 nm of 1 . 75 or greater ) . The integrity of the total RNA was determined to be acceptable for all samples ( RNA Integrity Numbers measured with the Agilent 2100 Bioanalyzer RNA 6000 Nano assay were greater than 9 . 0 ) . A custom Ziplex TipChip microarray containing oligonucleotide probes of between 35 and 50 bp for 32 genes was used to profile differences in gene expression between the LCL samples . Total RNA ( 500 ng ) from 108 independent samples was amplified and biotin labeled with the Illumina Totalprep RNA amplification kit ( Ambion ) . The concentrations of the labeled aRNAs were determined by measuring the absorbance at 260 nm , and 3 µg was hybridized on the custom TipChip with the Ziplex Automated Workstation protocol ( Xceed Molecular ) . After hybridization , the Ziplex Automated workstation software automatically quantified spot intensities and reported background subtracted expression values . The Ziplex software automatically evaluated attributes of each spot to identify spots that did not conform to quality control criteria and reported the mean value of the duplicate spots of each probe that passed quality control . Two healthy controls and one severely affected CdLS proband with an NIPBL protein-truncating mutation ( G5483A ) were used . Cells were crosslinked with 1% formadehyde at 70%–80% confluency for 10 min , chromatin was then prepared after quenching with 125 mM glycine and ChIP was performed as described [57] using anti-hRAD21 polyclonal antibodies ( Abcam , ab992 ) . In brief , lysates from crosslinked cells were incubated with the antibodies and preabsorbed protein A Affiprep beads ( Bio-Rad ) for 14 h at 4°C and for 2 h at 4°C , respectively . After washing , the beads were incubated in the elution buffer ( 50 mM Tris , 10 mM EDTA , 1% SDS ) for 20 min at 65°C . The elutes were treated with proteinase K for 1 h at 37°C and followed by 65°C overnight incubation for crosslink reversal . The samples were then treated with RNase and phenol-chloroform purified for one time , and further purified using PCR purification kit ( Qiagen ) with 80 µl water used for the final elution . The eluted chromatin was amplified and labeled with biotin then hybridized to high-density oligonucleotide tiling arrays ( Human tilling 2 . 0R array , Affymetrix ) as described [58] . A sample of DNA prepared from whole cell extract ( WCE ) was prepared in the same way . ChIP and WCE samples were hybridized on arrays according to the manufacturer's instructions , two technique replicates were used for each sample . After scanning and data extraction , enrichment values ( ChIP/WCE ) were calculated by using the MAT algorithm [59] . MAT is designed for high-density oligonucleotide tiling-array analyses in higher eukaryotes that could reduce probe-specificity biases because of genome complexity or high GC content . The resulting MAT scores are proportional to the logarithm transformed value of the fold-enrichment of the ChIP-chip samples [59] . We mapped MAT scores to positions in human genome assembly Hg 18 ( NCBI Build 36 ) . Bandwidth , MaxGap , and MinProbe parameters were set to 250 , 1 , 000 , and 12 , respectively . The cutoff threshold of p-values was set to 1×10−6 , which was equivalent to MAT scores higher than 4 . 85 . FDRs were also calculated with every experiment less than 1% ( Figure S4A and S4B ) . BED files were created , data were visualized in the Integrated Genome Browser ( IGB ) ( http://www . affymetrix . com/support/developer/tools/download_igb . affx ) and University of California Santa Cruz ( UCSC ) genome browser custom track ( http://genome . ucsc . edu/ ) . ChIP was performed as described above using hRAD21 and control antibodies . ChIP-qPCR analysis was performed as previously described [10] . ChIP samples ( 2 µl ) were used for one 25-µl PCR reaction . Analyses by qPCR were performed using a Platinum SYBR Green qPCR SuperMix UDG ( Invitrogen ) on an ABI 7500 cycler . The results were presented as fold-enrichment over control ChIP . Gene expression microarray data were processed by DNA-Chip Analyzer ( dChip ) ( http://www . dchip . org ) using PM-only background subtraction and invariant set normalization . Differential gene expression between controls and CdLS probands was ranked by the ratio of between- and within-group variance ( F statistic ) . During nearest centroid classification , distance of testing samples to training group centroids was measured as their Pearson's correlation coefficient . Statistical analyses were performed within R software environment ( http://www . r-project . org ) . PCA and heatmap plots were generated by Spotfire DecisionSite version 9 . 1 . 1 ( Spotfire , Inc . ) . More details about data analysis are provided in Text S1 . Genomic sequences reported in this manuscript have been submitted to NCBI GEO ( http://www . ncbi . nlm . nih . gov/geo ) : gene expression data are under accession number GSE 12408 and ChIP-chip data are under accession number GSE 12603 . | Appropriate segregation of chromosomes to daughter cells depends upon proper cohesion of sister chromatids during mitosis . The multiprotein cohesin complex and its regulators are key factors in this process . Intriguingly , recent work has shown that the cohesin complex also has other cellular roles , including a role in regulating gene expression . Additionally , mutations in cohesin structural and regulatory components have been linked to human multisystem developmental disorders such as Cornelia de Lange Syndrome ( CdLS ) , but the role cohesin is playing in the pathogenesis of this disorder is unknown . To define the role that cohesin plays in regulating gene expression in human cells , we analyzed gene expression and genome-wide cohesin binding patterns in cells from normal subjects and from CdLS probands with mutations in the cohesin regulator NIPBL or in the cohesin structural component SMC1A . We found a strikingly conserved pattern of gene dysregulation in these different cell lines that correlates with disease severity and a significant correlation between gene dysregulation and cohesin binding around misexpressed genes . The observed pattern of binding and misexpression is consistent with cohesin having a putative role as a boundary/insulator interacting protein or transcription factor , the activity of which is disrupted in CdLS probands . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"genetics",
"and",
"genomics/genomics",
"genetics",
"and",
"genomics/gene",
"expression",
"pediatrics",
"and",
"child",
"health/child",
"development",
"genetics",
"and",
"genomics/chromosome",
"biology",
"genetics",
"and",
"genomics/genetics",
"of",
"disease",
"developmental",
"biology/molecular",
"development",
"genetics",
"and",
"genomics/epigenetics",
"developmental",
"biology/developmental",
"molecular",
"mechanisms",
"cell",
"biology/gene",
"expression",
"genetics",
"and",
"genomics/medical",
"genetics"
] | 2009 | Transcriptional Dysregulation in NIPBL and Cohesin Mutant Human Cells |
Parasite biology , by its very nature , cannot be understood without integrating it with that of the host , nor can the host response be adequately explained without considering the activity of the parasite . However , due to experimental limitations , molecular studies of parasite-host systems have been predominantly one-sided investigations focusing on either of the partners involved . Here , we conducted a dual RNA-seq time course analysis of filarial worm parasite and host mosquito to better understand the parasite processes underlying development in and interaction with the host tissue , from the establishment of infection to the development of infective-stage larva . Using the Brugia malayi–Aedes aegypti system , we report parasite gene transcription dynamics , which exhibited a highly ordered developmental program consisting of a series of cyclical and state-transitioning temporal patterns . In addition , we contextualized these parasite data in relation to the concurrent dynamics of the host transcriptome . Comparative analyses using uninfected tissues and different host strains revealed the influence of parasite development on host gene transcription as well as the influence of the host environment on parasite gene transcription . We also critically evaluated the life-cycle transcriptome of B . malayi by comparing developmental stages in the mosquito relative to those in the mammalian host , providing insight into gene expression changes underpinning the mosquito-borne parasitic lifestyle of this heteroxenous parasite . The data presented herein provide the research community with information to design wet lab experiments and select candidates for future study to more fully dissect the whole set of molecular interactions of both organisms in this mosquito-filarial worm symbiotic relationship . Furthermore , characterization of the transcriptional program over the complete life cycle of the parasite , including stages within the mosquito , could help devise novel targets for control strategies .
Human lymphatic filariasis ( LF ) is one of the most debilitating of the neglected tropical diseases , causing severe morbidity as a result of stigmatizing and disabling clinical manifestations ( e . g . , elephantiasis and hydrocele ) [1] , [2] . LF results from infection with several species of mosquito-borne filarial nematodes , including Wuchereria bancrofti , Brugia malayi , or Brugia timori; but W . bancrofti is responsible for 90% of LF infections worldwide . Mosquitoes belonging to a number of different genera , including Culex , Aedes , Mansonia , Anopheles , and Armigeres , can serve as competent vectors . In susceptible mosquitoes , ingested microfilariae ( mf ) traverse the midgut epithelium and migrate to the indirect flight muscles . Migration to the head and proboscis occurs after a series of larval molts to the infective third stage ( L3 ) , which occurs in thoracic muscle cells . The L3 is then passed to the vertebrate host when the infected mosquito takes a bloodmeal [3] , [4] . The ability of certain mosquito species to ingest mf of filarial worm parasites and to support their development after ingestion is an important determinant of filarial parasite transmission [5] . Interactions between mosquitoes and filarial worms can range from an almost benign commensal relationship between organisms [3] , to a fatal competition resulting in death of the host [6] , or to a fatal competition resulting in death of the parasite [7] . Larval development within individual flight muscle cells therefore implies a highly intimate interaction between host and parasite , and often times the number of L3s developing from ingested mf is not constant [3] , [8] , [9] . For mf to successfully reach the L3 stage , host tissue must provide all of the nutritional needs of the rapidly growing nematodes , while simultaneously surviving the mechanical damage caused by nematode migration and development . As a result , infection processes accompany histological changes in host muscle fibers , ranging from depletion of glycogen granules to swelling and disintegration of nuclei or mitochondria ( likely altering the metabolism of these organelles ) to a complete cellular degeneration following the exit of L3s [10] , [11] . It is well accepted that the study of inter-species interactions are fundamental to understanding the biology of vector-borne parasites [12]–[14] , and the filarial worms that cause LF are no exception . In humans , the interactions these parasites engage in with host tissue and the immune system dictate the pathobiology of LF [15] , [16] . And as mentioned , the interactions between filarial worms and their mosquito intermediate hosts determine the course of parasite development to the infective-stage . Previous efforts to understand factors that influence the developmental success of filarial worms in their mosquito vectors have led to a deeper appreciation of the complexity of the parasite-host interplay ( e . g . , [3] , [4] , [17] ) ; however , the inability to make quantitative , molecular-level observations of the parasite and the host , simultaneously , during infection continues to limit our ability to develop a coherent understanding of the fine-grained details mediating filarial worm-mosquito interactions . Past studies have been essentially one-sided investigations focusing on either of the partners . And , most have focused on host factors while failing to take parasite activity into explicit consideration . As a consequence , filarial worm molecular processes underlying development in , and interaction with , the mosquito host tissue remain poorly described . By its very nature , parasite biology cannot be understood without integrating it with that of the host , nor can the host response be adequately explained without considering parasite activity . Accordingly , we initiated transcriptome profiling studies to assess the whole set of molecular interactions of both organisms involved in the dynamic symbiotic processes of filarial worm development in the mosquito . Advances in sequencing-based approaches to quantitative transcriptome profiling ( RNA-seq ) facilitated direct , integrated analysis of mixed-species samples obtained from in vivo infections [18] , in which host tissue samples often contained minute quantities of parasite material . With the increase in sequencing depth , RNA-seq offered improved levels of sensitivity and dynamic range of detection , without the need of predefined species-specific probes ( or selective amplification of parasite RNA ) and reliance on hybridization of targeted oligonucleotides ( e . g . , northern blotting , RT-PCR and microarrays ) . Here , we present a time-resolved dual RNA-seq analysis of B . malayi-Aedes aegypti interactions , which investigated the temporal organization of transcriptional events in both nematode and mosquito tissue from the establishment of infection to after infective-stage parasites had completed development in the mosquito . We report that parasite gene transcription dynamics exhibited a highly ordered developmental program consisting of a series of cyclical and state-transitioning temporal patterns , and we contextualized this parasite developmental program in relation to the concurrent dynamics of the host tissue transcriptome . Furthermore , we critically evaluated parasite gene expression changes during the heteroxenous life-cycle of B . malayi by comparing mRNA abundance in larval stages within the intermediate host relative to developmental stages within the mammalian definitive host in order to provide information on the transcriptomic features underpinning the mosquito-borne parasitic lifestyle .
This study was carried out in strict accordance with recommendations set forth in the National Institutes of Health Guide for the Care and Use of Laboratory Animals . All animals and animal facilities were under the control of the School of Veterinary Medicine with oversight from the University of Wisconsin Research Animal Resource Center . The protocol was approved by the University of Wisconsin Animal Care and Use Committee ( Approval #A3368-01 ) . Ae . aegypti used in this study were maintained at the University of Wisconsin-Madison as previously described [19] . Ae . aegypti , black-eyed Liverpool ( LVP ) strain supports the development of B . malayi mf to L3 [20] . The RED strain is a multiple marker strain previously used in quantitative trait loci ( QTL ) mapping of filarial worm susceptibility to B . malayi [21] , [22] and does not support the development of B . malayi to L3 [23] . Three- to four-day-old mosquitoes were used for blood feeding and sucrose starved for 14 to 16 hours prior to this event . Mosquitoes were exposed to B . malayi ( originally obtained from the University of Georgia NIH/NIAID Filariasis Research Reagent Resource Center ) by feeding on ketamine/xylazine anesthetized gerbils , Meriones unguiculatus . The same animals were used for both biological replicates . Microfilaremias were determined , using blood from orbital punctures , immediately before each feeding and ranged from 90–204 mf per 20 µl of blood . Control mosquitoes were exposed to anesthetized , uninfected gerbils . Mosquitoes that fed to repletion were separated into 30×30×30 cm colony cages and maintained on 0 . 3 M sucrose in an environmental chamber at 26 . 5°±1°C , 75±10% relative humidity , and with a 16∶8 hour light∶dark photoperiod with a 90 minute crepuscular period at the beginning and end of each light period . The sample groups were defined by the time post exposure ( PE ) to the infective blood meal , and thoracic tissues were collected every 24 hours from mosquitoes exposed to an infected or uninfected blood meal . Thoracic tissues were collected for four ( RED ) and eight ( LVP ) consecutive days , and also from non-blood-fed mosquitoes at the time of exposure ( i . e . , time-zero sample ) . For each time point , 30 thoraces were collected per strain per condition . Dissected tissues were immediately flash-frozen in microfuge tubes on dry ice and stored at −80°C until RNA isolation . At 8 days PE , an additional 30 mosquitoes for each strain were dissected , and the head , thorax , and abdomen were examined microscopically as described previously [4] to estimate infection intensity and prevalence ( Table 1 ) . The bootstrap confidence interval of Efron and Tibshirani was used to calculate 95% confidence intervals for mean intensity [24] . A complete set of replicate samples was collected using a distinct generation of mosquitoes blood-fed on the same infected animals in an independent exposure to take into account stochastic variations . Total RNA was isolated from female Ae . aegypti thoraces using the MasterPure RNA Purification Kit ( Epicenter , Madison , WI ) , which included built-in Proteinase K and DNase treatments . A Labgen7b electric tissue grinder ( Cole-Parmer , Vernon Hills , IL ) was used for tissue homogenization , and RNA purity and concentration were determined spectrophotometrically ( NanoDrop ND-1000; Thermo Scientific , Waltham , MA ) . For all samples , RNA integrity was visually assessed by denaturing gel electrophoresis using GelRed staining ( Phenix Research Products , Candler , NC ) and further confirmed by Agilent 2100 bioanalyzer ( Agilent Technologies , Santa Clara , CA ) . Only quality intact RNA was used for RNA sequencing . Multiplex sequencing libraries were generated from 10 µg total RNA ( per library ) using Illumina's mRNA-seq sample prep kit and multiplexing sample preparation oligonucleotide kit ( Illumina Inc . , San Diego , CA ) following the manufacturer's instructions . Separate libraries were constructed for each replicate sample from infected mosquitoes . Equal amounts of total RNA were pooled from replicates prior to library construction for uninfected blood-fed and non-blood-fed samples . Cluster amplification was performed on the Illumina cBOT following the manufacturer's protocol . Sequencing-by-synthesis on the Illumina Genome Analyzer IIx generated 2×50 bp paired-end reads . CASAVA v1 . 7 was used for de-multiplexing , and sequence quality was assessed based on %GC content , average base quality and sequence duplication levels . To minimize the confounding effects of lane-to-lane or run-to-run technical variations , instead of sequencing each library using a single lane , libraries were sequenced over a number of lanes and runs by the use of Illumina's multiplexing capability [25] . The Illumina Genome Analyzer IIx platform allowed 12 different sequencing libraries to be multiplexed in a single lane , and 7–8 separate lanes were available during each sequencing run . Because the total number of unique libraries exceeded the maximum number of usable indices , each index was used to label multiple libraries according to the scheme outlined in Figure 1 . Generation of sequence reads using these combinations of libraries ( occupying 6 of the lanes during each run ) contributed to a more robust dataset for longitudinal as well as cross-sectional analyses . Sequence reads from technical replicates ( i . e . , reads from the same library but sequenced in different lanes and runs ) were combined per library prior to mapping as they shared the same insert-size distribution . Alignment to the combined genomes of Ae . aegypti [VectorBase: AaegL1] and B . malayi [GenBank:DS236884-DS264093] was performed using TopHat v1 . 0 . 14 , a splice junction mapper built upon the short read aligner Bowtie [26] , [27] . The pipeline utilized exon records in the genome annotation to build a set of known splice junctions for each gene model , complementing its de novo junction mapping algorithm . Low quality alignments with mapping quality scores less than five were removed before downstream analyses [28] , [29] . Reads aligned to exonic regions were enumerated for each gene model using the HTSeq package ( v0 . 4 . 7 ) in Python ( ( www-huber . embl . de/users/anders/HTSeq ) . A set of paired-end reads were counted as one fragment to represent a single sampling event , and reads overlapping more than one gene model were counted as ambiguous with the mode parameter set as “union” . In addition , previously published B . malayi sequence reads ( ArrayExpress accession number: E-MTAB-811 ) were aligned and quantified using an identical analysis pipeline to facilitate comparative analysis of mosquito vs . mammalian life-cycle stages [30] . The RNA-seq data were prepared according to “minimum information about a high-throughput sequencing experiment” ( MINSEQE ) recommendations , deposited in the Gene Expression Omnibus ( GEO ) database , and can be accessed via the web [accession number GSE53664] . Statistical analysis of count data was performed with edgeR in Bioconductor [31]–[33] . Genes were retained for analysis if read counts were greater than 0 . 5 counts per million ( cpm ) in at least two libraries and the overall within-replicate negative binomial dispersion was lower than 0 . 5 . Count data were normalized to account for compositional biases using trimmed mean of M ( TMM ) method , in addition to adjusting for differing library sizes [34] . Generalized linear models ( GLM ) with explanatory variables of host strain , infection status and time was fit to the count data for each mosquito gene , and the biological coefficient of variation ( BCV ) was inferred using the Cox-Reid profile-adjusted likelihood method [33] . For the analysis of filarial worm genes , only the host strain and time were included as factors in the GLM . Information sharing was used so that genes could take individual values for the BCV , but stabilized towards a common value [31] . To identify parasite and host genes that were differentially transcribed between any of the time points ( within each time-series ) , we used nested interaction models , and conducted an ANOVA-like test for any differences by testing for multiple coefficients being equal to zero . Using full interaction models , we then assessed host genes that responded differently at any of the time points in the infected host relative to the uninfected host . Similarly , we tested parasite genes for differential transcription at any of the time points in the susceptible host ( LVP ) relative to the refractory host ( RED ) . Adjusted p-values were reported after correction for multiple testing using the Benjamini and Hochberg method [35] . All data were transformed using a log2 fold-change , and because log fold-change ( logFC ) estimates for genes with small read counts can be highly variable , we moderated logFC estimates using the ‘preFC’ function in edgeR . Thus , undefined values were avoided and poorly defined fold-changes for low counts were shrunk towards zero . We then used logFC values between sequential time points for each gene as a basis for clustering . After median-centering , K-means clustering was performed using a ( uncentered ) Pearson correlation as a distance metric to group and summarize expression profiles into common temporal patterns . Figure of merit ( FOM ) plots were utilized in determining the appropriate number of clusters to ensure informative partitioning [36] . Using goseq in Bioconductor , Gene Ontology ( GO ) terms ( or KEGG pathways ) statistically over-represented in each cluster were compiled to help interpret the biological implications of the expression patterns [37] . To assist in interpretation and visualization , we systematically summarized significant GO terms into a representative subset using an algorithm implemented in REVIGO ( revigo . irb . hr ) that relies on semantic similarity measures ( 16 ) . GO annotation was retrieved from UniProtKB-GOA [38] and VectorBase [39] . MDS plot ( a type of unsupervised clustering ) was generated in edgeR to analyze sample relationships . Distances between each pair of RNA-seq profiles corresponded to the average ( root-mean-square ) of the top 500 largest absolute logFC between each pair of samples .
To study gene expression dynamics during filarial worm-mosquito interactions using RNA-seq , we collected thoracic tissue samples from Ae . aegypti ( LVP ) exposed to either a B . malayi-infected or uninfected bloodmeal . Thoracic tissue samples were collected every 24 hours after exposure for eight days , and also from non-blood-fed mosquitoes at the time of exposure . LVP is a genetically selected susceptible strain that supports the complete development of B . malayi to the infective stage [20] . In addition , a matching set of tissue samples was concurrently collected for four days using refractory Ae . aegypti ( RED ) to comparatively evaluate the transcriptome dynamics in an incompatible parasite-host association ( Table 1 ) . In this refractory mosquito , B . malayi mf traverse the midgut and become intracellular within the thoracic musculature , but after reaching their site for development , parasites fail to develop and die within a few days [23] . Poly ( A ) -selected mRNA samples isolated from these tissues were subjected to Illumina sequencing ( 38 libraries across 48 lanes ) , and the resulting paired-end reads ( 2×50 bp ) were aligned to the combined genomes of Ae . aegypti and B . malayi using a splice-junction aware aligner . Out of 1 . 17 billion sequenced reads , 987 million reads ( 84% ) mapped unambiguously to the reference genomes ( Figure 2 ) . In the susceptible mosquito ( LVP ) , the proportion of parasite reads relative to host reads increased from 0 . 1 to 9 . 0% during the course of infection , consistent with nematode growth from mf to L3 ( [40] and Figure 3 ) . Interestingly , these data suggested that during larval development leading up to the first molt ( days 1–4 ) the rate of increase in the parasite's total mRNA output substantially exceeded that of body growth . This burst of transcriptional activity coincided with the initiation of tissue and organ differentiation following recovery from a developmental arrest during the mf stage . In the refractory mosquito ( RED ) , the proportion of parasite reads remained at 0 . 1% and declined to lower levels by day 4 , consistent with the observed failure of parasite development and survival ( Figure 3 ) . Next , in both parasite and host , we examined genome-wide temporal changes in transcript expression patterns in an effort to better understand the dynamic progression of infection processes ( Figure 4 and 5 ) . Genes displaying statistically significant differences in mRNA abundance levels across time points were identified ( negative binomial generalized linear model , likelihood ratio test , p<0 . 01 and fold-difference >2 ) . These “non-flat” profiles were grouped and summarized into common temporal patterns using k-means clustering . Mean expression changes between sequential time points were computed and visualized to better evaluate the temporal dynamics across clusters in reference to specific time periods . In addition , Gene Ontology ( GO ) terms statistically over-represented in each cluster were compiled to help interpret the biological implications of expression patterns ( Figure 5 ) . An overall assessment of the transcriptional dynamics in B . malayi during successful development from mf to L3 indicated a progressive and complex turnover in transcriptional contents ( Figure 4 and 5 ) . Abundant transcripts at day one diminished over time , and new sets of transcripts emerged as the parasite developed . Although some expression clusters displayed patterns of gradual incremental changes , a more prevalent cluster type was marked by patterns of disproportionately larger change at specific time intervals followed by transition to a new steady level . Collectively , these state-transitioning patterns form a series of transcriptional “waves” , characteristic of developmental programs imposing order on cellular biogenesis [41] . Cyclical patterns also were evident , consistent with gene transcription associated with recurrent processes underlying nematode molting [42] . The thoracic tissue of the mosquito in which filarial worms developed also underwent extensive transcriptional changes ( Figure 4 and 5 ) . Comparisons of infected and uninfected host profiles indicated that normal physiological activities of the mosquito , most notably those related to blood-feeding responses , were accountable for a large proportion of these time-dependent expression changes ( [43] and Figure 6 ) . Concordantly , overall transcriptome dynamics were considerably higher during the first two days after blood feeding . Infection-induced gene expression changes , on the other hand , exhibited distinct temporal kinetics that appeared to reflect closely the histological phenotypes previously described in filarial worm infected Ae . aegypti muscle tissues ( e . g . , [10] , [11] ) . In the following sections , we describe and discuss the temporal coordination of parasite and host gene expression along the developmental timeline of filarial worm-mosquito interactions . After ingestion with the bloodmeal , mf traverse the midgut epithelium within two hours and begin their migration through the hemocoel to the thoracic flight muscles [19] . Myofibrils are parted to form tunnels as larvae enter host cells , within which they lie uncoiled and parallel to muscle fibers [10] . Over the next two days , mf transform into shorter , sausage form larvae ( L1 ) . This process involves rearrangement and growth of preexisting microfilarial structures , as well as extensive cuticular reorganization [44] . RNA-seq expression profiles of B . malayi indicated that , between days one and two , transcriptional induction was prominent among gene clusters P1 , P5 and P6 ( Figure 4A ) . GO categories associated with glycolysis , mitochondrial hydrogen-transporting ATP synthase , signal peptidase complex , DNA replication , and phosphoric diester hydrolase activity were overrepresented in these clusters . These data suggest that filarial worms respond transcriptionally to their changing metabolic and energy needs in the intracellular environment of their new host during re-initiation of larval development . Induction of genes involved in DNA replication likely is important for supporting increased cell division required for tissue differentiation . During this period , a decreasing trend was observed in clusters P13 and P14 , and in the latter group , the pattern extended to day three . Genes with nucleic acid or zinc ion binding activity were highly represented in these clusters , many of which are C2H2-type zinc finger proteins that may serve as trans regulators of gene expression [45] . In the mosquito host , a proportion of both parasitized and non-parasitized muscle fibers show signs of structural abnormality . Within the first two days after infection , nuclear and mitochondrial enlargement occurs , and the number of affected fibers , the proportion of organelles involved , and the degree of damage progressively increases as larval development proceeds . Although the proportion is very low , severe damage and cellular degeneration , likely caused by migrating mf , also appear early in infection , but without a progression in the number of affected fibers ( see [11] ) . The mosquito host response profile , representing transcript levels of infected animals relative to the uninfected controls , indicated that gene clusters HR1 and HR2 ( Figure 4A ) , containing groups of stress response genes , such as small heat shock proteins with alpha-crystallin domains , displayed impulse-like patterns with acute dynamics during early infection . The distinct temporal pattern of induction among these genes , peaking at days one , six and eight , suggested a close link between the kinetics of the transcriptional response and the degree of cellular damage and mechanical disruption the host experiences as parasites penetrate into and out of muscle fiber , or actively ingest host cell contents . Middle to late L1 development , from day three to the first molt that occurs between day four and five , is characterized by numerous mitotic divisions , lengthening of the body , and differentiation of internal structures , including a well-defined intestine and a divided-type esophagus ( the anterior region muscular , the posterior glandular ) that is formed around the pharyngeal thread of the mf [46] . B . malayi gene clusters P7 and P8 ( Figure 4A ) , enriched with genes implicated in ion channel activity and transmembrane transport , exhibited transcriptional increase between days two and four , and their abundance levels were maintained on subsequent days . In addition , transient transcription increases were observed among clusters P2 , P3 and P9 ( Figure 4A ) , in which GO categories involved in calcium ion binding , response to stress , serpin activity , cuticle , metallopeptidase and steroid hormone receptor activity were overrepresented . Interestingly , these clusters showed a second peak during second stage larvae ( L2 ) development , forming a cyclical pattern with respect to molting events . By filtering on the direction ( increase or decrease ) and magnitude of transcript abundance changes over time points , we identified genes showing strong periodic patterns where transcript levels oscillated between high levels during intermolt periods and low levels during ecdysis ( Figure 4B ) . Given our sampling rate , we could discern three groups with distinct kinetics , reaching their maximum levels at different times ( C1–3 ) . Genes with pulsatile transcription dynamics identified in the present study included a number of structural and regulatory components ( such as cuticular collagens , cuticle-digesting proteases and nuclear receptor transcription factors ) that have been predicted to participate in various aspects of the molting process [42] , [47] . B . malayi , like all nematodes , progresses through its life stages via molts , each of which involves synthesis and secretion of a new cuticle , followed by separation and shedding of the old cuticle . From intracellular signaling to extracellular execution , molting requires a series of complex molecular reactions under tight spatiotemporal regulation at the level of the hypodermis . Also , equally critical to successful molting is a precise coordination of tissue development throughout the animal [48] . In this context , the specific ordering of gene clusters with cyclical patterns is highly intriguing and informative , shedding light on the temporal organization of molecular events underlying the periodic episodes of molting in relation to the progressive life stage transitions . In addition to the initial wave of acute reactions to the invading parasites , the mosquito host's transcriptional responses to developing larvae also contained relatively moderate but sustained temporal patterns of induction , as illustrated in cluster HR4 ( Figure 4A ) . These expression changes , extending mostly over the six to seven day period , likely reflect long-term chronic effects of filarial worm infection , part of which may represent a compensatory host response to energy and nutrient imbalances caused by intracellular parasite development . Overrepresented GO terms in this cluster included oxidoreductase activity , purine nucleotide binding , phosphoenolpyruvate carboxykinase activity , cAMP biosynthesis process , and gluconeogenesis . Because parasite-host interactions inherently involve two interconnected biological systems with a net flow of energy and nutrients , a metabolic perturbation likely is inevitable in the host [49] . It has been reported that in Ae . aegypti , filarial worm-infected muscle fibers show a large decrease in the amount of glycogen granules [44] , and our data point to a transcriptional induction in the pathway of gluconeogenesis , by which cells synthesize glucose from metabolic precursors , such as glycogen . Data further suggest that this shift in metabolic state likely involved regulation at the level of cAMP synthesis and phosphoenolpyruvate carboxykinase activity , the latter of which controls a rate-limiting step of the pathway . Another notable feature of the infected host tissue transcriptome was the observed expression changes in glutathione transferase and glutathione peroxidase , whose main functions are to inactivate toxic products of oxygen metabolism . Induction of these antioxidative enzymes could be functioning to enhance protection from oxidative damage , or alternatively , this also may change the redox state of the host cell environment to favor parasite survival [50] . By day five , the first molt is complete , which marks the transition between L1 and L2 . At this time , L2s begin feeding through their open stoma and newly developed pharynx with a complete cuticularized lumen . Flight muscle mitochondria of the mosquito appear within the larval esophagus and midgut after six days of development , indicating active tissue ingestion [51] . During the following days , the larvae elongate and the gut undergoes further development . The rectum , however , remains closed with an anal plug , preventing egress of larval gut contents into host tissue . This could represent one of the mechanisms by which filarial worms restrict trauma to the host cell during their intracellular development [52] . The lumen of the rectum is then formed within the anal vesicle just before the molt to the third stage [46] . Between days five and six , the genital primordium is formed , the position of which differs in males and females; it is at or just behind the esophago-intestinal junction in males , and at the midesophageal level in females [53] . During the middle to late L2 stage , the body wall consists of cuticle , chords ( differentiated small dorso-ventral chords and broad lateral chords ) and muscle components . A slight loosening of the cuticle at the head and tail is typically observed at day seven . By day eight , most larvae have either completed the second molt or are in the process of molting . Within a day or two , L3 migrate from the thoracic muscles through the head to the labium of the proboscis , from which they exit the mosquito during blood-feeding to infect the mammalian host . During L2 development in B . malayi , a diverse set of genes in clusters P2 , P3 and P9 , with oscillatory expression patterns , exhibited their second peak of transcription ( Figure 4A ) . Clusters P4 , P10 and P11 , on the other hand , showed state-transitioning patterns with transcriptional induction between days five and seven . These likely represent specific processes that were initiated during the L2 developmental period . Overrepresented GO categories in these clusters included cysteine-type peptidase activity , structural constituent of cuticle and glycolysis . Subsequently , between days seven and eight , transcript levels increased in cluster P12 , in which gene sets associated with cuticle components , transmembrane transport and chloride transport were enriched . Together , these clusters encompassed a large array of functional categories , as well as genes with unknown function . Although the magnitude of change was comparatively small , clusters P15 and P16 showed a decreasing trend over the period of L2 development ( Figure 4A ) . Enrichment of genes implicated in ribosome , translation and nucleic acid binding in these clusters suggested that genes involved in basal cellular activity were expressed in lower levels during late L2 and L3 compared to those at earlier stages ( Figure 5 ) . Overall , these data suggest a considerable change in the organism's transcript composition and a relatively uneven distribution of transcript abundances ( as supported by the decrease in Shannon Diversity Index [54] between days six and eight; Figure 7 ) in the second stage during which high levels of cellular differentiation and tissue development occur . Nevertheless , such interpretation requires caution because our temporal data are confounded with variations in transcript detection limits , due to the differences in sequence sampling depth over time ( Figure 3 and 7 ) . The mosquito transcriptional response during L2 development leading to L3 emergence was largely characterized by further induction of stress response genes of the heat shock protein family , many of which were responsive at the onset of infection ( clusters HR1 and HR2 ) . These may play a crucial role in the repair process of the host cell by serving as molecular chaperones , and thereby possibly suppressing or delaying necrosis during intracellular parasite development [55] . One striking observation from past histological studies is that , despite the large increase in parasite size and host tissue consumption , almost all infected cells harboring live larvae do not undergo cellular breakdown or degeneration until larval development is complete and the L3s migrate out of the flight muscle fibers [10] , [11] . It is thought that some physical or chemical factor associated with the terminal phase of larval development causes significant damage above a tolerable threshold , initiating necrosis . It remains unknown whether filarial worms actively modulate host cell survival in an effort to conserve their habitat . In Ae . aegypti about 5–15% of all fibers eventually degenerate completely , which represents a permanent loss of contractile capacity , likely contributing to decreased flight activity and longevity [56]–[58] . B . malayi gene expression changes discussed thus far underlie successful developmental progression to the infective stage , which is dependent on the mosquito host tissue environment . Changing the genetic background of the host to an incompatible strain ( RED ) results in a failure of the parasite to develop and survive . Using a negative binomial generalized linear model , we assessed time dependent transcriptional changes between days one and four that were different in B . malayi depending on host environment , i . e . , compatible ( LVP ) versus incompatible ( RED ) . The most significant transcriptional alterations , as judged by statistical evidence , are presented in Figure 8 ( p<0 . 05 ) . Our data indicated that transcripts encoding hAT family dimerization domain containing protein , transmembrane protein , Leucine Rich Repeat family protein , dehydrogenases , and spectrins , among others , showed an increasing trend during successful parasite development . In Caenorhabditis elegans , a mutation in β-spectrin resulted in retarded growth and paralysis , suggesting that it is required for normal nematode development [59] . Of particular interest is the presence of putative transcriptional regulators in this list , such as nuclear hormone receptor and C2H2 type Zinc finger protein [60] . Although their specific function in B . malayi has not been elucidated , these could be involved in the control of key developmental pathways that are initiated during early larval development in the mosquito . Among the transcripts that showed an increasing trend in non-developing or dying worms , was a calpain gene that encodes a member of the Ca2+-dependent cysteine protease family . Calpain activation is an integral component of necrosis , and has been implicated in apoptotic cell death [61] , which is consistent with the refractory host environment of the RED strain preventing filarial worm development , and the resultant death of the parasite ( Table 1 ) . However , because of the extensively divergent genetic backgrounds between RED and LVP , the host genetic factors that confer this lethal environment could not be reasonably inferred from the between-strain differences in the host tissue transcriptome . To further investigate the transcriptomic features of this heteroxenous parasite in the context of host transitions , we extended our analysis to include our previously reported RNA-seq profiles ( see [30] ) and comparatively analyzed the mosquito stages ( L1 , L2 , and L3 ) relative to the mammalian stages ( egg , microfilaria , L4 , adult male , and adult female ) . To facilitate this comparison , read count data from individual libraries were combined per stage to construct expression profiles enriched for ( or representative of ) each life-cycle stages ( Figure 9A ) . Sample relations based on multi-dimensional scaling ( MDS ) revealed a biologically plausible pattern where expression profiles of successive stages were overall more similar to each other . From these profiles , genes differentially transcribed in the mosquito stages relative to the mammalian stages were identified ( p<0 . 01 ) . Of the 413 significant genes , 200 genes displayed transcriptional induction predominantly during the mosquito stages ( Figure 9B ) . Overrepresented among these genes were GO terms , such as proteolysis , cell adhesion , oxidation-reduction process , and malate metabolic process , as well as several Molecular Function terms that appear consistent with the above-mentioned Biological Process terms ( p<0 . 01; Dataset S1 ) . Peptidases and peptidase inhibitors , including cathepsins , serpins and cystatins were also strongly represented , which was in agreement with previous studies giving support to their possible role in essential developmental processes or specific interactions with the host [62] , [63] . Enrichment analysis using KEGG pathway further highlighted pyruvate metabolism , citrate cycle , and arachidonic acid metabolism ( p<0 . 01; Dataset S1 ) . A closer look at individual metabolic genes indicated a highly specific induction among the key components of anaerobic mitochondrial pathways , such as phosphoenolpyruvate carboxykinase ( GenBank:Bm1_25195 ) , malic enzyme/malate dehydrogenase ( GenBank:Bm1_04060 , GenBank:Bm1_08150 , GenBank:Bm1_46465 , GenBank:Bm1_53540 ) , and pyruvate dehydrogenase ( GenBank:Bm1_26945 ) [64] , [65] . Adult filarial worms are homolactate fermenters that primarily produce lactate via cytosolic pathways when glucose is in excess [66] . Our data suggested that anaerobic mitochondrial metabolism involving malate dismutation is likely activated during parasite development in the mosquito , increasing the total yield of ATP per mole of glucose metabolized . A developmental regulation in energy metabolism as such could represent an important adaptive strategy for survival when glucose concentrations become limiting in the host milieu [67] . Of the 200 genes associated specifically with parasite stages in the mosquito , 51 have been either predicted to be secreted or shown to be secreted at some stage in the life-cycle ( Dataset S1 ) [53] , [68] . These molecules , including protease inhibitors , carbohydrate-binding proteins , and the abundant larval transcript ( ALT ) proteins , possibly constitute an important part of the parasite-host interface , some of which could potentially impact nematode survival and progression of infection . Transcriptomic approaches have been instrumental in studying the developmental regulation of gene expression underlying filarial worm life-cycle progression [30] , [69]–[72] . Advances in high-throughput nucleotide sequencing now enable us to interrogate the in vivo transcriptome dynamics of this metazoan parasite during its obligatory intracellular developmental phase in the mosquito host , which is critical for the maintenance of the disease cycle and transmission . Integrative analysis of the host transcriptome in a dual RNA-seq approach provides a high-resolution overview of the parasite-host system in which each partner's transcriptional state is dependent on the other partner . Studies of this nature will grow increasingly important in investigating the unique and common biological mechanisms that enable parasites to maintain their symbiotic association with the host , as well as the host strategies to counter parasitic infections , within a unified framework . | In a parasitic relationship , both host and parasite genotypes influence the parameters of their relationship . Previous studies examining host-parasite systems have examined the effects of the genotype of the host or the parasite on the relationship , but due to limitations of technology , have rarely examined interactions between the two genotypes . Here , we utilized a dual RNA sequencing ( RNA-seq ) approach to examine the entirety of the known transcriptomes and their interactions in the dynamic process of filarial worm development in the mosquito . In addition , the unprecedented sequencing depth achieved with this technology allowed us to compare , for the first time , parasite gene expression of larval developmental stages within the intermediate host with those life cycle stages found within the mammalian definitive host . These data provide a strong foundation for understanding how Brugia malayi interacts with its vector's transcriptome temporally during its complex life cycle and also simultaneously provides information on how Aedes aegypti responds to filarial worm infection . These data are extremely valuable for future studies of the underlying mechanisms of this mosquito-filarial worm relationship . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"and",
"Discussion"
] | [
"genome",
"expression",
"analysis",
"medicine",
"and",
"health",
"sciences",
"genomics",
"pathology",
"and",
"laboratory",
"medicine",
"entomology",
"host-pathogen",
"interactions",
"genome",
"analysis",
"transcriptome",
"analysis",
"genetics",
"biology",
"and",
"life",
"sciences",
"microbiology",
"computational",
"biology",
"pathogenesis",
"zoology",
"parasitology",
"helminthology"
] | 2014 | Dual RNA-seq of Parasite and Host Reveals Gene Expression Dynamics during Filarial Worm–Mosquito Interactions |
Diversity-generating retroelements ( DGRs ) are in vivo sequence diversification machines that are widely distributed in bacterial , phage , and plasmid genomes . They function to introduce vast amounts of targeted diversity into protein-encoding DNA sequences via mutagenic homing . Adenine residues are converted to random nucleotides in a retrotransposition process from a donor template repeat ( TR ) to a recipient variable repeat ( VR ) . Using the Bordetella bacteriophage BPP-1 element as a prototype , we have characterized requirements for DGR target site function . Although sequences upstream of VR are dispensable , a 24 bp sequence immediately downstream of VR , which contains short inverted repeats , is required for efficient retrohoming . The inverted repeats form a hairpin or cruciform structure and mutational analysis demonstrated that , while the structure of the stem is important , its sequence can vary . In contrast , the loop has a sequence-dependent function . Structure-specific nuclease digestion confirmed the existence of a DNA hairpin/cruciform , and marker coconversion assays demonstrated that it influences the efficiency , but not the site of cDNA integration . Comparisons with other phage DGRs suggested that similar structures are a conserved feature of target sequences . Using a kanamycin resistance determinant as a reporter , we found that transplantation of the IMH and hairpin/cruciform-forming region was sufficient to target the DGR diversification machinery to a heterologous gene . In addition to furthering our understanding of DGR retrohoming , our results suggest that DGRs may provide unique tools for directed protein evolution via in vivo DNA diversification .
Diversity-generating retroelements ( DGRs ) have been identified in numerous bacterial phyla [1] , [2] . Although most DGRs are bacterial chromosomal elements , they are prevalent in phage and plasmid genomes as well . The prototype DGR was identified in a temperate bacteriophage , BPP-1 , on the basis of its ability to switch tropism for different receptor molecules on host Bordetella species [3] . Tropism switching is mediated by a phage-encoded DGR which introduces nucleotide substitutions in a gene that specifies a host cell-binding protein , Mtd ( major tropism determinant ) , positioned at the distal tips of phage tail fibers . This allows phage adaptation to the dynamic changes in cell surface molecules that occur during the infectious cycle of its bacterial host [3] . Comparative bioinformatics predicts that all DGRs function by a fundamentally similar mechanism using conserved components ( [1]; Gingery et al . , unpublished data ) . These include unique reverse transcriptase ( RT ) genes ( brt for BPP-1 ) , accessory loci ( avd or HRDC ) , short DNA repeats , and target genes that are specifically diversified [1]–[4] . As illustrated by the BPP-1 DGR shown in Figure 1A , diversity results from the introduction of nucleotide substitutions in a variable repeat ( VR ) located at the 3′ end of the mtd gene [1]–[4] . Variable sites in VR correspond to adenine residues in a homologous template repeat ( TR ) , which remains unchanged throughout the process [1]–[4] . Transcription of TR provides an essential RNA intermediate that is reverse transcribed by Brt , creating a cDNA product which ultimately replaces the parental VR [4] . During this unidirectional retrotransposition process of mutagenic homing , TR adenines are converted to random nucleotides which subsequently appear at corresponding positions in VR [1]–[4] . Adenine mutagenesis appears to occur during cDNA synthesis and is likely to be an intrinsic property of the DGR-encoded RT [4] . Located at the 3′ end of VR is the IMH ( initiation of mutagenic homing ) region , which consists of at least two functional elements: a 14 bp GC-only sequence [ ( GC ) 14] which is identical to the corresponding segment of TR , and a 21 bp sequence containing 5 mismatches with TR that determines the directionality of information transfer [1] . Using a saturating co-conversion assay , we have precisely mapped a marker transition boundary that appears to represent the point at which 3′ cDNA integration occurs and information transfer begins [4] . This maps within the ( GC ) 14 element and we previously postulated that it represents the site of a nick or double-strand break in the target DNA [4] . If true , the resulting 3′ hydroxyl could serve to prime reverse transcription of the TR-derived RNA intermediate in a target DNA-primed reverse transcription ( TPRT ) mechanism [4]–[7] . cDNA integration at the 5′ end of VR requires TR/VR homology and may occur via template switching during cDNA synthesis [4] . There are 23 adenines upstream of the ( GC ) 14 element in the BPP-1 TR , each of which is capable of variation [3] . The theoretical maximum DNA sequence diversity is ∼1014 , which translates to a maximum protein diversity of nearly 10 trillion distinct polypeptides at the C-terminus of Mtd . For Mtd and other DGR-diversified proteins , co-evolution has resulted in the precise positioning of TR adenines to correspond to solvent exposed residues in the ligand binding pockets of variable proteins [8] , [9] . As implicated in Figure 1A , mutagenic homing occurs through a “copy and replace” mechanism that precisely regenerates all cis-acting components required for further rounds of diversification [4] . This allows the system to operate over and over again to optimize ligand-receptor interactions . The goal of this study was to characterize requirements for target site recognition by the BPP-1 DGR . Along with insights into the mechanism of mutagenic homing , our results reveal engineering principles that allow DGRs to be exploited to diversify heterologous genes through a process that is entirely contained within bacterial cells .
5′ and 3′ boundaries of the BPP-1 DGR target sequence were delineated using a PCR-based assay that specifically detects VR sequences that have been modified by DGR-mediated retrohoming [4] . The system consists of a donor plasmid ( pMX-ΔTR23-96 , Figure 1B ) carrying avd , a modified TR containing a 30 bp tag ( TG2 ) , and brt co-expressed from a BvgAS-regulated promoter [4] , and a recipient prophage genome deleted for avd , TR , and brt ( BPP-1ΔATR , Figure 1C ) . TR retrotransposition from the donor plasmid to the recipient prophage VR creates a “tagged” VR that can be detected using primer pairs specific for the tag and VR-flanking sequences ( P1/P4 and P2/P3 in Figure 1B; Table 1 ) . Controls include the demonstration that homing products are Brt-dependent and contain mutagenized adenines . An advantage of this assay is that it does not require infectious phage particle formation and consequently allows manipulation of sequences that are required for Mtd function . Deletions were introduced into VR and adjacent sequences in BPP-1ΔATR lysogens ( Figure 1C and Figure S1 ) and the abilities of mutated prophages to serve as recipients in retrohoming assays were measured ( see Materials and Methods ) . As shown in Figure 1D , sequences upstream of VR were dispensable for DGR homing ( lanes 4&13 ) . A deletion mutation that truncates the first 20 bp of VR still supported homing , although at a decreased level ( lanes 5&14 ) . Sequence analysis of homing products for this mutant suggested that 5′ cDNA integration occurred at cryptic sites within the truncated VR , although 3′ cDNA integration occurred in a normal manner ( Figure 1D , lanes 5&14; Figure S2 ) . At the 3′ end , homing was highly dependent on a 35 bp region located downstream of VR ( lanes 6&15 vs . lanes 7&16 ) . This implicated sequences with 8 bp inverted repeats that could potentially form a hairpin structure in ssDNA or a cruciform structure in dsDNA as a possible determinant of DGR target function ( Figure 1C ) . Additional analysis showed that deletion of sequences immediately downstream of the stem was well tolerated ( 3′Δ58 , Figure 1C and 1E ) , while further deletions at the 3′ end ( 3′Δ68 ) reduced target function to essentially non-detectable levels in homing assays . In the experiments in Figure 1 , homing products were not detected using a donor plasmid expressing enzymatically inactive Brt ( BrtSMAA , in which the active site motif YADD is replaced by SMAA; [1] , [3] , [4] ) , and sequence analysis of products generated with primer sets P1/P4 and P2/P3 demonstrated transfer of the TG2 tag from TR to VR . Adenine mutagenesis was observed in ∼53% of clones containing P1/P4 products and ∼32% of clones containing P2/P3 products ( data not shown ) , which had 3 and 2 TR adenine residues available for mutagenesis , respectively . These observations indicated that true DGR homing products were being detected . Equivalent amounts of template phage DNA , as measured by quantitative PCR , were included in each experiment ( lanes 19–27 , Figure 1D; lanes 17–24 , Figure 1E ) . We next determined whether the primary sequence or the secondary structure of the putative hairpin/cruciform located downstream of VR is important for function . To disrupt the structure , 7 consecutive residues proximal to the loop on the 3′ half of the stem were changed to their complementary residues ( StMut , Figure 2A ) . The resulting mutant was essentially unable to support DGR homing at a level that could be detected in PCR-based assays ( lanes 3&9 , Figure 2B ) . Complementary substitutions were subsequently introduced to the 5′ half of the stem to generate StRev ( Figure 2A ) . If the primary sequence is important , the StRev recipient should remain non-functional . Alternatively , if the structure of the stem is the critical element , restoring base pairing interactions might restore DGR target function . As shown in Figure 2B ( lanes 5&11 ) , this appears to be the case , as the StRev mutant regained DGR homing activity . Homing products were verified by sequencing and adenine mutagenesis was observed ( Figure S3 ) . Phage tropism switching assays provide a quantitative measure of DGR function [1] , [3] , [4] . Although the evolution of new ligand specificities is an inherently stochastic process , the frequency at which it occurs reflects the combined efficiencies of retrohoming and adenine mutagenesis . In Figure 2C , tropism switching was measured using BPP-1ΔATR or mutant derivatives complemented with plasmid pMX1 , which provides avd , TR and brt in trans ( see Materials and Methods ) . The StMut mutation resulted in over a 1000-fold decrease in tropism switching , which was restored to near WT levels by the StRev allele . Sequence analysis of VR regions in phages with switched tropisms ( 5 random clones each ) confirmed adenine mutagenesis in every case ( Figures S4 , S5 , S6 ) . Taken together , these data argue that the ability to form a hairpin or cruciform structure , as opposed to the primary sequence of the inverted repeats , is a critical determinant of target site recognition . The residual tropism switching activity of StMut phage suggests that hairpin/cruciform-independent pathways may exist , although they operate at a much lower efficiency . To determine if the hairpin/cruciform structure can form in vitro , supercoiled plasmids carrying WT or mutant BPP-1 DGR target sequences were isolated and treated with phage T7 DNA endonuclease I , followed by primer extension with 5′ end-labeled primers to identify specific cleavage sites [10] , [11] . T7 DNA endonuclease I is a structure-specific enzyme that resolves DNA four-way ( Holiday ) junctions and has previously been used to identify DNA hairpin or cruciform formation [10] , [11] . As shown in Figure 3 , cleavage sites were detected on both DNA strands in the hairpin/cruciform structure , with major cleavage sites at or near the four-way junction . Minor cleavage sites were also detected at or near the loop , as T7 DNA endonuclease I also has some activity on single-stranded DNA [12] . T7 endonuclease I cleavage at the hairpin/cruciform region requires structure formation , as plasmids containing a disrupted stem ( StMut ) were not cleaved in the corresponding region . Linearization of plasmids containing the WT sequence eliminated cleavage , suggesting that negative supercoiling is required for hairpin/cruciform formation [13] , [14] . These results demonstrate that hairpins can form on either strand of the target DNA . Although it is likely that they form simultaneously on both strands to create cruciforms , this is not directly addressed by enzyme cleavage assays , hence the hairpin/cruciform designation . We next determined whether the orientation of the target sequence relative to the phage genome is important for DGR retrohoming . In the experiment in Figure 4A , a segment of the BPP-1ΔATR prophage that includes VR and its flanking sequences was inverted , and PCR-based DGR homing assays were performed with donor plasmid pMX-ΔTR23-96 . DGR homing into the inverted target occurred at a level comparable to that of the WT control ( Figure 4B ) , and sequence analysis indicated that normal homing products were produced ( Figure S7 ) . These results show that the polarity of phage replication is not important for DGR homing , and that the hairpin/cruciform structure functions in a manner that is independent of its orientation relative to the leading or lagging strands formed during DNA replication . Inverted repeats are nearly always found downstream of VR sequences in target genes [Gingery et al . , unpublished data] , as illustrated by the phage DGR sequences shown in Figure 5 . These elements display a striking pattern of similarity , suggesting they have conserved and important functions . In each case , hairpin/cruciform structures with 7–10 bp GC-rich stems and 4 nt loops can potentially be formed . Although stems are always GC-rich , their sequences differ , while loops are more conserved with the consensus sequence ( 5′GRNA3′ , with R = A or G , N = any nucleotide ) in the sense strand . The exact distance between the hairpin/cruciform structures and the 3′ ends of their respective VRs appears to be quite flexible . We took advantage of the BPP-1 DGR system to test the relevance of these patterns of conservation , with the goal of generating a more comprehensive understanding of parameters important for target site recognition . We first studied requirements for stem length and sequence and found that although minor changes are tolerated , the WT configuration appears to be optimized for BPP-1 DGR function . Of the stem length variants in Figure 6A , extensions are better tolerated than deletions . Removal of 2 , 4 or 6 bp proximal to the loop results in markedly decreased activity in both PCR-based homing ( Figure 6B ) and phage tropism switching assays ( Figure 6C ) , to levels similar to those observed with the StMut allele in which the stem is completely abolished ( Figure 2A ) . Insertion of 2 bp next to the loop had little effect on activity , while longer insertions gradually decreased target site function . Keeping the length of the stem constant , a sequence change in the middle of the stem that converts 4 GC base pairs to AT base pairs ( StAT , Figure 6A ) greatly reduced , but did not eliminate function . We next tested the effects of altering the sequence and size of the loop using the mutant constructs shown in Figure 7A . Substituting CTTT for the consensus loop sequence GAAA , or increasing the size of the loop by as little as 2 nt , decreased activity in PCR-homing ( Figure 7B ) and tropism switching assays ( Figure 7C ) to near background levels . Based on these experiments , it appears that an 8–10 bp GC-rich stem is optimal for BPP-1 DGR homing , and that both the size and sequence of the 4 bp loop are critical for function . Our results correlate with the patterns of conservation shown in Figure 5 . In the experiments in Figure 8 , we tested the effects of altering the position of the hairpin/cruciform with respect to the 3′ boundary of VR and probed sequence requirements for the intervening region . SpM4 ( Figure 8A ) , in which the 4 residues in the spacer were switched to the complementary nucleotides , retained WT activity ( Figure 8B ) . In contrast , deletion of the spacer ( SpD4 ) resulted in a significant decrease in target function . The SpM4 and SpD4 mutations eliminate the mtd stop codon and generate non-infective phages , obviating the ability to measure tropism switching . Nonetheless , their relative levels of activity were readily apparent in PCR-homing assays . Expansion of the spacer was tolerated to a greater extent than deletion . SpI3 , which has a 3 bp insertion in the spacer ( Figure 8A ) , showed no significant defect in PCR-homing or phage tropism switching assays ( Figure 8B and 8C ) , but longer insertions gradually decreased target site function . The SpI6 insertion , which increases the distance between the hairpin/cruciform structure and the 3′ end of VR by 6 bp , retains a measurable level of activity . We took advantage of this and used a marker coconversion assay ( Figure S8; [4] ) to determine the relationship between the position of the hairpin/cruciform structure and the site at which information transfer initiates . As summarized in Figure 8D , our coconversion assay measured transfer of nucleotide polymorphisms from tagged TR donors to a recipient VR carrying the SpI6 mutation using PCR-based homing assays ( data not shown ) . With the WT recipient , a coconversion boundary occurs between positions 107 and 112 , and this was interpreted as representing the site at which TR-derived cDNA synthesis initiates [4] . As shown in Figure 8D , the coconversion boundary remains essentially unchanged in the SpI6 mutant . Although the position of the hairpin structure affects the efficiency of DGR homing , it does not determine the site at which cDNA is integrated at the 3′ end of VR . To determine if the results presented here complete our understanding of DGR-encoded requirements for retrohoming to a target gene , we applied them as engineering principles in an attempt to construct a functional , synthetic , TR/VR system . For a DNA sequence to serve as a recipient VR , three conditions must be met . First , it must be adjacent to an IMH region with functional ( GC ) 14 and 21 bp elements at its 3′ end [1] , [4] . Second , the IMH region must be followed by inverted repeats capable of forming a hairpin/cruciform structure of appropriate size , composition and distance from IMH . And finally , sufficient VR/TR sequence homology must be provided to allow efficient upstream ( 5′ ) cDNA integration . In recent studies we have shown that although short stretches of nucleotide identity ( ≥8 bp ) between the TR-derived cDNA and VR target sequences are sufficient to complete the homing reaction , homing efficiency is increased with longer ( ≥19 bp ) stretches of homology [4] . With these parameters in mind , we tested our ability to engineer the BPP-1 DGR to target a heterologous reporter gene ( aph3′Ia; [15] ) which provides facile detection of targeting events by antibiotic selection . The recipient VR-KanS cassette shown in Figure 9A contains an aph3′Ia kanamycin resistance ( KanR ) allele with a 3′ deletion that renders it nonfunctional by removing coding sequences for 6 essential C-terminal residues . The truncated gene was placed immediately upstream of IMH , followed by the hairpin/cruciform-forming inverted repeats from the BPP-1 DGR . Transcription is directed by the native aph3′Ia promoter . The donor plasmid expresses avd , brt , and one of two engineered TRs ( TR-Km1 , TR-Km2 ) from the Pfha promoter . Both TRs contain the intact 3′ end of the aph3′Ia open reading frame , followed by two consecutive stop codons and sequences 97–134 from the 3′ end of the BPP-1 TR . For TR-Km2 , the aph3′Ia fragment is also flanked , at its 5′ end , by the first 22 residues of the BPP-1 TR . DGR-mediated retrotransposition from the donor TR constructs to the VR-KanS recipient should regenerate a full-length aph3′Ia gene conferring KanR . We first tested whether targeting can occur in the context of a replicating phage . BPP-1ΔATR*KanS carries the VR-KanS cassette inserted between attL and bbp1 on the left arm of the prophage genome [16] , along with a deletion of avd , TR and brt and a series of synonymous substitutions in IMH to inactivate the mtd VR ( Figure 7A ) . B . bronchiseptica RB50 carrying the TR-Km1 or -Km2 donor plasmid , or derivatives with a null mutation in brt , were infected with BPP-1ΔATR*KanS and targeting efficiencies were determined by infecting RB50 with progeny phages and measuring relative numbers of KanR lysogens . KanR lysogens were readily detected when targeting occurred from Brt+ TR donors , but not Brt− donors ( Figure 9B ) , and sequence analysis showed that KanR resulted from the regeneration of full-length aph3′Ia alleles which often contained mutations at positions corresponding to adenines in donor TRs ( Figures S9 and S10 ) . It is interesting to note that the TR-Km1 donor was significantly more efficient than TR-Km2 . This suggests that the majority of cDNAs are extended to the 5′ termini of these short synthetic TRs , and target ( VR ) homology to the extreme 3′ ends of the extension products may be advantageous for cDNA integration . We also tested the ability to target the VR-KanS cassette when present on a resident prophage in the bacterial chromosome or on a plasmid . In the experiment in Figure 9C , RB50/BPP-1ΔATR*KanS lysogens were transformed with donor plasmids under conditions that suppress Pfha promoter activity . Following a 6 hr pulse of Pfha induction , cells were plated under promoter-suppressing conditions on media with or without kanamycin . In Figure 9D , a similar protocol was used to target a VR-KanS cassette carried on a medium copy number plasmid in RB50 cells containing a TR donor plasmid , but no other phage sequences . In both experiments , KanR colonies were readily detected when targeting occurred from Brt+ , but not Brt− TR donors , and sequence analysis showed characteristic patterns of adenine mutagenesis ( Figures S11 , S12 , S13 , S14 ) . Taken together , our results demonstrate the ability to engineer a VR/TR system that targets a heterologous reporter gene on a phage , plasmid or bacterial genome . The data in Figure 9D show that no BPP-1 phage products , other than those encoded in the DGR , are required for mutagenic retrohoming .
Understanding DGR target site recognition requires a precise definition of cis-acting sequences important for retrohoming . Our analysis of the boundaries of the BPP-1 DGR target showed that sequences upstream of VR are dispensable , as predicted by previous results [4] . More importantly , we show that homing is facilitated by an element downstream of VR , beyond the point at which TR/VR homology ends . Sequence analysis , mutagenesis , and structure-specific nuclease assays demonstrated that GC-rich inverted repeats directly following VR form a hairpin/cruciform structure that plays a critical role in retrohoming . Highly similar elements are present in analogous locations in many phage- or prophage-related DGRs ( Figure 5 ) , and hairpin/cruciform structures are predicted for the majority of DGRs that naturally reside on bacterial chromosomes and plasmids as well [Gingery et al . , unpublished data] . We propose that DNA hairpin formation near the 3′ end of VR is a conserved requirement for DGR-mediated retrohoming . For the BPP-1 DGR target , the 8 bp stem appears to function as a structure that is dependent on nucleotide composition but not sequence . In contrast , the loop of the hairpin/cruciform structure is constrained in size and sequence and conforms to the consensus , 5′-GRNA , derived from comparisons with other phage-related DGRs . This suggests that loop sequence and size may be important for stabilizing the hairpin/cruciform structure [17] , or for creating a strand bias in DNA cleavage by a host-encoded endonuclease . It is also possible that the loop is in direct physical contact with a critical component , such as Brt , Avd , a TR-containing RNA transcript , or other parts of the DGR target . By testing the effects of length and sequence variations between the hairpin/cruciform and VR , we found that distance is an important parameter , although some flexibility exists . Extending the spacer by 6 bp did not shift the marker coconversion boundary in the ( GC ) 14 region during DGR homing [4] , showing that the position of the hairpin/cruciform does not determine the site at which 3′ cDNA integration occurs . DGRs are evolutionarily related to group II introns [1] and it is interesting to note that a subset of these retroelements , the group IIC introns , also target motifs with stem-loop structures [18]–[20] . In nature , group IIC introns are often found to be located short distances downstream of sequences encoding known or predicted factor-independent transcription terminators , which are composed of GC-rich stems with loops of varying sizes followed by poly-uridine stretches [18]–[20] . Using an in vitro mobility assay , Robart et al . [19] have shown that reconstituted ribonucleoprotein particles from the Bacillus halodurans B . h . I1 group IIC intron recognize structures in ssDNA that correspond to RNA hairpins formed during transcription termination . As observed with the BPP-1 DGR , the B . h . I1 mobility reaction was highly dependent on stem formation but not absolute sequence [19] . Stems shorter than 9 bp had significantly reduced activities in in vitro mobility assays , a longer stem ( 14 bp ) retained function , and the efficiency of targeting correlated with GC content and predicted stem stability [19] . In contrast to our observations with the BPP-1 DGR , alterations in loop sequence had little effect on B . h . I1 mobility in vitro [19] . The adaptation of group IIC introns to recognize and insert downstream of factor-independent transcriptional terminators was proposed to provide a selective advantage by limiting their expression , avoiding the interruption of essential coding sequences , and facilitating horizontal spread as intrinisic terminators are common and conserved in bacteria [19] . For DGRs , we speculate that the ability to target sequences upstream of terminator-like stem-loop structures may have played a role in directing their sequence diversification capabilities to the 3′ coding regions of target genes . The TPRT model for DGR homing postulates that cDNA synthesis initiates with a nick or double-strand break in the IMH ( GC ) 14 sequence , providing a primer for reverse transcription of a TR-containing RNA transcript [4] . Analogous to target recognition by group IIC introns , the hairpin/cruciform structure may serve as a recognition element for a retrohoming complex that includes trans-acting DGR-encoded factors . A DNA endonuclease that might be responsible for cleavage awaits identification , and possibilities include Avd , Brt , a TR-derived catalytic RNA , or an unidentified host factor . It is also possible that the DNA hairpin/cruciform actively promotes single- or double-strand breaks . If DNA repair synthesis extends to the ( GC ) 14 region , the elongating antisense strand could then be used for cDNA priming . DNA breaks at the hairpin/cruciform structure could be created by an endonuclease that cleaves the single-stranded loop , or by a structure-specific enzyme similar to T7 endonuclease I [21] . Since DNA cruciforms are structurally similar to Holiday junctions , host-encoded recombination proteins that function in resolving recombination intermediates could be involved [22] . The cDNA priming mechanism of the BPP-1 DGR appears to be different from that of mobile group II introns that lack a DNA endonuclease activity in their intron-encoded proteins [23]–[25] . Reverse transcription in retrohoming and ectopic transposition of these elements is proposed to be primed by either the leading or lagging strand during DNA replication , and strong strand-specific biases are observed [23]–[25] . Our observation that the BPP-1 DGR target sequence is orientation-independent suggests that DNA replication polarity does not play a significant role in cDNA priming . Although our results to date are consistent with TPRT , further studies are required to definitively characterize the mechanism of cDNA initiation and integration at the 3′ end of VR and to determine the precise role of the hairpin/cruciform structure in the retrohoming process . The broad distribution of DGRs in nature attests to their utility , and prospects for adapting these elements for protein engineering applications are compelling . Our results demonstrate that the region containing the ( GC ) 14 and 21 bp sequences in IMH , and an adjacent hairpin/cruciform , is sufficient to direct the DGR mutagenic homing machinery to a heterologous target gene through appropriate engineering of a cognate TR . Using similar design principles we have successfully targeted a tetracycline resistance determinant as well ( HG and JFM , unpublished data ) . For DGRs to be useful tools , it will be necessary to engineer their activity to allow efficient and controlled diversification . Having defined the DGR-encoded cis- and trans-acting factors required to diversify heterologous sequences , efforts to optimize their activities can now proceed in an informed and comprehensive way . It will also be important to determine the effects of TR/VR size , composition , and position relative to cis-acting DGR elements , on the efficiency of diversifying heterologous sequences . In preliminary experiments , insertions of moderate size ( up to ∼200 bp ) at position 84 in the BPP-1 TR ( 134 bp ) are transferred to VR and mutagenized at adenines , suggesting that sequences of >300 bp could be diversified by an engineered system ( LVT , HG and JFM , unpublished data ) . In addition to providing prodigious levels of diversity , mutagenic homing is a regenerative process that allows DGRs to operate through unlimited rounds to optimize variable protein functions [4] . This may be particularly advantageous for directed protein evolution since desired traits can be selected and continuously evolved in iterative cycles , without the need for library construction or other interventions , through a process that takes place entirely within bacterial cells .
B . bronchiseptica strains RB50 , RB53Cm , RB54 and ML6401 have been described [16] . The BPP-1ΔATR lysogen was constructed from ML6401 , an RB50 strain lysogenized with phage BPP-1 , by deleting sequences from avd position 48 to position 882 of brt . Target region deletions/insertions and hairpin/cruciform modifications were introduced into the BPP-1ΔATR lysogen through allelic exchange [1] , [4] and are diagramed in the figures . The BPP-1ΔATR* lysogen contains multiple silent mutations at both the 5′ and 3′ ends of VR to inactivate it as a DGR target . It was used as the parental strain to create the BPP-1ΔATR*KanS lysogen , in which the KanR gene aph3′Ia has sequences encoding the C-terminal 6 amino acid residues truncated and is placed upstream of IMH and the hairpin/cruciform structure as a reporter for heterologous gene targeting . The aph3′Ia allele also contains an AAA to CGC substitution resulting in K260R . The VR-KanS reporter cassette was inserted between attL and bbp1 of the phage genome . Phage BPP-1ΔATR and its various derivatives were produced from the above lysogens . Plasmid pMX-ΔTR23–96 has TR positions 23–96 deleted and replaced by a 30 bp PCR tag as in pMX-ΔTR23–84 [4] . Its RT-deficient derivative contains the YMDD to SMAA mutation at Brt positions 213–216 [3] , [4] . Plasmids pMX1 and pMX1SMAA were used for phage tropism switching assays and have previously been described [4] . pUC-StWT is a pUC18-based plasmid containing the WT BPP-1 DGR target from position −6 upstream of VR to position +82 downstream of VR . pUC-StMut is its derivative with 7 residues in the 3′ half of the stem , proximal to the loop , mutated to their complementary nucleotides . Plasmids pMX-TRC85T , pMX-TRC91T , pMX-TRC97T , pMX-TRC100T , pMX-TRC105T , pMX-TRC107T , pMX-TRC109T , pMX-TRC112T , pMX-TRC115T , pMX-TRC120T and pMX-TRC125T have been previously described [4] . Plasmids pMX-Km1 and pMX-Km2 were constructed from pMX-ΔTR23–96 for KanR gene targeting , both containing the last 36 bp of aph3′Ia . The 36 bp sequence and its following two stop codons replace TR positions 1–96 in pMX-Km1 and TR positions 23–96 in pMX-Km2 . Plasmid pHGT-KanS contains the VR-KanS cassette described above and was used as the recipient plasmid for KanR targeting . The plasmid also carries a tetracycline resistance gene . Phage production for DGR functional assays was carried out by either single-cycle lytic infection or mitomycin C induction from lysogens as previously described [4] , except for minor modifications as noted . For single-cycle lytic infection , B . bronchiseptica RB50 cells transformed with appropriate donor plasmids were grown overnight at 37°C in Luria-Bertani ( LB ) media containing 25 µg/ml of chloramphenicol ( Cam ) , 20 µg/ml streptomycin ( Str ) , and 10 mM nicotinic acid to modulate to the Bvg− phase and prevent transcription from the Pfha promoter . An amount of cells equal to 1 ml of culture ( OD600 = 1 . 0 ) was pelleted , rinsed , and resuspended in 2 . 5 ml Stainer Scholte ( SS ) medium [26] containing 25 µg/ml Cam and 20 µg/ml Str ( SS+Cam+Str ) . Cultures were grown for 3 hr at 37°C to modulate bacteria to the Bvg+ phase and activate Pfha promoter expression . An aliquot of 500 µl from each culture was used for OD600 measurement and cell number calculation . Phage particles were added to the rest of the culture at a multiplicity of infection of ∼2 . 0 . Following 1 hr incubation at 37°C for phage absorption , infected cells were pelleted and resuspended in 1 ml of fresh , prewarmed SS+Cam+Str media and incubated at 37°C for 3 hr post phage addition to allow completion of a single cycle of phage development . Progeny phages were harvested following chloroform extraction . For phage production from lysogens , RB50 derivatives carrying appropriate prophages and donor plasmids were grown and modulated to the Bvg+ phase as in single-cycle lytic infections . Phage production was induced with 0 . 2 µg/ml mitomycin C for 3 hr at 37°C . Progeny phages were harvested by chloroform extraction . Phage tropism switching and DGR homing assays have been previously described [4] . Plasmids containing the WT BPP-1 DGR target and the StMut mutation were isolated from E . coli DH5αλpir cells using the QIAprep Spin miniprep kit ( Qiagen ) . Plasmids were linearized by digestion with BglI as indicated . To analyze hairpin/cruciform structure formation in supercoiled or relaxed DNAs , 0 . 5 µg of supercoiled or linearized plasmids were treated with 10 units of T7 DNA endonuclease I ( New England Biolabs , Ipswich , MA ) for 40 minutes as in Miller et al . [11] . The reactions were terminated by phenol-chloroform-isoamyl alcohol ( 25∶24∶1 ) extraction and DNAs were precipitated with ethanol . T7 DNA endonuclease I cleavage sites were determined by primer extension with 5′-end 32P-labeled primers using Vent ( exo- ) DNA polymerase ( New England Biolabs , Ipswich , MA ) as in Miller et al . [11] , except that 5% DMSO was added for GC-rich templates . Primer extension products were resolved on 6% polyacrylamide/8 M urea gels , alongside Sanger sequencing ladders generated with the same labeled primers and a plasmid template containing the WT target . To target the KanR gene on a replicating phage , BPP-1ΔATR*KanS phage particles were used for single-cycle lytic infection of RB50 cells transformed with appropriate donor plasmids , similar to phage production by single-cycle lytic infection described above . Progeny phages were titered and ∼1011 pfu of different phages were added to 25 ml RB50 cells ( OD600 = 1 . 2 ) in SS+Str media for 8 . 0 hr to reestablish lysogens . Cells were pelleted and resuspended in 5 ml LB and serial dilutions were plated on LB+NA+Str and LB+NA+Str+Kan ( 50 µg/ml ) to determine KanR gene targeting frequencies . Lysogen reestablishment efficiencies ranged from 60% to 100% based on PCR analysis of 10 colonies each picked on LB+NA+Str plates using phage specific primers . KanR targeting efficiency for each donor plasmid was determined as the ratio of colony forming units ( cfu ) on LB+NA+Str+Kan plates to those on LB+NA+Str , calibrated with the lysogen reestablishment efficiency for that sample . To target the KanR gene on a prophage in the bacterial chromosome , RB50 cells lysogenized with phage BPP-1ΔATR*KanS were transformed with appropriate donor plasmids . Starting cultures were grown overnight in LB+NA+Str+Cam as described above . An amount of cells equal to 1 ml of culture ( OD600 = 1 . 0 ) was pelleted , rinsed , and resuspended in 2 . 5 ml SS+Cam+Str and grown at 37°C for 6 hours . Serial dilutions were plated on LB+NA+Str and LB+NA+Str+Kan ( 50 µg/ml ) to determine KanR gene targeting frequencies . KanR targeting efficiencies were determined as relative numbers of KanR cells as above . To target the KanR gene on a plasmid , the recipient plasmid pHGT-KanS and appropriate donors were transformed into RB50 cells and analyzed similarly . Tetracycline was added to 5 . 0 µg/ml for recipient plasmid maintenance . | Diversity-generating retroelements function through a unique , reverse transcriptase–mediated “copy and replace” mechanism that enables repeated rounds of protein diversification , selection , and optimization . The ability of DGRs to introduce targeted diversity into protein-coding DNA sequences has the potential to dramatically accelerate the evolution of adaptive traits . The utility of these elements in nature is underscored by their widespread distribution throughout the bacterial domain . Here we define DNA sequences and structures that are necessary and sufficient to direct the diversification machinery to specified target sequences . In addition to providing mechanistic insights into conserved features of DGR activity , our results provide a blueprint for the use of DGRs for a broad range of protein engineering applications . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"biotechnology",
"bacteriology",
"organismal",
"evolution",
"microbial",
"mutation",
"genetic",
"mutation",
"retrotransposons",
"microbiology",
"mutation",
"molecular",
"cell",
"biology",
"microbial",
"evolution",
"dna",
"dna",
"structure",
"applied",
"microbiology",
"transposons",
"biology",
"evolutionary",
"genetics",
"molecular",
"biology",
"mutagenesis",
"biochemistry",
"adaptation",
"nucleic",
"acids",
"genetics",
"evolutionary",
"biology",
"genomic",
"evolution",
"bacterial",
"evolution",
"evolutionary",
"processes",
"genetics",
"and",
"genomics"
] | 2011 | Target Site Recognition by a Diversity-Generating Retroelement |
Nutritional interventions targeting the critical growth and development period before two years of age can have the greatest impact on health trajectories over the life course . Compelling evidence has demonstrated that interventions investing in maternal health in the first 1000 days of life are beneficial for both mothers and their children . One such potential intervention is deworming integrated into maternal postpartum care in areas where soil-transmitted helminth ( STH ) infections are endemic . From February to August 2014 , 1010 mother-infant pairs were recruited into a trial aimed at assessing the effectiveness of maternal postpartum deworming on infant and maternal health outcomes . Following delivery , mothers were randomly assigned to receive either single-dose 400 mg albendazole or placebo . Participants were followed-up at 1 and 6 months postpartum . There was no statistically significant difference in mean weight gain between infants in the experimental and control groups ( mean difference: -0 . 02; 95% CI: -0 . 1 , 0 . 08 ) at 6 months of age . Further , deworming had no effect on measured infant morbidity indicators . However , ad hoc analyses restricted to mothers who tested positive for STHs at baseline suggest that infants of mothers in the experimental group had greater mean length gain in cm ( mean difference: 0 . 8; 95% CI: 0 . 1 , 1 . 4 ) and length-for-age z-score ( mean difference: 0 . 5; 95% CI: 0 . 2 , 0 . 8 ) at 6 months of age . In a study population composed of both STH-infected and uninfected mothers , maternal postpartum deworming was insufficient to impact infant growth and morbidity indicators up to 6 months postpartum . Among STH-infected mothers , however , important improvements in infant length gain and length-for-age were observed . The benefits of maternal postpartum deworming should be further investigated in study populations having higher overall prevalences and intensities of STH infections and , in particular , where whipworm and hookworm infections are of public health concern . ClinicalTrials . gov ( NCT01748929 ) .
Worldwide , more than 200 million children under five years of age fail to reach their full developmental potential [1] . Suboptimal nutrition and growth during fetal and early childhood development are widely regarded as having adverse long-term consequences , including higher susceptibility to infections , decreased learning potential , lower economic productivity , and greater risk of mortality [2] . The most critical period of pre- and postnatal development ( coined the first 1000 days of life ) is a time when interventions can have the greatest impact on preventing negative health and economic outcomes over the life course [3 , 4] . Given the intrinsic relationship between maternal and child health , the nutritional status of mothers is considered to be a pivotal driver of infant growth , development , and survival [5 , 6] . To date , most studies have focused almost exclusively on maternal micronutrient deficiencies [7] and have not adequately addressed the multifactorial causes of maternal malnutrition , including the role of maternal infection in future infant growth and morbidity . The soil-transmitted helminth ( STH ) infections ( Ascaris lumbricoides ( the roundworm ) , Trichuris trichiura ( the whipworm ) and the hookworms , Necator americaus and Ancylostoma duodenale ) are considered the most prevalent parasitic infections of humans , affecting more than 1 . 45 billion people worldwide [8] . The global significance of STHs lies in their high burden of disease , attributed to blood loss , anemia , and malabsorption of nutrients that can cause or further exacerbate nutritional deficiencies [9] . Sequelae of chronic infection with STHs include fatigue , loss of appetite , growth faltering , and impaired cognition [9–11] . While much of the current research and focus of large-scale control programs have targeted preschool and school-aged children , women of reproductive age ( WRA ) remain a largely neglected group [12] despite their underlying poor iron and micronutrient status due to inadequate diet and high risk of blood loss due to menstruation , pregnancy , and childbirth [11 , 13 , 14] . While species-specific estimates are not available for WRA as a whole , it is estimated that , at any one time , as many as 37 . 7 million WRA are infected with hookworms alone [15] . The World Health Organization ( WHO ) recommends that single-dose deworming ( i . e . , single tablets of 400 mg albendazole or 500 mg mebendazole ) be given to WRA , including pregnant women ( after the first trimester ) and lactating women , in endemic areas where the prevalence of infection exceeds 20% [14] . Among the 112 countries considered endemic for STH infections [16] , no coverage estimates are available for deworming in WRA ( i . e . , the proportion of WRA reached by the deworming intervention ) , and few countries include this risk group among other deworming activities [17] . The evidence base on deworming in WRA is limited , and largely focuses on deworming during pregnancy . Randomized controlled trials ( RCTs ) and observational studies conducted in pregnant women have been largely inconclusive , providing mixed evidence on the benefits of deworming on maternal anemia , low birthweight , and perinatal mortality [15 , 18 , 19] . The uptake and health benefits of deworming during pregnancy may not be optimal because in many low-and-middle-income countries antenatal care attendance is low and reinfection rates can be high . The early postpartum period presents an innovative opportunity to reach WRA because deworming can be easily integrated into routine care , and many low-and-middle-income countries have , or are currently adopting and actively promoting , pro-hospital and health centre delivery policies . While maternal deworming will be of direct benefit to the infected woman ( by curing or reducing her burden of STH infection ) , the unique interface between mother and child during lactation suggests that benefits may also accrue to the newborn infant . To address this research gap , we designed an RCT aimed at comparing the effectiveness of single-dose albendazole vs . placebo administered to women following hospital delivery on infant growth and morbidity in an area of Peru known to be highly endemic for STH infection .
This study is a randomized , double-blind , placebo-controlled trial conducted in the Amazon region of Iquitos , Peru . A recent meta-analysis estimated that a quarter of the population in Peru was infected with STHs between 2005 and 2012 [20] . In Iquitos , the prevalence of infection has been reported to be much higher than the country average with prevalences > 80% in school-aged children [21] , > 40% in preschool-aged children [22] , and > 90% in pregnant women [23] . Iquitos was chosen as the study site because of its high STH endemicity , a high proportion of women delivering in hospital , and because , at the time of the trial , there were no routine deworming programs targeting WRA . A detailed description of study procedures can be found in the trial protocol [24] . Briefly , beginning in February 2014 , mother-infant pairs were enrolled into the trial using a two-stage approach . First , from pregnancy registries accessed in health centre records , women who were in their third trimester of pregnancy were visited in their homes by research personnel in order to assess eligibility using the following inclusion criteria: planned delivery at Hospital Iquitos “César Garayar García”; and intention to remain in the study area for a 24-month period . Women were ineligible to participate in the trial based on the following exclusion criteria: confirmed pregnancy of multiples; and inability to communicate in Spanish . Following a detailed explanation of the trial , women and their partners were asked to provide informed consent and , if this was obtained , a baseline questionnaire was administered . Upon arrival at the Hospital Iquitos “César Garayar García” for delivery , women were approached by research assistants and asked to re-confirm their consent to participate in the trial . Following delivery , mothers and newborns were assessed for the following exclusion criteria: stillborn; serious congenital abnormality or serious medical condition ( e . g . , exposure to HIV infection ) ; gestational age < 32 weeks; Apgar score < 4 at 5 minutes ( Apgar score is a scored evaluation of the physical condition of an infant immediately following delivery . The infant's heart rate , respiratory effort , muscle tone , reflex irritability , and color are assessed by clinical personnel [25] . Each component is scored from 0 to 2 and then the scores are summed , giving a total score between 0 and 10 . ) ; mother or baby transferred to another hospital; and hospitalization of mother or baby for > 3 days . In the case that women presented at the hospital for delivery prior to being visited in their home by a research assistant , every effort was made to recruit mothers into the trial as soon as possible before or after delivery ( i . e . , within 10 hours of birth ) . A computer-generated randomization schedule was prepared prior to recruitment by a statistician not otherwise involved in the trial , using a random number sequence according to simple randomization with a 1:1 allocation ratio . Group assignments were concealed in opaque sequentially-numbered envelopes according to the randomization schedule . Envelopes were stored in a secure temperature-regulated pharmacy . A print copy of the randomization schedule was stored in a sealed envelope under lock-and-key at the research office in Canada ( Research Institute of the McGill University Health Centre ) . Initially , the randomization sequence had been planned as a permuted block design with randomly varying block sizes of 6 and 8 with a 1:1 allocation ratio . Due to a miscommunication , the statistician prepared a simple randomization sequence with a 1:1 allocation ratio . With a large sample size and a similar rate of recruitment over time into the two intervention groups , either randomization sequence would be expected to produce balanced intervention groups , as was the case in this trial ( see Table 1 ) . Eligible , consenting women were visited at bedside prior to hospital discharge and were administered the tablet in the next sequentially-numbered envelope . Participants , research personnel , and data analysts were blinded to group assignment . Mothers randomized to the experimental group received single-dose 400 mg albendazole and mothers randomized to the control group received single-dose placebo . The albendazole tablets were manufactured by GlaxoSmithKline Inc . and donated for this trial by WHO . The placebo tablets were manufactured to be identical to the albendazole tablets in all aspects , including size , taste , color , and markings and were manufactured by Laboratorios Hersil in Lima , Peru . Both groups received the current standard of routine postpartum care from hospital personnel . At 6 months postpartum , all mothers were offered deworming as part of the trial protocol , if they were not newly pregnant ( i . e . , in the first trimester of pregnancy ) . The required sample size was estimated for the primary outcome of mean infant weight gain between birth and 6 months of age . An estimate of mean weight gain was obtained from recent data on children between 5 and 7 months of age residing in the study area ( i . e . , 4 . 24 kg with a standard deviation of 1 . 014 kg ) [26] . The trial sample size was computed using a two-sided independent t-test , with a significance level of 0 . 05 , a power of 0 . 80 , a standard deviation of 1 . 014 kg , a minimum detectable difference in weight gain of 0 . 2 kg , and taking into account an attrition rate of 20% . Based on the above specifications , a total of 1010 participants was calculated to be the sample size needed to declare that intervention groups would be different if , in fact , there would be a true between-group difference of 0 . 2 kg or more in weight gain between birth and 6 months of age . Sample size calculations were carried out using PS Power and Sample Size Calculations version 3 . 0 ( Dupont and Plummer , 2009 ) . Based on the number of deliveries at the study hospital in 2011 , recruitment was expected at the rate of 300 mother-infant pairs per month , suggesting a total recruitment period of approximately 3 . 5 months . The pre-specified primary outcome measure was mean infant weight gain between birth and 6 months of age . Pre-specified secondary infant outcome measures were length and head circumference ( HC ) gains , derived growth indices ( i . e . , weight-for-age ( WAZ ) , weight-for-length ( WFL ) , length-for-age ( LAZ ) , HC-for-age ( HCAZ ) , and MUAC-for-age ( ACAZ ) ) , prevalence of underweight , wasting and stunting , and occurrence of infant morbidity ( i . e . , occurrence of hospitalizations since birth , and incidence of diarrhea , respiratory problems , fever , and ear infections ) at 1 and 6 months postpartum . Secondary maternal health outcomes ( i . e . , STH infection and intensity , anemia , fatigue ) , and breast milk outcomes ( i . e . , milk quality and quantity ) are reported separately . This article presents infant health outcomes up to 6 months postpartum ( at which time the primary outcome was measured ) . However , follow-up for the trial will continue until the 24-month time point . Ethics approval for this trial was obtained from the research ethics committees of the Asociación Civil Impacta Salud y Educación ( Peru ) , the Instituto Nacional de Salud ( Peru ) and the McGill University Health Centre ( Canada ) . As per Peruvian Institute of Health guidelines , all women and their partners provided written informed consent for participation in the trial . In the case that a woman or her partner was under the age of 18 years , written assent was obtained , and informed consent was obtained from their parent , guardian or spouse/partner over the age of 18 years . Women who did not have a partner , or whose partner was absent for an indefinite period of time ( e . g . , death , separation , etc . ) , were asked to sign a sworn statement to declare the father’s absence , in accordance with Peruvian ethics guidelines . An independent DSMC comprised of three international experts was put in place to monitor trial progression and ensure participant safety . At pre-specified time points ( i . e . , after 50% and 100% of recruitment , and after the completion of the 1 and 6-month study visits ) , the DSMC reviewed all SAEs and approved trial continuation . The trial was registered prior to commencement with ClinicalTrials . gov ( NCT01748929 ) , and the trial protocol is available as a published manuscript [24] . The Consolidated Standards of Reporting Trials ( CONSORT ) checklist can be found in S1 Checklist .
Recruitment into the trial took place between February and August 2014 . Of the 2134 women and their partners who were approached to participate in the trial , 1010 were enrolled and randomized ( Fig 1 ) . The remaining 1124 mother-infant pairs did not participate because: the mother did not meet the pre-delivery eligibility criteria ( n = 752 ) , either mother or infant did not meet the post-delivery eligibility criteria ( n = 134 ) , either mother or her partner declined to participate ( n = 175 ) , or the sample size was reached before enrollment could take place ( n = 63 ) . Those who declined to participate were more likely to live in urban areas ( OR: 2 . 8; 95% CI: 1 . 4 , 5 . 5 ) , and be primigravida ( OR: 1 . 6; 95% CI: 1 . 1 , 2 . 4 ) . All 1010 mother-infant pairs enrolled into the trial were randomized to the experimental ( n = 510 ) or control ( n = 500 ) groups . Baseline maternal and infant characteristics for the two intervention groups are summarized in Table 1 . Groups were similar in terms of maternal and infant characteristics . However , some differences between groups were found in household characteristics ( i . e . , access to potable water in the home and the type of housing structure ) . In the hospital , 64 participants ( 12 . 5% ) in the experimental group provided a stool specimen that was analyzed immediately using the Kato-Katz technique . Those who provided a stool specimen for analysis in the hospital were similar in terms of baseline characteristics to those who did not provide a specimen . Due to the RCT design , prevalences and intensities of infection were expected to be similar between intervention groups . Uncorrected and corrected prevalences of STH infection , along with intensities are presented in Table 2 . The proportion of women testing positive for any STH infection was 48 . 4% ( 31/64 ) , of which 29 . 7% ( 19/64 ) tested positive for A . lumbricoides , 26 . 6% ( 17/64 ) tested positive for T . trichiura , and 6 . 3% ( 4/64 ) tested positive for hookworm . Only 7 women had co-infections: 2 with all 3 species and 5 with A . lumbricoides and T . trichiura . Bayesian methods that adjusted for imperfect sensitivity and specificity produced corrected prevalences of 31 . 4% ( 95% BCI: 13 . 7% , 49 . 4% ) for A . lumbricoides , 18 . 5% ( 95% BCI: 1 . 0% , 43 . 5% ) for T . trichiura , and 28 . 7% ( 95% BCI: 1 . 9% , 88 . 6% ) for hookworm . The mean eggs per gram ( epg ) values for each helminth species were of low intensity according to WHO thresholds [40] . During pre-recruitment , 450 participants ( 44 . 6% ) , of the total study population of 1010 participants provided a stool specimen that was stored and analyzed following the 6-month study visit using the direct smear and ethyl-ether concentration techniques . The baseline prevalence of any STH infection was 32 . 0% ( 144/450 ) , and the species-specific prevalences of A . lumbricoides , T . trichiura , and hookworm infections were 24 . 7% ( 111/450 ) , 8 . 9% ( 40/450 ) , and 2 . 7% ( 12/450 ) , respectively . Prevalences of T . trichiura and hookworm were similar between intervention groups . A small difference between groups was found in the proportion of participants infected with A . lumbricoides ( experimental group: 29 . 2%; control group: 20 . 1% ) . Because these baseline prevalences , although known to be underestimates due to the limitations of the diagnostic techniques used on the fixed and stored stool specimens , originate from the entire study population , data analyses investigating effects between the intervention groups use these prevalences ( the prevalences based on the Kato-Katz technique originating only from the experimental group ) . Follow-up for the 1-month study visit took place between March and August 2014 , and follow-up for the 6-month study visit took place between August 2014 and February 2015 . A total of 968 ( 95 . 8% ) infants completed both their 1 and 6-month study visits . Of the 42 infants who were lost-to-follow-up , 7 ( 0 . 7% ) were visited only at baseline , 31 ( 3 . 1% ) were visited at 1 month only , and 4 ( 0 . 4% ) were visited at 6 months only . The causes of attrition were: emigration from study area ( n = 35 ) , infant death ( n = 5 ) , maternal death ( n = 1 ) , and temporary withdrawal ( n = 1 ) . The mean number of days between the baseline and first follow-up visit was 32 . 0 days ( ±3 . 6 ) and the mean number of days between the baseline and second follow-up visit was 186 . 7 days ( ±13 . 3 ) . Follow-up rates were similar in the intervention groups at both follow-up time points . Overall , the majority of baseline characteristics were similar between participants who remained in the trial at 6 months and those who were lost to follow-up ( S1 Table ) . However , infants who missed their 6-month study visit had mothers who were more likely to be single ( RR: 2 . 6; 95% CI: 1 . 3 , 5 . 3 ) . The proportion of women who reported having received deworming outside of the trial protocol was 3 . 1% ( 15/490 ) in the experimental group and 2 . 7% ( 13/478 ) in the control group . Of the 144 mothers who tested positive for infection with any STH infection ( i . e . , using the direct smear and ethyl-ether concentration techniques ) at baseline , 142 completed their 1-month study visit and 139 completed their 6-month study visit . Weight , length and HC gains , as well as z-scores for WAZ , LAZ , HCAZ , and ACAZ between intervention groups at 6 months postpartum are compared in Table 3 . The primary outcome , mean infant weight gain between birth and 6 months of age , was similar between intervention groups ( 4 . 3 kg ±0 . 04 vs . 4 . 4 kg ±0 . 04 ) . There were no differences between groups in terms of the secondary anthropometric outcomes in unadjusted or adjusted ITT analyses ( Table 4 ) . Results were consistent in complete-case analysis ( S2 Table and S3 Table ) and per-protocol analysis ( S4 Table and S5 Table ) . In ad hoc subgroup analyses restricted to the 139 mothers who were found to be positive for any STH infection at baseline ( i . e . , using the direct smear and ethyl-ether concentration techniques ) and who were visited at 6 months postpartum , infants in the experimental group had greater growth in terms of mean length gain in cm ( mean difference: 0 . 8; 95% CI: 0 . 1 , 1 . 4 ) and length-for-age in z-score ( mean difference: 0 . 5; 95% CI: 0 . 2 , 0 . 8 ) ( Table 5 ) , but had similar prevalences of underweight and stunting ( S6 Table ) at 6 months postpartum . Deworming did not have a statistically significant effect on growth outcomes measured at 1 month postpartum ( S7 Table and S8 Table ) . There was no statistical evidence that the effect of deworming on anthropometric outcomes differed between boys and girls , or varied according to birthweight or birth length . The occurrence of hospitalization , and the incidence of diarrhea , respiratory problems , and fever , as assessed at 6 months postpartum are shown in Table 6 . The incidence of ear infection did not differ between intervention groups but was too low ( < 1% ) to perform statistical modelling . No significant difference was observed in indicators of infant morbidity between intervention groups . Results were consistent in complete-case analysis ( S9 Table ) , per-protocol analysis ( S10 Table ) and ad hoc analyses restricted to mothers who tested positive for STH infection at baseline ( S11 Table ) . Moreover , deworming did not have a statistically significant effect on infant morbidity measured at 1 month postpartum ( S12 Table ) . Between baseline and the 6-month study visit , 65 SAEs were reported in infants . The frequency of SAEs was similar between intervention groups , with 34 ( 6 . 7% ) occurring in the experimental group , and 31 ( 6 . 2% ) occurring in the control group . Of the 5 infant deaths that were reported over the 6 months of follow-up , 2 occurred in the experimental group and 3 occurred in the control group . No SAE was found to have been related to the administration of albendazole to the mothers .
This is the first trial assessing the effect of maternal postpartum deworming on infant and maternal health outcomes . We were unable to demonstrate an overall effect of maternal postpartum deworming on infant growth or morbidity indicators up to 6 months of age within the total study population . The relatively low prevalence and intensity of STH infection in this population may have reduced the ability of detecting a benefit due to effect dilution ( i . e . , the trial population comprising both STH-infected and uninfected mothers ) . When limiting the analyses to mothers who were infected with any one of the STH species at baseline , there was evidence of a benefit in terms of infant growth ( i . e . , mean length gain , LAZ ) at 6 months postpartum . Results were consistent in unadjusted and adjusted analyses . Although no previous studies of deworming have been conducted in lactating women , trials have been carried out in pregnant populations . There are inconsistencies in study findings among published reports of trials of deworming in second and third trimester pregnant women where infant growth was an outcome measure . In a trial conducted in Uganda between 2003 and 2005 , treatment with albendazole in the second and third trimester of pregnancy showed no benefit in terms of mean birthweight or the proportion of infants born with low birthweight [41] . In contrast , in a trial of single-dose mebendazole and daily elemental iron conducted in Peru , a beneficial effect on the proportion of infants born with very low birthweight ( <1500 g ) was observed [17] . In the present study , the prevalence ( < 50% ) and intensity of STH infection was low , and therefore closer to the baseline STH infection profile in the previous Uganda trial ( 68% prevalence ) compared to the previous Peru trial ( 91% prevalence ) . A recent Cochrane review of published trials on deworming during the second and third trimester of pregnancy found no overall effect on maternal anemia , low birthweight , preterm birth , or perinatal mortality [42] . However , there was important heterogeneity in baseline STH infection and interventions ( i . e . , combination of anthelminthic drugs and micronutrient supplementation ) among included trials that limits the appropriateness of pooling study results . Moreover , subgroup analyses restricted to those women infected with STH infections at baseline were not performed . The low STH prevalence in the current trial may account for the fact that an effect was not observed in the total study population ( which included both infected and uninfected mothers ) , but that effects were demonstrated when the analyses were restricted to STH-infected mothers at baseline . Our study has several strengths . First is the minimization of measured and unmeasured confounding by external factors achieved through the RCT design . The sample size was large and recruitment of the study population was completed within a period of six months . A high follow-up rate was maintained throughout the trial , minimizing the potential for bias caused by differential loss to follow-up . In-depth training and standardization of anthropometric measurements ensured a high degree of accuracy and precision of outcome ascertainment , thus reducing the potential for measurement error . While deworming medication is readily available for purchase over-the-counter in the study area , compliance with the study protocol was high and non-differential between intervention groups . The consistency of our results between ITT , complete-case , and per-protocol analyses demonstrates the robustness of study findings . Results from the present study may be generalizable to other populations of lactating women in STH-endemic areas with similar prevalence and intensity profiles , and where deliveries in health facilities are actively promoted . The limitations encountered in the conduct of this trial include a lower than anticipated baseline prevalence of STH ( considerably lower than the 91% reported in Larocque et al ( 2006 ) from the same study area ) [23] . This could be the reason why a benefit of deworming in the total study population was not detected , while ad hoc subgroup analyses in STH-infected women at baseline demonstrated a positive effect . Collecting stool specimens from women prior to enrolment was challenging and resulted in a less than optimal yield . WHO recommends use of the Kato-Katz technique on fresh stool specimens for the assessment of STH prevalence and intensity . However , due to ethical constraints , specimens collected from those participants randomized to the placebo group could not be immediately analyzed by this technique . The low number of available fresh stool specimens analyzed by the Kato-Katz technique may have affected the accuracy of STH parameter estimation at baseline . The storage of specimens until the end of the 6-month follow-up visit ( i . e . , > 1 year after collection ) likely affected the integrity of specimens and the ability to detect parasites , especially hookworm eggs [43 , 44] . Additionally , the direct smear and ethyl-ether concentration techniques have substantially lower test parameters compared to the Kato-Katz technique , and are unable to accurately assess the intensity of infection , thereby not allowing for the classification of morbidity thresholds established by WHO [40] . This likely resulted in some misclassification of baseline STH infection status , and likely also underestimated the effect of deworming in the analyses restricted to STH-positive women at baseline . Lastly , infant morbidity indicators were based on maternal reporting . We have no reason to believe that reporting differed by intervention group , although misclassification in morbidity status may have reduced the observed differences in morbidity outcomes between the two intervention groups . It has been well established that only individuals infected with STHs will benefit from deworming and , as a result , WHO only recommends the implementation of deworming programs in areas where the prevalence of infection exceeds 20% [14] . This recommendation is based on the fact that infections of moderate/heavy intensity are generally not found at prevalences lower than 20% [45] . Many argue that in order to evaluate the impact of deworming interventions , analyses must be restricted to those who were infected at baseline , and thus could benefit from treatment [30] . However , new trends in ethics guidelines prevent the detection of STH infection at baseline and subsequent randomization of infected individuals to placebo control groups , thereby withholding treatment from those in need . Without accurate individual participant data on baseline infection status , subgroup analyses restricted to infected populations are limited . For these reasons , the standard parallel RCT design may not be adequate to evaluate the effectiveness of deworming . There is a need for developing more novel research designs , such as stepped-wedge RCTs ( i . e . , cluster RCTs in which cross-over from control to intervention is randomized until all clusters receive the intervention [46] ) with careful consideration of secular trends . Overall , this is the first trial to provide rigorous empirical evidence on the benefits of maternal postpartum deworming . Study findings support WHO recommendations to include lactating women in deworming programs by demonstrating that this strategy is operationally feasible , culturally acceptable , and safe . Future research is needed , not only on the biological mechanisms underpinning the potential link between maternal postpartum deworming and benefits to the infant , but also in study populations having higher prevalences and intensities of STH infections , especially T . trichiura and hookworm infections which have a direct effect on anemia . The postpartum period is an ideal time to reach women periodically during their reproductive years because they are easily accessible , especially in areas where hospital-based deliveries are promoted , and where deworming can be easily integrated into standard postpartum care . Targeting treatment to the groups at highest risk of STH infection may produce the greatest health impacts . | Worldwide , over one billion people are infected with intestinal worms ( roundworms , whipworms , and hookworms ) . In worm-endemic areas , women of reproductive age are a high risk group for infection because of their poor nutritional status and increased physiological needs during pre-pregnancy , pregnancy , and lactation . To measure the effect of providing mothers with deworming treatment soon after delivery , we conducted a trial in 1010 mother-infant pairs . Mothers were randomly assigned to receive either a single-dose deworming tablet or a placebo tablet . Mothers and their infants were visited in their homes at 1 and 6 months following delivery . At the 6-month time point , among all mother-infant pairs , we could not detect an effect of deworming on infant growth or morbidity . We did , however , observe that , among women who were infected with intestinal worms at baseline , deworming had a beneficial effect on important infant growth outcomes . The potential benefit of maternal postpartum deworming in populations with higher prevalences and intensities of intestinal worms , particularly where infections with whipworm and hookworm predominate , warrants further investigation . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"invertebrates",
"children",
"medicine",
"and",
"health",
"sciences",
"maternal",
"health",
"obstetrics",
"and",
"gynecology",
"helminths",
"tropical",
"diseases",
"hookworms",
"social",
"sciences",
"anthropology",
"parasitic",
"diseases",
"animals",
"health",
"care",
"age",
"groups",
"women's",
"health",
"infants",
"pregnancy",
"neglected",
"tropical",
"diseases",
"families",
"morbidity",
"anthropometry",
"people",
"and",
"places",
"helminth",
"infections",
"health",
"statistics",
"physical",
"anthropology",
"trichuriasis",
"population",
"groupings",
"biology",
"and",
"life",
"sciences",
"soil-transmitted",
"helminthiases",
"organisms"
] | 2017 | A Double-Blind Randomized Controlled Trial of Maternal Postpartum Deworming to Improve Infant Weight Gain in the Peruvian Amazon |
Numerous experimental vaccines have been developed to protect against the cutaneous and visceral forms of leishmaniasis caused by infection with the obligate intracellular protozoan Leishmania , but a human vaccine still does not exist . Remarkably , the efficacy of anti-Leishmania vaccines has never been fully evaluated under experimental conditions following natural vector transmission by infected sand fly bite . The only immunization strategy known to protect humans against natural exposure is “leishmanization , ” in which viable L . major parasites are intentionally inoculated into a selected site in the skin . We employed mice with healed L . major infections to mimic leishmanization , and found tissue-seeking , cytokine-producing CD4+ T cells specific for Leishmania at the site of challenge by infected sand fly bite within 24 hours , and these mice were highly resistant to sand fly transmitted infection . In contrast , mice vaccinated with a killed vaccine comprised of autoclaved L . major antigen ( ALM ) +CpG oligodeoxynucleotides that protected against needle inoculation of parasites , showed delayed expression of protective immunity and failed to protect against infected sand fly challenge . Two-photon intra-vital microscopy and flow cytometric analysis revealed that sand fly , but not needle challenge , resulted in the maintenance of a localized neutrophilic response at the inoculation site , and removal of neutrophils following vector transmission led to increased parasite-specific immune responses and promoted the efficacy of the killed vaccine . These observations identify the critical immunological factors influencing vaccine efficacy following natural transmission of Leishmania .
Leishmania are obligate-intracellular protozoan parasites that establish infection in mammalian hosts following transmission to the skin by the bite of an infected Phlebotomine sand fly [1] . Different Leishmania species are associated with a spectrum of clinical outcomes in humans , including fatal , disseminated infection of the spleen and liver following infection with L . donovani , and self-curing cutaneous lesions associated with L . major and other cutaneous strains . Healed cutaneous lesions often result in a permanent scar that has been shown to harbor low numbers of parasites over the long term [2] . While this chronic , sub-clinical state can serve as a long-term reservoir for disease , it also maintains powerful protective immunity for the host , as individuals with healed primary lesions are highly resistant to re-infection , and complete elimination of a primary infection in animal models results in susceptibility to reinfection [3] , [4] . Deliberate needle inoculation with viable parasites in a selected site , referred to as “leishmanization , ” has been employed extensively as a live “vaccine” in people for generations , and is highly effective against natural exposure [5] , [6] , [7] , [8] . However , due to reports of adverse reactions at the site of inoculation , quality control issues , and concerns over causing serious disease in immuno-compromised individuals , leishmanization has fallen out of favor [8] , [9] . Employing the mouse model of L . major infection , numerous non-living [10] , [11] , [12] , [13] , [14] , [15] and live-attenuated [13] , [16] , [17] , or DNA-based [10] , [18] vaccine formulations have been developed as alternatives to leishmanization , which in many cases have conferred relatively long-term protection against experimental needle challenge [10] , [11] , [12] , [18] . In contrast , non-living vaccines , including formulations similar to those shown to work effectively in mice against needle challenge [11] , [13] , have yet to confer significant protection against natural exposure in people , despite the generation of measurable cell-mediated immunity [9] , [19] , [20] , [21] , [22] , [23] , [24] , [25] , [26] , [27] , [28] . This contradiction between the results in humans and animal trials suggests that the correlates of vaccine efficacy developed mainly from the mouse model , namely the generation of Th1 responses and the reduction of lesion size and/or parasite number following needle challenge , may not adequately define the requirements for protection against natural transmission . Observations by Rogers et al . [29] , in which vaccination with soluble leishmanial antigen plus IL-12 delayed the onset of progressive lesions following needle , but not infected sand fly challenge in BALB/c mice , support this suggestion . In addition to the delivery of infectious stage parasites into the dermis , sand flies also deposit pharmacologically active saliva , which aids in blood feeding , and egest parasite-released glycoconjugates , which accumulate behind the mouthparts in infected flies and form a promastigote secretory gel ( PSG ) . These molecules have been shown to enhance the severity of disease when co-administered with infectious stage parasites [30] , [31] , [32] , [33] . We have recently reported that sand fly transmission induces a qualitatively unique inflammatory response at the localized bite site that includes a dynamic recruitment of neutrophils , and that these neutrophils markedly enhance the ability of parasites to establish primary infection [34] . Thus , an analysis of the influence of sand fly transmission on vaccine efficacy is likely to be highly relevant to the generation of a Leishmania vaccine that is effective in people .
Healed primary L . major infection initiated by needle inoculation of mice has been extensively employed as a model that mimics the clinical practice of leishmanization . Mice with resolved primary lesions harbor L . major specific CD4 T cells that simultaneously produce IFN-γ , TNF-α , and IL-2 effector cytokines and mount powerful protective immunity at a site of needle re-challenge , resulting in the rapid control of parasite growth [13] , [35] . In order to characterize the protective immune response following natural transmission , 4 P . duboscqi sand flies , infected with L . major ( L . m . -SF ) , were allowed to feed on the ears of C57BL/6 mice with a healed primary lesion in the footpad . Under these conditions , a median of 2 flies will show evidence of blood engorgement , thereby ensuring parasite transmission to a sufficient number of ears to conduct the experiment , while at the same time more faithfully replicating natural transmission , which likely occurs following exposure to a single infected fly . At 1 and 3 days following exposure to the infected flies , a slight but significant increase in infiltrating CD4 T cells was found in the ears of healed mice relative to fly challenged , naïve , age-matched controls ( AMC ) ( Figure 1A ) . At 7 days post-challenge , the number of infiltrating CD4 cells in the healed mice was dramatically increased relative to controls . In order to determine if parasite antigen was required to mediate this recruitment , healed mice were also exposed to uninfected sand fly bites ( SF ) . Both infected or uninfected bites recruited equivalent numbers of T cells at day 3 post-bite , however , parasite antigen appeared necessary for the dramatic increase observed on day 7 ( Figure 1A ) . Remarkably , Ag re-stimulation of dermal derived cells revealed Leishmania-specific IFN-γ producing CD4+ T cells at the challenge site within 24 hours , a response that gradually increased to 17% of the total CD4 T cell population at 7 days ( Figure 1B ) , correlating with a >100 fold reduction in parasite numbers in the skin ( Figure 1C ) . Antigen re-stimulation of T cells from the ears of healed mice exposed to uninfected sand fly bites also revealed the presence of L . m . -specific IFN-γ producing CD4+ T cells ( Figure 1B ) , suggesting that a functional property of these effector cells is their ability to rapidly migrate to sites of tissue inflammation whether antigen is present or not . Vaccination with autoclaved L . major ( ALM ) , or a recombinant leishmania protein , plus CpG oligodeoxynucleotides ( ODN ) has been shown to effectively protect against needle challenge with L . major in mice [11] , [13] . We therefore employed ALM+CpG to test the efficacy of a non-living vaccine against natural transmission . Mice vaccinated with ALM+CpG three times s . c . in the footpad at two week intervals , along with age-matched naïve controls and mice with healed primary lesions , were exposed to the bites of 4 infected sand flies twelve weeks following the last vaccine injection . Four weeks following infected sand fly exposure or needle inoculation , coincident with the time of peak parasitic load in naïve mice , parasite burden in the ear dermis was assessed . Mice with healed primary lesions again dramatically controlled parasite growth following exposure to the bites of infected sand flies ( Figure 2A ) . In contrast , ALM+CpG vaccination conferred no protection against transmission by sand fly bite , despite conferring strong protection against needle inoculation . Ear lesion measurements obtained 4 weeks after infection also revealed a compromised benefit of the ALM+CpG vaccine against sand fly challenge ( Figure S1 ) . Note that despite the comparable parasitic loads in naïve mice following sand fly or needle challenge , the pathology associated with transmission by bite was far more severe . The respective doses of the fly versus needle inocula did not appear to be a factor in the different outcomes of infection in the ALM+CpG vaccine as naive mice infected via needle or sand fly bite contained similar numbers of parasites in the challenge sites at 4 wks post-infection . In order to address the issue of dose more directly , and to determine if sand fly-derived parasites might be more virulent than those obtained from culture , ALM+CpG vaccinated and healed mice were challenged by infected sand fly bite or by inoculation with a five-fold higher dose of metacyclic promastigotes purified from the midguts of sand flies harboring 14 d , mature infections . Based on previous observations [36] , 5×103 sand fly derived parasites are within the projected upper range of the variable doses transmitted following exposure to 4 infected sand flies . Mice with healed primary lesions were again powerfully protected against both needle and sand fly challenge ( Figure 2B ) , and the ALM+CpG vaccinated mice maintained their immunity against the higher dose , sand fly-derived , needle inoculum , demonstrating a 100-fold decrease in parasite load . Importantly , these mice again failed to demonstrate any protection against sand fly transmitted infection as measured by either parasite load ( Figure 2B ) , or a significant reduction in lesion scores ( Figure S2 ) . These results strongly suggest that transmission of L . major by sand fly bite , rather than an inherent difference in the dose or virulence of sand fly derived parasites , is responsible for the inability of ALM+CpG vaccinated mice to protect against natural challenge . Kinetic analysis of the immune response among groups of mice challenged by the bite of infected sand flies in Figure 2A revealed that healed mice mounted a rapid and robust L . m . -specific response , while ALM+CpG vaccinated mice mounted a much weaker response , as determined by intracellular staining of dermal- derived CD3+ T cells for IFN-γ ( Figure 2C ) or TNF-α ( Figure 2D ) . These different effector cell frequencies were reflected in the levels of IFN-γ secreted by ear-derived cells , as detected by ELISA ( Figure 2E ) . Previous observations suggest that CD4+ T cells capable of producing multiple cytokines in response to antigen stimulation are more effective at protecting against disease [13] . In agreement with these studies , we found that a large proportion of L . m . -specific T cells in the healed mice produced IFN-γ and TNF-α simultaneously at day 7 ( Figure 2F ) . These results emphasize the correlation between an early response and parasite clearance following sand fly transmission , and explain why ALM+CpG vaccinated mice were unable to control sand fly transmitted infection as compared to healed mice . We were also interested to understand why the delayed appearance of the Th1 effector response in ALM+CpG vaccinated mice was sufficient to protect against needle challenge but not sand fly challenge . At 4 wks post-infection , despite enhanced numbers of lymphocytes ( Figure S3 ) and increased levels of parasite antigen ( Figure 2A ) at the site of infected sand fly bite versus needle inoculation in ALM+CpG vaccinated mice , we observed a decrease in the frequency of both IFN-γ+ ( 6 . 8% versus 11% ) and TNF-α+ ( 3 . 7% versus 9 . 3% ) L . m . -specific T cells ( Figure 2 , C and D ) , as well as reduced levels of secreted IFN-γ ( Figure 2E ) . Thus , conditions in the bite site appear to compromise the activation and/or effector function of the memory response generated by the killed vaccine . We have recently demonstrated that the early host response to sand fly bites is associated with a unique and prolonged recruitment of neutrophils into the localized bite site , resulting in the formation of a “neutrophil plug” , and that the presence of neutrophils during the initiation of infection promotes the establishment of sand fly transmitted disease [34] . In order to explore the possibility that the host inflammatory response to sand fly bite is responsible for the failure of ALM+CpG vaccination to protect against sand fly transmitted infection , we first compared the inflammation induced by sand fly versus needle inoculation of L . major . When ear dermal cells from naive ( Figure 3A ) and ALM+CpG vaccinated mice ( Figure 3B ) were analyzed for the presence of neutrophils over the first week of infection , both needle and sand fly inoculated ears revealed a significant recruitment at 24 hours , although sand fly bitten ears had significantly greater numbers ( p = 0 . 001 ) . Importantly , only sand fly bitten ears maintained the neutrophilic infiltrate at the inoculation site at 3 and 8 days post-inoculation ( Figure 3 , A and B ) . Of note , a very transient neutrophilic response was also observed in the ears of the sham-transmitted mice , elicited by manipulation of the ear dermis during exposure to the transmission apparatus . Examination of cells derived from the ears of the needle or sand fly challenged mice shown in Figure 2A , revealed that only sand fly inoculation maintained recruitment of large numbers of neutrophils at the inoculation site at 1 and even 4 wks post-infection ( Figure 3 , C and D ) , which at least in the case of the naïve mice , was not explained by differences in the parasitic load . Analysis of all CD11b and Ly-6G/C ( Gr-1 ) expressing cells reveals that increased numbers of neutrophils were also associated with large numbers of CD11b+Gr-1int macrophages/monocytes ( Figure 3E ) . In order to visualize neutrophil recruitment and maintenance at individual sites of L . m . inoculation over time , we employed 2-photon intra-vital microscopy ( 2P-IVM ) in conjunction with a red fluorescent protein-expressing strain of L . m ( L . m . -RFP ) [36] , and naïve mice expressing enhanced green fluorescent protein ( eGFP ) under the control of the endogenous lysozyme M promotor ( LYS-eGFP mice ) [37] . As previously reported , the GFPhi cells in these mice are neutrophils [34] , [37] , and accumulate within both needle and sand fly inoculation sites shortly after infection ( Figure 3F , 2 hours; and Video S1 and S2 ) . The sand fly inoculation site is distinguished by an especially tight co-localization of RFP+ parasites and GFPhi neutrophils , which form a plug delineating the site of proboscis penetration . ( Figure 3F , 2 hours; Video S1 and S3 ) . While neutrophils were maintained at the site of parasite deposition by sand fly bite ( Figure 3F , Video S4 ) this co-localization was rapidly lost at the site of needle inoculation , and the majority of neutrophils present in the field of view at later times were within blood vessels ( Figure 3F and Video S5 ) . We explored the possibility that neutrophil depletion might rescue the ability of the killed vaccine to confer protection against sand fly transmitted infection . As neutrophils are important for the early establishment of sand fly transmitted infections , their depletion at the time of challenge would , as previously shown [34] , promote early resistance and compromise infection even in the naïve mice . Thus , the mice were left untreated for the first 3 . 5 days following sand fly transmission , then treated on days 3 . 5 , 9 , and 14 , with a neutrophil depleting Ab [38] , [39] or control IgG to mimic the loss of neutrophils observed following needle inoculation , but not sand fly transmission . Analysis of CD11b+Ly-6G+F4/80− neutrophils and CD11b+Ly-6G−F4/80+ macrophages/monocytes at the site of infection 6 days post-transmission revealed that the neutrophil depletion was both specific and efficient ( Figure 4 , A and B ) . At 2 weeks post-transmission , the neutrophil depletion promoted stronger Ag-specific IFN-γ and TNF-α responses in the ALM+CpG vaccinated mice ( Figure 4C ) . More importantly , the neutrophil depletion enhanced the efficacy of the killed vaccine . Analysis of extensive data pooled from three independent experiments revealed that on day 28 post-transmission , the neutrophil depleted , ALM+CpG-vaccinated mice showed a highly significant reduction in parasite load compared with neutrophil depleted , naïve mice ( p<0 . 0001 ) , as well as control treated , ALM+CpG vaccinated mice ( p = 0 . 002 ) , and indistinguishable from that in healed animals ( Figure 4D ) . The enhanced parasite clearance in neutrophil depleted , ALM+CpG vaccinated mice was associated with a significant reduction in lesion size compared with neutrophil depleted , naïve mice ( p<0 . 0001 ) and control treated , ALM+CpG vaccinated mice ( p = 0 . 01 ) ( Figure 4E ) . Importantly , the neutrophil-depleted , naïve controls did not exhibit lower parasite loads compared with their control treated counterparts , suggesting the effect of neutrophil depletion after the initial establishment of infection , and during the extended period of neutrophil recruitment following transmission by bite , was specific to the vaccine setting .
The generation of a safe , non-living , prophylactic vaccine against leishmaniasis has been largely unsuccessful , a failure that is not explained by the lack of available target antigens with the potential to confer a protective response [40] . Failed human trials reported in the 1990s employing ALM+BCG were particularly perplexing as the same or a similar vaccine has been shown to work well as immunotherapy to hasten cure in patients with active disease [41] , [42] . Furthermore , it elicits detectable parasite-specific IFN-γ production and leishmanin skin-test conversion in at least a proportion of recipients [20] , [21] , [22] , [23] , [24] , [25] , [26] , [27] , [28] , and similar vaccine formulations , including the ALM+CpG vaccine employed in this study , have been shown to be highly effective against needle challenge in mouse models [11] , [13] . The results reported here suggest that the killed vaccines failed in people because , while generating some correlates of immunity that may provide adequate defense against a needle inoculum , failed to generate and/or maintain the rapid , robust response at the site of secondary challenge induced by leishmanization that is required to prevent disease following delivery of parasites by sand fly bite . The protective response in healed mice is likely associated with the speed with which effector cells appear at a site of tissue damage , irrespective of the presence of parasites ( see Figure 1 ) . Following encounter with antigen in the inoculation site , these cells might then provide an immediate burst of effector cytokines , and counteract early on the down-modulatory environment created by the highly localized , neutrophil-dominated , response to sand fly bite . In contrast , and despite the ability of the killed vaccine plus CpG to generate multi-functional , effector T cells protective against needle challenge [13] ( see also Figure 4 ) , these cells are not present in adequate numbers and at sufficiently early time points to protect against sand fly transmission . This point is emphasized by the presence of similar numbers of neutrophils in both ALM+CpG vaccinated and healed mice one week following exposure to infected sand flies ( Figure 3E ) , yet only healed mice were protected . Both the rapidity of the effector response in healed mice , as well as the fact that these cells were recruited by uninfected sand fly bites , suggests that these cells are not derived from a “central” memory population , that would require antigen encounter and several rounds of division in the DLN before gaining effector function [43] . More likely , the rapid appearance of these cells in the challenge site reflects a pre-existent , tissue-seeking effector population , undergoing constant renewal by the persistence of viable organisms in the healed mice [3] , [4] . Further understanding of the effector population maintained by persistent infection , including the role of CD8+ cells , is likely to be highly informative to strategies of successful vaccination [44] . A critical question remains the mechanism by which neutrophil persistence following sand fly transmission inhibits parasite elimination in ALM+CpG vaccinated mice . Phagocyte clearance of apoptotic neutrophils during the resolution of inflammation has a known inhibitory effect on macrophage functions [45] , and DC functions are similarly impaired following uptake of apoptotic neutrophils [46] . Thus , infected macrophages and DC persistently exposed to apoptotic neutrophils at the site of sand fly bite are likely to be refractory to activation signals , inhibiting both the killing and APC functions of these cells . This is especially relevant to sand fly transmission where the association between neutrophils and macrophage/monocytes , as well as dendritic cells , is highly localized at the sand fly bite site . Apoptosis of infected neutrophils has been readily captured by 2P-IVM [34] . The maintenance of neutrophils at sand fly bite sites is likely the result of conditions leading to their protracted recruitment , as opposed to prolongation of their life span in the skin [47] . Thus , while the initial recruitment of neutrophils may be driven primarily by tissue injury , their continued presence is likely influenced by PSG or salivary components , that are themselves chemotactic or that initiate the inflammatory cascade [33] . The results reported here represent the first determination , so far as we are aware , of the factors influencing the efficacy of protective immunity generated by different vaccine formulations against sand fly challenge , and may be relevant to the conditions that modulate vaccine induced immunity to other vector borne pathogens . Beyond emphasizing the somewhat obvious importance of using natural challenge models to evaluate experimental vaccines against leishmaniasis , the results provide a more stringent set of screening criteria that might be used to predict vaccine success against fly challenge , relating to the rapid appearance of multifunctional effector cells within the challenge site . Pertinent to our findings are those of Rogers et . al . who demonstrated that vaccination of BALB/c with glycoconjugates derived from PSG diminishes disease severity following sand fly challenge [29] . Collectively , these findings should be especially informative for ongoing and future clinical development of “second-generation” Leishmania vaccines [48] , and reinforce the rationale for inclusion of molecules specific to natural transmission , such as selected components of sand fly saliva or promastigote-secretory gel , in an anti-Leishmania vaccine [49] .
Female C57BL/6 mice were obtained from Jackson Laboratories . C57BL/6 LYS-eGFP knock-in mice [37] were a gift from T . Graf ( Albert Einstein University , NY ) and were bred at Taconic Laboratories through a contract with the NIAID . Mice were maintained at a NIAID animal care facility under specific pathogen-free conditions . All animal experiments were performed under a study protocol approved by the NIAID Animal Care and Use Committee . All experiments were carried out using the L . major Friedlin strain obtained from the Jordan Valley NIH/FV1 ( MHOM/IL/80/Friedlin ) . In some experiments , a stable transfected line of FV1 L . major promastigotes expressing a red fluorescent protein was employed , as described previously [36] . Briefly , the DsRed gene was amplified by PCR employing the pCMV-DsRed-Express plasmid ( BD Biosciences/Clontech ) as a template and cloned into the SpeI site of the pKSNEO Leishmania expression plasmid . FV1 promastigotes were transfected with the resulting expression plasmid construct [pKSNEO-DsRed] and selected for growth in the presence of 50 µg/ml Geneticin ( G418 ) ( Sigma ) . L . major or L . major-RFP were grown at 26°C in medium 199 supplemented with 20% heat-inactivated FCS ( Gemini Bio-Products ) , 100 U/ml penicillin , 100 µg/ml streptomycin , 2 mM L-glutamine , 40 mM Hepes , 0 . 1 mM adenine ( in 50 mM Hepes ) , 5 mg/ml hemin ( in 50% triethanolamine ) , and 1 mg/ml 6-biotin . Infective-stage metacyclic promastigotes were isolated from stationary cultures ( 4–6 day-old ) by negative selection of non-infective forms using peanut agglutinin [50] ( PNA , Vector Laboratories Inc ) . In some experiments metacyclic promastigotes of L . major were isolated from sand flies on day 14 following infection with L . major , as previously described [36] . Briefly , Infected flies were killed , dissected aseptically , and the stomodeal valve and anterior gut of each fly was transferred into Dulbecco's modified Eagle's medium ( DMEM ) . The guts were macerated briefly using a plastic pestle , spun twice to remove the debris , and washed once in DMEM followed by metacyclic promastigote isolation as described above . Mice were subsequently infected with the specified number of parasites in the ear dermis by intra-dermal ( i . d . ) injection using a 29 ½ GA needle in a volume of 10 µl unless specified otherwise . Analysis of protective immunity in mice with a healed primary lesion was carried out using animals that had been infected 16–20 weeks previously with 104 L . major metacyclic promastigotes in the left hind footpad by sub-cutaneous injection using a 29 ½ gauge needle in a volume of 40 µl . Autoclaved Leishmania antigen ( ALM ) plus CpG oligodeoxynucleotides ( ODN ) vaccination was performed in a manner similar to that published previously [11] . Briefly , B6 mice were injected subcutaneously in their left hind footpad with 50 mg of clinical grade ALM , prepared from whole cell heat-killed L . major promastigotes ( WHO ) plus 50 µg of CpG ODN sequence 1826 ( Coley Pharmaceutical Group ) , graciously provided by Dr . P . Darrah ( VRC/NIH ) , using a 29 ½ gauge needle in a volume of 40 µl , three times , at 2 week intervals . Transmission of L . major parasites was performed as described [34] , [36] . Briefly , 2–4 day old P . duboscqi ( Mali colony ) female sand flies were infected via feeding through a chick skin membrane on heparinized mouse blood containing L . major or L . major-RFP amastigotes or promastigotes . After 14–15 days , individual flies were transferred to plastic vials covered at one end with nylon mesh . Mice were anesthetized by intraperitoneal injection of 30 µl of ketamine/rompin ( 100 mg/ml ) . Specially designed clamps were used to bring the mesh end of each vial into contact with the ear of an anesthetized mouse , allowing flies inside the vial to feed on the ear skin for a period of 2 to 3 hours in the dark . In some experiments mice were exposed to empty vials . The number of flies with blood meals was employed as a means of checking for equivalent exposure to potential transmission by sand fly bite among animals in different treatment groups . The median number of flies with blood meals in vials with 4 flies was 2 . Ear tissue was prepared as previously described [34] . Briefly , the ventral and dorsal sheets of needle or sand fly inoculated ears were separated , deposited in DMEM containing 100 U/ml penicillin , 100 µg/ml streptomycin and 0 . 2 mg/ml Liberase CI purified enzyme blend ( Roche Diagnostic Corp . ) , and incubated for 2 hours at 37°C and 5% CO2 . Digested ear sheets were subsequently homogenized for 3 minutes using the Medicon/Medimachine tissue homogenizer system ( Beckton Dickinson ) . Individual retromaxillary ( ear ) lymph nodes were removed , and mechanically dissociated using tweezers and a syringe plunger . Single cell suspensions of tissue homogenates were then filtered using a 70 µm-pore size Falcon cell strainer ( BD Biosciences ) . Mice were sacrificed and single cell suspensions from the ear dermis were obtained as described above . Cells were incubated without fixation with an anti-Fc-γ III/II ( CD16/32 ) receptor Ab ( 2 . 4G2 , BD Biosciences ) in RPMI without phenol red ( Gibco ) containing 1 . 0% FCS for 10” followed by incubation for 20” with a combination of 4 or 6 of the following anti-mouse antibodies: PE-Cy7 or APC anti-CD11b ( M1/70 BD Biosciences ) ; Per-CP Cy5 . 5 anti-Gr-1 ( Ly6G/C ) ( RB6-8C5 , BD Biosciences ) ; FITC or PE anti-Ly6G ( 1A8 , BD Biosciences ) ; PE anti-CD11c ( HL3 , BD Biosciences ) ; Per-CP Cy5 . 5 anti-CD11c ( N418 , BioLegend ) ; APC anti-F4/80 ( BM8 , eBioscience ) , FITC anti-I-Ab ( AF6-120 . 1 , BD Biosciences ) ; or Alexafluor-700 anti-mouse MHC II ( M5/114 . 15 . 2 , eBioscience ) . The isotype controls employed were rat IgG1 ( R3-34 ) and rat IgG2b ( A95-1 ) . The data were collected and analyzed using CELLQuest software and a FACScalibur or FacsDIVA software and a FacsCANTO flow cytometer ( BD Biosciences ) . Gating of ear-derived cells was carried out as described previously [34] . Ears were analyzed individually , or pooled with ears from the same group , as indicated in the text . Whole ear single-cell suspensions in RPMI 1640 containing 10% FCS , 10 mM Hepes , L-glutamine , and penicillin/streptomycin , obtained as described above , were incubated at 37°C in 5% CO2 for 16–18 hours in flat-bottom 48-well plates with 2 . 5×105 BMDCs , with or without 50 mg/ml freeze-thaw Leishmania antigen prepared from L . major V1 stationary phase promastigotes , in a final volume of 1 ml . During the last 5–6 hours of culture Brefeldin A ( Golgiplug; BD Biosciences ) was added to block golgi transport according the manufacturers' instructions . Following in vitro culture , cells were washed and stained with anti-Fc III/II ( CD16/32 ) receptor Ab ( 2 . 4G2 ) for 10 minutes in RPMI without phenol red containing 1 . 0% FCS , followed by PE-Cy7 or PE-Cy5 anti-mouse CD4 ( RM4-5 ) for 15 minutes . In some experiments cells were also stained with FITC anti-TcR β ( 145-2 C11 ) . Cells were then fixed with BD Cytofix/Cytoperm ( BD Biosciences ) and stained with anti-Fc III/II ( CD16/32 ) receptor Ab ( 2 . 4G2 ) followed by a combination of the following anti-mouse antibodies: PerCP-Cy5 . 5 anti-CD3 ( 145-2C11 ) , FITC- , APC- , or AlexaFluor 700 anti-IFN-g ( XMG1 . 2 ) , and FITC or PE anti-TNF-α ( MP6-XT22 ) . Intracellular staining was carried out for 30 minutes on ice . All antibodies were acquired from BD Biosciences . For each sample , greater then or equal to 4000 CD4+CD3+ cells were collected using a FACS Caliber or FACS Canto flow cytometer and analyzed using either Cell Quest Pro or FACS Diva Software , respectively ( BD Biosciences ) . For measurement of IFN-γ in culture supernatants , pooled , single-cell suspensions of ear tissue as described above were incubated in triplicate at 37°C in 5% CO2 for 72 hours in 96-well round bottom plates with 2 . 5×105/ml BMDC with or without freeze-thaw Leishmania antigen in a total volume of 200 ml . Following incubation , the concentration of IFN-γ in the culture supernatant was determined by ELISA according the manufactures instructions ( eBioscience ) . Parasite titrations were performed as previously described [31] . Briefly , tissue homogenates were serially diluted in 96-well flat-bottom microtiter plates containing biphasic medium , prepared using 50 µl NNN medium containing 20% of defibrinated rabbit blood and overlaid with 100 µl M199/S . The number of viable parasites in each ear was determined from the highest dilution at which promastigotes could be grown out after 7–10 days of incubation at 26°C . Because individual sand flies , or more then one sand fly may deposit parasites in more than one location , sand fly bitten ears often have more then one lesion . Total lesion diameter was determined by measuring the diameter of individual lesions using a caliper and in cases where there was more then one lesion per ear the diameters were added together . Two photon intravital imaging and image analysis was performed as described previously [34] . Briefly , anesthetized mice were imaged in the lateral recumbent position that allowed the ventral side of the ear pinna to rest on a coverslip . A strip of Durapore tape ( 3 M ) was stuck to a bench top several times ( to ensure that subsequent removal would not cause undue damage ) and placed lightly over the ear pinna and affixed to the imaging platform in order to immobilize the tissue . Care was taken to minimize pressure on the ear . Images were acquired using an inverted LSM 510 NLO multiphoton microscope ( Carl Zeiss Microimaging ) enclosed in an environmental chamber that was maintained at 30°C . This system had been custom fitted with 3 external non-descanned PMT detectors in the reflected light path . Images were acquired using either a 20×/0 . 8 air objective or a 25×/0 . 8 NA water immersion objective . Fluorescence excitation was provided by a Chamelon XR Ti:Sapphire laser ( Coherent ) tuned to 920 nm for eGFP excitation . Voxel dimensions were 0 . 64×0 . 64×2 µm using the 20× objective and 0 . 36–0 . 51×0 . 36–0 . 51×2 µm using the 25× objective . Raw imaging data were processed with Imaris ( Biplane ) using a Gaussian filter for noise reduction . All images are displayed as 2D maximum intensity projections . Movie files of 3-dimentional images were generated using Imaris . Animals were treated with three 0 . 5 mg injections of a neutrophil depleting ( NIMP-R14 ) or control ( GL113 ) IgG antibody , i . p . , on days 3 . 5 , 9 , and 14 following sand fly transmission . The first dose of antibody was delayed until 3 . 5 days after exposure to infected sand fly bite as earlier observations demonstrated that L . m . infection is established in macrophages at this time [34] . Antibody treatments were spaced 5 days apart as preliminary experiments suggested excessive administration of the NIMP-R14 antibody , such as on successive days , led to depletion of cell types other then neutrophils . Success and specificity of depletions were determined as described in the text . The NIMP-R14 hybridoma was a gift from Dr . Y . Belkaid ( NIAID ) . Parasite loads in the ears of mice transmitted with L . major by infected sand fly bite do not follow a Gaussian distribution . This is likely the result of variability in the infectious burden and feeding behavior of individual , infected , sand flies [36] . Therefore , data sets were compared using a nonparametric Mann Whitney test . Mann Whitney calculations were done using Prism 4 ( Graphpad Software , Inc . San Diego , CA ) . In Figure 4 , D and E , parasite loads and lesion size were compared using an exact stratified Wilcoxon rank sum test , stratified by experiment in order to allow pooling of experiments as described in the text . The stratified Wilcoxon calculations were done in StatXact 8 Procs ( Cytel , Inc . , Cambridge , MA ) . Comparisons in which the data represented replicate samples were carried out using t-tests . All p-values are two-sided . | The generation of vaccines that protect against intracellular pathogens such as malaria , human immunodeficiency virus and leishmaniasis have met with limited success . A perplexing aspect of this failure as it relates to leishmaniasis is the knowledge that individuals typically get the disease only once , and that individuals who are experimentally infected with cultured parasites are protected against sand fly transmitted infection , thereby providing a “gold standard” for vaccine design . Many engineered , non-living vaccines have been developed to mimic the immune response observed in protected individuals and some of these have been shown to provide excellent protection against needle inoculation of Leishmania parasites in mice . However , very similar vaccine formulations adapted for use in people have failed to protect against natural exposure to infected sand fly bites . In the present study , we attempt to reconcile these long-standing differences , and to provide the critical correlates of immunity that will predict vaccination success against natural exposure . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"infectious",
"diseases/neglected",
"tropical",
"diseases",
"infectious",
"diseases/skin",
"infections",
"infectious",
"diseases/protozoal",
"infections",
"immunology/immunity",
"to",
"infections",
"public",
"health",
"and",
"epidemiology/immunization"
] | 2009 | Vector Transmission of Leishmania Abrogates Vaccine-Induced Protective Immunity |
Dengue is a common and growing problem worldwide , with an estimated 70–140 million cases per year . Traditional , healthcare-based , government-implemented dengue surveillance is resource intensive and slow . As global Internet use has increased , novel , Internet-based disease monitoring tools have emerged . Google Dengue Trends ( GDT ) uses near real-time search query data to create an index of dengue incidence that is a linear proxy for traditional surveillance . Studies have shown that GDT correlates highly with dengue incidence in multiple countries on a large spatial scale . This study addresses the heterogeneity of GDT at smaller spatial scales , assessing its accuracy at the state-level in Mexico and identifying factors that are associated with its accuracy . We used Pearson correlation to estimate the association between GDT and traditional dengue surveillance data for Mexico at the national level and for 17 Mexican states . Nationally , GDT captured approximately 83% of the variability in reported cases over the 9 study years . The correlation between GDT and reported cases varied from state to state , capturing anywhere from 1% of the variability in Baja California to 88% in Chiapas , with higher accuracy in states with higher dengue average annual incidence . A model including annual average maximum temperature , precipitation , and their interaction accounted for 81% of the variability in GDT accuracy between states . This climate model was the best indicator of GDT accuracy , suggesting that GDT works best in areas with intense transmission , particularly where local climate is well suited for transmission . Internet accessibility ( average ∼36% ) did not appear to affect GDT accuracy . While GDT seems to be a less robust indicator of local transmission in areas of low incidence and unfavorable climate , it may indicate cases among travelers in those areas . Identifying the strengths and limitations of novel surveillance is critical for these types of data to be used to make public health decisions and forecasting models .
The global incidence of dengue has increased 30-fold between 1960 and 2010 [1] , with a recent study estimating that there are now 70–140 million cases per year [2] . Dengue is caused by infection with any of the four dengue virus ( DENV ) serotypes; the symptoms often include high fever , intense joint and muscle pain , headaches , and skin rash . Some infections result in more serious illness including hemorrhagic symptoms and death [3] . Endemic in many Asian and Latin American countries , dengue has become a leading cause of hospitalization and death among children in these regions [4] and contributes to substantial economic loss for governments and households [5] . Despite the health and economic impacts of dengue , population-level control methods are limited , resource intensive , and largely ineffective to date . Real-time dengue surveillance , therefore , is critical for identifying areas where transmission is ongoing or likely to occur so that interventions can be optimized . Traditional , healthcare-based , government-implemented dengue surveillance has several shortcomings . Often , it takes weeks to aggregate surveillance data and publish related reports . This lag in part reflects the time needed to collect and aggregate data at different scales , from practitioners up to the Ministry of Health level , but it can also be delayed or interrupted due to lack of resources and bureaucratic or political changes [6] , [7] . Meanwhile , as global Internet use has increased , novel disease monitoring tools based on health-related search queries have emerged . Google Dengue Trends ( GDT ) was developed by aggregating historical logs of anonymous online Google search queries associated with dengue using the methods developed for Google Flu Trends , a tool created to estimate influenza rates [8] . Google queries have shown to be a close proxy for national-level dengue surveillance in multiple countries [9] , [10] . And because data are collected and processed in near real-time , these tools produce surveillance data much faster than traditional systems [8] , [11] , [12] . While GDT has this significant advantage and well-demonstrated large-scale accuracy , it remains unclear how well it works at smaller scales where dengue transmission may be more heterogeneous . Dengue transmission dynamics are sensitive to the environmental factors that affect the vector mosquitoes [13] . Temperature increases can decrease the length of the gonotrophic cycle [14] , increase the feeding frequency [15] , increase the rate of mosquito development , and reduce the length of the DENV incubation period within the mosquito [16] , [17] . Mosquito survival also increases with temperature , but at a certain point , high temperatures can also lead to high mosquito mortality [14] , [18] , [19] . Precipitation is also important to the spatial and temporal spread of the mosquito vector [20]–[24] . Lastly , human behavior and habitat modification can contribute to DENV transmission dynamics: the use of screens or air conditioning can reduce human-vector contact [13]; water storage and trash disposal practices are important determinants of larval habitat availability [25]; and a high human population density provides more transmission opportunities [26] . Therefore , information about relevant environmental conditions can contribute to identifying the dengue risk . Mexico provides a unique setting to assess the value of GDT data; the climate varies widely across the country , dengue is endemic in many areas yet largely absent in others , and approximately 36% of the population has Internet access [27] . Here , we explore the relationship between GDT data and traditional surveillance data for 17 states in Mexico and use climate and socio-demographic data to investigate geographic variation in GDT accuracy .
The GDT index was developed as a linear model to predict reported dengue incidence from dengue-related Internet search patterns [9] . Specifically , it incorporates weekly query volume for key terms ( normalized to overall search volume ) and uses the historical relationship between those terms and reported cases to linearly predict ( nowcast ) dengue activity . We downloaded weekly GDT data for 2003–2011 for Mexico as a country and for the available years in that time range ( 2–8 years ) for the 17 individual states with available data: Baja California , Chiapas , Colima , Distrito Federal , Estado de Mexico , Jalisco , Morelos , Nayarit , Nuevo LeÓn , Oaxaca , Quintana Roo , Sinaloa , Sonora , Tabasco , Tamaulipas , Veracruz and Yucatan [28] [9] . To create a monthly GDT variable , we averaged GDT across all weeks beginning in each month . Traditional monthly dengue surveillance data for the same time period - 2003–2011 - were obtained from the Mexican Secretariat of Health ( http://www . epidemiologia . salud . gob . mx/anuario/html/anuarios . html ) [29] , Long-term ( 1941–2005 ) mean annual precipitation ( millimeters per year ) and mean , minimum , and maximum temperature ( °C ) data were obtained for each state from the Mexican Secretariat of the Environment and Natural Resources ( SEMARNAT ) ( smn . conagua . gob . mx ) . State-level socio-demographic data were obtained from the Mexican National Institute of Statistics and Geography ( INEGI ) ( www . inegi . org . mx/ ) . The socio-demographic data included the most recent data available for the following variables: the population size and density per kilometer ( 2010 ) , the percentage of the population under the age of 15 ( 2010 ) , the number of doctors per 100 , 000 residents ( 2008 ) , the percentage of the population with access to drinking water ( 2006 ) , the percentage of the population with municipal sewage ( 2008 ) , the percentage of the population with Internet access ( 2008 ) , and the average household income in pesos ( 2010 ) . The data for precipitation , population size , population density , and average yearly dengue cases were log transformed to reduce skewing . To quantify the accuracy of GDT relative to reported dengue cases , we used Pearson correlation to assess linear correlation because GDT was designed as a linear predictor of dengue incidence . We estimated the association between GDT and the traditional surveillance data at the national level and for each state , and calculated coefficients of determination ( R2 ) to assess the proportion of dengue incidence variance captured by the GDT data . We then logit-transformed R2 and used Gaussian regression to assess the association between each climate and socio-demographic variable and the variability in state-level correlations between GDT and traditional surveillance data . The Akaike's Information Criterion ( AIC ) was applied to compare the fit for each of the different models . All calculations were performed in R version 2 . 14 ( http://www . r-project . org/ ) .
A total of 352 , 093 dengue cases were reported in all of Mexico from 2003–2011 . Figure 1 shows the national-level monthly GDT index compared to the monthly reported cases . These data show a pattern of seasonal outbreaks , generally peaking between August and November , and substantial variation in incidence between seasons . The Pearson's correlation coefficient between GDT and reported dengue cases was 0 . 91 over the 9 years , indicating that GDT captured approximately 83% of the variability in the national surveillance data . Correlation between monthly GDT and traditional surveillance data , however , varied between states . The coefficient of determination , R2 , varied from 0 . 01 in Baja California to 0 . 88 in Chiapas . Despite the presence of GDT data for the Distrito Federal , the biggest metropolitan area of the country , R2 could not be calculated because there were no reported cases during the study period . Figure 2A shows the coefficients of determination for this relationship in each state . In general , there was a stronger correlation in the southern and western coastal states , with the exception of Baja California . State-level correlation between GDT and case data was strongest in the states with high annual dengue incidence ( Table 1 , Figure 3A ) . States with higher average mean temperature , maximum temperature , and precipitation had significantly higher correlation between GDT and dengue case numbers ( Figure 3B–D , Table 1 ) . States with lower average household income , a greater proportion of youths in the population , and less internet access tended to have higher correlations , but these associations were not statistically significant ( Table 1 ) . We investigated models incorporating combinations of these variables . A model incorporating maximum temperature , logged precipitation , and the interaction of those two variables described 81% of the variance compared to 67% for the model with only dengue incidence and reduced the AIC from 43 to 39 ( Table 1 , Table 2 ) . Adding socio-demographic factors to this model did not improve the fit . Next , we used this climate-based model to predict the correlation between GDT and case data for all the states , including those where GDT data are not available ( Figure 2B ) . There was general agreement between observed ( Figure 2A ) and estimated correlation ( Figure 2B ) . Furthermore , the model predicts that for states with higher incidence such as Guerrero , where GDT is not available , GDT may in fact be a good indicator of dengue . However , in states with lower dengue incidence and cooler temperatures , like Chihuahua , GDT may not be an accurate indicator of dengue incidence . Overall , the results show that GDT is a better indicator of real-time incidence in states with high incidence and climate conditions that favor transmission .
At the national level , we found that the official case reports correlated well with GDT . Yet , the correlation between GDT and reported cases varied substantially from state to state , with stronger correlation in states with higher dengue incidence . Climate plays a key role in determining the geographic range and activity of the mosquitoes that transmit DENV . We found that in states with warmer temperatures and greater precipitation , such as Chiapas and Jalisco , GDT was strongly correlated with reported dengue incidence . The role of climate in DENV transmission , however , is complicated by other biological and socio-demographic factors [20] . Here , however , we did not find that socio-economic factors had a strong influence on the accuracy of GDT . This is particularly important because GDT relies on internet searches , and internet access can vary widely in different settings . We found that Internet access from home was not associated with GDT accuracy , suggesting that even with Internet access in the 30% range , search query data may be robust enough to capture population-level disease dynamics . Internet access will likely only increase in the future , leading to the possibility that greater data flow will improve the accuracy of measures such as GDT . While it is possible that income or internet access do affect GDT accuracy in Mexico , their importance may be overshadowed and confounded by climate , the strongest determinant in our analysis . Our intention was to identify relatively static characteristics that relate to the potential utility of tools like GDT . As such , we used covariate data from the single , most recent year or long-term averages . Future work will build on these findings to determine how temporal variation in relevant covariates may be combined with GDT to improve dengue prediction . Using the climate-based model , we predicted the utility of GDT for the states where the GDT data are not available . For example , in Guerrero , where GDT is currently not available , our model suggests that it would provide a robust estimate of dengue incidence . Yet , for states where dengue cases are rarer , such as in Chihuahua , the predicted utility of GDT is low . In these areas , where GDT appears to be a poor indicator of local transmission levels , it may nonetheless be a good indicator of some level of health-related activity such as travelers becoming sick in endemic areas , returning home , and searching for dengue information on the Internet . This information would be useful for those interested in estimating local disease burden if not local transmission intensity . Thus , GDT may provide different value in distinct climatic or socio-economic contexts . Dengue transmission patterns are highly variable and difficult to predict; timely dengue surveillance methods like GDT are needed to keep pace with the spread of the disease . We found that GDT is accurate in areas of high incidence and favorable vector climate conditions . While it appears to be a less robust gauge of local transmission in areas of low incidence and unfavorable climate , it may indicate infections among travelers . As the burden of dengue increases and traditional surveillance efforts struggle to keep pace , novel surveillance tools like GDT can provide timely information to public health officials and contribute to real-time predictive models . | Dengue is a common and growing problem worldwide . Delays in traditional surveillance systems limit the ability of public health agencies to identify and respond to dengue outbreaks efficiently . Internet search queries provide near real-time indicators of infectious disease activity and have proven effective for monitoring disease activity in some countries , but have not been assessed on smaller geographic areas . We compared Google Dengue Trends data for 17 states in Mexico to traditional surveillance data from those states . We found that the utility of Google Dengue Trends at the state-level is highly variable and depends on climatic conditions supporting dengue virus transmission . Novel surveillance tools like Google Dengue Trends can provide timely information to public health agencies , but to be useful on a local scale , they must be considered within the local context of dengue transmissibility . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"medicine",
"disease",
"mapping",
"infectious",
"diseases",
"mosquitoes",
"climatology",
"epidemiological",
"methods",
"meteorology",
"infectious",
"disease",
"epidemiology",
"epidemiology",
"earth",
"sciences",
"atmospheric",
"science",
"dengue",
"fever",
"neglected",
"tropical",
"diseases",
"travel-associated",
"diseases",
"vectors",
"and",
"hosts",
"economic",
"epidemiology",
"infectious",
"disease",
"modeling",
"disease",
"informatics"
] | 2014 | Evaluation of Internet-Based Dengue Query Data: Google Dengue Trends |
Preterm birth is a leading cause of morbidity and mortality in infants . Genetic and environmental factors play a role in the susceptibility to preterm birth , but despite many investigations , the genetic basis for preterm birth remain largely unknown . Our objective was to identify rare , possibly damaging , nucleotide variants in mothers from families with recurrent spontaneous preterm births ( SPTB ) . DNA samples from 17 Finnish mothers who delivered at least one infant preterm were subjected to whole exome sequencing . All mothers were of northern Finnish origin and were from seven multiplex families . Additional replication samples of European origin consisted of 93 Danish sister pairs ( and two sister triads ) , all with a history of a preterm delivery . Rare exonic variants ( frequency <1% ) were analyzed to identify genes and pathways likely to affect SPTB susceptibility . We identified rare , possibly damaging , variants in genes that were common to multiple affected individuals . The glucocorticoid receptor signaling pathway was the most significant ( p<1 . 7e-8 ) with genes containing these variants in a subgroup of ten Finnish mothers , each having had 2–4 SPTBs . This pathway was replicated among the Danish sister pairs . A gene in this pathway , heat shock protein family A ( Hsp70 ) member 1 like ( HSPA1L ) , contains two likely damaging missense alleles that were found in four different Finnish families . One of the variants ( rs34620296 ) had a higher frequency in cases compared to controls ( 0 . 0025 vs . 0 . 0010 , p = 0 . 002 ) in a large preterm birth genome-wide association study ( GWAS ) consisting of mothers of general European ancestry . Sister pairs in replication samples also shared rare , likely damaging HSPA1L variants . Furthermore , in silico analysis predicted an additional phosphorylation site generated by rs34620296 that could potentially affect chaperone activity or HSPA1L protein stability . Finally , in vitro functional experiment showed a link between HSPA1L activity and decidualization . In conclusion , rare , likely damaging , variants in HSPA1L were observed in multiple families with recurrent SPTB .
Preterm birth ( PTB ) , defined as birth before 37 completed weeks of gestation , is a major global public health concern . Worldwide , over 15 million infants ( more than one in ten babies ) are born preterm and of those , more than one million die from complications related to preterm birth each year [1] . Preterm birth and its complications are the leading cause of neonatal deaths and have become the major cause of death among children under five years old [2] . Moreover , preterm infants are at increased risk , not only of short-term complications but also of life-long disabilities , such as respiratory and cognitive disorders [1] . Preterm birth also increases the risk of adult-onset disorders , such as obesity , diabetes and cardiovascular diseases [3 , 4] . Currently , there is no generally effective method for prevention of preterm delivery . The majority ( ~70% ) of preterm births occur after spontaneous onset of labor , with or without preterm prelabor rupture of the membranes ( PPROM ) [5] . Most spontaneous preterm births ( SPTBs ) are idiopathic [1 , 5]; however , recurrence of preterm birth among mothers and within families indicates that genetic factors may be important . Genetic factors are estimated to account for 25–40% of the variation in birth timing [6] , with the maternal genome playing the major , but not only , role in predisposition to preterm birth [7–11] . Despite many studies of the genetics of SPTB [6 , 12 , 13] , only a few variants have been robustly associated with this outcome [14] , and their functional implications are unclear . Previous genome-wide association studies ( GWAS ) of SPTB have involved common variants , but they explain only a small portion of the genetic risk . The role of rare variants in SPTB has been essentially unexplored . Whole exome sequencing ( WES ) in families offers a comprehensive method to identify rare variant associations with disease , including almost complete coverage of the protein coding regions of the genome . Even though studies of rare variants underlying Mendelian disorders have revealed novel genes [15 , 16] , using WES to study complex multifactorial syndromes remains a challenge [17] . Previous sequencing studies of PTB [18] or PPROM [19 , 20] have focused only on a set of candidate gene regions and , consequently , have missed the majority of the coding regions of the genome . In contrast to whole genome sequencing , WES is more cost effective and has the advantage of providing more easily interpreted results . We performed a WES study using families under the hypothesis that familial recurrence is influenced by rare variants with large individual effects on SPTB susceptibility . Such an approach has the potential of identifying genes containing rare variants shared in these multiplex families , as well as genes in pathways common across families . This method applies a hypothesis-free testing approach to identify potentially novel candidate genes for SPTB .
Seventeen mothers from seven northern Finnish multiplex families ( Discovery cohort ) and an additional 192 mothers from 95 Danish families ( Replication cohort ) were sequenced using WES . The pedigrees of the multiplex Finnish families are shown in S1 Fig . For the Discovery cohort , all samples ( except one that was excluded from subsequent analyses ) passed the quality control parameters used for the clinical exome sequencing at the CMH; quality control cutoffs were 85% reads aligned , 80% aligned with alignment quality of 20 or greater . For these samples , the mean and median heterozygous/homozygous variant ratios were 0 . 765 and 0 . 748 , respectively . Prior to variant filtering in Ingenuity , the mean/median numbers of nucleotide variant calls per individual were 318 , 767/326 , 474 ( Discovery cohort ) and 221 , 682/184 , 381 ( Replication cohort ) . This difference in variant calls between the Discovery and Replication populations is likely due the fact that populations were sequenced using different Next Generation Sequencing platforms , Illumina for Discovery cohort and Complete Genomics for Replication cohort , and their respective primary quality control measures and variant calling methods were thus different . Mean transition/transversion ( Ti/Tv ) ratios were 2 . 2 and 2 . 0 for Discovery and Replication exomes , respectively . An overview of the WES workflow is presented in Fig 1 . Three software programs ( Ingenuity Variant Analysis , Varseq and the CMH Variant Warehouse ) were used to assess common shared ( by affected mothers per family ) rare variants . Only those variants that passed the prioritizing steps with at least two of the annotating software tools were considered valid and are described below ( summarized in S4 Table ) . The benefits of comparing data obtained from multiple software is that it minimized the possibility of picking up falsely called variants that passed quality control filters only by one software . This approach resulted in a total of 844 variants in the Discovery population . For the Replication population , we combined and compared the shared rare variants passing the annotation and prioritizing steps of Ingenuity Variant Analysis and Varseq; a total of 8431 variants passed the filters of both software tools . The CMH Variant Warehouse was not available for the Replication set . For both populations , variants were categorized as loss of function , moderate , or other , according to their predicted consequences , i . e . pathogenicity ( S5 Table ) . We further compared the list of variants resulting from the family-based analyses ( as described above ) between the Discovery and Replication populations . Numbers of common genes and variants for both populations are shown in S2 Fig . There were 72 rare variants that were found in both populations in 72 genes ( S6 Table ) . Rare single nucleotide variants from HSPA1L [heat shock protein family A ( Hsp70 ) member 1 like] , identified by the Discovery Ingenuity pathway analysis , were further investigated using imputed GWAS data that also included variants with MAF <1% . In the Discovery set , variants in AR , NCOA3 and NCOR2 were either CAG repeat length polymorphisms , in-frame deletions or insertions , respectively , and were , therefore , not investigated in the GWAS datasets . Three independent GWAS datasets were used , one of general European ancestry containing more than 40 , 000 mothers of live births ( 23andMe dataset ) and two from Northern Europe containing 4 , 600 and 600 mothers ( Nordic and northern Finnish datasets , respectively ) . In the large 23andMe preterm birth GWAS dataset , the minor allele of rs34620296 in HSPA1L , which is in the glucocorticoid receptor signaling pathway , was found to be more common in cases than in controls ( case frequency 0 . 0025 vs . control frequency 0 . 0010 , p = 0 . 002; Table 3 ) . This association was also significant for gestational age as a continuous trait ( gestational age as weeks; p = 0 . 0016 , effect -0 . 8238 , standard error 0 . 2608 ) . The HSPA1L variants from the Discovery ( rs34620296 and rs150472288 ) and the Replication ( rs482145 , rs139193421 ) analyses are listed in detail in Table 3 . In the two smaller GWAS datasets , however , these four HSPA1L variants were absent or not significant . Lack of significance may be due to smaller numbers of individuals , especially in cases . Sanger sequencing confirmed the genotypes of the two rare HSPA1L missense variants ( rs34620296 and rs150472288 ) in the samples from the Discovery cohort . The rare HSPA1L variants were observed in a total of six mothers from four unrelated families . Additional family members with available DNA were sequenced for these variants . Interestingly , in two of the families , female carriers of the maternally inherited rs34620296 minor T-allele were born preterm , whereas in the other two unrelated families the male carriers of maternally inherited rs150472288 minor T-allele were born preterm . However , numbers of minor allele carriers are too small for any definite gender related conclusions . Pathogenicity predictions for rs34620296 and rs150472288 derived from the Discovery cohort as well as for rs482145 and rs139193421 from the Replication cohort were assessed using in silico tools SIFT and PolyPhen-2 , and all these variants were predicted as damaging and probably/possibly damaging , respectively ( Table 4 ) . In addition , MutationTaster and MutationAssessor predicted all four variants as disease causing and predicted functional ( high ) , respectively . According to the Combined Annotation Dependent Depletion ( CADD ) score ( >20 ) , all of these variants , except for rs139193421 , are among the top 1% of deleterious variants in human genome ( Table 4 ) . To assess potential consequences of these variants on transcriptional activity , we evaluated them for evidence of histone modification or DNase I hypersensitivity . In silico tools HaploReg 4 . 1 and/or RegulomeDB showed that all four variants were in regions that had histone marks , as well as strong transcriptional regulatory signatures in various cells of the immune system , especially in T lymphocytes from peripheral blood ( Table 4 ) . Evidence of an active transcription start site was predicted in the HeLa-S3 Cervical Carcinoma Cell Line for rs34620296 and in foreskin fibroblast primary cells for rs482145 ( Table 4 ) . Further evidence of active DNA accessibility ( DNAse ) was found in ovarian tissue for rs150472288 , and in psoas muscle tissue for rs139193421 ( Table 4 ) . There was also evidence of a transcriptional effect of rs34620296 and rs150472288 ( Discovery ) in ovary and fetal adrenal gland . Together these results from HaploReg 4 . 1 and RegulomeDB provide evidence for the potential involvement of HSPA1L variants in the endocrine system , as well as in the adaptive immune cells . These variants could , therefore , have a role in the etiology of SPTB . We further investigated putative effects of HSPA1L rs34620296 on protein structure . This variant was selected due to its association with SPTB in the large 23andMe GWAS dataset . This variant causes an amino acid change from Alanine to Threonine at position 268 ( Ala268Thr ) . According to the NetPhos 3 . 1 in silico prediction , Ala268Thr generates an additional phosphorylation site next to an existing phosphorylation site ( T267-p ) ( S3 Fig ) . Furthermore , Ala268Thr is near an adenosine triphosphate ( ATP ) nucleotide-binding site located downstream at position 270−277 ( Fig 2A ) . Gain of phosphorylation may cause changes in binding energy , modulate physio-chemical properties or stability kinetics and dynamics of the protein functions such as strength of protein-protein interactions [22] . To investigate the possible effects of the missense variant on protein structure , the reference HSPA1L protein structure and a structure including the Ala268Thr variant were compared simultaneously using UCSF Chimera . There was not a visible change in the overlaid protein structures ( Fig 2B ) . Instead , there was a slight change in the chemical bond lengths ( ≥0 . 002Å ) of the adenosine diphosphate ( ADP ) -ligand binding amino acid side chains at positions Glu270 , Arg274 and Asp368 , shown in the 3D model of the HSPA1L ( Fig 2C ) . This may be due to the change from a small size , and hydrophobic , ( Ala ) to medium size , and polar , ( Thr ) residue . Such a change in the amino acid side chains could affect the binding efficiency of the ADP molecule . To further explore possible underlying biological functionality , we investigated the tissue expression established via HSPA1L , along with AR , NCOA3 and NCOR2 , using HumanBase ( http://hb . flatironinstitute . org ) . HSPA1L was expressed in placental tissue with reasonable confidence ( 0 . 65 ) , and in ovarian ( 0 . 57 ) and fetal tissues ( 0 . 48 ) as well as in uterus ( 0 . 29 ) . For HSPA1L , AR , NCOA3 and NCOR2 together , the average expression confidence was high in placenta ( 0 . 74 ) , ovary ( 0 . 70 ) , fetus ( 0 . 65 ) , and moderate in uterus ( 0 . 29 ) , indicating high confidence for expression in female reproductive system overall ( S4 Fig ) . To determine whether the HSPA1L Ala268Thr ( rs34620296 ) variant alters activity of the GR signaling pathway , we analyzed the consequences of glucocorticoid exposure during decidualization . Human endometrial stromal fibroblasts were transfected with plasmids containing either WT or Ala268Thr cDNA , or with empty vector serving as control . The cells were treated with decidualization media for 72h in a presence of glucocorticoids ( 100nM dexamethasone ) as a surrogate of stress . Protein levels of HSPA1L and GR , as well as mRNA levels of Wnt Family Member 4 ( WNT4 ) were measured . Cells transfected with the WT HSPA1L-pcDNA3 . 1 trended to greater increases in cytosolic HSPA1L protein content than those transfected with the Ala268Thr HSPA1L-pcDNA3 . 1 ( mean ± SEM; 1 . 272 ± 0 . 142 vs . 0 . 893 ± 0 . 146 , respectively , p = 0 . 09 ) ( Fig 3 ) . Furthermore , the Western blot analysis showed that the relative cytosolic protein levels of GR differed significantly between the WT and Ala268Thr groups with more GR present in the WT group than in the Ala268Thr group ( mean ± SEM; 1 . 309 ± 0 . 099 vs . 0 . 993 ± 0 . 096 , respectively , p = 0 . 04 ) ( Fig 3; numerical data available in S7 Table ) . Next , we determined the relative gene expression of WNT4 by qPCR . WNT4 is a critical decidualization target found in the recent GWA study [14] associated with gestational length . Increased expression of WNT4 was observed in the WT group , whereas , the Ala268Thr group was less able to activate the WNT4-signaling pathway leading to a lower expression of WNT4 ( p = 0 . 04 ) .
To move beyond traditional case-control GWAS and family-based linkage studies , we performed a case-only whole exome sequencing study designed to investigate the burden of rare variants in families with recurrent SPTB . Whole exome sequencing enables the discovery of rare , putatively functional variants associated with the etiology of complex disease on a gene-by-gene or a pathway-by-pathway basis , and enrichment in multiplex families provides a means to filter large-scale sequencing data . Comparisons of mothers with recurrent preterm deliveries identified the glucocorticoid receptor signaling pathway as a candidate for mediating the risk of SPTB . Specifically , within this pathway , likely pathogenic missense variations in HSPA1L were found among four unrelated Finnish families ( rs34620296 and rs150472288 ) , and within Danish sister pairs ( rs482145 and rs139193421 ) . Notably , the rs34620296 minor allele variant was observed at a higher frequency in cases than controls in a very large 23andMe GWAS set . These variants were also identified via bioinformatics analyses as likely affecting either protein function or expression . Further functional evidence linked HSPA1L activity and decidualization . HSPA1L is a member of the Hsp70 superfamily and is near HSPA1A and HSPA1B within the major histocompatibility complex class III region on chromosome 6 . The HSPA1L protein ( also known as Hsp70-hom ) is ~90% identical to HSPA1A and HSPA1B , also known as Hsp70-1 and Hsp70-2 , respectively [23 , 24] . Heat shock proteins ( HSPs ) are highly conserved cellular defense mechanisms for cell survival and are present in all cell types in all organisms . Some HSPs are expressed constitutively , while others are stress-induced ( e . g . heat , hypoxia , oxidative stress , infection and inflammation ) [25 , 26] . Intracellular HSPs act as molecular chaperones and , together with co-chaperones , stabilize existing proteins against aggregation , mediate folding of newly translated proteins , and assist in protein translocation across intracellular membranes [25 , 27] . HSPs are categorized into families according to their approximate molecular weight; of which Hsp70 ( a group of proteins sized approximately 70 kDa ) is the best characterized . Potential involvement of stress-induced HSPA1A in adverse pregnancy outcomes , including preeclampsia and PTB , has previously been suggested [28 , 29] . Although , studies of the role of HSPA1L and HSPA1L in pregnancy are lacking , there is some evidence of involvement in adverse pregnancy outcomes such as preeclampsia [30] . The rare HSPA1L missense variants observed in our study , are in the nucleotide-binding domain ( NBD ) , except the rs482145 , which is in the substrate-binding domain ( SBD ) ( Fig 2 ) . ATP binds to the NBD , which is followed by the exchange from low-binding affinity ATP state to high-binding affinity ADP state [29 , 31 , 32] . We showed that the non-synonymous variant rs34620296 ( Ala268Thr ) generates an additional phosphorylation site near the nucleotide-binding site . It showed a modest change in the binding efficiency at this site , which could affect the interaction with ADP or HSPA1L stability itself , as suggested by our transfection studies . In agreement with our findings , a previous study of Caucasian patients with inflammatory bowel disease found that rare mutations in HSPA1L were significantly enriched in patients but absent in healthy controls [33] . Interestingly , one of the associated rare variants was Ala268Thr , and further in vitro biochemical assays of the recombinant HSPA1L showed reduced chaperone activity with this variant [33] . There is also evidence that possibly connects inflammatory bowel disease to adverse perinatal outcomes [34] . Additionally , a previous SPTB study in African Americans found a common nonsynonymous HSPA1L variant , rs2075800 , to associate with SPTB [35] . Furthermore , a meta-analysis of previously PTB associated genes linked HSPA1L and SPTB using Ingenuity Pathway Analysis [36] . Due to a very low incidence of the rare HSPA1L variants associating with SPTB in our study , the anticipated attributable risk in the population level is probably small . However , the identification of the damaging alleles may facilitate the identification of causative pathways . For instance , interaction between Hsp70 and Hsp90 chaperones as well as their co-chaperones is essential in the maturation and inactivation of nuclear hormone receptors ( e . g . glucocorticoid , androgen , estrogen and progesterone receptors ) [37 , 38] . In the absence of its ligand , glucocorticoid receptor ( GR ) is bound to a complex constituting of Hsp40 , Hsp70 and Hsp90 chaperones; this complex keeps the GR in a ligand-receptive conformation but remaining transcriptionally inactive until ligand binding [38] . As shown previously [33] , rare HSPA1L variants can cause partial loss of HSPA1L chaperone activity , and therefore , altered function or expression . Altered function of the chaperones can compromise the stability of the GR complex , leading to an accumulation of partially unfolded proteins that are prone for aggregation and degradation events [37] . Glucocorticoids , steroid hormones that mainly signal through the GR , have anti-inflammatory and immunosuppressive actions . Glucocorticoid signaling communicates with estrogen signaling pathways to tightly regulate the pro- and anti-inflammatory milieu in reproductive tissues [39] , and progesterone signaling , via nuclear GR , mediates anti-inflammatory and immunosuppressive effects in genital tract during pregnancy [40 , 41] . Sustaining a pregnancy is a complex interplay and balance between the innate and adaptive immune cells in the reproductive tissues and at the maternal-fetal interface . Imbalance between the inflammatory cells can cause a breakdown of maternal-fetal tolerance leading to activation of labor ( both term and preterm ) . An untimely stimulus ( e . g . stress , infection or inflammation ) together with impairments in the glucocorticoid receptor signaling pathway could impose an inadequate response against inflammation or stress . This can elicit a shift from an anti-inflammatory to pro-inflammatory microenvironment , causing a premature activation of labor initiating signals resulting in preterm birth [42 , 43] . Possible limitations of our study are that the Discovery and Replication populations were sequenced using different Next Generation Sequencing platforms , and primary quality control measures and variant calling methods were thus different . In addition , Next Generation Sequencing generates an enormous amount of data , which could lead to many sequencing artifacts that may be misidentified as variants . We attempted to minimize these artifacts by applying a variety of quality control filters and using a large internal control population to detect potential sequencing or annotation errors . We also compared the results of variant annotation and prioritizing filters from three different software tools to ensure reproducible results . Furthermore , reported variants were confirmed by Sanger sequencing . Another possible limitation was that our study did not include unrelated control samples . This limitation has been partly overcome with the use of additional large GWAS datasets including control samples . In conclusion , whole exome sequencing of families with recurrent occurrence of SPTB enables identification of rare alleles influencing the predisposition to SPTB . Among the individual genes , two minor alleles of HSPA1L had a strong association to SPTB in multiplex Finnish families and the association of a specific minor allele was confirmed in a large GWAS set . Furthermore , this variant was associated with altered modification and function of the protein . Overall , our data suggest the need for precise regulation of steroid signaling in mediating birth timing .
Written informed consent was obtained from all individuals participating in this study , and the study was approved by the Ethics committees of the participating centers: Oulu University Hospital ( 78/2003 , 73/2013 ) , University of Southern Denmark ( NVK#1302824 ) , and University of Iowa ( IRB#200608748 ) . Individuals in the large European American GWAS were research participants of 23andMe , Inc . , a personal genetics company . All 23andMe participants provided informed consent and participated in the research online , under a protocol approved by the external AAHRPP-accredited IRB , Ethical & Independent Review Services ( E&I Review ) . DNA samples of the 17 individuals from Discovery population were extracted from whole blood and saliva samples using standard methods [45] . Although , using DNA from both blood and saliva samples , there were no major difference in the overall sequencing metrics ( alignment metrics or total number of variants ) between the sample types . DNA samples were subjected to exon specific next generation sequencing performed at the Center for Pediatric Genomic Medicine , Children’s Mercy Hospital ( CMH; Kansas City , MO ) . Exome samples were prepared with the Illumina Nextera Rapid Capture Exome kit according to the manufacturer’s protocols as described previously [47] . Sequencing was performed on Illumina HiSeq 2500 instruments utilizing v4 chemistry with 2 x 125 nucleotide sequences . Sequence data were generated with Illumina RTA 1 . 18 . 64 . 0 and bcl2fastq-1 . 8 . 4 , and aligned against the reference human genome ( GRCh37 . p5 ) using bwa-mem [48] , and variant calls were made using the Genome Analysis Toolkit ( GATK ) [49] version 3 . 2–2 using previously described methods [50] . Duplicate reads were identified and flagged with the Picard MarkDuplicates tool . Realignment of reads around known indels was performed with the RealignerTargetCreator and IndelRealigner , and variants were called on individual samples using the HaplotypeCaller modules of the GATK . In addition , whole exome sequencing was performed on 192 affected individuals from 95 Danish families ( Replication set ) . Exome capture of the samples were carried out with the BGI Exon Kit following manufacturer’s protocols ( BGI , Shenzhen , China ) . DNA libraries were generated using combinatorial Probe Anchor Ligation ( cPAL ) technology , and 35 base paired end reads were generated from 500 bp genomic fragments . Whole exome sequencing was performed using the Complete Genomics platform ( BGI ) and using the manufacturer’s pipeline . Reads were aligned against the National Center for Biotechnology Information ( NCBI ) build 37 reference human genome . The variant call files ( VCF ) , containing the variant call results , generated by CMH and BGI were analyzed using Ingenuity Variant Analysis Software ( Qiagen , Germany ) and Golden Helix VarSeq Software v . 1 . 2 . 1 ( Bozeman , MT ) for both the Discovery and Replication population sets . Variants were filtered based on variant quality control measurements , frequency and predicted pathogenicity , as well as a dominant inheritance model . In the Ingenuity Variant Analysis , low quality variants ( read depth <15 and call quality <20 ) were removed . Furthermore , we only included rare variants ( i . e . MAF <1% in the 1000 Genomes Project , ExAC or in European American population in NHLBI ESP exomes ) and variants that would likely have functional effect ( i . e . variants that are predicted by SIFT or PolyPhen-2 as damaging or likely damaging , listed in Human Gene Mutation Database , or associated with gain or loss of function of a gene ) . In VarSeq , variants with read depth <15 and genotype quality score <20 were excluded . Only rare variants in Europeans ( MAF <1% or absent; 1kG Phase 3: Variant frequencies 5 , GHI Jan 2015 ) and missense or loss-of-function variants were included for analyses . Further filtering was applied to the data obtained from VarSeq . Variant quality was enhanced by applying range criteria ( 0 . 3–0 . 85 ) for alternative allele frequency ( i . e . ratio of alternate allele read depth / alternate allele read depth + reference allele read depth ) ; variants outside this range were excluded . Allele frequencies were searched from the Sequencing Initiative Suomi ( SISu ) database ( www . sisuproject . fi ) for variants originating from the Finnish mother data , and from the Exome Aggregation Consortium ( ExAC ) database ( http://exac . broadinstitute . org/ ) for Danish sister pair data . For these variants , a MAF cut-of value <1% in Finnish general population ( SISu ) or European “non-Finnish” ( ExAC ) population was used for Finnish or Danish mothers , respectively . The SISu database was used for Finnish mothers to exclude rare variants that are enriched in Finnish general population compared to the rest of the Europeans . For the Discovery samples , we also used allele frequency calculations derived from Center for Pediatric Genomic Medicine’s CMH Variant Warehouse database ( http://warehouse . cmh . edu ) including ~3900 individuals previously sequenced at the center [50] . Pathogenicity was categorized according to the American College of Medical Genetics [21] as 1; previously reported to be disease-causing , 2; expected to be pathogenic ( loss of initiation , premature stop codon , disruption of stop codon , whole-gene deletion , frame shifting indel , and disruption of splicing ) , and 3; unknown significance but potentially disease-causing ( nonsynonymous substitution , in-frame indel , disruption of polypyrimidine tract , overlap with 5' exonic , 5' flank , or 3' exonic splice contexts ) . Only variants that fit one of these criteria ( 1−3 ) were included for analyses . Rare and novel variants with relatively high frequency in this internal control population were also excluded as they were thought to be technical artifacts . In Ingenuity Variant Analysis , a dominant inheritance model ( including gain of function variants , and all heterozygous , compound heterozygous , haploinsufficient , hemizygous , and het-ambiguous variants ) was used to investigate predisposing variants that are inherited in the families . When analyzing affected mothers as a group , rare variants in genes that were common for a proportion of all cases were investigated . Whereas in family specific analyses , only variants that were shared by the affected individuals within each family were included . Ingenuity Variant Analysis provides a list of most significant pathways calculated specifically for each filtering output . P-values were calculated according to Fisher’s exact test assessing overlap enrichment of dataset-variant genes relative to known phenotype-implicated genes . Here , only pathways with p<0 . 01 were included for further analyses . To investigate rare variants in genes arising from the whole exome data in a larger population setting including controls , we used available sources of preterm birth GWAS data . GWAS data from a large cohort , identified among 23andMe’s research participants , included 43 , 568 mothers of general European ancestry [14] and meta-analysis data including 4 , 632 mothers from three independent Nordic ( Finnish , Danish , and Norwegian ) birth cohorts of European ancestry [51] . In addition , a set of GWAS data from a total of 608 mothers passing quality control measures was available . This set included mothers with spontaneous preterm deliveries and mothers with term deliveries originating exclusively from northern Finland . Genotyping was performed with Illumina Human CoreExome chip , followed by prephasing and imputation procedures with ShapeIT2 [52] and IMPUTE2 [53] . Association analysis was performed using SNPTEST v . 2 . 5 . 2 [54] . Since WES methodologies are associated with significant false positive rates , the presence of interesting variant findings from WES analyses was confirmed using Sanger sequencing . Samples were sequenced using capillary electrophoresis with ABI3500xL Genetic Analyzer ( Applied Biosystems , CA ) in Biocenter Oulu Sequencing Center , University of Oulu , Oulu , Finland . Details of the PCR primers and reaction conditions are available upon request . The possible functional effect of the rare HSPA1L variants ( rs34620296 and rs150472288 from Discovery analyses as well as rs482145 and rs139193421 from Replication analyses ) were investigated using in silico prediction tools such as SIFT , PolyPhen-2 , MutationTaster and MutationAssessor . These pathogenicity predictions were annotated using Varseq , whereas CADD scores to identify pathogenic and deleterious variants were obtained from Ingenuity . Variants with CADD score >20 are amongst top 1% of deleterious variants in human genome [55] . RegulomeDB ( http://www . regulomedb . org/ ) was used to investigate variant locations for e . g . chromatin state activity . In addition , we used HaploReg v4 . 1 ( http://archive . broadinstitute . org/mammals/haploreg/haploreg . php ) to assess whether variants are located within regions that show evidence for promoter or enhancer activity ( i . e . presence of histone modification marks H3K4me1 and H3K27ac that are associated with enhancer regions , or H3K4me3 and H3K9ca that are associated with promoter regions ) , as well as for DNase I hypersensitivity in human tissues and cell line samples . We further investigated the potential effects that missense variation rs34620296 ( Ala268Thr ) could have on protein sequence or structure . We used NetPhos 3 . 1 [56] to investigate possible changes in phosphorylation events in HSPA1L sequence . NetPhos 3 . 1 predicts serine , threonine or tyrosine phosphorylation sites in amino acid sequences of eukaryotic proteins . Evidence of being a phosphorylation site is given when the score is above the threshold ( 0 . 5 ) . To investigate the possible effects of missense variant in protein structure , the reference protein structure and the modified protein structure , including the missense variant , were compared . Original ( UniProtKB: P34931 ) and modified ( Ala268Thr ) amino acid sequences were submitted to SWISS-MODEL ( https://swissmodel . expasy . org/ ) for protein modeling . Resulting protein models were compared simultaneously using UCSF Chimera . Molecular graphics and analyses were performed with the Chimera-1 . 11 . 2 . package ( https://www . cgl . ucsf . edu/chimera/ ) . Chimera was developed by the Resource for Biocomputing , Visualization , and Informatics at the University of California , San Francisco ( supported by NIGMS P41-GM103311 ) [57] . | Preterm birth is the leading cause of infant mortality , and prematurity is further associated with serious morbidities in later life . Genetic and environmental risk factors play a role in the susceptibility to preterm birth . Despite numerous studies , the genetic basis for preterm birth remains poorly defined . We investigated the presence of rare , possibly risk associated nucleotide variants in mothers with spontaneous preterm births ( SPTB ) . The first set of mothers with family history of recurrent preterm births was of northern Finnish origin . An additional set of mothers ( sister pairs , both giving birth preterm ) of European origin was also studied . Whole exome sequencing identified multiple rare , likely damaging HSPA1L variants in several families affected by SPTB , and this gene was associated with the glucocorticoid receptor signaling pathway . Potential involvement of one of the HSPA1L variants in SPTB was further supported by large GWAS dataset . In addition , this variant alters protein post-translational modification potential , and thus may affect protein stability and its function as a chaperone . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"phosphorylation",
"genome-wide",
"association",
"studies",
"medicine",
"and",
"health",
"sciences",
"cellular",
"stress",
"responses",
"maternal",
"health",
"obstetrics",
"and",
"gynecology",
"engineering",
"and",
"technology",
"cell",
"processes",
"hormones",
"preterm",
"birth",
"women's",
"health",
"pregnancy",
"genome",
"analysis",
"industrial",
"engineering",
"quality",
"control",
"protein",
"structure",
"estrogens",
"pregnancy",
"complications",
"birth",
"heat",
"shock",
"response",
"computer",
"and",
"information",
"sciences",
"proteins",
"molecular",
"biology",
"biochemistry",
"cell",
"biology",
"post-translational",
"modification",
"genetics",
"software",
"engineering",
"biology",
"and",
"life",
"sciences",
"genomics",
"software",
"tools",
"computational",
"biology",
"macromolecular",
"structure",
"analysis",
"human",
"genetics"
] | 2018 | Whole exome sequencing reveals HSPA1L as a genetic risk factor for spontaneous preterm birth |
Chromosome ends are known hotspots of meiotic recombination and double-strand breaks . We monitored mitotic sister chromatid exchange ( SCE ) in telomeres and subtelomeres and found that 17% of all SCE occurs in the terminal 0 . 1% of the chromosome . Telomeres and subtelomeres are significantly enriched for SCEs , exhibiting rates of SCE per basepair that are at least 1 , 600 and 160 times greater , respectively , than elsewhere in the genome .
Chromosome ends participate in frequent recombination events . Human subtelomeric regions have undergone multiple interchromosomal exchanges during meiosis , giving rise to the highly duplicated structures proximal to telomere repeats [1] . Meiotic recombination maps of the human genome show an increase in recombination rate at the most distal markers , especially in males [2 , 3] . Other observations suggest that mitotic recombination might also be elevated at chromosome ends . In senescent cells , telomeres and subtelomeres are enriched for double-strand-break–binding proteins such as γ-H2AX [4] . Telomeric recombination is elevated in telomerase-negative cancer cells that follow the alternative lengthening of telomeres ( ALT ) pathway , generating chromosomes with highly variable telomere lengths [5–7] . Based on these somatic observations and the evolutionary dynamics of human subtelomeres , we hypothesized that subtelomeres might undergo high rates of mitotic sister chromatid exchange . Sister chromatid exchange ( SCE ) is a mechanism that resolves replication-dependent double-strand breaks and is thus an indicator of DNA damage and repair . Using a novel fluorescent method to detect SCEs anywhere between two chromosome ends ( chromosome orientation fluorescence in situ hybridization [CO-FISH] [8] ) , Cornforth and Eberle ( 2001 ) observed more SCEs than found using classical harlequin SCE techniques [9] . They attributed this excess to the most distal ≤7 Mb of the chromosomes . We have adapted CO-FISH to specifically measure SCEs in the telomeres , subtelomeres , and body of chromosomes ( Figure 1 ) and find that SCEs are highly concentrated within the most distal 100 kb .
We measured SCE in three terminal intervals of the chromosome , the distal 10 Mb ( probe X ) , the subtelomere ( the terminal ~110 kb , probe Y ) , and the most distal ~ 10 kb including the telomere ( probe Z ) . Three separate experiments were conducted by cohybridizing a probe specific for the telomere-repeat sequence , ( TTAGGG ) n , with one of the three internal probes ( X , Y , or Z ) ( Figure 2 ) . A change in the chromatid position of the internal probe relative to the telomere-repeat probe indicates an SCE in the interval between the signals of the two probes ( Figure 1 ) . For example , Probe Z lies in the most distal part of the subtelomere just proximal to the telomere-repeat array , so it monitors SCE events that occur in the <10-kb interval between its green signal and the orange signal produced by the telomere-repeat probe at the same chromosomal end . In the native state , these signals reside on opposite chromatids , due to the opposing 3′–5′ orientations of the two probes at this chromosomal end . If an SCE occurs between the target sequences such that the bulk of the telomere-probe signal is transferred to the other chromatid , the signals of both probe Z and the telomere probe will lie on the same chromatid . Probe Z also monitors SCE events that occur between it and the telomere-probe signal at the far end of the chromosome; we refer to this interval as the body of the chromosome . These events are recognized by the shift from the native configuration ( i in Figure 1 ) to the configuration diagrammed in ii , Figure 1 . SCEs that occur precisely within the sequences targeted by either probe will go undetected unless sufficient sequence is transferred to the opposite chromatid to be made visible by FISH . For example , since the telomeric arrays of human cells are relatively short ( <10 kb ) , many SCEs that occur within the telomere proper are missed because too little telomeric sequence is transferred ( or left behind ) to produce signals on both chromatids at the same chromosome end . We rarely detected such double telomere-probe signals in normal ( GM08729 ) cells ( 0 . 4% of labeled chromosome ends on average , Text S1 ) . These double signals could represent telomeric SCEs , but since they are relatively rare in normal cells and might also arise artifactually due to incomplete degradation of the newly synthesized DNA strand , we chose to omit this small group of chromosomes from our estimates of terminal SCEs to be conservative . Our telomeric and subtelomeric probes each hybridize to multiple chromosomes at approximately the same distance from the ends , whereas the probe assaying the distal 10 Mb hybridizes only to the long arm of Chromosome 15 ( Figure 2 and Table S1 ) . We counted the number of SCE events in the body of the chromosome and in each of the three terminal intervals in a lymphoblastoid cell line . Pooling SCE data from all chromosomes with the requisite signals ( internal probe + telomere probes on both p and q ends ) , we find that the frequency of chromosomes with a single terminal SCE in the most telomeric interval ( Z ) , subtelomere ( Y ) , and distal 10 Mb ( X ) is 0 . 6% , 1 . 3% , and 1 . 4% , respectively . Thus , the majority of SCEs in the last 10 Mb of a chromosome are confined to the distal 110 kb . Of all the SCE events we observed , 17% ( 50/291 ) occurred in the last 110 kb and 16% ( 20/124 ) occurred in the last 10 kb of the chromosome , far greater proportions than would be expected if SCEs were distributed uniformly along the chromosome . The frequency of observed SCEs in the terminal intervals implies rates of 2 , 100 × 10−9 SCE/ ( bp × cell generation ) , 380 × 10−9 SCE/ ( bp × cell generation ) , and 3 . 3 × 10−9 SCE/ ( bp × cell generation ) in the terminal 10 kb , 110 kb and 10 Mb , respectively ( Text S1; Table S2 ) . Recent studies in budding yeast have shown that subtelomeric sequences direct nuclear organization [10] , and that alterations in nuclear organization can affect the frequency of double-strand-break repair at subtelomeres [11] . Thus , we speculated that different human subtelomeres might have variable rates of SCE . However , our analysis of 16 different subtelomeres did not detect a chromosome-specific difference in terminal SCE rates ( Table S1 ) . The chromosome ends we sampled have undergone multiple interchromosomal exchanges during evolution giving rise to a patchwork of subtelomeric duplications . Therefore , we also measured SCEs in 7q , a chromosome end relatively devoid of duplications [1] . The rate of SCE in the last 50 kb of 7q is 420 × 10−9 SCE/ ( bp × cell generation ) , not significantly different from the rate of SCE on chromosome ends with more extensive interchromosomally duplicated content . The frequency of chromosomes with an SCE in the body of the chromosome outside of the terminal interval varies among the four experiments due to the differences in the DNA contents of the analyzed chromosomes ( Figure 2 and Table S1 ) . On average , the rate of SCE in the body of the chromosome is 1 . 3 × 10−9 SCE/ ( bp × cell generation ) ( Text S1 ) . Note that this rate for the body of the chromosome is inflated because it encompasses events occurring in the terminal interval of one chromosome end , since our internal probes usually only mark a single end . As in previous studies [9 , 12] , our measurements of SCEs in 15 different chromosomes show that the frequency of total SCEs increases linearly with chromosome size ( Figure S1 ) . Interestingly , in each of the four experiments , we found a significant number of chromosomes with two SCEs—one in the terminal interval and one in the remainder of the chromosome ( Figure 2 ) —over what would be expected if SCEs were independent events ( p < 0 . 0001; Text S1 , 2 × 2 contingency table ) . We also found an excess of cells with more SCEs than expected from the overall average SCE rates ( unpublished data ) . Thus , SCEs appear to cluster . We also measured terminal SCE in a SCE-sensitized background . Cells from patients with a mutation in the gene encoding the Bloom syndrome protein , BLM , have significantly more SCEs along the length of the chromosomes than do normal cells [13] . BLM physically interacts with the telomere-binding protein TRF2 in HeLa cells and unwinds telomere duplexes in vitro [14] . Thus , we considered the possibility that BLM could play a role in the hyper-recombination occurring at chromosome ends . We measured SCE in the two most terminal intervals of chromosomes in cells from a patient with Bloom syndrome ( Figure 2 ) . The very high rates of SCE in the body of chromosomes in these cells saturates the CO-FISH assay; chromosomes with an even number of SCEs in the body of the chromosome will have the same probe configuration as the native state , whereas chromosomes with any odd number of SCEs in the body will be indistinguishable from chromosomes with one SCE in the body . In the Bloom cells , the frequencies of the two configurations are equivalent , due to the multiplicity of events they represent . In contrast , the frequency of chromosomes showing terminal SCEs in Bloom cells was very similar to the frequency observed in normal cells ( Figure 2 ) . Thus , cells lacking functional BLM do not have a proportional increase in terminal SCE . Increased rates of exchange within the telomere-repeat array have been found in cancer cells following the ALT pathway [7 , 15 , 16] . ALT cells show an abundance of double telomere signals in CO-FISH assays , suggesting an increase in telomeric SCE and/or interchromosomal exchanges between telomere-repeat arrays on different chromosomes . To determine if subtelomeres were also involved in the terminal exchanges in ALT cells , we applied our subtelomeric CO-FISH assay to an ALT cell line ( WI38 VA13/2RA ) . As in previous studies [7] , we found an increase in double telomeric-probe signals in ALT chromosomes ( 18% versus 0 . 4% in normal ) , but we did not find double internal probe signals ( probe Y ) indicative of subtelomeric interchromosomal exchanges . The lack of subtelomeric interchromosomal exchanges in ALT cells is consistent with experiments using a non-native subtelomere [17] . Also , rates of SCE in the terminal 110 kb , exclusive of the telomere array itself , were not elevated in ALT cells relative to normal cells ( unpublished data ) , although these SCEs could be detected only in the chromosomes without double telomere signals at the end carrying the subtelomeric probe . These data suggest that the ALT pathway uses telomeres and not subtelomeres as a substrate for chromosome-end maintenance .
Chromosome ends are an extremely dynamic part of the genome . CO-FISH allows us to measure SCE anywhere along the entire length of the chromosome , from telomere to telomere . We find that SCE is highly concentrated at the very ends of chromosomes , as over 15% of all mitotic SCEs occur in a region roughly 0 . 1% of the chromosome's length . The most distal ~10-kb regions show the greatest density of SCEs , at 2 , 100 × 10−9 SCE/ ( bp × cell generation ) . When we subtract these most telomeric SCEs from the number of SCEs in the last 110 kb of the chromosome , we find a rate of 210 × 10−9 SCE/ ( bp × cell generation ) in the subtelomere alone . Both regions exhibit rates of SCE much greater than the rate elsewhere in the genome , approximately 1 . 3 × 10−9 SCE/ ( bp × cell generation ) . Our comparison of SCE rates is conservative , as the latter rate calculation does not correct for the excess of SCEs at the unanalyzed chromosome end . We do not find a significant difference in the frequency of SCEs at different chromosome ends , suggesting that terminal location alone may be sufficient for increased SCE . The 160-fold and at least 1 , 600-fold enrichment of SCE in subtelomeres and telomeres , respectively , suggests that chromosome ends are subject to more double-strand breaks during replication and/or that they are more likely to be repaired by SCE than more internal regions of chromosomes . These data indicate that human subtelomeres are hotbeds of DNA repair and exchange during mitosis and complement earlier findings of high rates of recent meiotic exchange at chromosome ends [1 , 18] . While most exchanges between sister chromatids probably leave their DNA sequences unaltered , SCE is known to mediate somatic changes in the length of D4Z4-repeat arrays in the 4q subtelomere; severe shortening of this array causes facio-scapulo-humeral dystrophy [19 , 20] . Thus , subtelomeric SCE is a common event that can have pathologic consequences .
Human lymphoblastoid cell lines GM08729 and GM16375 were obtained from the Coriell Institute for Medical Research ( http://www . coriell . org ) and grown in RPMI media supplemented with 10% fetal bovine serum , penicillin/streptomycin , and L-glutamine . SV40-transformed human fibroblast cell line WI38 VA13/2RA was obtained from the American Type Culture Collection ( http://www . atcc . org ) and grown in alpha-MEM supplemented with 10% fetal bovine serum , penicillin/streptomycin , and L-glutamine . Cells were harvested and prepared for CO-FISH as described by Cornforth and Eberle ( 2001 ) . Figure 1 diagrams the CO-FISH procedure . Briefly , cells were synchronized at the G1/S boundary by serum starvation , released into S phase , and treated with 30 μM BrdU , allowing the newly replicated DNA strands to incorporate BrdU . Mitotic cells were harvested and dropped on slides as described [21] . After 1–7 d of storage at room temperature , slides were treated with 0 . 5 mg/ml RNase A for 10 min at 37 °C , followed by rinsing in PBS . Next , slides were treated with 0 . 5 μg/ml Hoechst 33258 ( Sigma ) for 15 min and then exposed to 365-nm UV light for an additional 30 min . We used a TL-33E transilluminator ( UVP , Incorporated; http://www . uvp . com ) , which operates at an intensity of 91 , 000 Joules/m2 . Slides were washed in PBS and then chromosomes were digested with 100 μl of 3 U/μl Exo III ( Fermentas , http://www . fermentas . com ) for 5 min at room temperature . UV exposure followed by exonuclease digestion degrades BrdU-incorporated DNA strands , generating single-stranded chromatids . Finally , slides were ethanol dehydrated at room temperature by successively incubating them in 70% , 80% , 95% , and 100% ethanol for 2 min at each concentration , and allowed to air dry . Slides were denatured and hybridized to CO-FISH probes as described below . Single-stranded probes were constructed from four different genomic locations ( X , Y , Z , and 7q ) . Each CO-FISH probe is a pool of several single-stranded plasmids with inserts cloned from a particular region of the genome . Probe X is composed of eight plasmids , cloned from bacterial artificial chromosome ( BAC ) RP11-24J19 , which span approximately 50 kb of sequence about 10 Mb from the end of the long arm of Chromosome 15 ( March 2006 University of California Santa Cruz ( UCSC ) browser [http://www . genome . ucsc . edu] coordinates chr15:90227457–90279975 ) . It is not duplicated on other chromosomes . Probe Y is composed of four plasmids , cloned from P1-derived artificial chromosome RP5-855D21 , which span approximately 20 kb of duplicated sequence located about 110 kb from the end of the short arm of Chromosome 8 ( March 2006 UCSC browser coordinates chr8:102784–121678 ) . Probe Z is located at the most distal region of many chromosomes , just proximal of telomere repeats [1] . The five plasmids in probe Z were cloned from BAC RP11-395L14 and span approximately 25 kb of sequence ( March 2006 UCSC browser coordinates chr2:114050604–114075702 ) . This BAC originates from the ancestral site of the telomere-telomere fusion on Chromosome 2 , which contains sequences paralogous to subtelomeric sequences . The sequences of probes Y and Z are known from FISH and genome-sequence analyses to be duplicated on at least nine and thirteen chromosome ends , respectively , although the number and location varies among individuals ( [1] , and unpublished data ) . The eight plasmids comprising the 7q probe lie about 50 kb from the terminus of the long arm of Chromosome 7 . Cloned from BAC RP11-1112M14 , the 7q probe set spans approximately 78 kb ( March 2006 UCSC browser coordinates chr7:158701668–158780077 ) . Each plasmid insert was TA-cloned from a PCR product in F′ E . coli ( TOPO TA cloning kit; Invitrogen , http://www . invitrogen . com ) . Single-stranded DNA was generated by infecting F′ cultures with M13K07 helper phage according to the manufacturer's instructions ( New England Biolabs , http://www . neb . com ) . 1 μg of single-stranded DNA was digested with 5 U of DNase I for 1 min at room temperature ( New England Biolabs ) and then end-labeled with biotin-16-ddUTP according to manufacturer's instructions ( Roche Applied Science , http://www . roche . com ) . Slides were treated to generate single-stranded chromatids , as described in [9] and outlined above . CO-FISH probes were denatured , hybridized to slides , and detected with avidin-FITC as described [21] . Following biotin detection , slides were hybridized with 10 μl of a 0 . 5 μg/ml telomere peptide nucleic acid probe ( Cy3-[C3TA2]3 ) and washed as described [22] . Slides were covered with 20 μl of antifade solution with DAPI ( Vectashield; Vector Laboratories , http://www . vectorlabs . com ) , allowing for identification of the chromosomes from their banding patterns . Signals were examined using a Zeiss fluorescence microscope ( http://www . zeiss . com ) equipped with a cooled CCD camera , Chroma Technology spectral filters ( http://www . chroma . com ) , and MacProbe image analysis software ( Applied Imaging Corporation , http://www . aicorp . com ) .
The National Center for Biotechnology Information ( NCBI ) ( http://www . ncbi . nlm . nih . gov ) accession numbers for the genomic information discussed in this paper are BAC RP11-24J19 , AC104236; PAC RP5-855D21 , AC004908; BAC RP11-395L14 , AL078621; and BAC RP11-1112M14 , AQ747375 and AQ747941 . | The ends of chromosomes are evolutionarily dynamic and structurally unusual parts of the human genome . Arrays of telomeric repeats cap each end and protect chromosomes from degradation and end-to-end fusions . Just inside the telomeres are patchworks of larger DNA segments duplicated on different chromosome ends . These subtelomeric duplications reflect the high frequency with which DNA breaks in these regions were healed by interchromosomal repair processes during recent primate evolution . In this study , we asked if chromosomal ends are also unusually susceptible to replication-induced DNA breaks and repair during mitotic division of somatic cells . We employed a specialized fluorescent technique to measure sister chromatid exchange ( SCE ) specifically in telomeres and subtelomeres , as such events would be missed by standard SCE-detection methods . We find extraordinarily high rates of SCE in these terminal regions: over 15% of observed SCEs occur in just 0 . 1% of the genome . Thus , chromosome ends are hotspots of DNA breaks and recombinational repair in mitosis , as shown previously in meiosis . The enrichment of DNA breaks at chromosome ends contributes to normal variation , chromosome evolution , and chromosome rearrangements leading to disease . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods",
"Supporting",
"Information"
] | [
"primates",
"oncology",
"homo",
"(human)",
"genetics",
"and",
"genomics"
] | 2007 | Elevated Rates of Sister Chromatid Exchange at Chromosome Ends |
Apolipoprotein L-I ( apoL1 ) is a human-specific serum protein that kills Trypanosoma brucei through ionic pore formation in endosomal membranes of the parasite . The T . brucei subspecies rhodesiense and gambiense resist this lytic activity and can infect humans , causing sleeping sickness . In the case of T . b . rhodesiense , resistance to lysis involves interaction of the Serum Resistance-Associated ( SRA ) protein with the C-terminal helix of apoL1 . We undertook a mutational and deletional analysis of the C-terminal helix of apoL1 to investigate the linkage between interaction with SRA and lytic potential for different T . brucei subspecies . We confirm that the C-terminal helix is the SRA-interacting domain . Although in E . coli this domain was dispensable for ionic pore-forming activity , its interaction with SRA resulted in inhibition of this activity . Different mutations affecting the C-terminal helix reduced the interaction of apoL1 with SRA . However , mutants in the L370-L392 leucine zipper also lost in vitro trypanolytic activity . Truncating and/or mutating the C-terminal sequence of human apoL1 like that of apoL1-like sequences of Papio anubis resulted in both loss of interaction with SRA and acquired ability to efficiently kill human serum-resistant T . b . rhodesiense parasites , in vitro as well as in transgenic mice . These findings demonstrate that SRA interaction with the C-terminal helix of apoL1 inhibits its pore-forming activity and determines resistance of T . b . rhodesiense to human serum . In addition , they provide a possible explanation for the ability of Papio serum to kill T . b . rhodesiense , and offer a perspective to generate transgenic cattle resistant to both T . b . brucei and T . b . rhodesiense .
Normal human serum ( NHS ) is able to kill T . b . brucei , but not T . b . rhodesiense and T . b . gambiense . The lytic factor was identified as being apoL1 [1] , [2] . This protein is associated with HDL particles that are efficiently taken up by the parasite through specific binding to a haptoglobin-hemoglobin surface receptor , due to the simultaneous presence of haptoglobin-related protein ( Hpr ) acting as a ligand in these particles [3] . Trypanosome lysis results from anionic pore formation by apoL1 in the lysosomal membrane of the parasite [4] . Resistance to lysis has only been studied in case of T . b . rhodesiense , where it was shown to depend on a parasite protein termed SRA [5] . As synthesis of SRA only occurs after transcriptional activation of a given Variant Specific Glycoprotein ( VSG ) gene expression site from a repertoire of 10–20 candidates , T . b . rhodesiense clones can be either sensitive or resistant to NHS depending on which expression site is active [2] , [5] . The mechanism by which SRA inhibits the activity of apoL1 is unclear . Direct coil-coiling interaction between the C-terminal α-helix of apoL1 and the N-terminal α-helix of SRA was demonstrated in vitro , but in vivo only evidence for tight co-localization between the two proteins was obtained [1] . Total deletion of the C-terminal helix appeared to confer toxic activity to recombinant apoL1 on T . b . rhodesiense , suggesting that in vivo SRA neutralizes apoL1 through interaction with its C-terminal domain [1] . However , the trypanolytic effect of the deleted apoL1 was weak and incomplete [1] . Moreover , data obtained following transgenic expression of a similarly truncated apoL1 in mice suggested that its trypanolytic potential was lost in vivo [8] . Therefore , additional work was required to evaluate if the interaction observed in vitro was relevant for the in vivo mechanism of T . b . rhodesiense resistance to human serum . While human serum is unable to kill T . b . rhodesiense , the serum of some African primates like Papio sp . was equally active on both T . b . rhodesiense and T . b . brucei [6] , [7] . The phenotype of trypanolysis by Papio serum strikingly resembled that induced by human serum , as it was also dependent on HDL particles and was similarly inhibited by competing amounts of haptoglobin [7] . Therefore , it could be envisaged that in Papio , an apoL1-like protein unable to interact with SRA would be responsible for the trypanolytic activity of these primates . We used this working hypothesis as a base to construct human apoL1 mutants . We analyzed the effects of various deletions and mutations in the C-terminal domain of apoL1 on the trypanolytic potential of this protein against T . b . brucei and T . b . rhodesiense . The results confirmed the interaction model presented in [1] , as well as the role of this interaction in resistance to apoL1 in T . b . rhodesiense . In accordance with these findings , Papio-like apoL1 mutants able to efficiently kill both T . b . brucei and T . b . rhodesiense were identified .
As illustrated in Fig . 1A , apoL1 contains three functional domains responsible for its ionic pore-forming capacity , addressing to biological membranes and interaction with SRA , from N- to C-terminus respectively [4] . In E . coli , the pore-forming activity of apoL1 can be measured by the reduction of cell viability following IPTG-induced expression of the apoL1 gene from plasmid constructs [4] . In this system apoL1 expression closely mimicked the toxicity of the isolated pore-forming domain of bacterial colicin A fused to the signal peptide pelB , including depolarization of the plasma membrane [4] . As a control , deletion of the essential helix 9 of the pore-forming domain completely prevented this activity ( Fig . 1B ) . Although the pore-forming domain of apoL1 appeared to have access to the inner membrane of E . coli , it did not appear to be secreted as neither apoL1-expressing cells nor culture medium from these cells killed wild type E . coli when mixed with these cells ( data not shown ) . The toxicity exhibited by the apoL1 pore-forming domain was clearly dependent on acidic pH , since it occurred when E . coli was grown in LB medium at pH 6 , but not at pH 7 . 5 ( Fig . 1B ) . In addition to its effect on E . coli , the pore-forming activity of apoL1 is also responsible for the ability of this protein to kill trypanosomes , although the trafficking and intracellular targeting of the toxin are obviously very different between the two systems [4] . Addition of recombinant apoL1 to T . b . brucei resulted in efficient killing of the parasite , as measured after overnight incubation ( Fig . 1C ) . The trypanolytic activity of apoL1 is known to be inhibited by the T . b . rhodesiense protein SRA [1] , [5] . In order to analyze the mechanism by which SRA neutralizes apoL1 pore-forming activity , we constructed a bicistronic plasmid co-expressing the two proteins in E . coli , under the dependence of two inducible T7 transcription promoters ( pCDF-DUET; Fig . 2A ) . In this system apoL1 was tagged at the C-terminus with V5 and 6-His peptides , whereas SRA was provided with a C-term S tag . Following induction of expression , apoL1 was purified by binding to nickel beads , and after elution it was revealed using anti-apoL1 antibodies . As shown in Fig . 2A , apoL1 was clearly detected in the nickel-bound fraction . When co-expressed with apoL1 , SRA was also present in this fraction as revealed by anti-S tag antibodies . However , in the absence of apoL1 , SRA was no longer found in the bound fraction ( Fig . 2A ) . These results indicated that in E . coli , co-expression of apoL1 and SRA results in the formation of a complex associating these two proteins . Consistent with the lysosomal localization of this complex in T . brucei [1] , the formation of the apoL1-SRA complex appeared to be favoured at acidic pH as its amount was strongly reduced upon E . coli lysis at different pH values between pH 6 and pH 7 ( Fig . 2A ) . Significantly , the pore-forming activity of apoL1 was strongly inhibited upon co-expression of SRA ( Fig . 2B ) . Therefore , in the E . coli expression system apoL1 appeared to be inactivated following direct interaction with SRA . We evaluated the level of interaction of SRA with different mutants of apoL1 by measuring the relative amounts of either protein bound to nickel beads . More precise measurement of this interaction using plasmon resonance was impossible , due to the propensity of both proteins to stick to various matrices ( data not shown ) . We generated apoL1 variants deleted of either one of the three functional domains ( del 1 , del 2 and del 3 from N- to C-term , see Fig . 1A ) . The presence in each case of an N-terminal bacterial signal peptide ( pelB ) allowed the determination of the pore-forming potential of the variants in E . coli irrespective of the deletion of the membrane-addressing domain [4] . As expected [4] , only deletion of the N-terminal domain ( del 1 ) resulted in the loss of the pore-forming activity ( Fig . 2B ) . Co-expression of SRA with del 2 resulted in strong inhibition of the killing activity like it was observed with wild-type apoL1 , whereas SRA only mildly affected the activity of del 3 ( Fig . 2B ) . In accordance with these data , SRA was found to bind to del 1 and del 2 , but not to del 3 ( Fig . 2C ) . Altogether these data confirmed that SRA interacts with the apoL1 C-terminal domain , and revealed that this interaction inhibits apoL1 activity independently of the presence of the membrane-addressing domain . Recombinant forms of the three apoL1 variants del 1 , del 2 and del 3 were produced in E . coli and tested for their trypanolytic potential . As expected , del 1 and del 2 were inactive ( Fig . 1C ) . Surprisingly , despite the full conservation of its intrinsic pore-forming activity ( Fig . 2B ) , del 3 was also inactive ( Fig . 1C ) . These findings confirmed a recent report [8] , but contrasted with our previous work where recombinant del 3 expressed in Chinese hamster ovary cells was found to kill NHS-sensitive trypanosomes [1] . The C-terminal domain of apoL1 , as well as that of the other apoL family members , is characterized by the presence of a leucine zipper ( Fig . 3A ) . Different mutants affecting this zipper were generated ( Hel 1: L378/382/385S; Hel 2: L378/382/385A; Hel 3: L378P ) . As shown in Fig . 3A , in each case hydrophobic cluster analysis predicted a strong reduction of hydrophobic interaction potential [9] , [10] . This was confirmed by the prediction of interaction energy using measurement of the mean force potential according to the apoL1-SRA interaction model presented in [1] , [11] . This energy was decreased in the mutant sequences ( Table 1 ) . As expected from this prediction , SRA interaction was reduced in the three mutants ( Fig . 3B , C ) . However , like in the case of the del 3 variant , these mutants also largely lost their trypanolytic potential although they conserved full pore-forming activity in E . coli ( Fig . 3D , E ) . Therefore , conservation of the C-terminal helix appeared to be necessary for the trypanolytic activity of recombinant apoL1 . As shown in Fig . 4A , the serum of Papio cynocephalus was equally able to lyze NHS-resistant and -sensitive T . b . rhodesiense clones , although it did not affect T . b . gambiense ( data not shown ) . As NHS or recombinant apoL1 cannot lyze T . b . rhodesiense [1] , the Papio serum must contain a trypanolytic factor different from apoL1 . However , several observations suggest that this factor actually resembles apoL1 . First , like apoL1 the Papio trypanolytic factor is bound to HDL particles , as Papio trypanolytic activity was present in a serum fraction binding to anti-apoA1 , the main constituent of HDLs [7] . Second , like apoL1 [4] , it was sensitive to inhibition by the anionic channel inhibitor DIDS ( Fig . 4A ) . Third , like apoL1 it was inhibited by competition with an excess of haptoglobin , suggesting its association with Hpr [7] ( Fig . 4B ) . Fourth , the cellular phenotype of trypanosome lysis by Papio serum , involving considerable swelling of the lysosome , was indistinguishable from that induced by NHS ( Fig . 4C ) . Finally , proteins isolated from Papio serum through binding to anti-apoA1 could be detected by anti-apoL1 antibodies ( Fig . 4D ) . However , as expected considering the lack of resistance of T . b . rhodesiense to Papio serum , the apoL1-like proteins of Papio were unable to bind to SRA ( Fig . 4D ) . In order to evaluate the possible presence of apoL1-like proteins in Papio sp . , we examined the currently available Papio anubis genomic sequence information . As shown in Fig . 5 , two apoL1-like sequences were identified , one of which was shorter due to C-terminal truncation . However , from the actual state of information the two genes appeared to be interrupted by frameshift mutations , and we confirmed this frameshift in reverse transcripts of these genes from RNA of either blood cells of P . cynocephalus or endometrium of P . anubis , using various combinations between 5 forward and 5 reverse primers from different regions ( Fig . 5; data not shown ) . So far we were unable to clone either gene with full coding capacity . We hypothesized that converting the C-terminal sequence of human apoL1 into those found in the two apoL1-like sequences of Papio anubis could impede the interaction of this protein with SRA and confer the capacity to kill T . b . rhodesiense . This involved the replacement of the 386–389 sequence NNNY into TKIQ ( TKIQ mutant ) , as well as the same replacement together with the removal of the 9 C-terminal amino acids ( delTKIQ mutant ) ( Fig . 6A ) . In addition to the latter mutant , we also generated similar mutants where the four C-terminal amino acids of the truncated version were intermediate between Papio and human apoL1s ( NKIQ , NNIQ , NNNQ , NNNY ) ( Fig . 6A ) . These changes did not strongly affect the hydrophobic cluster pattern of the C-terminal helix ( Fig . 6A ) , but they significantly reduced the predicted energy of interaction of apoL1 with SRA according to the model presented in [1] ( Table 1 ) . Accordingly , these apoL1 variants lost their capacity to interact with SRA ( Fig . 6B , C ) . All mutants conserved their pore-forming activity in E . coli ( Fig . 6D ) . As shown in Fig . 6E , the different Papio-like apoL1 variants efficiently killed both T . b . brucei and NHS-resistant clones of T . b . rhodesiense . However , they were unable to kill T . b . gambiense . We evaluated if the Papio-like apoL1 variants could exhibit trypanolytic activity in mice as they did in vitro . As shown in Fig . 7A , expression of apoL1 can optimally be detected in mice one day after hydrodynamic injection of 10 µg of pCDNA3 plasmid encoding the protein . Similarly , apoL1 variants could be detected one day post-injection of DNA , although the apoL1 mutants appeared to be less expressed than WT apoL1 ( Fig . 7B ) . Intraperitoneal inoculation of 106 trypanosomes from different T . brucei subspecies was performed at that day post-DNA injection . Infection by NHS-sensitive T . b . rhodesiense ETat 1 . 2S parasites was inhibited following expression of either WT or delTKIQ apoL1 , as determined by the measurement of the parasite number at the peak of parasitaemia ( Fig . 7C ) . As expected , transgenic expression of WT apoL1 did not confer protection against the NHS-resistant T . b . rhodesiense clone ETat 1 . 2R ( Fig . 7C ) . However , mice expressing the delTKIQ apoL1 variant could resist both trypanosome lines ( Fig . 7C ) . Although both WT and delTKIQ apoL1 almost completely erased the first peak of infection by ETat 1 . 2S parasites , infection reappeared afterwards and in both cases all mice died between days 12 and 15 post-inoculation ( data not shown ) . This infection pattern was also observed for delTKIQ-expressing mice infected with ETat 1 . 2R . These observations contrast with the complete trypanolysis observed in vitro with delTKIQ apoL1 , and likely result from the rapid reduction of apoL1 concentration in the transgenic animals ( Fig . 7A ) . The delTKIQ-expressing mice were also able to kill T . congolense ( Fig . 7C ) . Therefore , it can be predicted that transgenic cattle constitutively expressing this variant would resist infection by T . b . brucei , T . b . rhodesiense and T . congolense . In contrast , neither WT nor mutant apoL1 conferred protection against T . b . gambiense ( Fig . 7C ) .
Our data demonstrate that in E . coli , SRA inhibits the pore-forming activity of apoL1 through direct interaction with the C-terminal helix of this protein , and show that apoL1 variants unable to bind SRA through deletion or mutations of this helix could overcome this inhibition . It is particularly interesting to note that the apoL1 variant lacking the original membrane-addressing domain ( but containing a bacterial signal peptide to compensate this activity ) was still fully inhibited following co-expression of SRA in E . coli . This result indicates that the membrane-addressing domain is dispensable for the control of the pore-forming domain by SRA , suggesting that this control does not operate through refolding of the protein . A possible explanation would be that interaction with SRA prevents membrane targeting of apoL1 . Similarly to the situation in E . coli , apoL1 variants unable to bind SRA were found to kill NHS-resistant trypanosomes , confirming that the conclusions drawn in E . coli regarding the neutralization of the protein by SRA were also valid for trypanosomes . In particular , the fact that in E . coli SRA can directly inhibit the activity of apoL1 suggests that in trypanosomes , its effect on apoL1 would not necessarily operate through reorientation of apoL1 trafficking in the cell . However , a clear difference was observed between E . coli and trypanosomes . In the latter case , conservation of most of the C-terminal helix was required to keep the lytic potential intact , even for trypanosomes devoid of SRA . Altogether these conclusions essentially confirmed our earlier proposal [1] , although our message concerning the truncated apoL1 variant ending at S342 had to be corrected . In [1] , we showed data suggesting that this variant , equivalent to the del 3 apoL1 version studied in this work , was able to kill both T . b . brucei and T . b . rhodesiense . In contrast , in a recent report analyzing the trypanolytic potential of apoL1 expressed transiently in mice following hydrodynamic plasmid transfection , the del 3 variant was found to be no longer trypanolytic [8] . This controversy could be resolved in this study , where we confirmed the last observation . Thus , it would appear that in contrast to what occurs in E . coli , in trypanosomes the C-terminal domain of apoL1 is not completely dispensable for the killing activity of the protein . Given the obvious differences between the two biological systems this observation could be explained in many ways . In the E . coli system the membrane targeting is direct and entirely intracellular , and like experimentally observed with bacterial colicins the isolated pore-forming domain of apoL1 can exhibit full toxic activity provided a membrane-addressing peptide is present [4] . In trypanosomes the toxin must be taken up through endocytosis , which requires a more complex trafficking pathway , and the target is the lysosomal membrane instead of the plasma membrane . Among other possibilities the C-terminal helix could be required for proper trafficking and/or targeting to the lysosomal membrane , or its absence could result in improper folding of the protein . Regarding the possible influence of the purification procedure on the protein conformation , it should be noted that after reconstitution with lipid particles , recombinant apoL1 was found to kill trypanosomes with efficiency close to that of native apoL1 in normal human serum , suggesting that the full-size recombinant protein is not significantly denatured during purification [4] . Interestingly , it was through inspiration driven by sequence analysis of Papio apoL1-like genes that we were able to generate apoL1 variants unable to bind SRA but still able to efficiently kill trypanosomes . As expected , these variants killed NHS-resistant T . b . rhodesiense clones as well as NHS-sensitive T . b . rhodesiense clones or T . b . brucei , like occurs with Papio serum . This finding strongly suggests that in Papio serum similar apoL1 variants are responsible for the trypanolytic activity . However , we were unable to identify such proteins , as frameshift mutations apparently prevented the possible candidates to be functional . This result was repeatedly observed following RT-PCR amplification of transcripts from either blood cells or endometrium tissue , using various combinations of 5 forward and 5 reverse primers specific to the apoL1-like genes from the Papio anubis sequence database . Given our failure to detect any functional Papio apoL1-like sequence , we can only speculate about the nature of the trypanolytic factor in these organisms . Among other possibilities , apoL4 could replace apoL1 given the presence of a sequence encoding a putative signal peptide in this gene , as also occurs in humans [12] . However , it should be noted that this apoL4 variant was never detected in human serum so far . Thus , despite the identity between the trypanolytic phenotypes exhibited between Papio serum and the apoL1 mutant generated in this work , we cannot exclude that this mutant is without relevance concerning the genuine trypanolytic activity of Papio sp . The existence of baboon apoL1 variants able to resist SRA would be consistent with the recent proposal that variations of apoL sequences are frequent at sites of interaction with pathogen proteins [13] , although the apoL1 mutations/deletion described here were not described in this report . As is the case with Papio serum [14] , the Papio-like human apoL1 was unable to kill T . b . gambiense . This observation is in keeping with the fact that in this subspecies the mechanism of resistance to apoL1 must be different from that of T . b . rhodesiense , as SRA is absent from T . b . gambiense [15] . Therefore , it appears that resistance to NHS in T . b . gambiense is independent from the C-terminal domain of apoL1 . The results presented in this work are promising in terms of generating transgenic cattle able to resist infection by African trypanosomes in Eastern Africa . Indeed , we show that mice transiently expressing Papio-like human apoL1 variants resist not only T . b . brucei , but also NHS-resistant clones of T . b . rhodesiense and the cattle pathogen T . congolense . Moreover , given the sensitivity of T . evansi to apoL1 [16] this transgenic cattle should resist T . evansi as well , and similar prediction could be proposed for T . vivax . Finally , in view of these results it can be envisaged that understanding the mechanism of resistance of T . b . gambiense to NHS would allow us to generate mutant versions of apoL1 also able to kill this parasite .
This research was approved by the ethics committee of the Institute for Molecular Biology and Medicine ( IBMM ) . All mice were housed in our pathogen-free facility and the experiments were performed in compliance with the relevant laws and institutional guidelines ( license LA2500482 ) . The apoL1 coding sequence was amplified by PCR from the pcDNA3 . 1-apoL1-V5-His6 construct [1] with primers creating a 5′ NcoI site and ATG codon upstream from the E28 codon and a 3′ NotI site downstream from the His6 tag coding sequence . The PCR fragment was cloned in NcoI and NotI sites of the first polylinker of pCDF-Duet1 expression vector ( Novagen ) . Mutant versions of apoL1 were created by site-directed mutagenesis with Accuprime Pfx DNA polymerase ( Invitrogen ) and verified by sequencing . The complete SRA coding sequence was obtained by PCR from pTSA-Rib-SRA [5] with primers creating a 5′ Ecl136II and 3′ SalI sites . The DNA fragment was cloned in frame with the S-tag coding sequence in EcoRV and XhoI sites of the second polylinker of pCDF-Duet1 vector and pCDF-Duet apoL1-V5-His6 construct . The plasmid constructs were transfected into E . coli BL21 ( DE3 ) and expression of recombinant proteins was induced from an exponentially growing culture by 1 mM isopropyl β-D-1-thiogalactopyranoside ( IPTG ) overnight at 37°C . Bacteria were resuspended in 50mM Tris-HCl ( pH 9 . 2 ) and lysed by two passages in French Press . Cell debris were removed by centrifugation at 5000g for 10 min , and inclusion bodies were recovered from supernatant by centrifugation at 16 , 000g for 15 min . Purification was performed from inclusion bodies dissolved in 6 M guanidium chloride , 150 mM NaCl , 50 mM Tris-HCl ( pH 8 . 0 ) . Solubilized proteins were incubated for 1 h with Ni-NTA beads ( Qiagen ) , and the bound proteins were extensively washed with 4 M guanidium chloride , 50 mM Tris-HCl ( pH 7 . 5 ) , 150 mM NaCl , 20 mM imidazole , before elution with 4 M guanidium chloride in 0 . 2 M acetic acid . After dialysis against 20 mM acetic acid , the proteins were concentrated ( Vivaspin , Sartorius ) . Purity and concentration were estimated by SDS-PAGE and Coomassie Blue staining . Cultures of BL21 ( DE3 ) strains transfected with pCDF-Duet1 constructs were grown at 37°C in LB containing 1% glucose and 50 µg/ml streptomycin from freshly plated colonies until optical density at 600nm ( OD600 ) reached 0 . 7 to 0 . 8 . Serial dilutions were plated onto LB containing 50 µg/ml streptomycin and either 1% glucose ( non-induced control ) or 50 to 75 µM IPTG ( induction of recombinant proteins ) . Usually the pH of LB was 6 . 0 , but the effect of raising the pH to 7 . 5 was evaluated . Colonies were counted after overnight incubation at 37°C and results were expressed as the ratio of colonies with and without IPTG induction . These results reflect the pore-forming activity of apoL1 [4] . Cultures of BL21 ( DE3 ) strains transfected with pCDF-Duet1 constructs were grown at 37°C in LB containing 50 µg/ml streptomycin and 1% glucose from freshly plated colonies untill OD600 reached 0 . 7 to 0 . 8 . Cultures were centrifuged and bacteria pellet was resuspended in fresh medium without glucose and distributed in 3 flasks to perform induction and copurification in triplicate . Expression of recombinant proteins was induced by addition of 1mM IPTG overnight at 20°C and 80 rpm . Cell density was measured by OD600 and bacteria were centrifuged and resuspended in ( 10% of culture volume×OD600 ) in cold hypotonic buffer ( 50 mM MES pH 6 . 0 or pH 7 . 0 ) containing EDTA-free protease inhibitors ( Roche ) . Bacteria were lysed by Fast Prep 24 ( MP Biomedicals ) with 1/10 volume of glass beads ( Lysing Matrix B , MP Biomedicals ) , for 3×30 s at 6 m . s−1 . Cell lysate was complemented by 0 . 6M NaCl , 1% Triton X100 , 20 mM imidazole , and vortexed vigorously . Cell debris were pelleted by centrifugation for 15 min at 16 , 000 g , and the supernatant was applied onto Ni-NTA beads ( Qiagen ) equilibrated in the same buffer , for 2 h at 4°C . Fig . S1 shows a comparative evaluation of the yield of the different apoL1 variants and SRA in the supernatants . After binding the beads were washed with 20 volumes of beads with cold binding buffer and the bound proteins were eluted with 2 volumes of SDS-PAGE sample buffer . Supernatant and bound fractions were analyzed by western blotting . ApoL1 was revealed with anti-recombinant apoL1 rat serum and peroxydase conjugated anti-rat antibody . SRA-Stag was detected by anti-Stag monoclonal antibody ( Novagen ) and peroxydase conjugated anti-mouse antibody . Peroxydase activity was revealed by ECL ( Western Lighting Chemiluminescence Reagent PLUS , Perkin Elmer ) . Signals were quantified in supernatant and bound fractions using ImageQuant TL Software ( GE Healthcare ) . Since only a fraction of each supernatant and elution was loaded onto the gel , the values obtained were reported to the total volume of lysis or elution ( “supernatant” and “bound” lanes respectively ) . ApoL1 and SRA binding values to nickel were calculated separately and the binding of SRA to apoL1 was expressed as “SRA binding versus apoL1 binding” ratio . The ratio obtained for SRA binding to WT apoL1 was considered as 100% . Trypanosoma brucei brucei 328-114 [17] were grown in HMI9 supplemented with 10% foetal bovine serum , 10% Serum Plus [18] at 37°C in 5% CO2 . T . b . rhodesiense ETat 1 . 2 NHS-resistant ( R ) and NHS-sensitive ( S ) clones and T . b . gambiense LiTat 1 . 2 were grown in IMDM medium supplemented with 20% foetal bovine serum [19] . The parasites were diluted at a final concentration of 5 . 105 cells/ml , and recombinant WT or mutant apoL1 , in 20 mM acetic acid , was added to reach 10 to 30 µg/ml . Dilution of the parasite culture medium never exceeded 15% , and in control samples without apoL1 the medium was similarly diluted with 20 mM acetic acid . As no attempt was made to reconstitute these proteins into HDL particles , the formation of aggregates is likely . Living parasites were counted in duplicate under the microscope . When lysis was not complete after 8 h , parasites were counted in duplicate after 24 h incubation and cell survival was expressed as % of control . Lysis experiments with human and Papio serum , as well as imaging of living immobilised trypanosomes , were performed as described in [20] . Alignments were performed by CLUSTALW software and edited with GeneDoc ( http://www . psc . edu/biomed/genedoc ) . Hydrophobic Cluster Analysis ( HCA ) was used to compare the proteins at 2D level [9] , [10] ( http://mobyle . rpbs . univ-paris-diderot . fr/cgi-bin/portal . py ? form=HCA ) . To calculate the energy of interaction , we used an approach based on a non-local energy between peptides [21] . The mean force potentials used here are based on 500 proteins , using 7 different atomic types , which are polar and nonpolar hydrogens ( H ) , Csp2 and Csp3 , O , S and N . Different functions have been evaluated for all the possible atomic pairs . These energy functions have been estimated for distances going from 0 . 0 to 10 . 0 Å with a 0 . 1 Å precision . The hydrodynamic transfection of mice was performed according to the method described in [22] . Briefly , 2 ml of saline solution containing 10 µg of pcDNA3-1 or pcDNA3-1 WT apoLI-V5His/del3-V5His/delTKIQ-V5His were injected in less than 8 seconds in the vein of the tail of 8 weeks-old BALB/c female mice . At day 1 , an aliquot of blood was analyzed by Western blot for expression of apoL1 and mice were injected intraperitoneally with 106 trypanosomes . Apolipoprotein L-I , NM_003661 . | The serum protein apolipoprotein L-I ( apoL1 ) is responsible for human innate immunity against Trypanosoma brucei brucei , because this protein kills the parasite by generating ionic pores in the lysosomal membrane . Two T . brucei subspecies ( T . b . rhodesiense and T . b . gambiense ) can resist apoL1 and therefore , infect humans and cause sleeping sickness . In T . b . rhodesiense , resistance to human serum is linked to interaction of the Serum Resistance-Associated ( SRA ) protein with the C-terminal region of apoL1 . We show that mutations targeted to this region reduced its interaction with SRA while preserving the activity of the ionic pore-forming domain . While some mutants also lost their trypanolytic potential , C-terminal mutants inspired by apoL1-like sequences of Papio anubis conserved this activity , but also acquired the ability to efficiently kill T . b . rhodesiense , both in vitro and in mice . These findings demonstrate that interaction of SRA with the C-terminus of apoL1 inactivate this protein and is responsible for the resistance of T . b . rhodesiense to human serum . Moreover , they suggest that apoL1-like proteins could be responsible for the trypanolytic potential of Papio species . Finally , Papio-like human apoL1 mutants could be used to generate transgenic cattle that would resist both T . b . brucei and T . b . rhodesiense . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"infectious",
"diseases/neglected",
"tropical",
"diseases",
"biotechnology/applied",
"microbiology",
"microbiology/cellular",
"microbiology",
"and",
"pathogenesis",
"cell",
"biology/cellular",
"death",
"and",
"stress",
"responses"
] | 2009 | C-Terminal Mutants of Apolipoprotein L-I Efficiently Kill Both Trypanosoma brucei brucei and Trypanosoma brucei rhodesiense |
The Plasmodium vivax Cell-traversal protein for ookinetes and sporozoites ( PvCelTOS ) plays an important role in the traversal of host cells . Although essential to PvCelTOS progress as a vaccine candidate , its genetic diversity remains uncharted . Therefore , we investigated the PvCelTOS genetic polymorphism in 119 field isolates from five different regions of Brazilian Amazon ( Manaus , Novo Repartimento , Porto Velho , Plácido de Castro and Oiapoque ) . Moreover , we also evaluated the potential impact of non-synonymous mutations found in the predicted structure and epitopes of PvCelTOS . The field isolates showed high similarity ( 99 . 3% of bp ) with the reference Sal-1 strain , presenting only four Single-Nucleotide Polymorphisms ( SNP ) at positions 24A , 28A , 109A and 352C . The frequency of synonymous C109A ( 82% ) was higher than all others ( p<0 . 0001 ) . However , the non-synonymous G28A and G352C were observed in 9 . 2% and 11 . 7% isolates . The great majority of the isolates ( 79 . 8% ) revealed complete amino acid sequence homology with Sal-1 , 10 . 9% presented complete homology with Brazil I and two undescribed PvCelTOS sequences were observed in 9 . 2% field isolates . Concerning the prediction analysis , the N-terminal substitution ( Gly10Ser ) was predicted to be within a B-cell epitope ( PvCelTOS Accession Nos . AB194053 . 1 ) and exposed at the protein surface , while the Val118Leu substitution was not a predicted epitope . Therefore , our data suggest that although G28A SNP might interfere in potential B-cell epitopes at PvCelTOS N-terminal region the gene sequence is highly conserved among the isolates from different geographic regions , which is an important feature to be taken into account when evaluating its potential as a vaccine candidate .
Malaria is an infectious parasitic disease with high prevalence and morbidity . Globally , it is estimated that 3 . 2 billion people in 95 countries and territories are at risk of being infected and develop the disease . In 2015 , malaria caused an estimate of 438 , 000 deaths , mostly in African children [1] . Among the protozoa species causative of human malaria , Plasmodium vivax , although less prevalent than P . falciparum in absolute numbers , presents the world's largest spread , an increasing morbidity [2] and became the main cause of malaria outside Africa . In Brazil , although there are three species of Plasmodium that cause malaria ( P . falciparum , P . vivax and P . malariae ) , approximately 87% of the 142 , 000 cases reported in 2015 were caused by P . vivax [3] . Thus , it is extremely important to develop new methods and intervention strategies to block or reduce this transmission . Significant effort and progress on P . vivax control have occurred over the last years , but the understanding of P . vivax biology is still crucial to develop potential vaccines and to achieve the goal of eliminating malaria . The ability of the Plasmodium to recognize , and then invade hepatocytes or red blood cells , is central to the life cycle and also to the disease process . During the pre-erythrocytic stage , it is well established that Plasmodium sporozoites migrate through Kupffer cells and several hepatocytes before finally infecting a hepatocyte . Therefore , antigens located on the surface of the parasite or specifically in apical organelles of the parasite during this stage have been suggested as a target for a better understanding of Plasmodium lifecycle and , consequently possibly used as vaccine [4] . In this context , the Cell-Traversal protein for Ookinetes and Sporozoites ( CelTOS ) has been considered a new alternative for vaccine development [5 , 6] . This protein , secreted by micronemes , is important to the success of cell crossing by sporozoites and ookinetes , and also hepatocyte invasion carried out by sporozoites . Studies have shown that the disruption of the CelTOS gene encoding , in P . berghei , reduces the infectivity in the mosquito host and also the infectivity of the sporozoite in the liver , almost eliminating their ability to cell pass [7] . In addition , the CelTOS is necessary for the motility of the parasite in both the mosquito vector and the human host , being determinant for the success of malaria infections [8] . Recently , studies from Jimah et al . suggested that the CelTOS is responsible for breaking the cell membranes from the inside of infected human and mosquito cells to enable the parasites to exit and complete the traversal process ( Jimah et al 2016 ) . In relation to its potential as a vaccine candidate , antibodies against PfCelTOS were able to inhibit sporozoite traversal of hepatocytes [9] , and induce protection in animals [10] . In humans , PfCelTOS derivative peptides elicited proliferative and IFN-γ responses in ex vivo ELISPOT assays using peripheral blood mononuclear cells ( PBMCs ) from irradiated sporozoite-immunized volunteers [8] and recombinant PfCelTOS were recognized by naturally acquired antibodies from exposed populations living in highly endemic areas from Africa [11] . However , all those previous studies used CelTOS protein of P . falciparum and/or P . berghei . Despite the antigenic and immunogenic properties of PfCelTOS , there is only one recent finding concerning the antigenic potential of its counterpart in P . vivax , the PvCelTOS , whose naturally acquired antibodies were able to recognize the recombinant protein [12] . Although essential to the development of its potential as a vaccine candidate , there is no available published data on the identification of pvceltos gene in field isolates and the evaluation of its genetic diversity in endemic areas . In fact , the extensive genetic diversity in natural parasite populations is a major obstacle for the development of an effective vaccine against the human malaria parasite , since antigenic diversity limits the efficacy of acquired protective immunity to malaria [13] . Despite the genetic diversity , which is one of the most prominent features of P . vivax infections , there is also a paucity of information on celtos gene polymorphism . Such data have importance in documenting the parasite genetic diversity changes and contribute to malaria control interventions in the future . Therefore , we proposed to identify pvceltos gene isolates from different regions of Brazilian Amazon and to study the potential impacts of the genetic diversity of PvCelTOS in protein structures and potential epitopes through bioinformatics tools .
Most cases of malaria in Brazil are concentrated in the Amazon Region , an endemic area for the disease [14] . Therefore , the study was carried out in five different regions of Brazilian Amazon ( Fig 1 ) . A subset of 81 patients was analyzed out of 312 individuals previously evaluated by Cavasini et al ( 2007 ) [15] ( 21 individuals from Plácido de Castro , 9 individuals from Oiapoque , 25 individuals from Novo Repartimento and 26 individuals from Porto Velho ) and , additionally , blood samples were collected from 38 P . vivax infected individuals from Manaus . Thus , a total of 119 blood samples were used in this study . Plácido de Castro ( PLC ) , is a city 90 km far from the capital of the State of Acre , located in Western Brazilian Amazon , with a population of 17 , 334 thousand inhabitants ( 16% aged above 18 years , at 153 meters above sea level , with a territorial area of 2 , 047 , 000 km2 , latitude of -09° 58’ 29” and longitude of 67° 48’ 36” , where the main economic activities are cattle breeding , rubber agriculture and farming . Active malaria transmission takes place during all periods of the year . Oiapoque ( OIP ) , Amapá State , located in the Brazilian Eastern Amazon , a mining gold area , with 17 , 423 a thousand inhabitants , presenting latitude of 03° 49’ 58” and longitude of 51° 49’ 51” . Manaus ( MAO ) , the capital of Amazonas State , located in the Northern Region of Brazil , with a population of more than 2 million people . It is the most populous city of Amazonas state , presenting latitude of -03° 06’ 07” and longitude of -60° 01’ 3” . Novo Repartimento ( NR ) , is a city 600 km far from Belém , capital of the State of Pará , located in Brazilian Eastern Amazon , with 47 , 197 thousand inhabitants , at 460 meters above sea level , with a territorial area of 11 , 407 km2 , presenting latitude of 04° 19’ 5” and longitude of 49° 47’ 47” , whose main economic activities are cattle breeding , commerce of manufactured products and farming . It presents active malaria transmission from January to December , with around 2 , 000 heterochthonous and autochthonous cases . Porto Velho ( PVL ) , capital of the Rondônia State , located in Western Brazilian Amazon , with a population of 360 , 068 thousand inhabitants ( 16% aged above 18 years ) , at 85 meters above sea level , with a territorial area of 34 , 082 km2 , latitude of -08° 45’ 43” and longitude of 63° 45’ 43” , where the main economic activities are cattle breeding , rubber agriculture , wood exploration and farming . Active malaria transmission takes place during all periods of the year . The distances between the study sites are shown in Table 1 . All P . vivax participants were enrolled according to the following criteria: sought medical assistance for clinical malaria symptoms , presented uncomplicated malaria symptoms , were > 18 years of age , and had a positive P . vivax malaria diagnosis . Pregnant women , patients < 18 years of age , and P . vivax- and P . falciparum-infected individuals were excluded from the study . Thin and thick blood smears were examined for the identification of the malaria parasite by a technician experienced in malaria diagnosis from the Brazilian Malaria Health Services . Thick blood smears from all of the subjects were stained with Giemsa , and a total of 200 microscopic fields were examined under a 1 , 000-fold magnification . Thin blood smears of the positive samples were examined for species identification . To increase the sensitivity of parasite detection , molecular analyses using specific primers for genus ( Plasmodium sp ) and species ( P . falciparum and P . vivax ) were performed in all of the samples as previously described . Donors positive for P . vivax and/or P . falciparum at the time of blood collection were subsequently treated by the chemotherapeutic regimen recommended by the Brazilian Ministry of Health . The study protocol was approved by the Research Ethics Committee of each locality , which included obtaining the following patients’ written consents for research use of their blood samples: Belém ( Novo Repartimento/PA ) : 68473–970; Porto Velho ( CEPEN ) : 76812–329; Rio Branco ( Hospital Geral de Plácido de Castro/AC ) : 69928–000; Oiapoque ( Hospital Municipal do Oiapoque/AP ) : 68980–000; Manaus ( CEP-FIOCRUZ ) : 346–613 . Written informed consents were obtained from all adult donors or from the parents of donors in the case of children . All the procedures adopted in this study fully complied with specific federal permits issued by the Brazilian Ministry of Health . The DNA was extracted from blood samples using the QIAamp DNA blood midi kit ( QIAgen ) according to the manufacturer’s instructions and stored at -20°C until amplification . The pvceltos gene is conserved among different species of Plasmodium and to obtain that of P . vivax , specific primers were designed using standard gene sequences of P . vivax Salvador-1 strain from NCBI database with Accession Nos . AB194053 . 1 . All oligonucleotides were checked for specificity by using the Primer-BLAST tool provided by the National Center for Biotechnology Information ( http://www . ncbi . nlm . nih . gov/tools/primer-blast/ ) . The forward primer ( 5’-CCCCCAAAGGCAAAATGAACAA-3’ ) corresponded position 20 to 41 of the pvceltos gene sequence and the reverse primer ( 5’-AACTCATCTTCAGCTTCTTCCTC-3’ ) corresponded to position 569 to 547 . The specific primers were chemically synthesized to perform PCR reaction and DNA sequencing . The pvceltos gene was amplified in a conventional PCR method using the pair of primers PvCelTOS 5’—CCCCCAAAGGCAAAATGAACAA—3’ ( forward ) and PvCelTOS 5’—AACTCATCTTCAGCTTCTTCCTC—3’ ( reverse ) . Amplification of the pvceltos gene was conducted in a reaction volume of 25 μL using 1 μL of DNA , 10 pmol/μL of each primer and the Master Mix kit ( Promega ) containing Taq DNA polymerase , PCR buffer and 10 nmol of each deoxynucleotide triphosphate ( dNTP , Promega , Madison , WI USA ) . The conventional PCR reactions were carried out using a GeneAmp PCR system 9700 ( Applied Biosystem ) and the cycling conditions were as follows: one step at 95°C for 2 min . ; 30 cycles at 95°C for 1 min . , 57°C for 1 min . and 72°C for 1 min . ; and a last step at 72°C for 1 min . In all reactions two negative controls were used ( one without DNA and other with DNA extracted from in vitro culture of P . falciparum PSS1 strain ) and a positive control ( P . vivax-infected sample ) . To confirm the presence of DNA from the in vitro culture of P . falciparum and that the lack of amplification was due the specificity of the primers for PvCelTOS , we performed the amplification of the P . falciparum P126 gene fragment and electrophoresis as previously described [16] . Moreover , three P . vivax-infected samples from our study sites were randomly chosen . Five μL of PCR product were submitted to electrophoresis in 2% agarose gel ( Sigma ) in 1x TAE buffer ( 0 . 04 M TRIS-acetate , 1 mM EDTA ) in the presence of 10x GelRed nucleic acid stain ( Biotium ) and afterwards the products were visualized by ultraviolet ( UV ) illumination . Sizing of products was performed using a GeneRuler 100 bp Plus DNA Ladder ( Thermo Scientific ) . Then , PCR fragments were purified using the GE Healthcare Lifesciences kit according to the manufacturer’s protocol and sequenced . The specificity of the assay was confirmed by sequencing the PCR products from all positive samples using a Big Dye terminator sequencing kit ( Applied Biosystems ) following the manufacturer’s instructions . The DNA sequencing was carried out on the 3730xl DNA analyzer ( Applied Biosystems ) and the results were analyzed using DNASTAR's sequence alignment software to identify polymorphism relative to the Sal-1 reference sequence from NCBI . The 3D structure of PvCelTOS was predicted using the Robetta algorithm [17] . The amino acid sequence was retrieved from NCBI under Accession Nos . AB194053 . 1 . The Robetta is an automated algorithm for predictions of the 3D structure of proteins through ab initio and comparative modeling . The first step is the search for structural homologs using BLAST [18] or PSI-BLAST [19] . In the protein sequence , the target primary structure is broken down into separated domains , or independently folding units of proteins , by comparing the sequence to structural families in the Pfam database [20] . Domains with homolog structures follow a template-based modeling protocol . The final five structures are selected by taking the lowest energy models as determined by the Rosetta energy function . The electrostatic surface was calculated with the Adaptive Poisson-Boltzmann Solver ( APBS ) software [21] integrated with Pymol . The APBS software solves the Poisson-Boltzmann equation in order to describe electrostatic interactions between solute in aqueous solution . Continuous electrostatics plays a very important role in determining ligand-protein and protein-protein binding kinetics . The prediction of linear B-cell epitopes was carried out using the program BepiPred [22] . This software takes a single sequence in FASTA format input and each amino acid receives a prediction score based on Hidden Markov Model profiles of known antigens and incorporates propensity scale methods based on hydrophilicity and secondary structure prediction . For each input sequence the server outputs a prediction score . The positions of the linear B-cell epitopes are predicted to be located at the residues with the highest scores . In order to consider a given region as a valid linear B-cell epitope for PvCelTOS , the cut-off value of 0 . 35 was used to warrant similar values of specificity ( 0 . 75 ) and sensitivity ( 0 . 49 ) . Therefore , the epitope score represents the average of the scores of at least eight consecutive amino acids above the cut-off , and the sequences with higher mean values were chosen as potential linear epitopes . The differential binding of T-cell epitopes spanning the full PvCelTOS sequence were made on 4/18/2016 using the IEDB analysis resource Consensus tool [23] which combines predictions from ANN aka NetMHC ( 3 . 4 ) [24 , 25] , SMM [26] and Comblib [27] . Considering lengths of 9 mers , the prediction score of each length was evaluated against 26 of the most frequent HLA alleles ( HLA-A*01:01; HLA-A*02:01; HLA-A*11:01; HLA-A*23:01; HLA-A*25:01; HLA-A*26:01; HLA-A*30:01; HLA-A*31:01; HLA-A*32:01; HLA-A*68:01; HLA-B*08:01; HLA-B*15:01; HLA-B*18:01; HLA-B*35:01; HLA-B*38:01; HLA-B*39:01; HLA-B*40:01; HLA-B*46:01; HLA-B*48:01; HLA-B*51:01; HLA-B*53:01; HLA-B*57:01; HLA-B*58:01; HLA-C*04:01; HLA-C*05:01; HLA-C*07:01 ) . Peptides with median consensus percentile rank 20 . 0 as predicted binders and at least 60% of HLA binding frequency was considered potential T-cell epitopes . The one-sample Kolmogorov-Smirnoff test was used to determine whether a variable was normally distributed . Differences in proportions of haplotypes frequencies between studied localities were evaluated by the Fisher’s exact test using Prism 5 . 0 for Windows ( GraphPad Software , Inc . ) . A two-sided P value < 0 . 05 was considered significant . Sequences were aligned using CLUSTAL X2 and the number of segregation sites ( S ) , number of haplotypes , nucleotide diversity ( π ) and haplotype diversity were computed using DnaSP v5 [28] . The Tajima’s D test [29] for determining departure from the predictions of the neutral theory of evolution was also estimated with DnaSP v5 . The genetic differentiation between populations was investigated evaluating the rate of fixation ( FST ) by analysis of molecular variance ( AMOVA ) implemented in ARLEQUIN v3 . 5 . 2 . 2 [30] and significances were estimated using 10 , 000 permutations . The significance level was adjusted by Bonferroni correction for multiple tests .
In order to identify the gene encoding the PvCelTOS in isolates from Brazilian endemic areas , 119 blood samples from infected individuals living in the cities of Porto Velho , Plácido de Castro , Manaus , Novo Repartimento and Oiapoque had the DNA extracted and subjected to molecular diagnosis by PCR . The primers designed from the Primer-BLAST program and PCR analysis by agarose gel revealed the amplification in 100% samples . All field isolates presented only one type of fragment corresponding to 550 base pair ( bp ) . In addition to these samples , P . falciparum specimens were also tested , but proved negative for PCR amplification of the pvceltos gene ( Fig 2 ) . Therefore , the 119 samples from individuals infected with P . vivax amplified by PCR were subjected to sequencing reactions in order to screen the possible single nucleotide polymorphisms of the gene encoding the PvCelTOS . Standard gene sequences of P . vivax Salvador-1 ( Sal-1 ) encoding PvCelTOS were aligned to sequences from different regions of Brazilian Amazon isolates . Identification of variants and novel haplotypes was done and our interpretations were confirmed with available standard gene sequence of the P . vivax CelTOS in PubMed database . The polymorphism identification in the gene encoding the PvCelTOS from our studied regions revealed that all isolates had a high degree of similarity in relation to base pair alignments with the reference strain ( 99 . 3% ) . However , from the 550 bp sequenced and aligned , four nucleotide bases ( 0 . 7% ) presented mutations in specific bp positions ( 24 , 28 , 109 and 352 ) , shown in Table 2 . Interestingly , we did not detect point mutations in a single field or geographic area and all SNPs were present in at least two isolates and two sampling localities . Even with the high conservation degree of pvceltos gene sequence , 85% of the studied isolates presented at least one SNP in relation to the reference strain . As shown in Fig 3a , the synonymous mutation C109A was present in 82% field isolates and was significantly higher than all other 3 mutations ( p<0 . 0001 ) , while the other synonymous mutation C24A was the least frequent mutation . Two non-synonymous mutations , G28A and G352C , which represent the substitution of Glycine for Serine and Valine for Leucine , respectively , were also detected in frequencies of 9 . 2% and 11 . 7% , respectively . In addition , regarding the endemic areas studied , the higher frequency of C109A was maintained in all localities . Manaus presented the highest diversity , since we detected all four mutations among the 38 samples , while Porto Velho presented the lowest diversity , with only the synonymous mutation C109A . Lastly , in field isolates from Plácido de Castro , the non-synonymous SNP G352C was also significantly higher than C24A ( p = 0 . 0086 ) and G28A ( p = 0 . 0480 ) , while in all other localities this predominance did not occur ( Fig 3b ) . Only 18 isolates ( 15% ) maintained their sequences identical to the reference strain in positions 24 , 28 , 109 and 352 ( H1 = CGCG ) . Furthermore , the mutations resulted in nine different haplotypes ( H2 = AGCG; H3 = CACG; H4 = CGAG; H5 = CAAG; H6 = CGAC; H7 = AGAG , H8 = CAAC; H9 = AAAG ) , whose frequencies are shown in Table 3 . Among all field isolates studied the haplotype H4 presented the highest frequency and was significantly higher when compared to the reference H1 ( p<0 . 0001 ) . On the other hand H2 ( p<0 . 0001 ) , H3 ( p = 0 . 0002 ) , H5 ( 0 . 0328 ) , H7 ( p = 0 . 0028 ) , H8 ( p<0 . 0001 ) and H9 ( p<0 . 0001 ) presented significantly lower frequencies when compared to H1 . However , regarding these haplotypes obtained from human isolates from the Amazon regions , we could not determine a genetic structure based on the localities . Therefore , we observed that H1 and H4 were present in all studied localities while H2 , H8 and H9 were detected in only a single locality ( Manaus , Novo Repartimento and Oiapoque respectively ) . Even though the haplotypes could not be segregated according to their geographic origin , Manaus and Novo Repartimento presented the highest diversity of field isolates with six different haplotypes , while Porto Velho presented the lowest diversity , with only two haplotypes , which were common to all localities ( H1 and H4 ) . Interestingly , despite the difference in number of field isolates , Oiapoque presented a high diversity of pvceltos gene sequence with five haplotypes while only four different haplotypes were detected in Plácido de Castro ( Table 3 ) . Due to the very high similarity among sequences from different geographic origins and the consequent lack of phylogenetic signal , it was not possible to analyze the haplotypes in reliable clades . We sequenced pvceltos gene ( positions 19–569 ) of 119 samples collected from five regions of Brazilian Amazon . From the alignment with reference strain ( Sal-1 ) , four distinct SNPs were identified . Two SNPs were synonymous ( C24A and 109A ) and two were non-synonymous ( G28A and G352C ) . The nucleotide diversity ( π ) for pvceltos of 119 sequences analyzed was 0 . 00141 ± 0 . 00014 . The highest nucleotide diversity was observed in the Oiapoque group ( 0 . 00202 ± 0 . 00044 ) , followed by the Plácido de Castro group ( 0 . 00161 ± 0 . 00029 ) . Among all 5 populations , Porto Velho sequences displayed the lowest nucleotide diversity ( 0 . 00067 ± 0 . 00017 ) as expected , since only one SNP was detected in this group ( Table 4 ) . Similarly , parasites from Oiapoque presented the highest estimate of haplotype diversity ( Hd ) ( 0 . 806 ± 0 . 014 ) whereas parasites from Porto Velho showed the lowest Hd ( 0 . 369 ± 0 . 091 ) . Haplotype diversity was similar among the other studied areas ( Table 4 ) . The Tajima’s D test was performed to asses if there is selective pressure on the pvceltos gene . Although the Tajima’s D values ranged between -0 . 279 and 0 . 699 , tests showed no significant departures from neutrality in all studied areas , indicating no significant selection in the pvceltos gene ( Table 4 ) . Pairwise comparisons between each parasite population were performed using the FST statistics to check whether there was indicative of genetic differentiation between parasite populations , but all FST values were non-significant , suggesting lack of genetic differentiation between the studied populations ( Table 5 ) . The detected non-synonymous mutations characterized the specific amino acid substitutions in positions 10 ( Glycine for Serine ) and 118 ( Valine for Leucine ) . As observed in the protein sequence alignments , PvCelTOS also presented high amino acid sequence conservation degree , since only 24 isolates ( 19 . 2% ) presented non-synonymous mutations and had different sequences in comparison with the reference Sal-1 strain , whose frequency was significantly higher than all other protein sequences found in our field isolates ( 79 . 8%; p<0 . 0001 ) . Therefore , we subsequently aligned the protein sequence of these mutant field isolates in relation to other three hypothetical CelTOS protein derivatives from P . vivax genome data available in PubMed protein database ( Fig 4a ) . Only 13 isolates ( 10 . 9% ) presented sequences identical to Brazil I strain and none of our field isolates presented complete homology with North Korean and India VII strains , however both Asian strains also presented mutations in C terminal region at position 178 ( Lysine for Threonine ) that was not detected in our Amazon isolates . Interestingly , the N-terminal mutation at position 10 ( Gly10Ser ) was never detected in available sequences , but it was present in 9 . 2% of our field samples . Regarding the five regions studied , all isolates from Porto Velho presented full homology with Sal-1 amino acid sequence , while in other regions the frequencies of mutant sequences ranged from 21% to 44% ( Fig 4b ) . Noting the diversity identified following the pvceltos gene , our data indicate that it is limited in isolates from different regions of the Brazilian Amazon . However , these two non-synonymous mutations found may have an impact on the protein folding and also influence its potential as an epitope . Fig 5a depicts the electrostatic potential around the mutations . The region encompassing Arg37 shows a strong negatively charged surface . The Lys178 position showed the same negative pattern , while Val118 and Gly10 are positively charged . Pro8 region is mostly neutral . Arg37 and Val118 are part of a stable alpha helix structure , whereas Pro8 , Gly10 , and Lys178 belong to flexible loop structures . Also , all residues are exposed to the surface , except Arg37 which is hidden inside the negative pocket . As shown in Fig 5b four high scored potential linear epitopes with at least eight amino acids were identified in the entire protein sequence ( Lys6-Asn13; Gly38-Arg57; Ile136-Glu143 and Lys166-Ser191 ) . The prediction scores ranged from 0 . 97 to 1 . 17 and no immunodominant epitopes could be identified by this approach . Considering that two of non-synonymous mutations were inserted in predicted B-cell linear epitopes ( Gly10Ser and Lys178Thr ) , we analyzed the prediction scores of mutate epitopes . Interestingly , the C-terminal mutation Lys178Thr , observed only in Asian strains , North Korean ( KMZ96916 . 1 ) and Indian VII ( KMZ78086 . 1 ) , resulted in a slight increase of the predicted score; while the N-terminal mutation Gly10Ser , observed in our Brazilian isolates Ser10-Val118 and Ser10-Leu118 , resulted in a decrease of the predicted score for an epitope ( Fig 5b ) . On the other hand , the predicted TCD8+ epitopes were conserved among all known strains and isolates , once non-synonymous mutations were not observed inside these epitopes . Analyzing the full sequence of PvCelTOS , six TCD8+ predicted epitopes presented consensus score smaller than 20 and were predicted to be recognized by more than 60% of analyzed HLA ( Fig 5b ) . Among these epitopes , the sequence RVSEDAYFL ( PvCelTOSI83-L91 ) was considered a potential promiscuous TCD8+ epitope , since it was predicted as bonded by 81% of evaluated HLAs and presented a mean consensus score of 11 . 81 . However , the potential of predicted epitopes as target of immune response and the effects of mutations on immune response against PvCelTOS remain unexplored .
Cell-traversal protein for ookinetes and sporozoites ( CelTOS ) has been considered a potential novel alternative for a vaccine against malaria . Although the biological function is not completely elucidated , its pivotal role in the cell traversal of host cells in mosquito and vertebrate host is important to a successful hepatocyte traversal and infection . Immunologic studies have demonstrated that CelTOS is target of naturally acquired cellular [8] and humoral response in exposed individuals [9] . However , one of the major obstacles to malaria vaccine development is still the low efficiency in inducing protection , which , in part , can be explained by genetic polymorphisms encoding different proteins used as immunogens [31] . In this context , the genome sequence of various organisms and the advances in bioinformatics have revolutionized the field of vaccinology , allowing the identification of vaccine candidates presenting low antigenic variation . Actually , several studies concerning the genetic diversity of Plasmodium spp . have described P . vivax and the gene coding for antigenic determinants such as circumsporozoite surface protein ( CSP ) [32] , Merozoite Surface Proteins ( MSP ) [33] , Duffy Binding Protein ( DBP ) [34] and Apical Membrane Antigen-1 ( AMA-1 ) [35] . ( Reviewed by [36] ) . In fact , the genetic diversity of these proteins in hyperendemic areas has been described as a limiting factor for the rapid acquisition of protective immunity , and as a consequence for the development of an effective vaccine . Moreover , the antigenic polymorphism of P . vivax vaccine candidates has been little discussed in unstable transmission areas such as the Brazilian endemic areas . Thus , considering that the epidemiology of malaria in Brazil presents unstable transmission and the knowledge about the genetic polymorphism of PvCelTOS remains unknown , we aimed to identify the pvceltos gene in isolates from different regions of the Brazilian Amazon and to study the potential impacts of the genetic diversity of PvCelTOS in protein structures and predicted epitopes . The identification and evaluation of the genetic diversity of pvceltos gene in isolates from different geographic regions has not been previously studied and this was the first report . Despite the large distance among the studied localities and the possible existence of a gene flow of Plasmodium vivax genome among the studied populations which , associated with migration of people , could promote the gene flow of the parasite [37] , our first results showed that pvceltos gene is highly conserved , presenting only 4 SNPs along its entire sequence , 2 synonymous and 2 non-synonymous mutations . This high conservation degree was expected , once it has been shown that CelTOS amino acid sequence is partially conserved even among three different Plasmodium species ( P . vivax , P . falciparum and P . berghei ) [7] . In relation to specific P . vivax celtos gene , there is a paucity of information available . In fact , it was described for only four different strains used in complete genome studies: Sal-1 , Brazil I , North Korean and India VII . Therefore , even with the high conservation degree of pvceltos gene sequence in relation to the reference strain Sal-1 , all these strains also presented at least one SNP . In our studied isolates , the synonymous mutation C109A was predominant and significantly higher than all other 3 mutations found , while the other synonymous mutation C24A was the least frequent mutation . It is important to mention that this predominant mutation ( C109A ) is also present in human P01 strain , a new reference genome for P . vivax from an Indonesian clinical isolate [38] . Classically , synonymous changes were thought to have no effect on the protein and were called silent , however , recent studies show that even synonymous nucleotide changes can affect protein folding and function [39–41] ( Reviewed by [42] ) . Indeed , in most of the gene encoding proteins , the rate of synonymous substitutions is higher than the rate of non-synonymous substitutions , a condition known as purifying selection , and this has been demonstrated in other Plasmodium proteins , such as PfAMA-1 [43] . Interestingly , in relation to pvceltos we observed a perfect balance of synonymous and non-synonymous substitutions in the few polymorphisms found in all gene sequences among geographically distinct isolates . This balance and the low diversity observed could raise at least two hypothesis: firstly , a possible low selective pressure of the immune system against this antigen , which can be corroborated by recent findings from Longley and colleagues that demonstrated a low frequency of naturally acquired antibodies against PvCelTOS in comparison with other sporozoite antigens such as CSP [44]; secondly , the high importance of this protein in sporozoite and ookinetes traversal process could be a consequence of this high conserved profile observed in the sequences of our study . Therefore , aiming to evaluate the degree of diversity of PvCelTOS in different field isolates from Brazilian Amazon , we also compared the amino acid sequence of each field isolate with the reference strain ( Sal-1 ) and the three other hypothetical CelTOS protein derivatives from P . vivax genome ( Brazil I , North Korean and India VII ) . Curiously , our isolates presented higher similarity in relation to the reference strain than to Brazil I strain which presented identical sequences in only 13 isolates . Additionally , none of our field isolates presented complete homology with North Korean and India VII strains , both Asian strains presented a mutation in C terminal region at position 178 that was not detected in our Amazon isolates . Moreover , we observed an N-terminal mutation at position 10 ( Gly10Ser ) , which had never been detected in available sequences , but was present in 9 . 2% of our field samples , as isolates Ser10-Leu118 and Ser10-Val118 . This mutation was present in three distant study localities ( Manaus , Novo Repartimento and Plácido de Castro ) and it was more frequent than the sequence of Brazil I strain in Novo Repartimento and Plácido de Castro . Interestingly , although the distance from Novo Repartimento , Plácido de Castro and Manaus to Oiapoque could difficult the gene flow and thus explain the absence of this mutation in Oiapoque population , the low frequency of gene flow promoted by the distance would not be the reason for the absence of this mutation in populations; since Porto Velho , which is closer to Plácido Castro ( the locality with the highest frequency of this mutation ) , did not present this mutation . Unfortunately , due to this high similarity degree we could not determine a genetic structure based on the localities , and the sequences and haplotypes could not be eligible to construct a phylogenetic tree . However , it was possible to identify 9 different haplotypes of pvceltos among the 119 P . vivax field isolates from the Amazon regions that were analyzed . Regarding the pvceltos sequences , we observed that haplotype H1 and H4 were present in all studied localities , however haplotype H4 presented the highest frequency and was significantly higher when compared to the reference H1 . These findings suggest a global distribution of parasites containing similar pvceltos genotypes . Moreover , the existence of the same haplotypes in different malaria endemic areas will be important for the rationale of malaria vaccine designs . Like other antigens of pre-erythrocyte stage , the immunity focused on CelTOS depends on humoral and cellular immune responses [10] . Antibodies induced by immunization with P . berghei CelTOS were able to recognize live as well as fixed P . berghei sporozoites [10] and immunization with P . falciparum CelTOS elicits cross-species protection against heterologous challenge with P . berghei [9] . Despite this cross-species reactivity , the low degree of similarity between the P . falciparum and P . vivax CelTOS ( 63% ) , and the knowledge that the protection can be reduced by depleting T-cell subsets in immunized animals prior to the sporozoite challenge thus eliminating the contribution of cellular components in protection [10] , make crucial the evaluation of both arms of the adaptive response against PvCelTOS to validate it as a vaccine candidate . Additionally , studies based on the genetic diversity of P . falciparum merozoite surface proteins , have demonstrated that non-synonymous SNPs contribute to the variability of the parasite and provide escape from host immunity [45] . Thus , to assess the targets of immune response in PvCelTOS and evaluate the potential effects of non-synonymous mutations on immune response against PvCelTOS , we used in silico approaches to determine differences on predicted TCD8+ epitopes and linear B-cell epitopes among the reference strain ( Salvador-1 ) and mutant PvCelTOS . Firstly , four epitopes were predicted as linear B-cell epitopes on full sequence of PvCelTOS . Interestingly , non-synonymous mutations could modify the potential of these predicted epitopes , once the N-terminal and C-terminal described non-synonymous mutations ( Gly10Ser and Lys178Thr ) were inserted in predicted linear B-cell epitopes and affected its prediction score . We hypothesize that this finding could not justify the low frequency of responders observed in the unique work that evaluated the natural immune response against PvCelTOS on exposed individuals from Western Thailand [44] , but it could indicate the genetic diversity of Plasmodium vivax and therefore , its possible effects on immune response can be considered in future studies . Moreover , it has been demonstrated that few amino acid changes can prejudice the binding of peptides to MHC molecules , reduce recognition by T cells or generate antagonistic peptides that inhibit activation of specific T cells by the MHC-peptide complex ( Reviewed by [42] ) . Therefore , in relation to potential T-cell epitopes , six TCD8+ epitopes were predicted as hypothetical promiscuous epitopes , presenting an HLA binding frequency higher than 60% and a mean consensus rank smaller than 20 . Curiously , PvCelTOS has conserved TCD8+ epitopes among all different strains and isolates; once there are not non-synonymous mutations inserted on any predicted T-cell epitope . This finding allied to the showed cellular response to Plasmodium falciparum CelTOS in exposed individuals [8] supports the necessity to identify and validate PvCelTOS T-cell epitopes that could be interesting on new vaccine approaches . P . vivax displays almost twice as much genetic diversity as P . falciparum in terms of SNP diversity and gene family variability . This implies that the global population of P . vivax may have a capacity for greater functional variation , mainly in gene families associated with immune evasion and erythrocyte invasion . In summary , our findings in PvCelTOS indicate that the very low variations in gene sequences could suggest that this conservative profile is important to the parasite’s survival and transmission . Moreover , although some studies have shown the influence of positive natural selection on genetic variability of other P . vivax vaccine candidates such as PvAMA-1 , PvDBP and PvTRAP [46–48] , our epitope prediction results indicate that the few CelTOS polymorphism in P . vivax is not maintained by balancing selection related to avoidance of immune recognition by the human host . However , future investigations aiming the naturally acquired cellular and humoral immune response against PvCelTOS derived antigens are still needed to corroborate the potential of PvCelTOS as a vaccine candidate .
Plasmodium vivax pvCelTOS mRNA for Pv cell-traversal protein , complete CDs . Accession number: AB194053 . 1; S4 [Plasmodium vivax Sal-1] Accession number: XP_001617263 . 1; Hypothetical protein PVBG_00206 [Plasmodium vivax Brazil I] Accession number: KMZ84426 . 1; Hypothetical protein PVNG_01740 [Plasmodium vivax North Korean] Accession number: KMZ 96916 . 1; Hypothetical protein PVIIG_00773 [Plasmodium vivax India VII] Accession number: KMZ 78086 . 1 | Cell-traversal protein for ookinetes and sporozoites ( CelTOS ) presents a pivotal role in the cell traversal of host cells in mosquito and vertebrate hosts . For this reason , it has been considered a potential novel alternative for a vaccine against malaria caused by P . falciparum . However , little is known about its orthologous P . vivax CelTOS . Although the genetic diversity of this protein could be a limiting factor for acquisition of immunity and present implications for an effective vaccine development , it has never been explored . Thus , considering that the epidemiology of malaria in Brazil presents variable transmission rates and the knowledge on the genetic polymorphism of PvCelTOS remains unknown , we aimed to identify the pvceltos gene in isolates from five different regions of the Brazilian Amazon and to study the potential impacts of the genetic diversity of PvCelTOS in protein structures and predicted epitopes . Our findings indicate that PvCelTOS is an extremely conserved protein , presenting only four SNPs in the entire sequences of field isolates from Brazilian Amazon . The two non-synonymous mutations found in our field isolates presented no significant effect on the protein structure and a very low impact on potential T and B-cell epitopes indicated by our epitope prediction . Collectively , our data suggest that the small need to avoid the immune recognition by the human host and its importance on the parasite’s survival and transmission reflects a very conservative profile of pvceltos gene in field samples from Brazil and other endemic areas worldwide . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion",
"Genes",
"and",
"protein",
"sequences",
"used"
] | [
"parasite",
"groups",
"medicine",
"and",
"health",
"sciences",
"plasmodium",
"population",
"genetics",
"tropical",
"diseases",
"immunology",
"parasitic",
"diseases",
"parasitic",
"protozoans",
"genetic",
"mapping",
"parasitology",
"vaccines",
"preventive",
"medicine",
"apicomplexa",
"protozoans",
"molecular",
"biology",
"techniques",
"population",
"biology",
"vaccination",
"and",
"immunization",
"research",
"and",
"analysis",
"methods",
"sequence",
"analysis",
"public",
"and",
"occupational",
"health",
"sequence",
"alignment",
"bioinformatics",
"malarial",
"parasites",
"artificial",
"gene",
"amplification",
"and",
"extension",
"molecular",
"biology",
"haplotypes",
"polymerase",
"chain",
"reaction",
"heredity",
"database",
"and",
"informatics",
"methods",
"genetics",
"biology",
"and",
"life",
"sciences",
"malaria",
"evolutionary",
"biology",
"organisms"
] | 2017 | Plasmodium vivax Cell Traversal Protein for Ookinetes and Sporozoites (PvCelTOS) gene sequence and potential epitopes are highly conserved among isolates from different regions of Brazilian Amazon |
Single eukaryotic cells commonly sense and follow chemical gradients , performing chemotaxis . Recent experiments and theories , however , show that even when single cells do not chemotax , clusters of cells may , if their interactions are regulated by the chemoattractant . We study this general mechanism of “collective guidance” computationally with models that integrate stochastic dynamics for individual cells with biochemical reactions within the cells , and diffusion of chemical signals between the cells . We show that if clusters of cells use the well-known local excitation , global inhibition ( LEGI ) mechanism to sense chemoattractant gradients , the speed of the cell cluster becomes non-monotonic in the cluster’s size—clusters either larger or smaller than an optimal size will have lower speed . We argue that the cell cluster speed is a crucial readout of how the cluster processes chemotactic signals; both amplification and adaptation will alter the behavior of cluster speed as a function of size . We also show that , contrary to the assumptions of earlier theories , collective guidance does not require persistent cell-cell contacts and strong short range adhesion . If cell-cell adhesion is absent , and the cluster cohesion is instead provided by a co-attraction mechanism , e . g . chemotaxis toward a secreted molecule , collective guidance may still function . However , new behaviors , such as cluster rotation , may also appear in this case . Co-attraction and adaptation allow for collective guidance that is robust to varying chemoattractant concentrations while not requiring strong cell-cell adhesion .
Many individual cells , including white blood cells and bacteria , chemotax—sensing and following gradients of signals . Some cells , though , are not loners—they migrate collectively—and cells traveling in clusters and sheets during development must chemotax together . Many experiments [1–5] have shown that clusters can have capabilities that single cells lack: in particular , clusters of cells can follow a gradient even when single cells do not . How can cells work together to follow a gradient that each individual cell is incapable of sensing ? How can cells integrate data from across the cluster to improve their gradient sensing abilities ? Is cluster chemotaxis essentially different than single-cell chemotaxis ? The simplest possibility , that cells just spatially average the gradient signal acting independently on each of them and thereby achieve a more accurate sensing capability , is ruled out , at least for lymphocytes , by experiments that show clusters can travel in the direction opposite to that of single cells [1] . A different possible explanation relies on the qualitative idea of collective guidance [6] , in which a cluster of cells can gain a direction even though each of its individual cells senses only the level of signal , and not its gradient . To make this notion more quantitative , we have recently introduced such a model of collective guidance in the context of neural crest cells where the cluster’s directionality comes from a regulation of contact inhibition of locomotion ( CIL ) [7]; a related model was also proposed for clusters of lymphocytes [1] and extended for studying border cell migration [8] . However , our current understanding of collective guidance and how collective chemotaxis occurs without single-cell gradient sensing does not account for the possibility of response coordinated by chemical signaling between cells . Our minimal model of collective guidance posits that each cell reacts only to the local chemoattractant and the physical presence of its neighbors [7] . More complicated signal processing could take place on the cluster scale if cells use signaling molecules to communicate with each other to collectively process the information contained in the chemoattractant gradient , as was recently suggested to be the case in branching morphogenesis [4 , 9] . It is therefore important to ask: What experimental signatures would tell us if this were happening , and how would this signal processing change the efficiency of the cluster’s movement ? Can collective signal processing overcome shallow gradients seen in vivo ( e . g . [10] ) , amplifying differences in cluster behavior between the front and the back ? In minimal models of collective guidance [1 , 7] , the cluster moves by a tug of war , and is likely under tension . Nevertheless , collective chemotaxis can also occur in the absence of strong adhesion [2] . How does this happen ? We will address all of these questions in this paper . Our initial focus will be on understanding in vitro experiments in relatively controlled environments [1 , 2] , especially experiments on explants of neural crest cells , and using these results to develop a useful quantitative framework for the study of collective guidance more generally , including collective chemotaxis in vivo [11 , 12] . To understand why clusters of explanted neural crest cells chemotax where single cells do not [2] , we will analyze both short-range interactions between cells and long-range interactions mediated by chemical secretions . The primary short-range interaction between neural crest cells are cadherin-mediated adhesion and contact inhibition of locomotion ( CIL ) . CIL results in cells repolarizing away from each other after contact . CIL in tissues may be regulated by the type of cadherin expressed , as well as being linked to mechanical force between cells [13–16] . Many possible molecular mediators of CIL have been established , including the non-canonical Wnt-planar cell polarity pathway and ephrin signaling [17 , 18] . Within this paper , we will take a phenomenological approach to modeling CIL , describing its consequences rather than its molecular origin . We first study models of biochemical processing of the chemoattractant signal within the cell cluster , assuming strong cell-cell adhesions as in our earlier model [7] . We treat the possibility of gradient sensing via cell-cell communication , using a mechanism that allows adaptation , i . e . the cluster’s response becomes insensitive to the overall level of the signal S ( r ) . We do this using a local excitation , global inhibition ( LEGI ) scheme [19] . This model is supported by recent experiments on collective gradient sensing in branching morphogenesis , which identify gap-junction-mediated communication between cells as a critical aspect of collective gradient sensing [4 , 9] . We also consider the possibility of cluster-level amplification of a sensed gradient , where relatively small changes in the chemoattractant signal S ( r ) across the cluster are amplified into much larger changes in the response level . With both adaptation and a switch-like amplification , we find that clusters of an optimal size are more efficient at chemotaxing than either smaller or larger clusters . Amplification of the external signal allows clusters to develop a large velocity even in a shallow gradient . We argue , based on simple scaling principles , that sufficiently large clusters with only short-range adhesion undergoing collective guidance would be expected to either fragment or become increasingly slow . We then show that if the cohesion of a cluster is not controlled by local cell-cell adhesion , but rather by chemotaxis toward a secreted signal ( or “co-attraction” [20] ) , a cluster of cells can undergo collective guidance by regulation of CIL even if cells are not in continuous contact . We show how co-attraction and regulated CIL interact in order to create robust chemotaxis . In the presence of co-attraction , new behaviors , including persistent cluster rotation , may emerge . We provide an extensive characterization of the transition to rotation , and how rotation can alter the efficiency of gradient-sensing clusters .
We want to model the collective guidance of a cluster of cells exposed to a chemical gradient S ( r ) . We use the experiments of [2] on neural crest explants responding to Sdf1 gradients as a guide to determine the features we include as well as the model parameters , though we expect our results to be more generally applicable as well . There are four major elements of a model of this process: 1 ) single-cell dynamics , 2 ) physical interactions between cells and contact-range effects like contact inhibition of locomotion , 3 ) the response of the cells to the chemical S ( r ) , and 4 ) chemical communication and signaling between cells . We use a two-dimensional stochastic particle model to describe cells exposed to a chemical gradient S ( r ) . We describe each cell i with a position ri and a polarity pi . The cell polarity indicates the cell’s direction and propulsion strength , i . e . the velocity with which it would travel in the absence of additional forces; we thus define pi so that an isolated cell with polarity pi has velocity pi . The cell’s motion is overdamped , so physical forces like cell-cell adhesion and exclusion change the cell’s velocity—the velocity of the cell is pi plus the net force the other cells exert on it , ∑ j ≠ i F i j . We model chemically-induced effects like CIL as altering a cell’s biochemical polarity pi . Our model is then: ∂ t r i = p i + ∑ j ≠ i F i j ( 1 ) ∂ t p i = - 1 τ p i + σ ξ i ( t ) + β i ∑ j ∼ i r ^ i j + χ ∇ c ( r i ) | ∇ c | Θ ( | ∇ c | - g 0 ) ( 2 ) where Fij are intercellular forces , e . g . cell-cell adhesion and volume exclusion , and ξi ( t ) are fluctuating , temporally uncorrelated noise terms that are Gaussian with 〈 ξ μ i ( t ) ξ ν j ( t ′ ) 〉 = 2 δ μ ν δ i j δ ( t - t ′ ) , where the Greek indices μ , ν run over the dimensions x , y . The first two terms on the right of Eq 2 are a standard Ornstein-Uhlenbeck model [28 , 29]: pi returns to zero with a timescale τ , but is pushed away from zero by the fluctuating noise ξ ( t ) . This models a cell that has a motion that is only persistent over a time of τ . Throughout this paper , we choose our units to be defined by the typical parameters of neural crest cells . With this in mind , we take our length scale to be the typical equilibrium cell-cell separation and our time scale to be the relaxation time—this corresponds to setting the cell diameter to be unity and the relaxation time τ = 1 . To convert between these simulation units and real units , we use values estimated from the experiments of [2]: typical equilibrium cell-cell separation is 20 μm and the typical time over which a cell reorients is roughly 20 minutes , i . e . τ = 20 minutes in real units . Within the simulation units we have chosen , measured neural crest cell velocities are on the order of 1 , so we choose σ = 1 . This choice means that the root mean square speed of an isolated cell is 〈 | V | 2 〉 1 / 2 = 2 1 / 2 σ τ 1 / 2 ≈ 1 . 4 microns/minute , in good agreement with , e . g . [2] . When we include adaptation , we assume that the kinetics of Eqs 7 and 9 are fast compared with the dynamics of interest , and set them to their steady states , assuming k−R ≫ kR and thus Ri = Ai , ss/Ii ( t ) . We set the diffusion rate kD = 4 in our units , corresponding to a time for equilibration of a few minutes , consistent with experiments using FRAP to see equilibration of fluorescent dyes across gap junctions [39] . However , we note that this rate can depend on the identity of the inhibitor , and may also be regulated [40 , 41] . We set the rates of generation and decay of the inhibitor to be kI = k−I = 1; this is discussed more in the adaptation section . A complete list of parameters and their justifications is included in the Supplementary Information , Table S1 . We integrate Eqs 1 , 2 and 7–9 explicitly with an Euler-Maruyama integrator [42] . The time step varies: for rigid clusters with high adhesion , we choose Δt = 1 × 10−4 , and for co-attraction simulations we choose Δt = 1 × 10−3 . Further details about time step selection as well as source code are available in the Supplementary Information .
In our earlier paper [7] , we studied a minimal version of the model described above , with no co-attraction ( χ = 0 ) and no adaptation or amplification , i . e . β i = β ¯ S ( r i ) . We briefly note a few results from that paper here , as in some limits , our more complex model will reduce to this model . Under assumptions of cluster rigidity and slow reorientation , the mean drift of a cluster of cells obeying Eqs 1 and 2 is given by 〈 V 〉 c ≈ β ¯ τ M · ∇ S ( 10 ) with the approximation true for S ( r ) ≈ S0 + r · ∇S . 〈⋯〉c is an average over the fluctuating pi but with fixed configuration and orientation of cells ri . The matrix M depends only on the configuration of cells; formulas for many cluster shapes and sizes are given in [7] . Mean cluster velocity 〈Vx〉 saturates at large number of cells N . This arises because we have the difference in signal between the front and the back growing as the cluster radius ( ∼ N ) , while the perimeter of the cluster also grows as N . The force on the cluster then grows as N at large N , while the effective friction of the cluster grows independently with the number of cells , as N—hence the net velocity should behave as ∼N1/2 × N1/2/N ∼ 1 at large N . ( Similar scaling arguments are found for the circular cluster limit in [1] . ) As we move beyond the minimal model , these scaling assumptions may break down , and therefore larger clusters will not necessarily have saturating velocities . Ref . [7] also provides analytic results for the chemotactic index CI– a measure of the directionality of the cluster motion . This is commonly defined as the ratio of the distance traveled in the direction of the gradient ( the x direction ) to the total distance traveled . To clarify how we average over many realizations of a path , we define CI = 〈Vx〉 / 〈|V|〉 . Until this point , we have only looked at highly adherent , effectively rigid clusters . However , collective cell migration can also occur with a high degree of fluidity and cell-cell rearrangement [12 , 53–60] . In addition , we have until now assumed that the only attraction between cells is short-range , representing cell-cell adhesion . However , neural crest cells also attract one another through chemical secretions , which can control the extent of cluster directionality and cohesion [20 , 27]—and many other cell types also chemotax toward secretions [61 , 62] . We extend our model to allow for this possibility , and show that clusters of cells that cohere via co-attraction can also be directed by collective guidance . These clusters need not be rigid , and can have significant re-arrangement or even only transient contacts . In this section , we will treat clusters with co-attraction ( χ ≠ 0 ) , but assume only the minimal model of signal processing , with the CIL susceptibility β i = β ¯ S ( r i ) .
Other variants of stochastic particle models have been used to model collective cell migration , ranging from models that use single particles to represent cells [30 , 53 , 63 , 84–86] to those that use more detailed representations of cells with either multiple particles or additional details of cell shape [87–91] . Other techniques , such as the Cellular Potts Model [58 , 75 , 92] and phase field models [23 , 93 , 94] have also been developed to study collective cell migration with significantly greater levels of detail on the cell’s shape and its internal biochemistry . Because emergent collective guidance has had only limited quantitative models in the past [1 , 7] , we have chosen our cell models to be as minimal as possible , in an attempt to focus on the essential aspects of collective guidance . Earlier models have been created to study neural crest chemotaxis in vivo [81 , 83 , 95]; however , these have explicitly described chemotaxis arising from a “follow-the-leader” mechanism where single leader cells can sense a gradient [96] , rather than through the collective mechanism we study here , where individual cells need not sense the gradient level . We also mention that unlike many of the models discussed above , our model does not include an interaction designed to align a cell’s polarity with its neighbors’ motion [53 , 63 , 75] or its own velocity or displacement [30 , 58 , 87 , 88] , and these mechanisms are not necessary for the effects we describe here . Competition between the collective guidance mechanism and alignment mechanisms may be an interesting area for future study . Our stochastic interacting particle model is relatively simple , which allows us to in some cases derive analytic results [7] . Many extensions of this approach are possible . Our model could be developed further for more quantitative comparisons by careful measurement of single-cell statistics in or out of a chemoattractant gradient [28 , 97]; this could lead to nonlinear or anisotropic terms in Eq 2 . Our description of contact inhibition of locomotion has also assumed , for simplicity , that contact with both the front and back of the cell is inhibitory; other possibilities may alter the collective dynamics of the cell cluster [23] . Our main findings are: 1 ) Cluster velocity and chemotactic index may reflect internal signal processing , and provide an experimental window into these processes . 2 ) We expect sufficiently large clusters undergoing collective guidance to either become increasingly slow or break up . 3 ) Strong adhesion between cells is not necessary for collective guidance to function if cells chemotax to a secreted molecule . 4 ) A balance of this co-attraction and graded contact inhibition of locomotion are necessary for efficient chemotaxis . 5 ) Co-attraction may also induce cluster rotation , and we have explicitly characterized the transition to rotation . 6 ) The combination of cluster rotation and cluster chemotaxis may induce systematic drifts that depend on cluster rotation . | To get from one part of the body to another , single cells often follow chemical signals . Sometimes , though , isolated cells ignore these signals , but a group of cells still manages to travel in a directed way . How can this happen ? We argue that if the signal changes how the cells interact with their neighbors , a cluster of cells can detect signals single cells ignore . We use computational models to study how this can happen , and show that the speed of the cluster will depend on how the cells process the signal , as well as whether or not the cells are tightly connected to one another . We also show if the cells are only loosely connected , and are attracted to a secreted molecule , cell clusters may develop rotation and other effects that will change how effectively they can sense signals . | [
"Abstract",
"Introduction",
"Model",
"Results",
"Discussion"
] | [
"velocity",
"cell",
"physiology",
"cell",
"motility",
"medicine",
"and",
"health",
"sciences",
"classical",
"mechanics",
"engineering",
"and",
"technology",
"nervous",
"system",
"signal",
"processing",
"junctional",
"complexes",
"electrophysiology",
"neuroscience",
"cell",
"polarity",
"gap",
"junctions",
"stem",
"cells",
"phase",
"diagrams",
"developmental",
"neuroscience",
"computer",
"and",
"information",
"sciences",
"animal",
"cells",
"neural",
"crest",
"chemotaxis",
"physics",
"signal",
"transduction",
"cellular",
"neuroscience",
"neural",
"stem",
"cells",
"cell",
"biology",
"anatomy",
"synapses",
"data",
"visualization",
"physiology",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"physical",
"sciences",
"cell",
"signaling",
"neurophysiology",
"motion"
] | 2016 | Collective Signal Processing in Cluster Chemotaxis: Roles of Adaptation, Amplification, and Co-attraction in Collective Guidance |
Extra Cytoplasmic Function ( ECF ) σ factors are a diverse group of alternate σ factors bacteria use to respond to changes in the environment . The Bacillus subtilis ECF σ factor σV responds to lysozyme . In the absence of lysozyme , σV is held inactive by the anti-σ factor , RsiV . In the presence of lysozyme RsiV is degraded via regulated intramembrane proteolysis , which results in the release of σV and thus activation of lysozyme resistance genes . Signal peptidase is required to initiate degradation of RsiV . Previous work indicated that RsiV only becomes sensitive to signal peptidase upon direct binding to lysozyme . We have identified a unique domain of RsiV that is responsible for protecting RsiV from cleavage by signal peptidase in the absence of lysozyme . We provide evidence that this domain contains putative amphipathic helices . Disruption of the hydrophobic surface of these helices by introducing positively charged residues results in constitutive cleavage of RsiV by signal peptidase and thus constitutive σV activation . We provide further evidence that this domain contains amphipathic helices using a membrane-impermeable reagent . Finally , we show that upon lysozyme binding to RsiV , the hydrophobic face of the amphipathic helix becomes accessible to a membrane-impermeable reagent . Thus , we propose the amphipathic helices protect RsiV from cleavage in the absence of lysozyme . Additionally , we propose the amphipathic helices rearrange to form a suitable signal peptidase substrate upon binding of RsiV to lysozyme leading to the activation of σV .
One key survival feature of all living organisms is the ability to recognize changes in their environments and respond appropriately . One signal transduction system bacteria use to sense and respond to environmental stresses are Extracytoplasmic Function ( ECF ) sigma ( σ ) factors [1 , 2] . ECF σ factors are a diverse family of alternative σ factors that are responsible for transcribing a wide variety of genes in response to environmental signals , often extracellular stress [3 , 4] . One hallmark of ECF σ factors is the presence of an anti-σ factor that is responsible for sequestering σ activity in the absence of signal . When bacteria encounter a specific signal , the anti-σ factor is inactivated via one of three mechanisms: 1 ) degradation of the anti-σ factor [3–5] 2 ) conformational change of the anti-σ factor [6–9] or 3 ) a partner switching mechanism in which an anti-anti-σ factor binds the anti-σ factor [10–13] . All three mechanisms result in the signal-dependent release of the σ factor and subsequent transcription of genes involved with the stress response . Degradation of the anti-σ factor is frequently achieved by Regulated Intramembrane Proteolysis ( RIP ) [4 , 14 , 15] , a stepwise proteolytic cascade . In most cases , cleavage occurs extracellularly at site-1 followed by an intramembrane site-2 protease . Cytosolic proteases degrade the remaining portion of the protein , releasing the σ factor to interact with RNA polymerase [4 , 14 , 15] . Cleavage at site-1 is typically the rate-limiting step of this cascade and thus is the regulated step of ECF σ factor activation [16–19] . A well-studied example of such regulation is the ECF σ factor σE in E . coli which is responsible for responding to cell envelope stress and is held inactive by the membrane bound anti-σ factor RseA [18 , 20] . Unfolded outer membrane proteins bind to the PDZ domain of DegS leading to the activation of DegS and cleavage of RseA at site-1 [21] . This cleavage event initiates the RIP cascade where RseA is further degraded by the intramembrane protease RseP and cytosolic proteases resulting in the activation of σE [14 , 19 , 22] . The B . subtilis ECF σ factor σW is another example where the rate-limiting step of anti-σ factor inactivation is proteolytic cleavage at site-1 . RsiW is the anti-σ factor that holds σW in an inactive state in the absence of signal [17] . When cell envelope stresses are present , the site-1 protease PrsW cleaves RsiW and initiates a RIP cascade , resulting in the activation of σW [23 , 24] . In B . subtilis the ECF σ factor σV belongs to the ECF30 subfamily which is found primarily in low GC gram positive Firmicutes including C . difficile and E . faecalis [1] . σV is held inactive by its anti-σ factor RsiV and the inducing signal is C-type lysozyme [25] . Lysozyme is a component of the innate immune system that degrades peptidoglycan by cleaving β- ( 1 , 4 ) -linked N-acetylmuramic acid and N-acetylglucosamine [26–28] . σV is responsible for transcribing genes involved in lysozyme resistance including oatA which encodes an O-acetylase that adds an acetyl group to peptidoglycan increasing resistance to lysozyme cleavage [16 , 29–31] . The dltABCDE operon is also transcribed by σV and encodes genes responsible for modifying the charge of teichoic acid which presumably works to repel positively charged lysozyme [25 , 31–34] . In the absence of σV both C . difficile and E . faecalis become more sensitive to lysozyme [35 , 36] . Activation of σV occurs through RIP-mediated degradation of the anti-σ factor RsiV and occurs only in the presence of C-type lysozyme [16 , 37 , 38] . The current model of σV activation is shown in Fig 1A . Our lab has previously shown that signal peptidase is the site-1 protease responsible for removing the extracellular portion of RsiV [38] . After site-1 cleavage , the remaining transmembrane portion of RsiV is then degraded constitutively by the intramembrane metalloprotease RasP at site-2 . The remaining cytoplasmic portion of RsiV is presumably degraded by cytosolic proteases [16] . RIP-mediated cleavage of RsiV leads to the release of σV and subsequent transcription of σV-controlled genes resulting in increased lysozyme resistance for the bacterium . In bacteria , signal peptidases are associated with the secretion machinery and are critical for cell function [39 , 40] . Signal peptidases are typically responsible for cleaving secreted pre-proteins containing a signal peptide , thus releasing them from the membrane . Signal peptidase function is not known to be regulated and cleavage is believed to occur during or shortly after translocation [39–41] . The findings in B . subtilis then raise interesting questions: how is RsiV able to resist signal peptidase cleavage in the absence of lysozyme ? What changes occur once lysozyme binds to create a suitable substrate for signal peptidase ? Here we provide evidence for the presence of amphipathic helices in RsiV that protect it from signal peptidase cleavage in the absence of lysozyme . We identified the region required to protect fusion proteins from signal peptidase cleavage . We determined that disruption of the putative amphipathic helices by introducing lysine residues results in constitutive cleavage by signal peptidase . We show using a membrane-impermeable reagent that the hydrophobic face of the amphipathic helixes can only be labeled in the presence of lysozyme . We also propose a model to explain how RsiV and lysozyme binding leads to a suitable signal peptidase substrate .
We previously demonstrated that RsiV is proteolytically degraded in a RIP-dependent manner in the presence of lysozyme leading to the activation of σV [16 , 37] . The protease required for site-1 cleavage of RsiV is signal peptidase [37 , 38] . However signal peptidase is only able to efficiently cleave RsiV in the presence of lysozyme [38] . It has been proposed that in the absence of lysozyme , RsiV is in a conformation that is resistant to signal peptidase cleavage . Upon binding to lysozyme , RsiV is hypothesized to undergo a conformational change that allows signal peptidase access to the cleavage site , leading to degradation of RsiV and subsequent σV activation [42] . Site-1 cleavage of RsiV by signal peptidase occurs between residues 66 and 67 [37] . To identify the key residues of RsiV sufficient for protection from signal peptidase , we fused the signal peptide of RsiV ( amino acids 1–66 ) to GFP and expressed the fusion protein in B . subtilis . We predict that if the RsiV signal peptide is able to protect from signal peptidase cleavage , then the fusion of these residues to GFP would be sufficient to block secretion of GFP into the culture supernatant . We analyzed the RsiV1-66-GFP fusion by performing western blot analysis of the whole cells and concentrated culture supernatant using an anti-GFP antibody . We detected cleaved GFP in the culture supernatants from strains producing RsiV1-66-GFP suggesting the signal peptide alone is not sufficient to protect from signal peptidase ( Fig 2A ) . We then created constructs with various lengths of the RsiV N-terminus fused to GFP and expressed in B . subtilis to determine if additional residues after the signal peptide ( RsiV1-76-GFP , RsiV1-86-GFP , and RsiV1-96-GFP ) are sufficient to protect GFP from cleavage by signal peptidase . Cells producing these proteins were analyzed by western blot as described above . We found that cleaved GFP was present in the culture supernatant of strains producing RsiV1-66-GFP and RsiV1-76-GFP ( Fig 2A ) . However , we did not detect a cleaved GFP band in the supernatant for RsiV1-86-GFP , or RsiV1-96-GFP . Importantly , we were able to detect uncleaved RsiV-GFP in whole cells confirming they were produced at similar levels ( Fig 2A ) . These results suggest that residues 1–86 of RsiV are sufficient to protect from signal peptidase cleavage . Our data suggest that residues 1–86 are sufficient to protect the RsiV-GFP fusion proteins from signal peptidase cleavage , but residues 1–76 were not sufficient . To determine if the region after the signal peptidase cleavage site is required to protect RsiV from signal peptidase cleavage we constructed internal deletions of RsiV after the signal peptidase cleavage site; RsiVΔ67–76 and RsiVΔ67–86 . To monitor signal peptidase cleavage of RsiV in the presence and absence of lysozyme , the cell pellets and culture supernatants were collected and analyzed by western blotting with anti-RsiV antibodies . Similar to the previous assay , if a cleavage product is detected in the supernatant it indicates that RsiV is being cleaved in the absence of lysozyme . Consistent with previous observations , wild-type ( WT ) RsiV was cleaved only in the presence of lysozyme ( Fig 2B ) [37] . In contrast for both deletion constructs a cleavage product was detected in the culture supernatant in the absence of lysozyme ( Fig 2B ) . We constructed the same deletions in a reporter strain to measure their impact on σV activity . We measured σV activity using a PsigV-lacZ promoter fusion which is dependent upon σV for expression [25 , 31] . We found , as expected , that RsiVΔ67–76 and RsiVΔ67–86 had increased basal levels of PsigV-lacZ expression indicating increased σV activation ( Fig 2C ) . In addition , PsigV-lacZ was induced only ~2-fold by lysozyme when the region located after the cleavage site was deleted ( Fig 2C ) . We also performed a western blot of the pellet and supernatant from the strains used in the σV activity assay and found similar results to IPTG expression strains ( S4 Fig ) . WT RsiV was not cleaved while both RsiV deletion constructs were degraded in the absence of lysozyme ( S4 Fig ) . Taken together , these results suggest that residues within 67–86 are required for the protection of RsiV from signal peptidase cleavage in the absence of lysozyme . We propose that residues surrounding the cleavage site of RsiV to amino acid 86 contain a region that is important for blocking signal peptidase cleavage of RsiV in the absence of lysozyme . Based on our observation that residues 67–86 are necessary to protect against signal peptidase cleavage , we sought to characterize the region that is responsible for resistance to signal peptidase . An alignment of 203 RsiV homologs was used to create a sequence logo showing a high degree of homology for residues 61–84 ( S1 Fig ) [37 , 43 , 44] . The sequence logo also revealed an alternating pattern of hydrophobic and hydrophilic residues suggesting the presence of a potential amphipathic helix ( Fig 1B ) . We used the secondary structure prediction software PEP-FOLD3 to predict the structure of this region in B . subtilis and we identified a predicted structure with two α-helices separated by a loop ( Fig 1B ) [45 , 46] . Upon investigation of the predicted structure , we identified a hydrophobic and hydrophilic face for each helix suggesting the predicted secondary structure contains two amphipathic helices . Helical wheel projections of each helix were created using NetWheels [47] and further suggest the presence of putative amphipathic helices containing both a hydrophobic and hydrophilic face ( Fig 1B ) . The putative amphipathic helices are part of a domain of unknown function DUF4179 [48] . In the case of another DUF4179 member , the anti-σ factor BAS1627 from Bacillus anthracis the structure was solved ( 3FBQ ) [49] revealing two α-helices separated by a turn much like our predicted structure ( Fig 1B ) . Analysis of Pfam database revealed that the DUF4179 domain is largely restricted to the phyla Firmicutes ( Table 1 ) . RsiV and BAS1627 are members of the ECF30 family which are almost exclusively found in Firmicutes ( 48 ) and it appears the DUF4179 is restricted to these anti-sigma factor families . Our in-silico analysis of RsiV revealed potential amphipathic helices that we hypothesized could be responsible for blocking signal peptidase cleavage in the absence of lysozyme . To test this hypothesis , we sought to disrupt the amphipathic helices by introducing a positive charge into the hydrophobic face . We constructed strains in which a hydrophobic residue of Helix 1 ( M67 ) , Helix 2 ( I73 , I76 ) , or a residue directly after the predicted amphipathic helices ( I80 ) were each changed to a lysine residue . The resulting constructs were analyzed for σV activity using the PsigV-lacZ reporter . We found that a lysine substitution at each hydrophobic residue resulted in dramatic ( >100-fold ) increase in the basal level of PsigV-lacZ expression in the absence of lysozyme ( Fig 3A ) . When incubated with lysozyme ( 10 μg/ml ) , cells producing RsiVM67K showed only a modest ~2-fold increase while the remaining mutants showed no further increase ( Fig 3A ) . Thus , changing select hydrophobic residues to positively charged lysine residues results in constitutive σV activity which suggests that RsiV is being cleaved by signal peptidase in the absence of lysozyme . To ensure RsiV was being produced and to determine if it was being cleaved at site-1 by signal peptidase in the absence of lysozyme we measured RsiV levels both in the cells and in the culture supernatant by western blot using anti-RsiV antibodies . In each case we observed cleaved RsiV in the culture supernatants of cells producing RsiV with a lysine substitution but not WT RsiV ( Fig 3B ) . In addition , we observed increased levels of RsiV in whole cell fractions of the cells producing RsiV with lysine substitutions ( Fig 3B ) . This is likely due to increased activation of σV and thus increased expression of rsiV itself . These results suggest that changing the hydrophobic residues of these putative amphipathic helices to a positively charged residue leads to increased cleavage of RsiV in the absence of lysozyme . To ensure that these mutations did not simply lead to misfolded RsiV , we used Circular Dichroism spectroscopy ( CD ) to compare the secondary structure of RsiVA66W I76K and RsiVA66W I80K to WT RsiV and RsiVA66W . The addition of RsiVA66W was necessary to allow for purification of the proteins without cleavage by signal peptidase . CD spectra showed troughs at 208 and 222 nm indicating helical secondary structure in the protein . Moreover , the CD spectra of all the variants are very similar to WT , suggesting that the lysine mutations do not result in protein with any gross change in secondary structure ( S6 Fig ) . This supports a model in which an amphipathic helix protects RsiV cleavage in the absence of lysozyme and disruption of the amphipathic nature of these helices results in lysozyme independent cleavage of RsiV by signal peptidase . We hypothesize the hydrophobic face of the amphipathic helices are interacting with the cell membrane and this interaction is limiting signal peptidase access to the cleavage site . To test whether the hydrophobic residues are membrane embedded we used a Substituted Cysteine Accessibility Method ( SCAM ) assay which takes advantage of the fact that RsiV lacks endogenous cysteine residues . Before performing the SCAM assay , we first tested the effect of the various RsiV cysteine substitutions on σV activity by monitoring expression of the PsigV-lacZ promoter fusion in response to increasing concentrations of lysozyme . We found that the majority of cysteine substitutions did not alter activation of σV in response to lysozyme and behaved like WT RsiV ( Fig 4A ) . Only one cysteine substitution ( RsiVI73C ) led to increased σV activity in the absence of lysozyme ( 4A ) . Additionally , we monitored RsiV degradation by testing both the pellet and supernatant of these constructs for RsiV by western blot analysis . Similar to the σV activity assay , only RsiVI73C was cleaved in the absence of lysozyme ( Fig 4B ) . These results suggest that single cysteine substitutions of RsiV , except for RsiVI73C , do not impact σV activity and result in a functional RsiV . To perform the SCAM assay we constructed a strain producing 6xHis-RsiVA66W . The 6xHis-tag was added to allow for purification of RsiV mutant proteins after labeling . The A66W mutant was necessary to allow for purification of RsiV after lysozyme addition as we previously demonstrated that RsiVA66W blocks site-1 cleavage [37 , 38] . We then introduced cysteine substitutions at various positions along the putative amphipathic helices of RsiV ( Fig 1B ) . To label cysteine substitutions of 6xHis-RsiVA66W we used ( Na- ( 3-maleimidylpropionyl ) biocytin ( MPB ) [50 , 51] . MPB is a membrane-impermeable molecule with a biotin group that allows for detection using streptavidin conjugates and a maleimide group capable of covalently binding to cysteine [51 , 52] . Cells were either exposed to lysozyme ( 150 μg/ml ) or left untreated . MPB ( 50 μM ) was added to label extracellular cysteines . Next , RsiV was affinity purified using nickel resin . We then performed western blotting to analyze RsiV levels with anti-RsiV and MPB labeled RsiV was detected using Streptavidin IR680LT . As a negative control for MPB we included RsiVA66W which lacks cysteines and found that , as expected , we could not detect biotinylated RsiV . As a positive control , we introduced a cysteine at position 167 which is where RsiV interacts with lysozyme and we expect to be present extracellularly [42] . We found that RsiVA66W A167C was labeled when incubated in the presence and absence of lysozyme ( Fig 5 ) . We observed that hydrophilic residues ( Q65C , K69C , K78C , and E84C ) could be labeled in both the presence and absence of lysozyme indicating these residues are outside of the membrane and accessible to label ( Fig 5 ) . Conversely , hydrophobic residues ( M67C , I70C , V72C , I73C , I76C , and I80C ) do not label in the absence of lysozyme suggesting they are embedded in the membrane when lysozyme is not present . In the presence of lysozyme these same hydrophobic residues are now labeled , indicating they are accessible to the label . F82C is a hydrophobic residue that labels in both the absence and presence of lysozyme indicating that it is not embedded in the membrane ( Fig 5 ) . Interestingly , F82 is not predicted to be part of the α-helices ( Fig 1B ) and was previously observed to be part of the β-sheet in the RsiV-Lysozyme co-structure [42] . These results further suggest the presence of amphipathic helices that protect RsiV from signal peptidase cleavage in the absence of lysozyme . We have shown that changing hydrophobic residues of the amphipathic helix to lysine leads to constitutive cleavage of RsiV and that cysteine substitutions in the hydrophobic face are not accessible in the absence of lysozyme . To measure the effects of the lysine mutations on accessibility of the hydrophobic face we combined I80K with different hydrophobic cysteine substitutions ( V72C , I76C ) and analyzed these strains using the SCAM assay in the presence and absence of lysozyme . As we showed previously RsiVV72C and RsiVI76C did not label in absence of lysozyme and upon the addition of lysozyme they were labeled by MPB ( Fig 6 ) . Interestingly , the cysteines in RsiVV72C , I80K and RsiVI76C , I80K were accessible to MPB labelling in both the presence and absence of lysozyme ( Fig 6 ) . This suggests the hydrophobic face is no longer in the membrane when there is a lysine substitution . This result is consistent with the model of lysine substitutions in the hydrophobic face of the amphipathic helix displacing the helix from the membrane and leading to constitutive cleavage of RsiV and thus constitutive σV activation .
Our data suggests the presence of amphipathic helices that are required for the protection of RsiV from signal peptidase cleavage . The fusion construct RsiV1-86-GFP demonstrates that this region is sufficient to block signal peptidase cleavage , whereas fewer residues fail to protect . Deletions of this region lead to both constitutive cleavage of RsiV and constitutive σV activity showing that this region is necessary to resist signal peptidase cleavage . This study also found that disrupting the hydrophobicity of the amphipathic helices through lysine substitution abolished protection from signal peptidase cleavage . The sensitivity to signal peptidase cleavage is likely due to the lysine substitutions forcing the helices out of the membrane and is supported by the observation that a lysine substitution leads to accessibility of the hydrophobic face in the absence of lysozyme . The results of the cysteine labeling assay show that in the absence of lysozyme , residues on the hydrophobic face of the amphipathic helix are inaccessible to label . This suggests these helices are either membrane-associated , interacting with each other via the hydrophobic faces , or the hydrophobic residues are buried inside of RsiV . Our observation that RsiV1-86-GFP , which lacks the majority of extracellular RsiV , yet is protected from signal peptidase cleavage implies that the amphipathic helices are not interacting with RsiV . Thus , we favor a model where they interact with the membrane or with one another via their hydrophobic surfaces thus protecting them from signal peptidase cleavage . Our lab has previously shown that lysozyme binds to RsiV and we have obtained the crystal structure of this complex [42] . We have also shown that the binding of lysozyme to RsiV is necessary for signal peptidase to cleave RsiV [37 , 38 , 42] . Previously it has been hypothesized that the binding of lysozyme to RsiV induces a conformation change leading to the susceptibility of signal peptidase cleavage . We propose that binding of lysozyme to RsiV causes the amphipathic helices to no longer interact with the membrane , allowing signal peptidase access to the previously sequestered signal peptide cleavage site . This model is supported by the observation that residues which we presume to interact with membrane in the absence of lysozyme become accessible when lysozyme binds to RsiV . One major question remains: what drives α-helices from the membrane upon lysozyme binding ? Signal peptidases preferentially cleave flexible and accessible regions of a protein over ordered structures such as α-helices [53] . The co-structure of RsiV and lysozyme shows that residues 79–89 are in a β-sheet while residues 59–78 are disordered when bound to lysozyme [42] . In the absence of lysozyme , however , residue 80 is unable to be labeled by a membrane-impermeable dye , suggesting it is membrane embedded . The latter is supported by structural prediction programs which propose amphipathic helices for region 61–79 . Thus , one possible mechanism responsible for driving the amphipathic helices from the membrane is that lysozyme binding causes part of the helix to convert to a β-sheet . A structure of RsiV alone is needed to answer these questions . There are other reports of amphipathic helices controlling degradation or cleavage of proteins by signal peptidases . One example is the HIV glycoprotein 160 ( gp160 ) which is processed in the endoplasmic reticulum of infected cells and forms the soluble subunit gp120 which is involved in viral entry [54 , 55] . Once gp160 is translocated across the ER membrane , an α-helix surrounding the signal peptidase cleavage site interacts with the ER membrane and prevents cleavage until the protein is properly folded [56] . This α-helix works as a quality control mechanism to ensure proper folding prior to signal peptidase cleavage . In fact when the α-helix is disrupted , proper folding does not occur [56] . Another example of an amphipathic helix controlling degradation is the human protein squalene monooxygenase ( SM ) which is the second , rate-limiting step in cholesterol biosynthesis [57] . When cholesterol levels are low , the amphipathic helix of SM is associated with the ER membrane and SM is able to function normally . Once cholesterol levels rise , the amphipathic helix is disassociated from the membrane likely due to cholesterol thickening and flattening of the ER membrane . This disassociation leads to the degradation of SM by the ubiquitin–proteasome system and leads to a decrease in cholesterol biosynthesis [58] . These systems are similar to RsiV in that they all use a membrane-associated amphipathic helix to control the access of a protease to its cleavage site resulting in blocking or delaying cleavage . These observations may suggest a broader use of amphipathic helices in controlling the cleavage of other as yet unidentified signal peptidase substrates . The transmembrane segment and putative amphipathic helices of RsiV are part of a domain of unknown function , DUF4179 [48] . This domain is found primarily in Firmicutes and a crystal structure has been solved for a protein containing this domain . The crystal structure ( 3FBQ ) is the extracellular domain of an ECF anti-σ factor from B . anthracis [49] . The structure shows a region directly after the transmembrane domain that has two short α-helices similar to the predicted structure of the region surrounding the RsiV cleavage site . However , the rest of the protein is unrelated to RsiV . This structural information , combined with the results of our study , and the role of amphipathic helices controlling signal peptidase cleavage in other organisms raises the intriguing possibility that this could be a common motif in ECF signaling , playing an important role in protecting the anti-σ factor from degradation , and triggering the proteolytic cascade that releases the cognate σ-factor in the presence of the appropriate signal .
All B . subtilis strains ( Table 2 ) are isogenic derivatives of PY79 , a prototrophic derivative of B . subtilis strain 168 [59] . Plasmids used in this study are listed in Table 3 and were confirmed by DNA sequencing ( Iowa State DNA Sequencing Facility ) . B . subtilis strains were transformed using a one-step method to induce competence previously described by [60] . Plasmids used for transformation into B . subtilis were first created using site directed mutagenesis ( Supplemental Methods ) along with isothermal assembly [61] and propagated in E . coli Omnimax cells . Oligonucleotide primers used for PCR and site-directed mutagenesis are listed in S1 Table . Constructs were cloned into pDR111 in order to place rsiV variants under control of the IPTG-inducible hyper-spank ( Phs ) promoter . The vector , pDR111 was digested with HindIII and SphI and PCR products were introduced by isothermal assembly . All pDR111 constructs were transformed into CDE1563 ( ΔsigVrsiV::kn ) and the RsiV construct was introduced at the amyE gene on the chromosome . WT RsiV was created using CDEP3194/CDEP3195 ( chromosomal DNA ) . Rsiv-GFP fusion construct inserts were created using CDEP3865/Appropriate RsiV-GFP Reverse ( Rev ) primer ( chromosomal DNA ) and CDEP3868/Appropriate RsiV-GFP Forward ( For ) primer ( pCM11 template ) [62] . RsiV deletion constructs were created using CDEP3194/Appropriate RsiVΔ Rev primer ( chromosomal DNA ) and CDEP3195/Appropriate RsiVΔ For primer . SCAM constructs were created by first building 6xHis-rsiVA66W using primers CDEP3629/A66W Rev ( chromosomal DNA ) and CDEP3195/A66W For ( chromosomal DNA ) . The remaining SCAM constructs were created using 6xHis-rsiVA66W as template . The inserts were created with primers CDEP3629/Appropriate rsiV For primer and CDEP3195/Appropriate rsiV Rev . Constructs cloned onto pDG1664 were created by first digesting the vector with BamHI and EcoRI to allow for Isothermal Assembly of the PCR products . All pDG1664 constructs were transformed into CDE1936 ( ΔsigVrsiV::kn pyrD::PsigV-lacZ ( cat ) ) and the sigVrsiV constructs were introduced at the thrC gene on the chromosome . WT sigVrsiV was created using primers CDEP3533/CDEP3534 ( sigVrsiV template ) . Deletions constructs , lysine substitutions , and cysteine mutations in the reporter strain were created with the primers CDEP3533/Appropriate rsiV Rev and CDEP3534/Appropriate rsiV For ( sigVrsiV template ) . Constructs cloned onto pET21b were created by first digesting the vector with BamHI and NdeI to allow for Isothermal Assembly of the PCR products . All pET21b constructs were transformed into E . coli BL21λDE3 . ANC108 ( pET21b 6xhis-rsiV ) [38] was used as the initial template to create the constructs . First , we generated 6xhis-rsiVA66W using the primers CDEP3165/1562 and CDEP3267/CDEP1561 . We then used 6xhis-rsiVA66W as template to create 6xhis-rsiVA66W I76K and 6xhis-rsiVA66W I80K using the primers CDEP3165/Appropriate rsiV Rev and CDEP3267/Appropriate rsiV For . Antibiotics were used at the following concentrations: chloramphenicol , 5 μg/ml; MLS erythromycin plus lincomycin , 1 μg/ml and 25 μg/ml; kanamycin , 5 μg/ml; spectinomycin , 100 μg/ml; ampicillin 100 μg/ml . Isopropyl β-D-1-thiogalactopyranoside ( IPTG ) was used at a final concentration of 1 mM unless otherwise noted . Cell cultures were grown overnight in LB broth at 37°C and subcultured 1:100 in 5 mL LB IPTG ( 1mM ) . Cells were incubated at 37°C for ~3hrs to an OD600 of 0 . 8–1 and 1 mL of cell culture was centrifuged to obtain a pellet and supernatant . The proteins in the supernatant were precipitated with methanol chloroform [63] and the resulting precipitant was analyzed by western blotting . The culture pellet was resuspended in 100 μL LB with or without lysozyme ( 10 μg/mL ) and incubated at room temperature . After 10 min , 100 μL 2x Laemmli sample buffer was added to stop the reaction . Cells were sonicated with a Branson Sonifier 450 and analyzed by western blotting . Cells were grown in LB ( 1% NaCl ) to inhibit lysozyme activity [64] . ON cells were subcultured 1:20 and grown at 37°C for 2 hrs . These early log-phase cultures were then diluted to an OD600 of 0 . 05 and grown in 200 μL volumes in a round bottom 96-well plate in the presence of varied concentrations of lysozyme ( 0 , 5 , and 10 μg/mL ) . The plate was shaken at 1000 rpm at 37°C for 4 hrs . The cell density ( OD600 ) was determined by diluting 50 μL from each well into 150 μL LB in a 96-well flat bottom plate . The remainder of each culture was lysed by mixing with 6 μL 2% sarkosyl , 12 μL chloroform . Lysate ( 20 μL ) was added to 100 μL Z-buffer ( 60mM Na2HPO4 , 40mM NaH2PO4 , 10mM KCl , 1mM MgSO4 , 50mM 2-mercaptoethanol , pH 7 . 0 ) in a flat bottom 96-well plate . The reaction was started by the addition of 50 μL ONPG ( 10 μg/mL ) and color development followed at 405 nm , every 1 min for 30 min ( Tecan Infinite M200 Pro ) . β-galactosidase activity was reported as the rate of product development normalized to cell density . This protocol was adapted from [65] . Cells cultures were grown overnight in LB broth at 37°C and subcultured 1:100 in 50 mL LB IPTG . Cells were incubated at 37°C for ~3hrs to an OD600 of 0 . 8–1 and 15 mL cell cultures were centrifuged and pellets resuspended in 500 μL protoplast buffer ( 0 . 4 M sucrose , 10 mM KPi , 15 mM MgCl2 ) [66] with and without lysozyme ( 75 μg/mL ) and incubated at room temperature . After 10 minutes Na- ( 3-maleimidylpropionyl ) biocytin ( MPB , Invitrogen ) ( 50 μM ) was added to label cysteine residues . The reaction was quenched by the addition of 2-mercaptoethanol to a final concentration 20 mM . Cells were washed with protoplast buffer amended with 2-mercaptoethanol ( 20 mM ) . Labeled RsiV variants were then affinity purified as described below , with volumes adjusted to the smaller scale: 50 μL nickel resin , 500 μL lysis buffer , 500 μL wash buffer , and 50 μL elution buffer . The eluted protein was immediately mixed with 50 μL 2X Laemmli buffer and analyzed by immunoblotting . Samples were electrophoresed on a 15% SDS polyacrylamide gel ( Biorad ) and proteins were then blotted onto a nitrocellulose membrane ( GE Healthcare , Amersham ) . Nitrocellulose was blocked with 5% Bovine Serum Albumin ( BSA ) and proteins were detected with either 1:10 , 000 dilution anti-RsiV59-285 [16] or anti-GFP . Streptavidin IR680LT ( 1:2500 ) was used to detect two biotin-containing proteins , PycA and AccB , which serve as loading controls [67] . To detect primary antibodies , nitrocellulose was washed and incubated with 1:10 , 000 dilution of Goat anti-Rabbit IR800CW ( Li-Cor ) and imaged on an Odyssey CLx Scanner ( Li-Cor ) . All immunoblots were performed a minimum of three times with a representative example shown . Expression cultures of RsiV variants were prepared as previously described [37] . Briefly , 50 mL of LB Amp100 was inoculated 1:100 with overnight cultures of BL21 ( DE3 ) harboring the respective expression plasmids , grown shaking at 37°C to an OD600 of 0 . 8 , induced with 1 mM IPTG and shifted to 30°C . After 3 hours of growth , cells were pelleted by centrifugation . RsiV variants included a 6xHis-tag at the N-terminus and were affinity purified over Nickel resin ( Thermo HisPur Ni-NTA , Cat Nr 88222 ) . Batch purification was performed in 50 mM TrisHCl , pH 8 . 0 , 250 mM NaCl , all steps at 4°C . Lysis buffer , wash buffer and elution buffer included 10 mM , 20 mM and 250 mM imidazole , respectively . Nickel resin ( 500 μL slurry ) was equilibrated with lysis buffer . Cell pellets were lysed in 2 mL lysis buffer by sonication ( Branson Sonifier 450 , microtip , output 3 , three cycles of 30 pulses ) and the lysate was clarified by spinning at 30 min 16 , 000 x g . The supernatant was incubated with the prepared resin for 20 min , washed five times with 2 mL wash buffer and protein was eluted in two 1 mL fractions . The fractions were combined and dialyzed against 50 mM NaPi , pH 7 . 0 , 100 mM NaCl . CD spectroscopy to determine protein secondary structure was performed with about 10 μM protein in 50 mM NaPi , pH 7 . 0 , 100 mM NaCl . CD spectra were recorded from 260 nm to 190 nm with 1 nm data interval at 25°C in a Jasco J-815 CD spectropolarimeter . Data integration time was 2 seconds and the scanning speed was 100 nm/min . For comparison to WT protein , spectra were normalized to the signal at 220 nm . | Signal transduction involves ( i ) sensing a signal , ( ii ) a molecular switch triggering a response , and ( iii ) altering gene expression . For Bacillus subtilis’ response to lysozyme , we have a detailed understanding of ( i ) and ( iii ) . Here we provide insights for a molecular switch that triggers the lysozyme response via σV activation . RsiV , an inhibitor of σV activity , is cleaved by signal peptidase only in the presence of lysozyme . Signal peptidase constitutively cleaves substrates that are translocated across the membrane . A domain-of-unknown-function ( DUF4179 ) in RsiV contains the signal peptidase cleavage site , and protects RsiV from cleavage in the absence of lysozyme via amphipathic helices . In addition to RsiV , DUF4179 is found in an unrelated and uncharacterized anti-σ factor present in Firmicutes including within some clinically-relevant species . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"medicine",
"and",
"health",
"sciences",
"chemical",
"compounds",
"pathology",
"and",
"laboratory",
"medicine",
"enzymes",
"pathogens",
"bacillus",
"enzymology",
"microbiology",
"cell",
"disruption",
"organic",
"compounds",
"prokaryotic",
"models",
"membrane",
"receptor",
"signaling",
"experimental",
"organism",
"systems",
"basic",
"amino",
"acids",
"amino",
"acids",
"bacteria",
"bacterial",
"pathogens",
"cysteine",
"research",
"and",
"analysis",
"methods",
"specimen",
"preparation",
"and",
"treatment",
"stress",
"signaling",
"cascade",
"proteins",
"medical",
"microbiology",
"microbial",
"pathogens",
"chemistry",
"sulfur",
"containing",
"amino",
"acids",
"biochemistry",
"signal",
"transduction",
"mechanical",
"treatment",
"of",
"specimens",
"specimen",
"disruption",
"organic",
"chemistry",
"post-translational",
"modification",
"cell",
"biology",
"bacillus",
"subtilis",
"biology",
"and",
"life",
"sciences",
"proteases",
"physical",
"sciences",
"lysine",
"cell",
"signaling",
"signal",
"peptides",
"organisms",
"signaling",
"cascades"
] | 2018 | Bacterial sensing: A putative amphipathic helix in RsiV is the switch for activating σV in response to lysozyme |
The epigenetic regulation of gene expression by the covalent modification of histones is a fundamental mechanism required for the proper differentiation of germ line cells during development . Trimethylation of histone 3 lysine 9 ( H3K9me3 ) leads to chromatin silencing and the formation of heterochromatin by recruitment of heterochromatin protein 1 ( HP1 ) . dSETDB1/Eggless ( Egg ) , the ortholog of the human methyltransferase SETDB1 , is the only essential H3K9 methyltransferase in Drosophila and is required for H3K9 trimethylation in the female germ line . Here we show that Windei ( Wde ) , the Drosophila homolog of mouse mAM and human MCAF1 , is an essential cofactor of Egg required for its nuclear localization and function in female germ line cells . By deletion analysis combined with coimmunoprecipitation , we have identified the protein regions in Wde and Egg that are necessary and sufficient for the interaction between the two proteins . We furthermore identified a region of Egg that gets covalently modified by SUMOylation , which may facilitate the formation of higher order chromatin-modifying complexes . Together with Egg , Wde localizes to euchromatin , is enriched on chromosome 4 , and binds to the Painting of fourth ( POF ) protein . Our data provide the first genetic and phenotypic analysis of a mAM/MCAF1 homolog in a model organism and demonstrate its essential function in the survival of germ line cells .
The epigenetic regulation of gene expression by the modification of histone proteins is a very important mechanism to control the differentiation of many cell types during development . The N-terminal , outward protruding histone tails are targets of posttranslational modifications , such as acetylation , ubiquitination , phosphorylation and methylation . These histone modifications are supposed to act sequentially or in combination to form a histone code that can be deciphered by different chromatin-associated proteins to mediate changes in chromatin structure and transcriptional activity [1]–[3] . One of the best-studied histone modifications is the methylation of the histone 3 lysine residue 9 ( H3K9 ) , which generally correlates with transcriptional repression [4]–[7] . However , recent results also point to a function of H3K9 methylation in the dynamic regulation of transcription , since this histone modification has frequently been found in the chromatin of actively transcribed genes [8] . H3K9 can be mono- di- or trimethylated and it has been shown that promoter H3K9 trimethylation results in much stronger transcriptional repression than promoter H3K9 dimethylation [9] . Methylated H3K9 can recruit Heterochromatin Protein 1 ( HP1 ) [10]–[13] , a chromatin-associated protein that has been implicated in heterochromatin formation but may also function in the regulation of euchromatic genes [14] . HP1 is highly conserved from yeast to human and was first found in Drosophila as a suppressor of position effect variegation Su ( var ) 2–5 [15] , [16] . Several histone methyltransferases ( HMTs ) have been identified that specifically methylate H3K9 , the first being Su ( var ) 3–9 of Drosophila [17] , which is required for di- and trimethylation of H3K9 at the chromocenter [18] . Mammalian homologs of Su ( var ) 3–9 are predominantly associated with constitutive heterochromatin [19] , [20] and have been implicated in the regulation of telomere length [21] . G9a is a second H3K9 specific HMT which catalyzes mono- and dimethylation of H3K9 at euchromatic loci of mammalian cells [22] . G9a and its close relative GLP/Eu-HMTase1 form a heteromeric complex and appear to function cooperatively in the regulation of euchromatic genes [23] . A third class of H3K9 specific HMTs is represented by SETDB1/ESET [24] , [25] . SETDB1 can be recruited to euchromatin by binding to KAP1/KRAB-ZFP transcriptional repressor complexes and functions in gene silencing by local methylation of H3K9 [24] , [26] . In contrast to Su ( var ) 3–9 and G9a HMTs , recombinant GST-SETDB1 fusion proteins have little HMT activity in vitro [24] . This is most likely caused by the requirement for binding to mAM/MCAF1 , a protein copurifying with SETDB1 in mammalian nuclear extracts [9] . Knock-down of mAM by RNAi leads to an increase of H3K9me2 , caused by the failure of SETDB1 to convert H3K9me2 to H3K9me3 [9] . mAM can bind simultaneously to SETDB1 and to the methyl CpG binding protein MBD1 and thus may provide a link between DNA methylation at CpG dinucleotides and histone H3K9 methylation mediated by SETDB1 [27] , [28] . Knockout mice lacking the function of SETDB1 [29] , Suv39h1 and Suv39h2 [30] , G9a or its close relative GLP [23] , [31] are all embryonic lethal , albeit at different developmental stages , demonstrating that these enzymes are essential and apparently have non-redundant functions . Many proteins involved in transcriptional repression are either covalently modified by conjugation to the small ubiquitin-related modifier ( SUMO ) or they contain SUMO binding domains [32] . Binding to SUMO has been reported for both MCAF1 [33] and for SETDB1 [34] . It is generally thought that SUMOylation and binding to SUMO contributes to the efficient assembly of large protein complexes that allow the coordinated modification of multiple histone tail residues during the formation of heterochromatin . In Drosophila , only the SETDB1 homolog dSETDB1/Eggless ( Egg ) is essential for viability and fertility [35]–[38] , whereas mutants for Su ( var ) 3–9 [17] and G9a [39] are homozygous viable and fertile . In polytene chromosome squash preparations , Egg localizes to euchromatic regions and is strongly enriched on chromosome 4 [35] . egg mutants loose most of the H3K9 methylation marks as well as binding of HP1 on chromosome 4 , which is consistent with global changes in the transcription level of genes located on chromosome 4 that were observed in egg mutants [35] , [37] . Egg coimmunoprecipitates with the chromosome 4 associated Painting of fourth ( POF ) protein [37] , which is required for chromosome-wide transcriptional upregulation of genes on chromosome 4 [40] , [41] . Homozygous egg mutant females possess only rudimentary ovaries , due to massive apoptosis at early stages of oogenesis in somatic and germ cells [36] , [38] . H3K9me3 levels were strongly reduced in egg mutant germ line cells , particularly at the earliest stages of oogenesis in the germarium [36] , [38] . So far it was not known whether Egg requires a binding partner homologous to mammalian mAM/MCAF1 for its function . Here we show that Windei ( Wde ) , the Drosophila ortholog of mAM/MCAF1 precisely colocalizes with Egg in ovaries and binds to Egg in vivo . We furthermore show that Egg gets covalently modified by SUMOylation , which is a hallmark of many chromatin-associated proteins involved in transcriptional repression . Wde localizes to euchromatic regions of salivary gland polytene chromosomes , in particular to chromosome 4 , and associates with POF in a protein complex . We have generated null mutations in wde , which are homozygous lethal and can be fully rescued by a transgene encoding a GFP-Wde fusion protein . Surviving homozygous wde mutant females are sterile and possess only rudimentary ovaries . Loss of wde function in germ line clones eliminates nuclear localization of Egg , leads to the arrest of oogenesis before stage 10 and to subsequent degeneration of mutant egg chambers by apoptosis . Like egg mutant cells , germ line cells mutant for wde show strongly reduced H3K9 trimethylation . According to the indistinguishable subcellular localization and mutant phenotypes of the two interactors , we propose that Wde is an essential binding partner of Egg required for the conversion of H3K9me2 to H3K9me3 .
In human cells , conversion of dimethyl to trimethyl H3K9 by the histone methyl transferase SETDB1/ESET is greatly facilitated by binding of this enzyme to mAM/MCAF1 ( also called ATFa associated factor ) [9] . The Drosophila homolog of SETDB1/ESET called dSETDB1/Eggless ( Egg ) is essential for oogenesis [36] , [38] and for H3K9 trimethylation on chromosome 4 [35] , [37] . So far it was not known whether Egg activity requires a cofactor homologous to mAM . Database screening using the BLAST algorithm ( http://blast . ncbi . nlm . nih . gov/Blast . cgi ) revealed the existence of a single Drosophila homolog of mAM encoded by the CG12340 transcription unit located at position 47C1 on the right arm of the second chromosome ( Figure 1A ) . Due to its mutant phenotype ( see below ) we named this gene windei ( wde , german for wind egg ) and will use this name throughout the manuscript . wde encodes a strongly acidic protein ( pI = 4 , 55 ) of 1420 amino acids and a calculated molecular weight of 157 . 776 Dalton . Although the overall sequence identity between mAM and Wde is only 14 , 8% , the domain structure with an internal coiled-coil region and a C-terminal fibronectin type III repeat is identical ( Figure 1B ) . Within the fibronectin type III repeat , the sequence identity is 36% ( 57% similarity ) ( Figure 1C ) . In order to study the expression pattern and subcellular localization of Wde , we raised specific antibodies against two peptides corresponding to aa 70–84 and aa 1286–1301 . The specificity of the antibodies was tested in stainings of wild type , wde mutant and Wde overexpressing embryos ( Figure S1 ) and in ovaries containing wde mutant germ line clones ( Figure S3 ) . For all analyses shown here , we used antiserum affinity purified against the peptide coprresponding to aa 70–84 . Because mutant flies deficient for Egg , the potential binding partner of Wde , show severe defects during oogenesis [36] , [38] , we focused our analysis on the subcellular localization of Wde in ovaries . Wde was ubiquitously expressed both in the somatic follicle cells and in germ line cells at all stages of oogenesis ( Figure 2 ) . Wde was nuclear in interphase ( Figure 2B and 2F ) and localized in the cytoplasm in mitosis after nuclear envelope breakdown ( Figure S1K , Figure S1L , Figure S1M ) . Within the nucleus , Wde was not homogeneously distributed but showed a reproducible localization to subnuclear structures ( Figure 2F ) . This was particularly obvious in the oocyte nucleus and in the highly polyploid nurse cell nuclei ( Figure 2F and 2F''' ) . To determine more precisely to which structure Wde localized in the nucleus , we performed double stainings with an antibody against HP1 . HP1 is enriched in heterochromatin , in particular at the chromocenter , the centromeric heterochromatin in which all four chromosomes of Drosophila are attached to each other during interphase [42] . In the oocyte nucleus , Wde always was present in one or 2 very brightly staining dots that were in close apposition , but not colocalizing with the brightest spot of HP1 staining at the chromocenter ( Figure 2F–2H' ) . In nurse cell nuclei , Wde and HP1 colocalized to some extent , but there were also regions where only one of the two proteins was detectable ( Figure 2F–2H'' ) . The published localization pattern of Egg [36] was strikingly similar to that of Wde . In order to test whether the two proteins indeed colocalize , we generated transgenic flies expressing a full length Egg-RFP fusion protein , which resembled precisely the published localization pattern of Egg . Double stainings of endogenous Wde with Egg-RFP ( data not shown ) and of GFP-Wde with Egg-RFP revealed that both proteins colocalized exactly ( Figure 3 ) . This was also true for the prominent dots in the oocyte nucleus ( Figure 3A'–3D' ) . To test whether Wde also colocalized with POF , a known binding partner of Egg [35] , [37] , we coexpressed GFP-Wde and POF-RFP in germ line cells ( Figure 3E–3H ) . Both proteins colocalized precisely in the oocyte nucleus , showing that the prominent dot that stained for Wde , Egg and POF corresponds to the fourth chromosome . The precise colocalization of Wde with Egg and the fact that the mammalian homologs of Wde and Egg bind to each other [9] prompted us to test whether Wde and Egg associate in a protein complex . To that aim , we generated a series of full length and partially deleted GFP-Wde and Egg-HA fusion proteins ( Figure 4A and Text S1 ) for expression in Drosophila S2 cells . To test our anti Wde antibody for specificity in Western blots , we used extracts from wild type embryos and from embryos homozygous mutant for a null allele of wde ( see below ) . In wild type embryos , the antibody detected several bands with a molecular weight around 250 kD that were absent in extracts of homozygous mutant embryos ( Figure 4B ) . We then coexpressed full length GFP-Wde with full length Egg-HA . Coimmunoprecipitation with the anti Wde antibody , followed by Western blot with antibodies against GFP and HA showed that the antibody precipitated GFP-Wde and Egg-HA , demonstrating that both proteins were associated in a complex ( Figure 4C ) . The same result was obtained when anti-GFP antibody was used instead of the affinity-purified antiserum against Wde ( data not shown ) . To narrow down the regions of both proteins that were required for complex formation , we coexpressed different deletion constructs for both proteins ( Figure 4A ) and tested them by coimmunoprecipitation . These experiments revealed that a fragment of Wde containing the coiled-coil-region ( aa 842–907 ) was sufficient for coimmunoprecipitation of Egg ( Figure 4D–4F ) . The smallest fragment of Egg required for coimmunoprecipitation with Wde consisted of aa 366–521 ( Figure 4A , 4D–4F ) , a region that does not contain any known protein motif detected by the SMART algorithm ( http://smart . embl-heidelberg . de/ ) . Many proteins involved in transcriptional repression can either bind to SUMO and SUMOylated proteins or are covalently modified by SUMOylation . We noticed that full length Egg ran at a higher molecular weight in SDS-PAGE than predicted from its sequence ( Figure 4A ) . This was also true for all fragments of Egg that contained the N-terminal 202 aa ( Figure 4A ) , suggesting that this region is covalently modified . To test whether aa 1–202 of Egg are SUMOylated , we expressed this part of Egg fused to HA ( Egg-6-HA; Figure 4A ) in S2 cells , immunoprecipitated the protein with HA antibody and probed the Western blot with an antibody against SUMO ( Figure 4G ) . The immunoprecipitated 55 kD band corresponding to Egg-6-HA was clearly recognized by the SUMO antibody ( Figure 4G ) . In the reverse experiment , Egg-6-HA was detected in immunoprecipitates pulled down with the SUMO antibody ( Figure 4G ) , confirming that Egg-6-HA was modified by SUMOylation . To address the in vivo relevance of these observations , we stained egg chambers expressing Egg-RFP with antibodies against SUMO and HP1 ( Figure 4H–4K ) . Consistent with our tissue culture data , SUMO colocalized with the dot of Egg-RFP on the fourth chromosome in the oocyte nucleus ( Figure 4H–4K ) . To find out whether Wde is a chromatin-associated protein , we performed immunofluorescence stainings on squashed salivary gland polytene chromosomes . Wde was not present in significant amounts on the chromocenter , but intense staining was detectable on the fourth chromosome that was also stained by the HP1 antibody ( Figure 5B–5D ) . In addition , Wde was present in several bands in the euchromatic region of all chromosomes ( Figure 5B ) . The enrichment of Wde on the fourth chromosome was confirmed by double stainings of GFP-Wde and Painting of fourth ( POF ) , a protein that binds predominantly to the fourth chromosome of Drosophila melanogaster ( Figure 5F–5H ) [41] . POF staining overlapped with HP1 only on the fourth chromosome , but not on the chromocenter ( Figure 5J–5L ) . To test whether Wde and POF were associated with each other in a protein complex , we coexpressed full-length GFP-Wde and full length POF-HA in S2 cells . Upon coimmunoprecipitation using an antibody against GFP , both GFP-Wde and POF-HA were detectable in Western blots ( Figure 5M ) , demonstrating that both proteins were present in one complex . To analyze the function of wde in development , we generated a null mutation of wde ( wdeTD63 ) by FLP/FRT mediated recombination in trans of two P-elements flanking the wde locus on both sides ( Figure 1A; for details see Materials and Methods ) . Two additional mutant alleles ( wde00884 and wde06198 ) caused by insertion of the P-elements P{Epgy2}CG12340EY00884 and P{XP}CG12340d06198 , respectively , into the coding region of wde ( Figure 1A ) are predicted to result in premature termination of translation and are likely to be null alleles as well . Animals homozygous for any of the three mutant alleles or transheterozygous for any combination of the three alleles die at pupal stages . However , rare escapers were obtained by raising homozygous mutant larvae separated from their heterozygous siblings ( see Materials and Methods ) , which eliminates competition for food and allows the weak mutants to reach the adult stage . Adult homozygous mutant animals were very weak and survived only for few days . The ovaries of homozygous wde mutant females were tiny and did not develop to the stage when egg chambers bud off from the germarium ( Figure S2 ) . The lethality and the ovary phenotype of the wdeTD63 null allele was fully rescued by ubiquitous expression of the full length GFP-Wde fusion protein under control of daughterless::GAL4 using the UAS/GAL4 system ( data not shown ) , proving that the observed defects were due to mutation of wde and not to a second site mutation elsewhere in the genome . To analyze the requirement for wde during germ line development without affecting the function of wde in follicle cells , we generated germ line clones of the wde00884 and wdeTD63 alleles using the autosomal FLP-DFS technique [43] . wde00884 and wdeTD63 germ line clones did not produce any eggs , whereas control clones using the same FRT chromosome without a wde mutation produced eggs at the expected frequency ( data not shown ) . To analyze the oogenesis defect of wde mutants in more detail , we generated germ line clones marked by the absence of GFP expression using the FLP/FRT technique . While control clones showed robust nuclear Wde staining ( Figure S3C and Figure S3E ) , clones mutant for wde00884 completely lacked nuclear Wde staining ( Figure S3H and Figure S3J ) . This was also true for the alleles wde06198 and wdeTD63 ( data not shown ) . We noted that egg chambers with wde mutant germ line clones did not develop beyond stage 8 of oogenesis . Closer inspection of late stage egg chambers with wde germ line clones revealed that the nuclei of germ line cells were highly condensed and degenerated subsequently ( Figure 6A and 6D ) . This morphological feature is typical of cells undergoing apoptosis . Stainings of wde mutant ovaries with an antibody against the activated caspase Drice , a marker for apoptotic cells [44] , showed a strong increase of Drice staining in wde mutant germ line cells ( Figure 6C and 6D ) , demonstrating that wde is required for survival of germ line cells . Previous work on the function of Egg during oogenesis had shown that this enzyme is required for trimethylation of H3K9 , especially in the germarium [36] , [38] . To check whether wde mutant germ line cells also show reduced levels of H3K9me3 , we stained ovaries containing germ line clones of wde with an antibody against H3K9me3 . Whereas H3K9me3 levels were unaffected in germaria with control germ line clones ( Figure 6G and 6H ) , H3K9me3 staining was strongly reduced in clones of wde mutant germ line cells ( Figure 6K and 6L ) . The oogenesis phenotype of egg mutants has so far been only described for the rudimentary ovaries of homozygous mutant females [36] , [38] . To compare the mutant phenotypes of egg and wde mutants , we generated germ line clones for the egg1473 allele , which has a deletion of the SET domain and thus is nonfunctional with respect to its histone methyl transferase activity [36] . The reduction in H3K9me3 staining in egg1473 germ line clones was as strong as in wde mutant germ line clones ( data not shown ) , demonstrating that Wde is indispensable for H3K9 trimethylation by Egg . In general , the germ line clone phenotypes of wde and egg1473 mutants were indistinguishable with respect to apoptosis and the timing and severity of egg chamber degeneration ( data not shown ) . Moreover , germ line clones doubly mutant for wde06198 and egg1473 showed the same phenotype as either single mutant ( data not shown ) , indicating that both proteins function in the same process and are nonfunctional in the absence of their binding partner . To test whether Wde and Egg are dependent on each other for proper nuclear localization in germ line cells , we analyzed germ line clone ovaries with respect to the localization of both proteins . Wde was normally localized in germ line clones for egg1473 ( Figure 7C and 7D ) , whereas Egg-RFP was hardly detectable in germ line clones of the null allele wdeTD63 ( Figure 7K and 7L ) , in contrast to control germ line clones with the same FRT chromosome carrying a wild type copy of wde ( Figure 7G and 7H ) . These data clearly show that Wde is required for stabilization of Egg in germ line cells , but since egg1473 is not a protein null allele they leave open the question of whether Egg is also required for proper localization of Wde . To clarify this issue , we analyzed the subcellular localization of GFP-tagged Wde and RFP-tagged Egg in transfected S2 cells . When transfected alone , Wde localized to the nucleus ( Figure 8B ) and Egg to the cytoplasm ( Figure 8C ) . Cotransfection of full length Wde and Egg resulted in nuclear colocalization ( Figure 8D ) . A deletion analysis of Wde ( Figure 8A ) revealed that the C-terminal region of Wde is required and sufficient for its nuclear localization ( Figure 8G and 8I ) , and that the coiled-coil region is additionally required to recruit Egg to the nucleus ( Figure 8J ) . From these results we conclude that Wde can localize to the nucleus in the absence of Egg and that Wde is required for nuclear localization of Egg .
In this study , we have analyzed the function of Wde , the Drosophila homolog of mAM/MCAF1 , in development . Wde precisely colocalizes with Egg and the mutant phenotypes of wde and egg are indistinguishable , indicating that Wde is an indispensable binding partner of Egg required for trimethylation of H3K9 at euchromatic sites , in particular on the fourth chromosome . Functional data on mAM/MCAF1 have so far only been obtained by RNAi-mediated knock-down [9] , or by expression of mutated mAM/MCAF1 proteins in tissue culture cells [27] . The first study concluded that mAM/MCAF1 increases the enzymatic HMT activity of SETDB1 , in particular with respect to the conversion of H3K9me2 to H3K9me3 [9] . The second study showed that expression of mAM/MCAF1 mutated in its binding site for MBD1 interferes with recruitment of SETDB1 to chromatin [27] . Our study is the first using a null mutant of a mAM/MCAF1 homolog und our results clearly show the strict requirement for Wde for proper localization and in vivo function of Egg in germ line cells . It was shown before that mAM/MCAF1 and SETDB1 associate in a protein complex and that a short region of mAM/MCAF1 including the coiled-coil domain is sufficient for binding to SETDB1 [9] , [27] . We have confirmed and extended these observations by showing that a region including the coiled-coil domain of Wde is sufficient for binding to Egg and that a short region of Egg ( aa 366–521 ) devoid of any known protein motif is sufficient for binding to Wde . For mAM/MCAF1 it was proposed that its binding to SETDB1 alters the catalytic activity and substrate specificity of the histone methyl transferase domain , thus allowing efficient trimethylation of H3K9 [9] . While the same may be true for the Wde/Egg interaction , our results show that in the absence of Wde , Egg is hardly detectable in germ line cells , most likely because Wde is required to protect Egg from proteolytic degradation . Moreover , when Egg is overexpressed in the absence of Wde , it fails to localize to the nucleus , revealing an additional function for Wde in nuclear import of Egg . On polytene chromosomes Wde binds strongly to the fourth chromosome and to multiple euchromatic bands on all other chromosomes . Strong binding to the fourth chromosome has also been reported for Egg [35] and is consistent with the hypothesis that Egg may be specifically required for euchromatic H3K9 trimethylation on the fourth chromosome , which is not affected in Su ( var ) 3–9 and G9a mutants [18] , [39] . Two recent studies showed indeed that Egg specifically affects the transcription of loci located on chromosome 4 [35] , [37] . However , the two studies come to apparently contradictory results . While the first study [35] reported derepression of transgenes inserted on chromosome 4 in egg mutants , the second study [37] reported a general reduction of the transcription of genes on the fourth chromosome in egg mutants , measured in a microarray experiment . Nonetheless , the involvement of both Egg and Wde in the transcriptional regulation of genes on chromosome 4 appears very likely , since both Egg [37] and Wde ( this study ) bind to POF . We could not determine whether Wde and Egg bind to POF independently or sequentially , because we cannot exclude that the expression of endogenous Egg in S2 cells contributes to the binding of transfected Wde and POF . POF is a unique example of a protein that specifically associates with a single autosome , the fourth chromosome of Drosophila melanogaster [41] , [45] . In pof mutants , the transcription level of genes on the fourth chromosome is reduced , indicating that POF promotes transcription of genes on chromosome 4 [40] . On the other hand , the localization of POF to chromosome 4 is dependent on HP1 and vice versa , and there appears to be competition between these two proteins for binding to genes and their promoters on chromosome 4 [40] , [46] . These observations have led to the model that the activities of HP1 and POF have to be balanced in order to ensure transcription of genes on chromosome 4 at the right level [40] , [46] . We propose that Egg and Wde are part of this balancing mechanism because both proteins bind to POF and recruit HP1 by generating H3K9me3 marks on chromosome 4 . Two recent studies showed that Egg is required for the development of ovaries in Drosophila . Ovaries of homozygous egg mutant females are rudimentary and degenerate by apoptosis before egg chambers bud off the germarium [36] , [38] . We have confirmed this result and have shown that homozygous wde mutant females show exactly the same phenotype . From these observations it was not clear whether the function of Egg and Wde is required in the germ line cells , the somatic follicle cells , or both . To address this question , we eliminated the function of egg and wde in germ line cells by FLP/FRT mediated mitotic recombination . Egg chambers with egg or wde germ line clones did develop up to stage 8 of oogenesis , but subsequently degenerated due to apoptosis . Because the ovary phenotype of homozygous mutant egg and wde females was more severe than the germ line clone phenotype of mutants in both genes , we conclude that wde and egg may also be required for proper development of somatic follicle cells . It has been speculated that Egg may be dispensable for trimethylation of H3K9 at later stages of oogenesis because this function could be taken over by Su ( var ) 3–9 [38] . However , this hypothesis is not consistent with the different localization of the Wde/Egg complex and Su ( var ) 3–9 on salivary gland polytene chromosomes and with the different consequences of the respective mutations on H3K9 methylation in pericentric heterochromatin and euchromatin , in particular on chromosome 4 [18] , [35] , [37] . Furthermore , mutations in wde and egg lead to apoptosis of germ line cells , which obviously cannot be rescued by the presence of Su ( var ) 3–9 which is already expressed in the germ line at the time when apoptosis starts . Modification by SUMOylation and binding to SUMO is a common hallmark of many chromatin regulators involved in transcriptional repression [32] . Both mAM/MCAF1 and SETDB1 can bind SUMO and it has been suggested that this property is required for the recruitment of these proteins to promoters bound by transcriptional repressors such as KAP1 , Sp3 and MBD1 [33] , [34] , [47] , [48] . Our finding that Egg is itself modified by SUMOylation suggests that binding of additional chromatin modifiers to SUMOylated Egg may contribute to the efficient assembly of higher order chromatin repression complexes at specific euchromatic sites .
The following stocks were used in this study: P{EP}EP2024 ( Szeged Drosophila Stock Center ) , P{XP}CG12340d06198 , P{XP}d03942 , P{XP}d01917 ( Exelixis collection at Harvard ) , P{EPgy2}CG12340EY00884 ( #15045 ) , Df ( 2R ) 27 ( #8109 ) , daughterless-GAL4 ( #5460 ) , engrailed GAL4 ( #6356 ) , tubulin-GAL4 ( #5138 ) , mat67-GAL4 ( #7062 ) , P{w+FRTG13}GFP ( #5826 ) , P{neoFRT40A} P{w+FRTG13} ( #8217 ) , y w hsFlp; Sco/CyO ( #1929 ) ( Bloomington Drosophila stock center , stock # given in parentheses ) , egg1473 [36] . A chromosome doubly mutant for wde06198 and egg1473 was generated by meiotic recombination . The wdeTD63 null allele was generated by FLP/FRT mediated recombination in trans of the P{XP}d03942 and P{XP}d01917 P element insertions [49] . Expression of UASP-GFP-Wde , UASP-Egg-RFP , UASP-POF-RFP and of endogenous Wde from the P insertion P{EP}EP2024 [50] in transgenic flies was done with the UAS-GAL4 system [51] . Germ line clones for wde and egg were generated as described using a heat shock promoter driven flippase on the X-chromosome [43] . Transgenic fly lines for the constructs pUASP-GFP-Wde , pUASP-Egg-RFP and pUASP-POF-RFP were generated as described [52] . To obtain homozygous wde mutant adults , living embryos lacking GFP fluorescence derived from the CyO[twi::GFP] balancer chromosome were separated from their GFP positive siblings under a fluorescence stereo microscope and raised in separate vials . Antibodies against Wde were generated by immunizing two rabbits with the following peptides: DKPKKISDRERNPGS ( aa 70–84 ) and RSENTPPPASRLRYSH ( aa 1286–1301 ) . Final bleeds of both rabbits were pooled and affinity purified against the peptide corresponding to aa 70–84 ( Eurogentec , Seraing , Belgium ) . For immunohistochemical stainings of embryos , ovaries and salivary gland polytene chromosomes the following antibodies were used: rabbit anti Wde , affinity purified ( see above ) , 1∶5000 ( 1∶500 for polytene chromosomes ) ; rabbit anti POF , 1∶1000 ( 1∶400 for polytene chromosomes ) [41]; rabbit anti activated Drice , 1∶2500 [44]; rabbit anti SUMO , 1∶500 [53]; rabbit anti H3K9me2 , 1∶500 ( Upstate 07–441 ) ; rabbit anti H3K9me3 , 1∶500 ( Upstate 07–442 ) ; rabbit anti GFP , 1∶1000 ( Abcam ab65556 ) ; mouse anti GFP 3E6 , 1∶1000 ( Invitrogen ) ; mouse anti GFP , 1∶1000 ( Roche 11814460001 ) ; mouse anti HP1 C1A9 , 1∶25 ( DSHB ) ; mouse anti Orb 4H8 , 1∶25 ( DSHB ) . Secondary antibodies conjugated to Cy2 , Cy3 ( Jackson Laboratories ) and Alexa 647 ( Invitrogen ) were used at 1∶400 . DNA was stained with DAPI . Ovaries were fixed in 4% formaldehyde/PBS and stained according to standard procedures . Embryos were fixed and stained as described [54] . Polytene chromosomes were prepared and stained as described [55] . Images were taken on a Zeiss LSM 510 Meta confocal microscope and processed using Adobe Photoshop . Lysates of S2 cells were prepared in TNT buffer ( 150 mM NaCl; 50 mM Tris-Cl pH 8 , 0; 1% Triton X-100 ) supplemented with protease inhibitors ( Roche ) . Western blots and coimmunoprecipitations were done as described [56] . For Western blots , the following antibodies were used: rabbit anti Wde , affinity purified , 1∶1000; rabbit anti POF , 1∶3000 [41]; rabbit anti Sumo , 1∶5000 [53]; rabbit anti actin A2066 , 1∶1000 ( SIGMA ) ; mouse anti GFP , 1∶1000 ( Roche 11814460001 ) ; mouse anti HA 12CA5 , 1∶1000 ( Roche ) . | Germ line cells are the only cells in an organism that are able to transmit their genetic material to the next generation by forming eggs or sperm . They do not participate in the formation or function of tissues and organs and therefore show a unique pattern of transcription , with many genes being silenced that are only required for somatic functions . The covalent modification of histones by methylation , acetylation , and other mechanisms is crucial for these global alterations in the transcriptional program . Among the modifications involved in silencing of chromatin regions , methylation of histone 3 lysine 9 ( H3K9 ) is among the most important ones . Methylation of this residue in Drosophila is controlled by three different histone methyl transferases , but only one of these , dSETDB1/Eggless , is essential for viability and fertility of the fly . Here we describe an essential cofactor for dSETDB1/Eggless that is specifically required in germ line cells for their survival . This cofactor , that we called Windei , binds to dSETDB1/Eggless and recruits it to the nucleus . Null mutations in windei show strongly reduced trimethylation of H3K9 in germ line cells , demonstrating that Windei is one of the factors required for controlling chromatin organization in the germ line . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"developmental",
"biology/stem",
"cells",
"cell",
"biology/nuclear",
"structure",
"and",
"function",
"cell",
"biology/cellular",
"death",
"and",
"stress",
"responses",
"cell",
"biology/developmental",
"molecular",
"mechanisms",
"genetics",
"and",
"genomics/chromosome",
"biology",
"genetics",
"and",
"genomics/epigenetics",
"genetics",
"and",
"genomics"
] | 2009 | Windei, the Drosophila Homolog of mAM/MCAF1, Is an Essential Cofactor of the H3K9 Methyl Transferase dSETDB1/Eggless in Germ Line Development |
Hookworm infection is a major cause of disease burden for humans . Recent studies have described hookworm-related immunosuppression in endemic populations and animal models . A Tissue Inhibitor of Metalloproteases ( Ac-TMP-1 ) has been identified as one of the most abundant proteins released by the adult parasite . We investigated the effect of recombinant Ac-TMP-1 on dendritic cell ( DC ) and T cell function . Splenic T cells from C57BL/6 mice injected with Ac-TMP-1 showed reduced proliferation to restimulation with anti CD3 or bystander antigens such as OVA . Incubation of bone marrow-derived DCs with Ac-TMP-1 decreased MHC Class I and , especially , Class II expression but increased CD86 and IL-10 expression . Co-incubation of splenic T cells with DCs pulsed with Ac-TMP-1 induced their differentiation into CD4+ and , particularly , CD8+ CD25+Foxp3+ T cells that expressed IL-10 . These cells were able to suppress proliferation of naïve and activated CD4+ T cells by TGF-Β-dependent ( CD4+ suppressors ) or independent ( CD8+ suppressors ) mechanisms . Priming of DCs with non-hookworm antigens , such as OVA , did not result in the generation of suppressor T cells . These data indicate that Ac-TMP-1 initiates the development of a regulatory response through modifications in DC function and generation of suppressor T cells . This is the first report to propose a role of suppressor CD8+ T cells in gastrointestinal helminthic infections .
The human hookworms Necator americanus and Ancylostoma duodenale are directly transmitted nematode parasites of the small intestine , and the main species that cause human hookworm infection , a leading cause of iron-deficiency anemia and malnutrition with a prevalence of 600 million cases in the tropical developing world [1] . Though mortality is rare , the global burden of hookworm disease is high , with an estimated 22 million Disability-Adjusted Life Years ( DALYs ) lost each year [1] . These numbers alone outrank diseases such as African trypanosomiasis , dengue , Chagas' disease , schistosomiasis , and leprosy [2] . For many common helminthic infections , including ascariasis , trichuriasis , and schistosomiasis , the intensity of infection peaks during childhood and adolescence [3] . In contrast , there appears to be considerable variation in the age profile of hookworm infection . Although the hookworm burden may be heavy in children , especially those in sub-Saharan Africa [4] , [5] , the most commonly recognized pattern is a steady rise in the intensity of infection during childhood , with either a peak or a plateau in adulthood . This lack of exposure or age-related immunity indicates that hookworms can either evade or suppress host immune responses . Studies performed by us and others have confirmed that hookworm infections decrease the ability of the immune system to respond to hookworm and bystander antigens , as evidenced by decreased lymphocyte responses in hookworm-infected humans [6] , [7] , [8] , dogs [9] and hamsters [10] , [11] , as well as elevated serum IL-10 and immunosuppression in patients infected with N . americanus [12] , or infected and exposed to adult parasite extracts [13] . Chemotherapy against the parasite restores the immune response in humans [14] and increases the immunogenicity of anti-hookworm vaccines in hamsters [10] , [11] . Most of the pathology caused by the hookworm results from the adult stage of the parasite [15] , [16] . While feeding , adult worms release into host tissues a battery of pharmacologically and immunologically active molecules [17] . Work by several groups has begun to unravel the biochemical events linked to the resultant blood loss that develops as a consequence of parasite attachment [18] . Among the secreted antigens , a hookworm-secreted Tissue Inhibitor of Metalloproteases ( Ac-TMP-1 ) has been identified in A . caninum [19] and A . ceylanicum [20] as one of the most abundant proteins released by the adult parasite , at a rate of 40 ng/h [19] . In this report , we aimed to investigate the effect of the recombinant protein Ac-TMP-1 on dendritic cell function ( DC ) and generation of suppressor T cells . Splenic T cells from mice treated with Ac-TMP-1 exhibited decreased lymphoproliferative responses when restimulated ex vivo with Ac-TIMP or anti CD3 . To understand the mechanism behind this suppression of proliferation , we incubated bone marrow-derived dendritic cells ( DCs ) from C57BL/6 mice ( B/6 ) with Ac-TMP-1 , and discovered that DCs exposed to the hookworm antigen decreased expression of MHC Class I and II and increased expression of CD86 and IL-10 , as well as production of TGF-Β . Moreover , co-incubation of naïve splenic T cells with DC pulsed with Ac-TMP-1 induced their differentiation of T cells into IL-10 producing CD4+ and CD8+ CD25+Foxp3+ regulatory T cells that suppressed proliferation of both naïve and activated CD4+ T cells . Interestingly , neutralization of the cytokine TGF-Β reduced the suppressive ability of Ac-TMP-derived CD4+ T cell suppressors , but did not affect the ability of the CD8+ cells to suppress proliferation . Because CD4 , but particularly CD8+ T cells are abundant in the gut ( the site of hookworm infection ) we propose a novel mechanism of immunosuppression by a parasitic helminth .
Recombinant Ac-TMP-1 was kindly provided by Dr . Bin Zhan and Dr . Peter Hotez at The George Washington University . To generate the recombinant protein , a cDNA encoding a putative tissue inhibitor of metalloproteinase was cloned from an Ancylostoma caninum adult hookworm cDNA library by immunoscreening with anti-hookworm secretory products antiserum . The protocol of the cloning and protein expression is described in detail elsewhere [19] . C57BL/6 ( B/6 ) mice were purchased from Taconic ( Germantown , NY ) . All mice were maintained in the Baker Institute Animal Care Facility under pathogen-free conditions . All animal studies were approved by the Institutional Animal Care and Use Committee at Cornell University . We simulated continuous exposure by injecting 50 µg Ac-TMP-1 or ovoalbumin ( OVA ) to C57BL/6 mice intraperitoneally , every 2 days , for a total of 8 days . Two days after the last injection , spleens were collected , T cells purified with enrichment columns ( R& D Systems , Minneapolis , MN ) , labeled with CFSE as described [10] and restimulated with 50 µg Ac-TMP-1 or 5 µg/ml anti CD3 . Cells were harvested 5 days later . Proliferation was assessed by loss of CFSE staining . Bone marrow-derived DCs were cultured in the presence of 20 ng/ml GM-CSF and collected 6–8 days after culture . DCs were then plated in 6-well plates ( 106/well ) before Ac-TMP-1 was added to the wells . At different time points , brefeldin A ( 10 µg/ml ) was added for 6 h and DCs were then collected and fixed in 4% paraformaldehyde . Prior to staining , cells were incubated with an anti-Fcγ III/II receptor and 10% normal mouse serum ( NMS ) in PBS containing 0 . 1% BSA , 0 . 01% NaN3 . Cells were permeabilized and stained for the surface markers CD11c ( clone 223H7 ) , CD80 ( clone 16-10A1 ) , CD86 ( clone GL1 ) , MHC Class I ( clone 28-14-8 ) and MHC Class II ( clone M5/114 . 15 . 2 ) and for the cytokines IL-12p40/p70 ( clone C17 . 8 ) and IL-10 ( clone JES5-16E3 ) . Incubations were carried out for 30 min on ice . Unless specified , all antibodies were purchased from BD Biosciences or eBioscience . The data were collected using a FACScalibur flow cytometer and analyzed in CELLQuest software ( Becton Dickinson , San Jose , CA ) . For each sample , at least 30 , 000 cells were analyzed . ELISAs for the detection of murine TGF-Β were carried out using antibodies from R&D systems ( Minneapolis , MN ) following the manufacturer's instructions . Splenocytes were purified from spleens of naive B/6 mice by mechanical disruption . Prior to co-culture , red blood cells were lysed for 10 minutes with cold ACK lysing buffer . T cells were enriched using columns as above , and added to cultures containing bone marrow-derived DCs ( 5 T cells: 1 DC ratio ) that have been left unstimulated or had been treated with Ac-TMP-1 ( 50 µg for 16 h ) . Forty-eight hours after initiation of the co-culture , brefeldin A was added for 6 h; cells were then collected and fixed in 4% paraformaldehyde . Prior to staining , cells were incubated with an anti-Fcγ III/II receptor and 10% normal mouse serum ( NMS ) in PBS containing 0 . 1% BSA , 0 . 01% NaN3 . Cells were permeabilized with saponin and stained for the surface markers CD4 , ( clone RM4-5 ) , CD8 ( clone 53-6 . 7 ) , CD25 ( clone PC61 . 5 ) , the transcription factor Foxp3 ( FJK-16s ) , IFN-γ ( clone XMG1 . 2 ) and IL-10 ( clone JES5-16E3 ) . Incubations were carried out for 30 min on ice . All antibodies were purchased from BD Biosciences or eBioscience . The data were collected and analyzed by flow cytometry as described above . Suppressor cells ( CD4+ and CD8+ ) used for this assay were generated by incubation with bone marrow-derived DCs pulsed with 50 µg Ac-TMP-1 as described above . Following priming and seeding of DCs , they became adherent and cannot be easily collected from the experimental well . T cell purity was determined following collection by cytospins , Diff quick staining and quantitation under the microscope . The percentage of T cells collected from the wells containing DCs was >98% . Target cells ( CD4+ T cells ) were isolated from the spleens of B/6 using magnetic beads ( Miltenyi , Auburn , CA ) [10] . Naïve T cells were stained with CFSE and used immediately . Activated CD4 T cells were generated by culture with anti-CD3 ( 5 µg/ml ) for 3 days , when they were also labeled with CFSE . The CD4+ or CD8+ suppressor cells were treated with 0 . 8 µg/ml mitomycin C ( Sigma-Aldrich , St . Louis , MO ) to prevent their proliferation and added to the CFSE labeled target cells CD4+ T cells , at a 1∶1 ratio . To trigger proliferation of target cells , anti-CD3 ( 5 µg/ml ) and IL-2 ( 10 U/ml ) were added . Cells were harvested , and proliferation was measured at day 5 . Neutralization of IL-10 was performed employing anti-mouse IL-10 neutralizing antibody ( R&D systems ) at a concentration of 10 ng/ml . Neutralization of TGF-Β was carried out by adding anti-mouse TGF-Β neutralizing antibody ( R&D systems ) at a concentration of 25 µg/ml . Data are presented as mean±SD or SEM . Differences were analyzed for significance Student's unpaired , two-tailed t-test or ANOVA using Graph Pad Prism Software ( San Diego , CA ) . A P value less than 0 . 05 was used as the threshold for significance . Specific P values are indicated in each figure .
We wanted to explore the effect of Ac-TMP-1 exposure in a small animal model . The mouse is not a permissive model of hookworm infection , because the parasite cannot establish itself in the gut . Thus , we simulated continuous exposure by injecting 50 µg Ac-TMP-1 to B/6 mice intraperitoneally , every 2 days , for a total of 8 days . This regimen was substantiated by our knowledge of the parasite and the pharmacokinetics of the intraperitoneal route . An adult hookworm secretes 40 ng Ac-TMP-1/h [21] . Because adult infections typically range from 10–100 worms [22] , the level of Ac-TMP-1 in the interstitial fluid/serum at any given time should be 0 . 4–4 µg . Drugs given intraperitoneally have a half life in serum of 30–40 h , and a recovery of 1–10% of the original concentration . By injecting every 2 days , Ac-TMP-1 would always be at the maximum concentration in serum ( 0 . 5–5 µg ) , simulating infection conditions . As controls , we injected either PBS or the non-hookworm protein ovoalbumin ( OVA ) using the same regimen . Two days after the last injection , spleens were collected , T cells purified , labeled with CFSE and restimulated ex vivo with Ac-TMP-1 , OVA or anti CD3 for 5 days . Proliferation ( or lack thereof ) was assessed by determining the percentage of CFSE-positive cells . Figure 1A shows that unstimulated cells did not proliferate in culture ( 86–90% did not lose CFSE staining ) . Interestingly , mice did not proliferate in response to Ac-TMP-1 restimulation , including the mice that were primed with the antigen in vivo . As expected , T cells from control mice that had been injected with PBS or OVA proliferated in response to anti CD3 ( only 18–25% retained CFSE staining , P = 0 . 002 ) . Ex vivo proliferation to anti CD3 was decreased in mice treated with Ac-TMP-1 when compared to PBS-injected control animals ( 58 vs . 18% cells positive for CFSE , P = 0 . 002 ) . Most strikingly , proliferation to OVA was decreased in OVA-primed mice if cells were restimulated ex vivo in the presence of Ac-TMP-1 . This experiment was repeated 3 times and the average±SEM is shown in Figure 1B . These data indicate that in vivo treatment of mice with Ac-TMP-1 decreases the ability of their splenic T cells to initiate lymphoproliferative responses to the hookworm protein , or bystander antigens . In order to elucidate the mechanism ( s ) of immunosuppression by Ac-TMP-1 , we turned into in vitro models of DC-T cell interactions . To optimize in vitro conditions , bone marrow-derived DCs from B/6 mice were obtained and cultured in RPMI or in the presence of increasing doses ( 1–100 µg ) Ac-TMP-1 for 16 hours , when the cells were collected and fixed . The Mean Intensity of Fluorescence ( MFI ) in CD11c+ DCs expressing the co-stimulatory molecules CD80 and CD86 , as well as MHC Class I and II was analyzed by flow cytometry ( Figure 2A ) . In the presence of Ac-TMP-1 , MFIs for CD80 and CD86 were slightly increased , and the difference was statistically significant ( P = 0 . 02 ) for the latter activation marker if exposed to 50 µg Ac-TMP-1 . MFI values for MHC Class I decreased ( 78 in unstimulated cells vs . 51 in Ac-TMP-1-treated cells , although the difference was not statistically significant ) . In contrast , MHC Class II MFI was significantly downregulated ( P = 0 . 008 ) by incubation with 50 µg Ac-TMP-1 . Because this concentration was the dose at which the highest effect was observed , we generated a time-course curve in which the level of MHC Class II expression was detected at 1 , 6 , 16 and 72 h . In this experiment , we included a non-hookworm protein as a control , and incubated DCs with 50 µg/ml OVA . Figure 2B shows that the expression of the surface marker was already decreased since after 1 h incubation with the hookworm antigen and onwards . The downregulation of expression was statistically significant at 6 and 16 h post antigenic exposure when compared to RPMI-treated cells . No changes were observed in OVA-primed DCs . These results indicated that 50 µg Ac-TMP-1 was able to induce a biological effect on DCs in a period between 6–16 h . Intracellular staining for the cytokines IL-12p40/p70 and IL-10 was also determined in DCs stimulated with Ac-TMP-1 or OVA for 16 h ( Figure 2C ) . IL-12 expression was slightly increased following antigenic stimulation , although the difference with the RPMI-treated control was not statistically significant . In contrast , IL-10 expression was significantly increased ( P = 0 . 001 ) . Cytokine data assayed by ELISA confirmed the results ( not shown ) . Priming with OVA did not induce a cytokine response . Finally , the levels of the anti-inflammatory cytokine TGF-Β were determined in the DC supernatants by ELISA . The secretion of TGF-Β was increased in DC cultures incubated with Ac-TMP-1 , and unaffected by OVA . Together , these results indicate that in vitro treatment of bone marrow DCs with Ac-TMP-1 decreased their ability to present antigen and increased their ability to produce anti-inflammatory cytokines such as IL-10 and TGF-Β . For this experiment , splenic T cells were obtained from B/6 mice and co-cultured for 48 h with either unstimulated or Ac-TMP-1-treated bone marrow-derived DCs ( for 6 h with 50 µg Ac-TMP-1 ) . DCs alone produced negligible amounts of cytokines ( <2% , not shown ) . First , we determined the percentage of CD4+ and CD8+ T cells expressing the activation marker CD25 and the transcription factor Foxp3 ( expressed in regulatory T cells ) . Ac-TMP-1 priming increased Foxp3 expression in CD4+ T cells ( from 5% to 20% ) and especially , in CD8+ T cells ( from 6% to 56% ) ( Figure 3 ) , demonstrating that the hookworm antigen induced naïve T cells to become regulatory T cells . This finding was further confirmed by the study of cytokines . We determined the expression of IFN-γ and IL-10 in both the CD25+Foxp3− ( activated , non-regulatory T cells ) and CD25+Foxp3+ lymphocyte populations ( true regulatory T cells ) . Cytokine expression in unstimulated controls was <5% ( not shown ) . Ac-TMP-1 treatment induced CD4+ CD25+Foxp3− cells to express IFN-γ ( to 12% ) ; in contrast , IFN-γ expression in CD4+CD25+ Foxp3+ cells was only 4% . The frequency of both CD8+CD25+ Foxp3+IFN-γ+ and CD8+ CD25+Foxp3−IFN-γ+ cells was low ( 5% ) . Both CD4+ and CD8+ CD25+ T cells co-cultured with Ac-TMP-1 treated DCs expressed IL-10 ( 8% and 9% respectively ) . Interestingly , Ac-TMP-1-treated bone marrow-derived DCs induced the highest increase in IL-10 expression in CD4+ CD25+Foxp3+ cells ( to 16% ) and , most strikingly , in the CD8+CD25+Foxp3+ population ( 38% ) . These findings show that Ac-TMP-1-treated bone marrow-derived DCs selectively biased the differentiation of naive T cells , in particular CD8+ T cells , toward a regulatory phenotype via increased expression of the transcription factor Foxp3 and the cytokine IL-10 . For these experiments , target CD4+ T cells were purified from the spleens of naïve B/6 mice . Activated CD4+ T cells were generated by restimulation in vitro with anti-CD3 for 3 days . Both were labeled with CFSE , and plated . The suppressor T cells were generated by incubation of splenic naïve T cells with bone marrow-derived DCs pulsed with Ac-TMP-1 , OVA or RPMI ( unstimulated controls ) . Suppressor CD4+ and CD8+ T cells were then added to the target naïve or activated CFSE-stained CD4+ T cells . Co-cultures were then incubated with a mixture of anti CD3/IL-2 to enhance proliferation of target T cells in the presence or absence of neutralizing antibodies for IL-10 and TGF-Β . Negative proliferation was quantitated as % CFSE+ cells 5 days after initiation of co-culture . In the absence of suppressor T cells , >70% cells proliferated in response to antiCD3/IL-2 treatment ( Figure 4 ) ; the proliferation was unaffected by cytokine neutralization . Moreover , >70% of the naïve CD4+ T cells proliferated following co-culture with CD4+ and CD8+ T cells incubated with unstimulated DCs . Again , neutralization of IL-10 or TGF-Β did not produce any effect . Similarly , OVA-primed T cells were unable so suppress proliferation of splenic T cells . Ac-TMP-1 primed-CD4+ T cells were able to suppress the proliferation of naïve CD4+ T cells ( to 35% ) and activated CD4+ T cells ( to 25% ) , although the difference was not statistically significant when compared to unstimulated cells alone . Treatment of cultures with anti-IL-10 antibodies resulted in a decrease in the ability of Ac-TMP-1 primed CD4+ T cells to suppress proliferation ( from 35 to 48% CFSE+ cells , although the difference was not statistically significant . However , the decrease in suppressive ability in cultures where TGF-Β was neutralized was statistically significant ( P = 0 . 05 ) . Finally , CD8+ T cells primed with Ac-TMP-1 were more effective in suppressing T cell responses by significantly decreasing the ability of both naïve and activated CD4+ T cells to divide ( to 20% and 12% , respectively ) . The suppressive ability of these cells was not abolished by neutralization of either IL-10 or TGF-Β . These results demonstrate that DC priming with Ac-TMP-1 , but not other proteins , induced the generation of CD4+ and CD8+ suppressor T cells . In our system , CD8+ suppressor T cells were more efficient in reducing proliferation of both naïve and activated target CD4 T cells , and their ability to suppress was unaffected by neutralization of either IL-10 or TGF-Β , as opposed as the CD4+ suppressors , that required both cytokines , in particular TGF-Β .
Chronic infections with helminths have been suggested to induce suppressor cells by a variety of mechanisms . The published immunological and epidemiological data on hookworm infection in humans and animal models suggest that these parasites are particularly successful in establish chronicity and modulation [1] , [6] , [7] , [10] , [11] , [17] , [23] , [24] , [25] , [26] , [27] , [28] . Numerous helminth-derived proteins are though to contribute to the immunosuppression associated with these parasites [29] , [30] . Tissue inhibitors of metalloproteases themselves have proven to have immunomodulatory properties [31] , [32] , [33] , [34] . In this report , we have investigated the effect of the hookworm tissue inhibitor of metalloproteases Ac-TMP-1 , one of the most abundant proteins released by the parasite following establishment , on DC function and T cell differentiation . We have demonstrated that recombinant Ac-TMP-1 is able to induce bone marrow-derived DCs to downregulate MHC molecules and release anti-inflammatory cytokines . More importantly , DCs pulsed with Ac-TMP-1 promoted the development of regulatory T cells . Remarkably , CD8+ suppressor T cells were more abundant , more potent , and used different suppressive mechanisms than CD4+ T cells . It is considered that the immature developmental stages of DC differentiation produce tolerogenic DCs which in turn induce T cell anergy or regulatory T cells [35] . The controlled environment of the in vitro experiments performed by us revealed that bone marrow-derived DCs decreased their ability to present antigen ( by downregulating MHC Class I and , especially class II expression ) and increased their ability to produce the anti-inflammatory cytokines IL-10 and TGF-Β . This phenotype is consistent with the development of tolerogenic DCs [36] . Thus , the initiation of suppressive responses in hookworm infectious may be initiated by an increased frequency in the tolerogenic DC population in the sites where the antigen is released . How CD8+ suppressor T cells generate after the first interaction with DCs is still unknown . Whereas downregulation of MHC Class I has been implicated in the generation of suppressor T cells by some , others have proposed that their generation do not require MHC mechanisms , or that it may be caused by the recognition of other ligands , such as CD40L [37] , [38] , [39] , [40] . While our studies demonstrate the generation of a suppressive population of T cells , the exact mechanism whereby these cells arise deserves further investigation . Studies in human populations and animal models have suggested that adult hookworms are immunosuppressive . In fact , our experiments in animal models have revealed that peripheral blood or splenic cells from dogs and hamsters infected with the parasite do not proliferate in response to adult hookworm extracts [9] , [11] . Recently , the analysis of human responses to adult hookworm extracts demonstrate that restimulation of peripheral blood cells with adult proteins causes an increase in IL-10 production [13] . The in vivo studies presented here support this hypothesis and demonstrate that priming mice with the abundant adult extract protein Ac-TMP-1 results in a decrease their lymphoproliferative responses to TCR stimulation . More importantly , they also revealed that exposure to Ac-TMP-1 also diminished specific proliferation to bystander antigens , such as OVA , suggesting that the hookworm antigen is able to cause potent , generalized immunosuppression . This is further demonstrated by the fact that mice exposed to Ac-TMP-1 are unable to initiate lymphoproliferative responses to the antigen . Interestingly , dogs vaccinated against Ac-TMP-1 did not develop proliferative responses against the hookworm antigen ( unpublished ) . Our in vitro experiments attempt to begin to unravel the underlying mechanism of immunosuppression and revealed that priming DCs with Ac-TMP-1 induced the de novo generation of CD4+ , and more importantly , CD8+ Foxp3+ IL-10+ cells . Both cell populations are able to display suppressive functions when co-cultured with both CD4+ naïve and activated T cells . Interestingly , the mechanism of suppression by CD4+ suppressor T cells seems to be mediated by the release of anti-inflammatory cytokines , whereas CD8+ suppressor T cells do not require the presence of IL-10 or TGF-Β to suppress T cell proliferation , suggesting perhaps direct cell contact mechanisms . The existence of suppressor CD8 populations has been documented in different models [41] , [42] , although their role and mechanism of suppression remains poorly characterized . Whereas some authors postulate that their suppressive function is dependent on IL-10 production , some others demonstrate that their function is cytokine independent ( i . e . [38] , [39] ) . Although our data support the latter hypothesis , further experiments need to be performed to determine the mechanism of immunosuppression in Ac-TMP-1-primed suppressor T cells , and the relevance of such mechanism in infection models . Because the role of CD8+ suppressor T cells has been postulated as a very important homeostatic mechanism in the gut mucosa [43] , the infection site for hookworms , the demonstration of the role of CD8+ suppressor T cells in the regulation of gut immunity will be of importance not only to bring forward the role of these cells in the mucosal environment , but also to enhance the relevance of this effector population in the context of gastrointestinal parasitic infections . CD8 suppressor T cells have been generated in vitro in response to other extracellular nematodes such as Echinococcus multilocularis protoscoleces [44] , but this is the first report in which they have been implicated in the immune response against gastrointestinal nematodes . In summary , our data demonstrates that Ac-TMP-1 modulates the immune response of the host by more than one mechanism ( s ) . This hookworm molecule appears to induce de-activation of the DC and enhance IL-10 production , and to elicit the development of T cells with regulatory functions . These findings open the door to future studies to determine the nature of the interaction of the hookworm antigen with antigen presenting cells , as well as to investigate the relevance of suppressor T cells in helminthic infections , and their mechanism of suppression . | Chronic infections with helminths have been suggested to induce suppressor cells by a variety of mechanisms . The published immunological and epidemiological data on hookworm infection in humans and animal models suggest that these parasites are particularly successful in establishing chronicity and modulation . We have demonstrated that the recombinant form of a hookworm tissue inhibitor of metalloprotease ( Ac-TMP-1 ) induces bone marrow-derived DCs to downregulate MHC Class I and II and release anti-inflammatory cytokines such as IL-10 and TGF-Β . More importantly , DCs pulsed with Ac-TMP-1 promote the development of regulatory CD4+ and , especially , CD8+ T cells from naïve T cells , which are able to suppress proliferation of naïve and activated splenic CD4+ T cells; this suppression is mediated by TGF-Β for CD4+ suppressor cells , but it is independent of the cytokine for CD8+ suppressor cells . These studies initiate the first steps to investigate the relevance and nature of suppressor T cells in hookworm infections and their mechanism of suppression . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"immunology/immunomodulation"
] | 2009 | The Hookworm Tissue Inhibitor of Metalloproteases (Ac-TMP-1) Modifies Dendritic Cell Function and Induces Generation of CD4 and CD8 Suppressor T Cells |
Chronic wasting disease ( CWD ) , an environmentally transmissible , fatal prion disease is endemic in North America , present in South Korea and has recently been confirmed in northern Europe . The expanding geographic range of this contagious disease of free-ranging deer , moose , elk and reindeer has resulted in increasing levels of prion infectivity in the environment . Soils are involved in CWD horizontal transmission , acting as an environmental reservoir , and soil mineral and organic compounds have the ability to bind prions . Upper horizons of soils are usually enriched with soil organic matter ( SOM ) , however , the role of SOM in prion conservation and mobility remains unclear . In this study , we show that incubation of PrPCWD with humic acids ( HA ) , a major SOM compound , affects both the molecular weight and recovery of PrPCWD . Detection of PrPCWD is reduced as HA concentration increases . Native HA extracted from pristine soils also reduces or entirely eliminates PrPCWD signal . Incubation of CWD prions with HA significantly increased incubation periods in tgElk mice demonstrating that HA can reduce CWD infectivity .
Chronic wasting disease ( CWD ) , a fatal prion disease affecting free ranging white-tailed deer , mule deer , elk , moose and reindeer as well as farmed cervids , is the only spongiform encephalopathy ( along with scrapie ) that is environmentally transmitted . Currently , CWD-infected cervids are present in 3 provinces of Canada , 25 US states , South Korea , Norway and Finland , and its geographic range continues to expand among free-ranging cervids . Evidence suggests that horizontal transmission of CWD involves soils as an environmental reservoir of infectivity [1–4] . Soil-related CWD transmission in cervids may occur orally [5] , although intranasal and aerosol routes of transmission are also possible [6 , 7] . The CWD agent may enter the soil via alimentary secretions , blood , and decomposing infected carcases [8] . While prion infectivity persists in the soil for a few years in laboratory conditions [9] , infectivity of soils in CWD-endemic regions remains largely unknown . Different soil compounds , mineral and organic , can differentially bind prions and change their infective properties [1] . Montmorillonite ( mte ) mineral particles bind prions avidly and increase their infectivity [10]; kaolinite ( kte ) and quartz microparticles ( common soil minerals ) also may increase disease transmission [11] . Upper horizons of soils are usually enriched with soil organic matter ( SOM ) but the role of SOM in prion conservation and mobility remains unclear . SOM is defined as all biologically derived organic matter that resides within the soil matrix and is divided into living and non-living components [12] . Humus is the most important and specific part of SOM; it is a mixture of amorphous organic materials that contains identifiable biomolecules ( e . g . polysaccharides , lipids , proteins etc . ) and non-identifiable molecules ( humic substance ) [13] . Humus is an extraordinarily complex , molecularly flexible material that can be fractionated into specific humic substances: fulvic acids ( FA ) , humic acids ( HA ) , and humin . Humic acids are comprised of weak aliphatic ( carbon chains ) and aromatic ( carbon rings ) organic acids which are insoluble in water under acid conditions , but soluble in water under alkaline conditions . Humic acids have average molecular weights varying from 10kDa-103kDa for soil-derived material [14] . They are considered to be flexible linear polymers that exist as random coils with cross-linked bonds . In water solution , the HA are large , dynamic supramolecular associations , held together by hydrophobic interactions , which are easily disrupted and capable of exhibiting micellar properties [15] . Due to the wide variety of soil types , composition and amount of SOM differs between locations . For example , in western Canada , there are at least 12 great groups of soils [1] and each is comprised of different amounts and types of organic compounds . SOM content in the upper soil horizons vary from 0 . 5% in the surface horizon of Regosolic soils , to 30% in the LHF ( plant litter ) horizon of most soils in the northern part of Canada . The ratio of different humic compounds also varies across soil types . In soils where the vegetative ( biologically productive ) period is longer , the humus has more HA than FA ( e . g . in Chernozems ) compared to soils that are less biologically active ( due to non-optimal temperatures and moisture regimes ) , where the humus contents are mostly FA ( e . g . in Luvisols ) [16 , 17] . For Chernozemic soils of Northern American CWD-endemic regions , HA abundance was estimated between 0 . 75%-1 . 4% [18] . In Luvisolic and Brunisolic soils , the upper horizon contains high amounts of SOM , but insignificant amounts of HA [19]; for boreal region soils , the HA content is estimated as 0 . 1–0 . 7% [16] . Since most of SOM is concentrated in the surface soil horizon and HA is a major component of SOM , it likely plays an important role in the fate of shed prions . Although the absolute concentration of HA in soil is significantly less than the abundance of mineral particles , HA is more biologically and chemically active and can adsorb on mineral particles creating films on the surfaces and masking them . Analyses of HA-prion interactions , and their impact on CWD infectivity , are important factors in determining the fate of PrPCWD in soil environments . While a series of studies have investigated prion binding to soils varying in SOM content , the direct interaction of HA with PrPCWD has not been studied [20 , 21] . Studies using non-infectious recombinant PrP ( recPrP ) indicated that HA and other SOM compounds have a strong affinity to the recPrP [22–26] . However , due to difference in structure of infectious and non-infectious prions , they may interact differently with SOM compounds . It has previously been shown that hamster prions , when incubated with low concentrations of HA , remain infectious to hamsters with little biological impact on the incubation period [27] . The objective of this study was to determine , by examining a more complete range of HA concentrations , how interactions with HA affects PrPCWD and resulting infectivity .
The interactions between prions and HA were first analyzed using commercially available pure HA . HA levels used in this study mimicked levels naturally present in soils ( 1g L-1-25g L-1 ) . Humic acids and brain homogenates ( CWD infected and uninfected controls ) were incubated overnight . PrPCWD signal decreased with increasing HA concentration with the ~35 kDa molecular weight band ( diglycosylated PrP ) not visible at the highest concentration ( 25 g L-1 ) of HA ( Fig 1 ) . The quantity of PrP signal for Fig 1A declined to 33% following incubation with 2 . 5 g L-1 HA , and for 25 g L-1 HA the decline was to 5% ( Fig 1B ) . Due to the variability of HA amount and composition in soils , the effects of native soil HA on PrPCWD might vary . Native HA were extracted from different soils from boreal , prairie and mountainous regions of western Canada ( S1 Table ) . The amount of extracted HA varied amongst the soils . As expected , HA extracted from surface humic Ah horizon of Chernozemic soils had maximal concentrations of HA , 19–22 g L-1 , and the lowest concentrations of HA were in the eluvial horizon Ae of Luvisol from the boreal ecozone , 0 . 2 g L-1 . All soil-extracted HA decreased PrPCWD signal but to differing levels ( S2 Fig ) . Stronger signal was observed with lower concentrations of HA ( e . g . in the LF horizons of Luvisol and Brunisol ) , while the weakest signals were observed after incubation with higher concentrations of HA , ( similar to the results with commercially available pure HA ) . LF horizons have the highest amount of total organic carbon ( 34–38% ) but are a negligible source of HA because they are composed primarily of degraded materials; the non-decomposed organic fraction has minor chemical importance because its intact structure exhibits a relatively small surface area [28] . To determine if HA from different soils has similar effects on reducing PrPCWD signal , we normalized the HA concentration to 20 g L-1 . After incubation with normalized soil HA , PrPCWD signal was also reduced ( Fig 2 ) , and differences between HA in different soils was not observed . Pure HA included as a control at the same concentration ( 20 g L-1 ) showed a similar effect on PrP signal . In all cases , the ~30kDa PrP band was degraded , and 22–23 kDa protein remained . There are several potential explanations for the observed decline in PrPCWD signal upon incubation with humic acids . One possibility is that a component of SOM interferes with PrPCWD during immunoblotting which could alter migration , limit entry into the gel , or interfere with membrane transfer [27] . We found that PrPCWD migrated as expected ( 20–35 kDa ) with or without HA , thus altered migration during SDS-PAGE is not the reason for decreasing PrP signal . Staining the membranes with Coomassie Blue showed that the majority of proteins entered into the gel and did not remain in the wells ( S3 Fig ) , indicating there is no effect of entry into the gel . The observed loss of PrP signal with increasing HA concentration could also result from the encapsulation of PrPCWD impacting prion migration , alternatively HA may cause partial degradation of PrPCWD so that only a portion remains detectable . The exact mechanisms of these processes are still unclear [3] , but there is evidence to suggest that either is possible . Tomaszewski et al . [29] proposed negatively-charged HA could encapsulate positively-charged proteins and preserve their activity . Another study [25] has shown HA-like substances copolymerize with recPrP and irreversibly incorporate it into their structure , creating complexes which decrease the efficiency of recPrP recovery . The PrP insolubilization and co-precipitation by aggregation with HA without altering PrP secondary structure , which potentially can decrease prion detectability and reduce their bioavailability , also were discussed [26 , 27] . Similar prion interactions with another high-organic matrix ( e . g . compost ) revealed PrPCWD degradation following 230-days of composting [30] . We also assessed the effects of high concentration HA on PrPres ( S4A Fig ) . The results showed that HA degraded PrPres with mono- and unglycosylated forms degraded more rapidly . A similar effect of HA was found for deglycosylated PrPCWD ( S4B Fig ) . Incubation with 25 g L-1 HA decreased the intensity of the higher molecular band ( 25kDa ) but also affected the lower weight bands as shown by a decline in intensity of the 13-16kDa bands . Our initial bioassays challenged animals ( tgElk mice ) by the intracerebral route . Due to brain toxicity of HA , only the relatively low concentration ( 0 . 25 g L-1 ) was examined . Two CWD strains were studied , elk prions ( source: infected elk brain ) and white-tailed deer ( WTD ) prions ( source: tg mice ) . Control mice were incubated with uninfected NBH and NBH+0 . 25HA; they did not show clinical signs and were euthanized at the end of the experiment ( at 175dpi ) . Mice inoculated with 0 . 16% BH from CWD infected elk exhibited clinical symptoms earlier than tg-mice inoculated with 0 . 16% CWD infected elk BH + 0 . 25 g L-1 HA . The difference between treatments was 16 days ( 102±13 dpi vs 118±7 dpi ) suggesting a slight reduction in effective titer , but due to the overlap in incubation periods , the difference was not statistically significant . For WTD-CWD inoculum , the incubation periods were similar: 104±8 ( dpi ) without HA and 100±0 dpi with HA ( Fig 3A ) . This is supported by the results from immunoblotting where there is no reduction in PrPCWD intensity for these inoculums . Similar to the study of hamster prions/humic acids [27] , we identified a slight , biologically insignificant ( <1 log ) change in infectivity . To examine the impact of higher HA concentrations on CWD infectivity , mice were infected by intraperitoneal route . Elk prions were incubated with 1 g L-1 HA , 2 . 5 g L-1 HA and 25 g L-1 HA . Control reactions included uninfected brain homogenate as well as CWD-elk prions incubated without HA . Western blot analysis confirmed a loss of signal with increasing HA concentration ( Fig 3B ) . This result is similar to what we observed previously , where the strongest PrP signal was detected for the sample without humic acids , and the signal decreased with increasing HA concentration . Animal bioassay of the samples showed a concomitant decline in CWD infectivity with increasing concentration of HA . Significant differences ( p<0 . 001 ) between mice inoculated with CWD brain homogenate control ( no HA ) and infected brain homogenate treated with 1g L-1 HA ( Fig 3B ) ; the incubation period for BH ( control ) was 154±10 days post inoculation ( dpi ) , while for BH pre-incubated with 1 g L-1 HA it was 180±10 dpi . Only half of the mice inoculated with BH pretreated with higher HA concentrations ( 2 . 5 g L-1 and 25 g L-1 ) exhibited clinical signs within the 280-days ( starting at 167–188 dpi ) . We analyzed brains harvested from these mice and all were positive for PrPres ( after PK digestion ) by western blot . The remaining mice ( 5 out of 10 for 2 . 5 g L-1 and 5 out of 8 for 25 g L-1 ) were euthanized at 280 dpi without clinical signs . PrPres was not identified in their brains . Control animals inoculated with normal ( uninfected ) BH ( NBH ) pre-incubated with 25 g L-1 HA survived during the course of the bioassay ( >280 dpi ) . This bioassay clearly demonstrates a decline in CWD infectivity after incubation with higher concentrations of HA . Due to the complexity of soil , it is impossible to distinguish the influence of each compound separately because of their overlapping influence and sometimes different impact on prions . In soil surface horizons , mineral clay particles are usually covered with films of soil organic matter ( SOM ) ( or iron/manganese oxides , or carbonates depending on soil type ) . Thus detection of HA degrading ability on PrP is an important first step to understanding soil-prion interactions . Similar to previous studies [2 , 4 , 10 , 31 , 32] where prion interactions with pure minerals were investigated , our research will help to understand the complexity of soil-prion interactions . It is important to analyze how HA changes binding capacity of minerals and look in this direction for our future experiments . The adsorption of organic constituents by clay particles leads to the formation of organo-mineral complexes; they are abundant and common in soils , and adsorption of humic substances by clays has been extensively investigated [33 , 34] . Organic substances can bind to clays through a variety of mechanisms that depend on the properties of the organic compounds and the mineral surface . These interactions can change the adsorption capacity and the reactivity of minerals [35] . The surface chemistry of organo-mineral complexes is dominated by adsorbed organic matter that masks the properties of the supporting minerals to varying extents [36 , 37] . Therefore , the fate of prions in soil ( binding and mobility ) may be regulated primarily by interactions with organo-mineral complexes , and the effect of prion binding by organo-mineral complexes and its persistence in the soils needs additional study to determine its implications on the bioavailability of PrPCWD . This study shows that a common soil component , HA , altered the mobility and abundance of PrPCWD with a concomitant decline in CWD infectivity . A wide range of HA concentrations were tested , from 0 . 25 to 25 g L-1 , representing the spectrum of HA present in native soils . The lowest HA concentrations ( <1 g L-1 ) occur in boreal and tundra soils while high HA concentrations ( >20 g L-1 ) are present in soils of prairie grassland . We have shown that high concentrations of HA ( >2 . 5 g L-1 ) decrease both PrPCWD signal and prion infectivity . HA extracted from a variety of pristine soils also reduced PrPCWD signal , suggesting that a similar mechanism of prion-HA interaction can occur in different types of soils . The lack of detailed knowledge on the composition and structure of HA makes it difficult to identify the specific relationships between the structure and activity of these substances . Although we showed that native HA purified from different soils have a similar effect on prions , the soil diversity and complexity with varying mineral and organic compounds may affect the HA degradation ability , and enhance PrPCWD persistence in soil , and also may contribute to the migration of prions in the soil profile that could change prion bioavailability to grazing animals . This study further emphasizes the complexity of soil-prion interactions , where soil minerals bind prions and enhance infectivity while the organic compounds can degrade CWD-prions .
For HA-prion incubation experiments , we used commercially available HA ( Sigma-Aldrich , cat . #53680; referred to as pure HA ) , dissolved in deionized water at the following concentrations: 1 g L-1 , 2 . 5 g L-1 , and 25 g L-1 . These concentrations were chosen to reflect actual concentrations of HA in soils ( 1–2 . 5 g L-1 or less , northern boreal and tundra regions; 25 g L-1 , prairie Chernozemic soils ) . HA were also extracted from the surface horizons of 6 pristine soils: two Chernozems from the prairie region , a Luvisol and Brunisol from the boreal region , and a Brunisol from the mountainous region . Bulk soil samples from two upper horizons were collected at 6 sites in Alberta , Canada ( S1 Fig ) . Each site had native vegetation and soil profiles that were undisturbed . Soil samples were air-dried and sieved to collect material <2 mm . Selected soil characteristics are provided in S1 Table . Soil texture was determined by gravimetric method with hydrometer; mineral composition of clay fractions was detected by X-ray diffraction ( XRD ) analyses in EAS ( University of Alberta ) using the Rigaku Geigerflex powder diffractometer . The extraction followed the protocol recommended by the International Society of Humic Substances [38] . Briefly , soil was pre-incubated in 1N HCl followed by a multistep extraction procedure: ( i ) extraction with 0 . 1N NaOH at room temperature overnight; ( ii ) centrifugation to collect the supernatant; ( iii ) acidification of the supernatant with 2M HCl; ( iv ) precipitation of HA overnight; and ( v ) separation of HA ( precipitate ) from FA ( supernatant ) by centrifugation . The NaOH extraction followed by acidic separation was repeated until the solution was clear ( 2–3 more times ) to assure that all humic and fulvic acids were extracted . All centrifugation was carried out at 5000g for 10 minutes . Obtained HA pellets were combined and resuspended in water , resulting in varying concentrations of HA between collected soils . For incubation experiments with normalized HA concentrations , the HA pellets were weighed and re-suspended in water to adjust the concentration to 20 g L-1 . CWD agent was obtained from infected brain tissues of elk [39] , transgenic tgElk mice ( expressing 132MM elk PrP , #2045 from already-existing collection in CPPFD ) or infected transgenic tg33 mice ( expressing wild type white-tailed deer PrP , #1268 from previous existing collection in CPPFD ) . Uninfected controls were brains of uninfected transgenic tg33 and tgElk mice . Brain tissues were homogenized ( 10% w/v ) in water , and then were clarified at 800g for 5 min before experiments were initiated [4] . Identical amounts of 10% brain homogenate ( BHCWD , or uninfected: NBH ) were incubated with water ( control ) and HA ( 1g L-1 , 2 . 5 g L-1 and 25 g L-1 ) at 4°C . For all experiments ( unless otherwise specified ) , the BH and HA were incubated 24 hours . Following incubation , samples were analyzed by western blot: samples ( 10 μL ) were resolved on 12-well 12% NuPAGE bis-Tris gels ( Invitrogen ) , transferred to PVDF membrane and probed with anti-PrP antibody Bar 224 ( diluted 1:20 000; Bertin Pharma ) . Quantitative analyses of western blot images were performed using ImageJ software ( https://imagej . nih . gov/ij/index . html ) , which output the net intensity and area of each blot . Net intensities of the samples were normalized as a percent of the untreated controls run on the same gel . Proteinase-resistant PrPCWD was identified by digestion of 10 μL of brain homogenate with 3 . 5 μg of Proteinase K ( PK ) ( Roche ) for 45 minutes at 37°C in a volume of 50 μL ( 50 mg/ml PK final concentration ) . Digestion was terminated by addition of 10 μl of AESBF ( proteinase inhibitor , #A8456 Sigma-Aldrich ) . Deglycosylation of PrPCWD was performed with PNGase F kit ( #P0704S , BioLabs Inc . ) following manufacturer’s protocol ( https://international . neb . com/protocols/2014/07/31/pngase-f-protocol ) . All work with animals was performed in compliance with the Canadian Council on Animal Care Guidelines and Policies . All procedures involving animals were reviewed and approved by the Health Sciences Animal Care and Use Committee of the University of Alberta under protocol “Etiology and Pathogenesis of Prion Diseases” AUP # 914 . Infectivity of CWD incubated with HA was determined by intraperitoneal or intracerebral infection of tgElk mice . To test low concentration of HA , 44 tgElk mice were inoculated intracerebrally . The inocula ( 10% elk-CWD and WTD-CWD BH ) were incubated with 2 . 5 g L-1 HA , then diluted 10-fold to reach a non-toxic concentration of HA , pasteurized 10 min at 80°C , and 25 μL used to intracerebral inoculate tgElk mice . Intraperitoneal route was used to test higher concentrations of HA . For this experiment , BH ( 10% elk-CWD ) was incubated overnight with varying concentrations of HA ( 1g L-1 , 2 . 5 g L-1 and 25 g L-1 ) , pasteurized 10 min at 80°C , and 100 μL used to intraperitoneal inoculate tgElk mice . Equivalent amounts of BHCWD or uninfected BH ( incubated with 25 g L-1 HA ) were used as inoculate control . In both bioassays , mice were monitored daily for the onset of clinical symptoms and euthanized upon confirmed clinical disease . Brains from clinically positive mice and uninfected controls were analyzed for protease-resistant PrP ( PrPres ) by immunoblotting as described above . | Chronic wasting disease ( CWD ) is a contagious prion disease affecting several species of captive and wild cervids . Environmental prion contamination plays a major role in increasing incidence of CWD , with CWD infectivity being released into the environment by decaying carcasses , or shedding of biological fluids including urine , feces , and saliva . Horizontal transmission of CWD involves soils as an environmental reservoir of infectivity . Here , we tested the role of a soil organic matter compound , humic acid , for its ability to bind CWD prions and impact infectivity . A wide range of humic acid concentrations were examined representing the extensive spectrum of humic acid levels present in native soils . We found that incubation of CWD prions with high concentrations of humic acids ( >2 . 5 g L-1 ) decreases the both CWD-prion signal and infectivity , whereas lower levels of humic acids did not significantly impact protein stability or infectivity . Our study provides new insights into soil-prion interactions , prions persistence in soil , and their bioavailability to grazing animals . | [
"Abstract",
"Introduction",
"Results",
"and",
"discussion",
"Materials",
"and",
"methods"
] | [
"chronic",
"wasting",
"disease",
"animal",
"diseases",
"medicine",
"and",
"health",
"sciences",
"deer",
"canada",
"chemical",
"compounds",
"ruminants",
"prions",
"molecular",
"probe",
"techniques",
"geographical",
"locations",
"immunoblotting",
"vertebrates",
"animals",
"mammals",
"organic",
"compounds",
"north",
"america",
"animal",
"prion",
"diseases",
"molecular",
"biology",
"techniques",
"zoology",
"research",
"and",
"analysis",
"methods",
"infectious",
"diseases",
"zoonoses",
"proteins",
"chemistry",
"molecular",
"biology",
"people",
"and",
"places",
"biochemistry",
"eukaryota",
"organic",
"chemistry",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"amniotes",
"organisms",
"prion",
"diseases"
] | 2018 | Soil humic acids degrade CWD prions and reduce infectivity |
The double bromodomain and extra-terminal domain ( BET ) proteins are critical epigenetic readers that bind to acetylated histones in chromatin and regulate transcriptional activity and modulate changes in chromatin structure and organization . The testis-specific BET member , BRDT , is essential for the normal progression of spermatogenesis as mutations in the Brdt gene result in complete male sterility . Although BRDT is expressed in both spermatocytes and spermatids , loss of the first bromodomain of BRDT leads to severe defects in spermiogenesis without overtly compromising meiosis . In contrast , complete loss of BRDT blocks the progression of spermatocytes into the first meiotic division , resulting in a complete absence of post-meiotic cells . Although BRDT has been implicated in chromatin remodeling and mRNA processing during spermiogenesis , little is known about its role in meiotic processes . Here we report that BRDT is an essential regulator of chromatin organization and reprograming during prophase I of meiosis . Loss of BRDT function disrupts the epigenetic state of the meiotic sex chromosome inactivation in spermatocytes , affecting the synapsis and silencing of the X and Y chromosomes . We also found that BRDT controls the global chromatin organization and histone modifications of the chromatin attached to the synaptonemal complex . Furthermore , the homeostasis of crossover formation and localization during pachynema was altered , underlining a possible epigenetic mechanism by which crossovers are regulated and differentially established in mammalian male genomes . Our observations reveal novel findings about the function of BRDT in meiosis and provide insight into how epigenetic regulators modulate the progression of male mammalian meiosis and the formation of haploid gametes .
The bromodomain is a highly conserved motif that recognizes and binds to acetylated lysine residues , a key step in reading epigenetic marks . Among the bromodomain-containing proteins , the BET ( bromodomain and extra-terminal ) subfamily is characterized by the presence of two bromodomains ( hereafter referred to as BD1 and BD2 ) that bind to acetylated histones , and an extra terminal ( ET ) domain , which functions as a functional module for protein–protein interactions [1 , 2] . In mammals , there are four BET members , BRD2 , BRD3 , BRD4 and BRDT , and among them , BRD4 and BRDT are structurally more similar in that they also have a region of homology at the carboxyl-terminus , referred to as the C-terminal domain ( CTD ) . The BET proteins are epigenetic regulators with multiple functions in chromatin organization and transcriptional regulation . They direct the recruitment of diverse regulatory complexes to discrete regions in the genome , by mediating the tethering of protein complexes to acetylated histones and other proteins . In recent years , the BET proteins have received increasing attention due to their important implications in a wide range of human diseases [3] , and thus , for being therapeutic targets of BET protein inhibitors [4] . BRDT is expressed uniquely in the testis , in late prophase I spermatocytes and spermatids [5] . BRDT has been shown in the mouse model to be essential for the normal progression of spermatogenesis: loss of BD1 results in a truncated BRDT protein and produces defects in spermiogenesis [6] , with severely impaired chromatin organization in the spermatids [6 , 7] . Moreover , complete loss of BRDT function ( Brdt-/- ) produces an arrest of spermatogenesis in meiosis and the absence of post meiotic cells [8] . Studies in humans have identified polymorphisms in the BRDT gene that are significantly associated with impaired spermatogenesis and male infertility [9 , 10] , suggesting that BRDT may contribute to idiopathic male infertility and could also be a potential druggable target for male contraception [11] . Biophysical experiments have shown that BD1 of BRDT binds to nucleosomes through a simultaneous recognition of acetylated histone tails and DNA , and that the nonspecific interaction of BD1 with the DNA facilitates the recruitment of BRDT to bulk chromatin [12] . In meiotic and post-meiotic cells , BRDT regulates gene expression , possibly by forming complexes with the P-TEFb components , cyclin T1 and Cdk9 [8] . In post-meiotic cells , BRDT has also been shown to function in post-transcriptional regulation [13] , and allows the recruitment of chromatin remodeling complexes to promoters to regulate gene expression in a developmental-stage specific manner [14] . It has been increasingly proposed that epigenetic modifications play pivotal roles in modulating the progression of spermatocytes through the critical events of meiotic prophase I [15 , 16] . Chromatin structure and epigenetic modifications have been shown to define recombination hotspots in the genome [17 , 18] , recruitment of recombinases [19] , chromosomal interactions and segregation [20] , and silencing of the sex chromosomes in spermatocytes [21–23] . However , which proteins mediate the recognition of the epigenetic modifications in meiosis and how these readers contribute to modulate the progression of prophase I are not well understood . In the present study , we show that BRDT is an essential regulator of chromatin organization and reprograming during meiotic prophase . Loss of BRDT function disrupts the dynamics of chromatin modifications involved in maintaining the meiotic sex chromosome inactivation ( MSCI ) , affecting the synapsis and silencing of the X and Y chromosomes . We also demonstrate that BRDT modulates the global chromatin organization of spermatocyte chromosomes and the local histone modifications of the chromatin close to the synaptonemal complex ( SC ) . This function also influences the homeostasis of crossover ( CO ) localization and formation during pachynema , highlighting a possible epigenetic mechanism by which COs are regulated in mammalian genomes . Thus , BRDT may constitute a novel molecular pathway by which epigenetic regulators modulate the progression of male mammalian meiosis .
Although we have previously reported that the Brdt gene is highly expressed in meiotic prophase spermatocytes , to gain a better understanding of the role of BRDT in meiosis in particular , we examined the developmental stage and sub-cellular localization of BRDT protein during meiotic prophase I in detail . To this end , we immunolocalized BRDT protein along with SYCP3 , a marker of the axial elements ( AE ) of the synaptonemal complex ( SC ) , which permitted clear identification of the meiotic chromosomes and classification of the stages of prophase I [24] , in spermatocyte spreads from three 3 month-old wild type ( WT ) testes . BRDT protein was barely detectable , if at all , in leptotene and zygotene spermatocytes ( Fig 1A and 1G ) , but started to be detected at early pachynema , mainly in the chromatin of autosomes ( Fig 1B and 1G ) . By mid pachynema , BRDT was clearly detected throughout the chromatin of autosomes , but significantly less intensely in the sex chromosomes ( Fig 1C , inset; 1G , **p<0 . 01 ) . In late pachytene spermatocytes , BRDT protein was more robustly expressed as compared to earlier pachytene stages . That is , BRDT labeling appeared as a bright signal on the chromatin of autosomes and was slightly more intense in the X and Y chromosomes as compared to the sex chromosomes at early and mid pachynema ( Fig 1D , inset; 1G , *p<0 . 05 ) . This pattern persisted throughout diplonema , where BRDT was still strongly detected in the chromatin of autosomes and was significantly less intense in the chromatin of the X and Y chromosomes ( Fig 1E and 1G , **p<0 . 01 ) . At metaphase I , BRDT was barely observed in the chromatin , remaining primarily in the cytoplasm surrounding the chromosomes ( Fig 1F ) . ( Three 3 month-old WT mice , n = 10 leptotene , 10 zygotene , 20 early pachytene , and 30 mid pachynema , late pachynema , early diplonema and mid/late diplonema spermatocytes , respectively , per each mouse ) . In contrast to the defects in spermiogenesis seen in mice expressing a truncated protein lacking the BD1 of BRDT [6] , complete loss of BRDT function ( Brdt-/- mice ) results in a total absence of spermatids in adult testes [8] . Indeed , Brdt-/- testes exhibited a significant decrease in the number of cells bearing H3 phosphorylated at serine 10 ( H3S10Ph ) [8] , a mark associated with chromosome condensation during cell division [25] , suggesting that a blockage might be occurring at the end of the meiotic prophase [8] . To more specifically elucidate which meiotic cells are lost in the absence of BRDT , we performed TUNEL staining in testicular sections from three 2 month-old WT and Brdt-/- mice , respectively . In WT adult testis , few apoptotic spermatocytes were observed and they were mainly in stage XII and I tubules , and were identified as late pachytene and metaphase spermatocytes , and early pachytene spermatocytes and step 1 round spermatids , respectively ( S1 Fig ) . However , in Brdt−/− testes , TUNEL-positive spermatocytes were observed in all stages of tubules with pachytene and diplotene spermatocytes , with the most notably increased numbers in stages IX to XII ( p<0 . 001; S1 Fig ) and identified as mid pachytene to diplotene spermatocytes . The onset of apoptosis observed in these cells corresponded with the robust expression of BRDT normally observed in these meiotic stages . Interestingly , although it was previously reported that no post-meiotic spermatocytes were observed in the mutant testes [8] , our very detailed analysis revealed occasional apoptotic pro-metaphase and metaphase spermatocytes in stage XII tubules in the Brdt-/- mice ( S1 Fig ) . Clearly , however , the loss of BRDT largely affects the progression of late prophase I stages , confirming and extending previous observations [8] . To begin to elucidate which defects are triggering the apoptosis of Brdt-/- spermatocytes , we first analyzed the efficiency of early DNA repair events by immunodetection of γH2AX as a marker of double strand breaks ( DSBs ) , along with SYCP3 , in spermatocyte spreads of Brdt-/- and WT mice . In both WT and Brdt-/- spermatocytes , γH2AX localized throughout all chromatin in leptonema and early zygonema ( S2A and S2B Fig ) , indicating that normal DSBs are produced in the absence of BRDT . At early pachynema in both genotypes , γH2AX localization was restricted to a few foci in the chromatin close to the SC and extended over the chromatin of the sex chromosomes ( Fig 2A and 2C ) . By mid pachynema ( Fig 2B and 2D ) to the end of diplonema , γH2AX persisted only in the X and Y chromosomes , indicating that the DSBs were repaired . However , despite the normal pattern of distribution of γH2AX , we regularly observed pachytene spermatocytes with synapsis defects in the sex chromosomes ( Fig 2D ) . To gain insight as to the extent and temporal nature of the synapsis defects of the X and Y chromosomes , we quantified the frequency of fully unsynapsed X-Y throughout the distinct phases of pachynema . We classified the sub-stages of pachynema using the appearance and distribution of γH2AX throughout the chromatin and the status of synapsis of the autosomes as criteria which have been previously described [24] . In early pachytene Brdt-/- spermatocytes , 86 . 7% of the pseudoautosomal region ( PARs ) of the X and Y were synapsed and concomitantly , 13 . 3 ±4 . 3% of the sex chromosomes were fully unsynapsed ( Fig 2E , p<0 . 01 ) ( n = 99 spermatocytes from three 3 month-old Brdt-/- mice ) . This percentage increased to 18 . 5 ±6 . 7% by mid to late pachynema ( Fig 2E , p<0 . 001 ) ( n = 132 spermatocytes from three 3 month-old Brdt-/- mice ) . The number of unsynapsed X-Y present in pachytene Brdt-/- spermatocytes was significantly different as compared to that in WT pachytene spermatocytes ( Fig 2E , p<0 . 01 , p<0 . 001 for early and mid/late pachynema , respectively ) , suggesting that synapsis of the sex chromosomes is significantly compromised by depletion of BRDT . To evaluate if the regulation of DNA recombination by RAD51 was affected by depletion of BRDT , we next immunodetected RAD51 along with SYCP3 , in WT and Brdt-/- spermatocytes and quantified the number of RAD51 foci in each stage of prophase I . In zygonema , both genotypes had similar numbers of mean RAD51 foci ( 162 ±22 . 5 and 166 . 3 ±14 . 6 foci in WT and Brdt-/- , respectively ) ( S3A and S3B Fig ) . In early pachynema , RAD51 foci were reduced in both WT and Brdt-/- ( 31 . 2 ±8 . 2 and 27 . 7 ±9 . 3 foci , respectively; p = 0 . 1 ) and by mid pachynema , the number of foci decreased to 15 . 6 ±6 . 9 and 17 . 6 ± 9 . 2 foci respectively ( S3A and S3B Fig ) . By late pachynema and diplonema , RAD51 foci were rarely observed in WT and Brdt-/- spermatocytes ( S3A and S3B Fig , p = 0 . 1 ) ( n = 18 and 21 zygotene , 25 and 37 early pachytene , 100 and 200 mid and late pachytene WT and Brdt-/- spermatocytes , respectively , three 3 month-old mice per genotype ) . Although no differences were observed in the total number of RAD51 foci between WT and Brdt-/- spermatocytes , because defects in DSB formation and RAD51 localization in the PAR are associated with synapsis defects in the X and Y [26 , 27] , we quantified the number of spermatocytes with a RAD51 focus or foci in the PAR in WT and Brdt-/- zygotene/early pachytene and mid pachytene cells . We observed that RAD51 focus/foci were properly formed in the PAR of Brdt-/- spermatocytes , with no significant differences as compared to WT spermatocytes ( S3C Fig , p = 0 . 37 ) ( n = 78 and 75 zygotene/early pachytene , and 85 and 89 mid pachytene WT and Brdt-/- spermatocytes , respectively; three 3 month-old mice per genotype ) . Thus , the analyses of γH2AX and RAD51 indicated that formation of DNA DSBs and homologous recombination per se were unaltered in the absence of BRDT . These results further suggest that repair of DNA DSBs was unaffected in Brdt-/- spermatocytes and is not likely involved in the synapsis defects exhibited by the sex chromosomes . We next analyzed the dynamic pattern of chromosome synapsis throughout prophase I . Immunostaining Brdt-/- and WT spermatocyte spreads with SYCP3 and SYCP1 , a marker of the central element ( CE ) of the SC , revealed that the timing of the progression of pairing and synapsis in leptonema and zygonema did not differ between WT and Brdt-/- spermatocytes ( S2C and S2E Fig ) . At early pachynema , all homologous autosomes were completely synapsed throughout their entire length ( Fig 2F and 2H ) . The lack of differences in the synapsis of autosomes between WT and Brdt-/- spermatocytes persisted until late pachynema ( Fig 2G and 2I ) and by the diplotene stage , desynapsis of the autosomes proceeds with no significant differences between WT and Brdt-/- spermatocytes ( S2D and S2F Fig ) . In contrast to the normal synapsis exhibited by autosomes , the absence of BRDT gave rise to spermatocytes with synapsis defects of the X and Y chromosomes ( Fig 2I ) , in sub-stages of pachynema where these chromosomes are normally fully synapsed through the PAR ( Fig 2G ) . Taken together , these data show that the achievement and stabilization of the synapsis of the sex chromosomes , but not the autosomes , is affected in the absence of BRDT . In pachynema , the pairing and synapsis of the X-Y is thought to be maintained by formation of the MSCI [26] . Given the synapsis defects of the PAR observed in Brdt-/- spermatocytes , we next investigated whether the establishment of characteristic epigenetic modifications of MSCI might be affected in the absence of BRDT . As mentioned earlier , the temporal dynamics and localization of γH2AX is unaltered by depletion of BRDT , suggesting that initiation of MSCI may not be affected by BRDT ( Fig 2C and 2D ) . However , BRDT may influence the second phase of epigenetic modifications that maintain the MSCI during pachynema and diplonema . To determine if the progression and maintenance of MSCI is BRDT-dependent , we assessed the dynamics of appearance of distinct and temporally regulated histone modifications characteristic of MSCI [28] . We first analyzed the temporal localization pattern of H3 trimethylated at lysine 9 ( H3K9me3 ) , a histone mark associated with chromatin condensation and transcriptional repression , and quantified the signal intensity in the sex chromosomes at different stages of prophase I . In mouse spermatocytes , H3K9me3 localizes at the pericentric heterochromatin throughout prophase I and exhibits a specific localization pattern in the sex chromosomes [28] . As expected , in WT spermatocytes , H3K9me3 localizes in both X and Y in early pachynema and is then restricted to the Y chromosome in the transition from early to mid pachynema . H3K9me3 progressively decreases until it disappears from the sex chromosomes during late pachynema ( Fig 3A and 3E ) , but re-appears intensely in the unsynapsed chromatin of the sex chromosomes at diplonema ( Fig 3B and 3E ) . However , in Brdt-/- spermatocytes the pattern of H3K9me3 is altered in later stages of prophase I . That is , at mid pachynema , H3K9me3 signal in the X-Y was significantly lower than in WT spermatocytes ( Fig 3E , p = 0 . 0072 ) . Although in Brdt-/- late pachytene spermatocytes H3K9me3 disappears from the sex chromosomes in a similar temporal pattern as WT cells ( Fig 3C and 3E ) , the re-appearance of this histone mark in the X-Y at diplonema is notably weaker than that observed in WT diplotene spermatocytes ( Fig 3D ) and it never reached the normal levels that are observed in WT cells ( Fig 3E , p = 0 . 001 ) ( n = 10 and 17 early pachynema , 40 mid and 40 late pachynema , 30 early and 40 mid/late diplonema WT and Brdt-/- spermatocytes respectively per each mouse; three 3 month-old WT and Brdt-/- mice ) . We next examined the dynamics of localization and signal intensity of histone H3 monomethylated at lysine 4 ( H3K4me1 ) , another modification related to transcriptional repression [29] and which has been shown to appear in the sex chromosomes by the end of the pachytene stage [28] ( Fig 3F and 3G ) . Similar to H3K9me3 , depletion of BRDT resulted in a defective localization of H3K4me1 in the sex chromosomes at the end of pachynema and diplonema ( Fig 3H and 3I ) . Moreover , in Brdt-/- spermatocytes , the X and Y chromosomes exhibited significantly lower levels of H3K4me1 than those in WT spermatocytes during all late pachytene and diplotene stages ( Fig 3J , p = 0 . 001 ) ( n = 13 early pachynema , 30 mid pachynema , 40 and 50 late pachynema , 30 early and 30 mid/late diplonema WT and Brdt-/- spermatocytes respectively per each mouse; three 3 month-old mice per genotype ) . Analysis of H3K4me3 showed no observable differences between WT and Brdt-/- spermatocytes ( S4A–S4H Fig ) . During the course of our analysis , we observed that in Brdt-/- spermatocytes , H3K4me1 localized closer to the SC of autosomes and sex chromosomes ( Fig 3M–3N” ) , a pattern distinct from that seen in WT spermatocytes ( Fig 3K–3L” ) . That is , from late pachynema ( Fig 3M–3M” ) through diplonema ( Fig 3N–3N” ) , H3K4me1 was enriched in the chromatin loops closer to the SC in 76±11 . 8% of the spermatocytes analyzed ( n = 100 WT and 110 Brdt-/- spermatocytes; three 3 month-old mice per genotype ) . Moreover , quantification of H3K4me1 signal intensity in the overall chromatin ( not close to the SC ) of Brdt-/- autosome chromosomes evidenced a significant reduction of 1 . 9-fold as compared to WT autosomes ( p = 0 . 0021 ) ( S5 Fig ) . This suggested that BRDT could be involved with determining the correct localization pattern of H3K4me1 in distinct chromatin regions of autosomes and sex chromosomes during late prophase I stages . Although we do not exclude the possibility that the overall change in distribution of H3K4me1 in late prophase I Brdt-/- spermatocytes influences the reduction in its levels observed in the sex chromosomes , the higher fold-difference in H3K4me1 signal between WT and Brdt-/- sex chromosomes ( 5 . 1 , p = 0 . 0006 ) ( S5 Fig ) , compared to that in WT and Brdt-/- autosome chromosomes ( 1 . 9 , p = 0 . 0021 ) , suggested that failure of mechanisms specific to the MSCI might also be mediating the decrease of H3K4me1 in the X-Y in late prophase I Brdt-/- spermatocytes . The altered dynamics of histone modifications involved in the maintenance of MSCI during late prophase I stages raised the possibility that the temporal dynamics of localization of active transcriptional markers in the sex chromosomes is altered in Brdt-/- spermatocytes . To test this hypothesis , we first analyzed the localization of histone H3 acetylated at lysine 9 ( H3K9ac ) , which associates with transcriptionally active chromatin . H3K9ac localizes in the chromatin of autosomes from mid pachynema to diplonema and is barely detected in the chromatin of the sex chromosomes during these stages [28] ( Fig 4A and 4B ) . While the localization of H3K9ac in the autosomes is similar between WT and Brdt-/- pachytene and diplotene spermatocytes , its distribution in the X and Y chromosomes is altered ( Fig 4C and 4D ) . That is , there was a consistent presence of H3K9ac in the sex chromosomes in mutant pachytene and diplotene spermatocytes , whereas it was not detected in the WT . Quantification of the signal intensity of H3K9ac in the sex chromosomes revealed that its levels in Brdt-/- spermatocytes actually increased throughout late prophase I stages , in contrast to the decrease that was observed in the X and Y in WT spermatocytes ( Fig 4E , p = 0 . 05 and p = 0 . 001 ) ( n = 10 and 12 early pachynema , 24 and 25 mid pachynema , 30 late pachynema , 20 early and 40 mid/late diplonema WT and Brdt-/- spermatocytes , respectively per mouse; three 3 month-old mice per genotype ) . This result suggested that the transcriptional repression of the sex chromosomes is affected by depletion of BRDT . Indeed , immunolocalization of RNA pol II was barely detected in the sex body of WT spermatocytes [28] ( Fig 4F and 4G ) , but showed an intense signal in the XY chromatin of Brdt-/- late pachytene and diplotene spermatocytes ( Fig 4H and 4I ) . Quantification of the levels of RNA pol II protein signal in the sex chromosomes in Brdt-/- spermatocytes revealed that its levels were significantly higher than in WT XY chromatin ( 1 . 65 and 1 . 9 fold-change in mid and late pachynema , respectively , and 2 . 3 and 2 . 2 fold-change in early and mid/late diplonema , respectively; Fig 4J , p = 0 . 001 ) ( n = 15 early pachynema , 25 mid pachynema , 40 late pachynema , 25 early and 35 mid/late diplonema WT and Brdt-/- spermatocytes , respectively per mouse; three 3 month-old mice per genotype ) . Collectively , these results showed that BRDT is involved in regulating the timing of appearance and disappearance of epigenetic modifications linked to the MSCI in the X and Y chromosomes during late prophase I stages . Our results indicate that BRDT influenced epigenetic modifications associated with silencing of the sex chromosomes , suggesting that BRDT could influence the transcriptional repression of sex-linked genes during the MSCI . To test this hypothesis , we analyzed the expression of all genes located in the X chromosome by mining an available transcriptomic database of Brdt-/- spermatogenic cells at 17 and 20 dpp ( GEO number GSE39910 ) [8] . Importantly for our analyses , in these juvenile mice , the appearance of the phenotype of Brdt-/- mice is not yet visible , the number of spermatocytes is similar between WT and Brdt-/- , and the onset of apoptosis is not detected [8] . Among all the genes located in the X chromosome , we focused on genes that are transcribed only in testes , according to the mouse testes profile described by Schultz et al . [30] . Upon comparison with WT 17 and 20 dpp testes , the genes with a p-adjusted value below 0 . 005 and a fold-change of <1 . 5 were defined as differentially expressed genes ( DEGs ) . Based on these criteria , in 17 dpp ( Fig 5A and 5B; S3 Table ) , 525 genes showed no change in expression between WT and Brdt-/- testes , while only 5 were overexpressed in absence of BRDT . In 20 dpp testes ( Fig 5C and 5D; S3 Table ) , from 838 genes analyzed we found 163 genes with no change in expression and 675 DEGs . Among them , 128 genes were under expressed and 547 genes were overexpressed . This suggests that in testes containing predominantly pachytene and diplotene spermatocytes , in which normally the sex chromosomes are subjected to chromosome-wide silencing , 65 . 3% of the genes located in the X chromosome were significantly overexpressed in absence of BRDT . This overexpression pattern appears to be characteristic of X-linked genes , contrasting with the overall downregulation pattern observed in all autosomes in Brdt-/- 20dpp testes ( S6A Fig ) . We also note that most of the genes with no change in expression correspond to genes expressed in spermatogonia , which do not express BRDT and would not be expected to be affected by its loss . To confirm the overexpression observed in the X-liked genes , we used RNA isolated from WT and Brdt-/- spermatocytes to quantify by qRT-PCR the expression of six genes located in different positions within the X chromosome , which in the transcriptomic analysis were overexpressed: Ddx3x ( 8 . 17cM ) , Nxf2 ( 20 . 72cM ) , Fmr1 ( 38cM ) , Pet2 ( 39 . 95cM ) , Gm5072 ( 40cM ) and Huwe1 ( 68cM ) , and Scml2 ( 74cM ) . qRT-PCR showed that in Brdt-/- spermatocytes , the expression levels of Nxf2 , Fmr1 , Gm5072 , and Huwe1 were significantly higher than WT spermatocytes ( S6B Fig , * p<0 . 05 and *** p<0 . 001 ) . Interestingly , we noticed that genes located toward the distal end of the X chromosome , which contains the PAR , tended to be overexpressed . These gene expression studies show that BRDT influences the silencing of the sex chromosomes during meiosis . Given the altered pattern of H3K4me1 distribution observed during late prophase , we next asked if this reflected a change in the overall chromatin organization of the spermatocytes . To examine this possibility , we assessed chromatin accessibility in Brdt-/- and WT total testicular cells from 17 dpp by digestion with MNase . This age was selected because , as previously mentioned , testes at this age are highly enriched in pachytene spermatocytes and neither the number of meiotic cells nor the incidence of apoptosis present in the testis is significantly altered in the absence of BRDT [8] . We found that depletion of BRDT led to a more condensed chromatin structure in spermatocytes , with a large portion of the chromatin being inaccessible to MNase digestion ( Fig 6A ) . That is , compared to WT cells , the digestion of Brdt-/- cells generated more high molecular weight DNA , which ranged between 1500 and 4000 kb in size ( Fig 6A and 6B ) . There was a concomitant reduction of low molecular weight fragments , particularly the mono- , di- and tri-nucleosome fractions , indicating a decrease in the MNase cleavage between these nucleosome oligomers ( Fig 6A and 6B , p<0 . 001; four WT and Brdt-/- mice each were used per experiment , 3 experiments per digestion ) . A previous study reported that defects in chromatin organization can have pronounced effects in the length of the axis of a chromosome [31 , 32] . Consideration of this model led us to propose that depletion of BRDT might also alter the length of the axes of meiotic prophase chromosomes . To test this hypothesis , we measured and quantified the chromosome axis length in early , mid , and late pachytene , and early and mid/late diplotene WT and Brdt-/- spermatocyte spreads . At early pachynema , the lengths of autosomal axes were similar in WT and Brdt-/- cells ( average of 12 . 7 μm ±3 . 6 and 11 . 5 μm ±3 . 1 respectively ( Fig 6C , p = 3 . 2; n = 22 WT and 39 Brdt-/- spermatocytes , three 3 month-old animals per genotype ) . However , by mid and late pachynema , there was a significant shortening in the axial lengths in BRDT-deficient as compared to WT autosome chromosomes ( 15 ±4 . 8% and 19 . 5 ±2 . 9% respectively , **p<0 . 0028 , ***p<0 . 0001 , respectively; n = 92 and 45 mid pachytene , and 147 and 105 late pachytene WT and Brdt-/- spermatocytes , respectively; three 3 month-old mice per genotype ) . Furthermore , at early and mid/late diplonema , this shortening was even more pronounced , with mean axial lengths that were 30 . 5 ±4 . 2% and 32 ±3 . 4% shorter than observed in WT spermatocytes , respectively ( Fig 6C , ***p<0 . 0001; n = 38 and 53 early diplotene , and 158 and 65 mid/late diplotene WT and Brdt-/- spermatocytes , respectively; three 3 month-old animals per genotype ) . These results indicate that the chromatin organization mediated by BRDT influences the length of the chromosome axis in late prophase I stages . They also suggest that the meiotic events that are dependent on the context provided by the chromosome axis and chromatin loop size , such as CO formation [33] , could be affected by depletion of BRDT . Crossing over occurs and localizes in the context of both DNA and chromatin close to the SC . Given our observations of changes in the epigenetic organization of this chromatin as well as in the length of the chromosome axis in the absence of BRDT , we next determined whether CO formation during pachynema might also be altered , using immunolocalization of the late recombination marker MLH1 to foci . In WT spermatocytes , CO foci are found in all of the autosomes and in the PAR of the X and Y ( Fig 7A ) , with an average of 24 ±2 . 5 CO foci per WT spermatocyte ( Fig 7C; n = 45 spermatocytes , three 3 month-old WT mice ) . In contrast , depletion of BRDT resulted in a decrease in the number of MLH foci in the autosomes , with only 16 . 2 ±2 . 4 MLH1 foci per spermatocyte ( Fig 7C , p<0 . 0001; n = 39 spermatocytes , 3 three month-old Brdt-/- mice ) . Indeed , some chromosomes completely lacked MLH1 foci , and there were very few chromosomes with two foci per chromosome ( Fig 7B ) . Notably , the single MLH1 focus that is always produced in the PAR of the sex chromosomes decreased in frequency by 79 . 1 ±6 . 8% as compared to WT spermatocytes ( Fig 7D , p<0 . 0001; n = 45 cells per genotype , three 3 month-old WT and Brdt-/- mice ) . This is of interest in light of the previously mentioned increase in defects of synapsis of the sex chromosomes observed in Brdt-/- spermatocytes , as CO formation in the PAR is believed to stabilize the synapsis of the X and Y [34] . We then examined if the COs formed in Brdt-/- spermatocytes are properly distributed within the autosomes and if CO interference , which ensures that COs are well-spaced along the chromosomes , was affected by the changes in chromatin organization and chromosome axis length produced by depletion of BRDT . To this end , we determined the distribution of MLH1 foci along the SCs from WT and Brdt-/- spermatocytes using analytical procedures described by Anderson et al . [35] . This method reveals the position of each MLH1-positive focus within the SC as its distance from the centromere ( expressed as percent of the SC length ) , allowing the comparison of data from chromosomes of different lengths [35] . Thus , by dividing each SC in 10 equal length intervals ( from centromere to distal telomere ) ( Fig 7E ) , it is possible to localize each MLH1 focus in distinct SC intervals . The distribution of MLH1-positive foci within the chromosomes is then obtained by taking the sum of the number of MLH1 foci detected in each interval . In WT spermatocytes , for SCs with one MLH1 focus , 72 . 1% of the COs were mostly distributed between intervals 3 and 9 , with 19 . 1% ±3 . 4 and 17% ±2 . 6 of MLH1 foci concentrated in intervals 3 and 7 respectively , and 12 . 8% ± 3 . 1 and 12 . 8% ±2 . 9 localized in both intervals 4 and 8 respectively ( Fig 7F; n = 15 cells , from two 3 month-old WT mice ) . There was a very low frequency of COs at the ends of the chromosomes , in agreement with the reported negative effect of pericentromeric heterochromatin as well as telomeres on CO formation [36 , 37] . In Brdt-/- spermatocytes , however , there was a change in the position of the MLH1 foci along the SC , with the general distribution pattern shifted toward the distal end of the autosomes ( Fig 7F ) and the observation of most of the COs being distributed in fewer intervals as compared to WT . That is , 69 . 8% of the MLH1 foci were distributed between intervals 4 and 8; and the highest levels of foci were observed at intervals 5 ( 23 . 3% ±2 . 5 ) , 6 and 8 ( both with 16 . 6% ±1 . 9 and 16 . 6% ±2 . 7 respectively of the loci ) , rather than in intervals 3 and 7 in the controls ( n = 15 cells , from two 3 month-old Brdt-/- mice ) . Similar to WT , MLH1 foci were not found in the intervals closest to the centromere and telomeres . Despite the change of localization of MLH1 foci in the SCs , it was possible to observe that COs were still well-spaced along the chromosomes in Brdt-/- spermatocytes . That is , when two foci were present on the same SC , both WT and Brdt-/- cells exhibited foci evenly distributed along the chromosomes , with an average separation between foci of 58 . 3 ±9 . 1 and 61 ±3 , respectively . These observations suggest that BRDT may be involved in defining the formation and proper localization of COs in the chromosomes during pachynema , but it is dispensable for CO interference . Consistent with these observations , analysis of the few mutant spermatocytes that reached diakinesis/metaphase I revealed a significant decrease in chiasmata in Brdt-/- compared to WT spermatocytes ( 12 . 7 ±2 . 1 versus 22 . 9 ±1 . 4 respectively ) ( Fig 7I , p<0 . 0001 ) ( n = 4 and 6 , from two WT and two Brdt-/- 3 month-old mice , respectively ) . Furthermore , while in WT cells all chromosomes formed chiasmata , all Brdt-/- diakinesis/metaphase I spermatocytes examined exhibited achiasmate univalent chromosomes ( Fig 7G and 7H ) . Interestingly , qPCR analysis of genes that encode proteins that participate in establishing COs ( Mlh1 , Msh4 , Rnf212 , Cntd1 , Cdk2 , Dmc1 , Rad51 ) , were unaltered in Brdt-/- spermatocytes as compared to WT ( S7 Fig ) . This result suggests that the defects in CO formation exhibited by depletion of BRDT are not likely to have been produced by defects in the transcription of genes involved in the DNA repair pathway of COs . To determine if BRDT plays a role in promoting a regional chromatin organization in genomic regions containing CO sites , we took advantage of the unique characteristics of the PAR hotspot of the X chromosome and analyzed the chromatin organization of this region . The PAR hotspot was selected because it contains a well-defined cluster of overlapping hotspots that comprises 21 . 3 kb in length , at position 166 , 425 to 166 , 446 kb in the X chromosome [38] . The defined localization of the X-chromosome hotspots , together with the size of the PAR and the organization of the chromatin as short chromatin loops , generate a CO density that is at least 20 times higher than that found in autosomes [26] . This represents the ‘hottest’ cluster of DSBs in the mouse genome , and there is a CO formed in 99 . 9% of the cells [38] . This region was also selected because depletion of BRDT caused a significant reduction in the formation of the CO focus in the PAR ( Fig 7D ) and the histone modifications in the chromatin close to the SC of the X chromosome were altered ( Fig 3M and 3N ) . We therefore isolated chromatin from enriched Brdt-/- and WT spermatocyte fractions and performed a DNA protein occupancy mapping analysis , based on MNase digestion of the chromatin , to obtain mononucleosome fractions ( Fig 8A ) , followed by a nucleosome scanning assay ( NuSA ) , which utilizes tiled qPCR [39] . Using this approach , we mapped a 3 . 3 kb region within the 21 . 3 kb PAR hotspot to define the occupancy of proteins on the DNA ( Fig 8B ) . We observed that in WT cells , protein occupancy is distributed all along the DNA and locates in mainly twelve DNA sites within the analyzed PAR hotspot region , separated by an average of 261 ±56 . 5 bp ( Fig 8B and 8C ) . The higher peaks of protein occupancy ( R value >0 . 75 ) were found between 166 , 426 , 342 to 166 , 428 , 487 bp , suggesting strongly positioned proteins in this region . Moderate peaks ( R value 0 . 49–0 . 58 ) were found between 166 , 425 , 240 to 166 , 426 , 243 bp ( Fig 8B and 8C ) . This chromatin organization pattern changed in the absence of BRDT . For example , localization , position and DNA enrichment levels ( R value ) were changed in the 21 . 3 kb PAR hotspot . From 166 , 424 , 993 to 166 , 426 , 243 bp , we found four areas of protein occupancy separated by an average of 510 ±249 bp , in contrast to the six areas observed in WT chromatin ( Fig 8B and 8C ) . Moreover , in Brdt-/- chromatin , three of the four areas are not observed in WT chromatin ( Fig 8C , blue arrows ) and two of them corresponded to low protein occupancy areas ( R value 0 . 46 and 0 . 42 ) ( Fig 8C , green arrows ) . Also , from 166 , 426 , 515 to 166 , 428 , 487 bp , we observed a shift in the protection of some areas ( 166 , 426 , 515 bp and 166 , 427 , 205 bp ) and a lower protection as compared to WT regions ( Fig 8B and 8C ) . These data suggest that BRDT influences the normal chromatin architecture of the PAR hotspot .
The BET proteins are key epigenetic readers known to establish and modulate chromatin structure and organization , producing changes in the epigenomic landscape of cells [1 , 2] . In the mouse , all four BET genes are expressed at unique stages of spermatogenesis [5] , but their roles in this process are not well understood . The testis specific BET protein , BRDT , is expressed only in spermatocytes and spermatids , and its essential role in spermatogenesis has been clearly demonstrated by targeted mutagenesis in mouse models . Loss of the BD1 of BRDT impairs spermiogenesis [6] and complete loss of BRDT function blocks the progression of spermatocytes into the first meiotic division [8] , both resulting in complete male sterility . Although the BD1 has been implicated in mediating chromatin remodeling and mRNA processing during spermiogenesis [7 , 13] and BRDT has been implicated and changes in transcription during meiosis , the role of BRDT in specific meiotic functions is not well documented . Our detailed analysis of BRDT expression during meiotic prophase revealed its restriction to late prophase I stages , mid pachynema to late diplonema , in particular . Interestingly , in contrast to the retention of other BET family members ( BRD2 and BRD4 ) on mitotic metaphase chromosomes [40 , 41] , BRDT was not found to be significantly localized to meiotic metaphase I chromosomes . Indeed , the restricted expression pattern suggests that BRDT is dispensable for early meiotic events such as pairing and synapsis of autosomes , as well as for the global repair of DSBs . The restricted pattern also focused our attention on events that occur at these stages and impact the genetic and epigenetic stability of the sperm: namely , the formation and localization of COs , the behavior of the X and Y chromosomes , and the MSCI . The significant percentage of sex chromosomes in complete asynapsis during pachynema that we observed suggests that the absence of BRDT impacts the synapsis process in the sex chromosomes , which could result from a failure to form stable synapsis or a premature desynapsis of the sex chromosomes prior to the end of pachynema . Our data further indicate that the asynapsis of the sex chromosomes is not a result of defective pairing of the X and Y , or lack of formation of DSBs in the PAR , as RAD51 foci are properly formed in the PAR of Brdt-/- spermatocytes . Alternatively , the defective synapsis of the PAR could be a consequence of defects in maintenance of the MSCI and the significant decrease of the CO formation in the PAR , both events that have been suggested to stabilize the synapsis of the X and Y in pachynema [26] . Efficient MSCI is important for normal progression of prophase I and failure of MSCI can lead to apoptosis via the pachytene checkpoint [42] . The spreading of γH2AX to the entire chromatin and the normal dynamics of RAD51 in the sex chromosomes suggest that BRDT is dispensable for the initiation of the MSCI . However , the defective establishment of repressive histone modifications in the X and Y indicate that BRDT is key for the epigenetic remodeling of the sex chromosomes , which is necessary for the maintenance of MSCI . It is thus possible that the failure of the sex chromosomes to achieve or maintain a proper synapsis and to maintain MSCI might trigger the robust apoptosis seen in late pachytene and diplotene spermatocytes in the absence of BRDT . Spermatocytes that do reach stable synapsis of the PAR are then able to progress until the end of diplotene , but are characterized by defective epigenetic changes in maintenance of MSCI of the X and Y as observed with H3K9me3 , H3K4me1 and H3K9ac . Whether BRDT has a direct or indirect role in maintaining the MSCI remains to be determined . Indeed , as BRDT is also involved in transcriptional activation during meiosis [8] , we do not discard the possibility that BRDT could play an indirect role in the MSCI through transcription of other factors involved in epigenetic reprograming of the chromatin . Recent studies have shown that X-linked genes expressed in germ cells play key roles in regulating male fertility and that alteration in their expression or mutations in the genes may be associated with infertility in men [43] . We showed that BRDT regulates the silencing of the X-linked genes and its depletion produces an overexpression of 65 . 3% of these genes . In fact , our data showed that Nxf2 and Tex11 , two genes which have been shown to be essential for the regulation of meiosis and spermatogenesis , are overexpressed in Brdt-/- spermatocytes . NXF2-deficient mice exhibit defects in spermatogonia proliferation as well as defective spermatid formation , resulting in male infertility [44] . TEX11 promotes chromosome synapsis and interestingly , regulation of CO formation [45] . This is of particular interest as the phenotypes of Tex11-/- and Brdt-/- spermatocytes overlap in some features , such as asynapsis of the PAR , decrease in CO number , and defective chiasmata formation . This raises the possibility of a molecular link between TEX11 and BRDT and/or that the overexpression of TEX11 could potentiate the phenotype of Brdt-/- spermatocytes . Our results further suggest that BRDT is also essential to maintain proper higher order chromatin structure during meiosis . That is , we observed that BRDT promotes proper histone modifications in the chromatin close to the SC and an overall open chromatin configuration in spermatocytes . The overall chromatin condensation observed in Brdt-/- spermatocytes is indeed consistent with the overall downregulation exhibited by autosomal genes that we observed and is in agreement with the previous report which showed a massive downregulation of genes in 20dpp testes in the absence of BRDT [8] . In addition , it has been shown that the structural maintenance of chromosomes influences the size of the chromatin loops located close to the SC [32 , 46] and that variation in the length of the SC changes the number and length of DNA loops that are anchored to the SC [26 , 31] . This is relevant because it is in the context of the chromatin loops that DSBs , recombination , and COs occur . Indeed , longer SC lengths have been correlated with higher numbers of COs per chromosome [35] . We thus speculate that the reduction in the length of the chromosomal SCs observed during mid and late pachynema contributes to the reduced number and altered distribution of COs within the chromosomes in Brdt-/- spermatocytes . COs are also influenced by the local and higher order chromatin organization of the chromosomes [47] . In this context , the modulation of BRDT in local and global chromatin configuration and epigenetic marks might also be contributing to the successful formation of the COs in mammalian male meiosis . The defects in CO formation observed in Brdt-/- spermatocytes could be involved in triggering the apoptosis of spermatocytes at late pachynema . For those spermatocytes that achieve late diplonema to the metaphase 1 transition , defective COs producing achiasmatic chromosomes which could activate the metaphase checkpoint and trigger another wave of apoptosis of Brdt-/- spermatocytes [48] . As mentioned earlier , our analyses revealed that early recombination pathway events appear normal , and the fact that no chromosome fragmentation was observed suggests that DSBs are properly repaired in Brdt-/- spermatocytes . However , the reduction in number and abnormal localization of MLH1 foci indicate that BRDT is likely involved with CO formation in meiosis . One possibility is that BRDT modulates the maturation of the designated COs , by ensuring that the machinery that either promotes or antagonizes COs is properly recruited to the CO sites within the chromosomes [49] . Transcriptional analysis of genes encoding CO-related proteins revealed no changes in their expression , suggesting that BRDT is probably not involved in the regulation of expression of key factors that promote maturation and establishment of the COs . Whether BRDT has a direct role in recruiting CO complex proteins or indirect role by providing an epigenetic and chromatin structure suitable for the recruitment of the CO machinery proteins remains to be determined . An additional observation about the function of BRDT in crossing over is that it appears to affect the proper localization/distribution of MLH1 foci within autosomes . The few MLH1 foci that are formed in absence of BRDT are distributed in regions that are different than those in WT chromosomes . However , CO interference is not affected in the chromosomes as our results also show that the distance between two crossover foci , when formed in Brdt-/- spermatocytes , is conserved independent of the change in SC length . This finding raises an interesting issue , as it has been shown that CO interference is part of a pattern shaped by the SC [31] . CO interference is not confined uniquely to the CO , but is related to SC nucleation and structure , which impacts the threshold-based designation and spreading interference process of the final COs [50] . Indeed , the structure of the SC provides a platform that promotes not only CO formation but also its inhibition , which limits the number of CO [51] . BRDT has no detectable role in SC formation per se; hence , the spreading pattern that COs might have within the SC , which determines CO interference , is unchanged by depletion of BRDT . Even changes in SC length at later prophase I stages are still not enough to modify this pattern . However , and despite the compromised position of CO sites in the chromosomes SC , the final formation and localization of a CO is produced by other mechanisms in which the chromatin and epigenetic landscape might play a determinant role . We propose that BRDT is a key element in regulating these mechanisms . These data indeed support previous results showing that CO formation is independent of CO interference [52] . Our findings concerning the role of BRDT in controlling CO formation and distribution in mammalian male meiosis provides a link between a male-specific epigenetic factor and the establishment of COs in specific regions of the chromosomes . Interestingly , they further suggest a possible mechanism by which epigenetic factors differentially expressed in males and females are potential mediators of the sexual dimorphism in recombination and CO patterns that occurs in mammalian meiosis [53] . The finding that lack of BRDT produces multiple alterations in different processes during prophase I , as well as the reported misregulation of gene expression such as of Ccna1 , a key regulator of the diplotene to metaphase I progression [8] , suggests there may be a combinatory set of defects which causes a progressive cell death in response to different checkpoints during prophase I and metaphase I . In conclusion , we have observed important functions for BRDT in regulating chromatin configuration and reprograming essential for MSCI and formation and localization of CO in spermatocytes , key meiotic processes that determine the genetic and epigenetic homeostasis of the gamete and the variability among individuals . Our observation of failure of these meiotic functions of BRDT provides insight into the infertility in the Brdt mouse models [6 , 8] and in individuals reported as carrying polymorphisms in the BRDT gene [9 , 10] , as well as the possibility that BET protein inhibitors might be considered as a non-hormonal approach to male contraception [11] .
All experiments involving mice were approved by the Columbia University Institutional Animal Care and Use Committee and performed in accordance with the National Institutes of Health guidelines for the care and use of animals . B6JTyr;B6N-Brdttm1a ( EUCOMM ) Wtsi/Wtsi ( hereafter named Brdt-/- ) mice were obtained from the International KnockOut Mouse Consortium , Welcome Trust Sanger Institute , UK . Construction of the vector and characteristics of the mutation were previously described by Skarnes et al . [54] and Gaucher et al . [8] . B6JTyr;B6N mice were used as wild type ( WT ) strain . Testes from three 2 month-old WT and Brdt-/- mice were fixed in 10% formalin overnight at 4°C and then washed in 70% ethanol . Fixed and paraffin-embedded tissues were sectioned at 4 μm and serial sections were obtained . One section was used for staging the mouse seminiferous epithelium , by staining with hematoxylin along with immunostaining for γH2AX to identify and classify spermatocytes according to the different stages of prophase I . The serial section was then counterstained with hematoxylin along with TUNEL staining , performed using an in situ cell death detection kit ( Roche Diagnostics ) as previously described [55] . Only clearly stained cells were considered as apoptotic and only tubules cut perpendicular to the length of the tubule ( round tubules in sections ) were evaluated . 200 tubules per testis were analyzed per each genotype and the number of spermatocytes in each stage of the seminiferous tubule and the TUNEL-positive cells per cross-sectioned tubule were counted . Results were expressed as the percentage of TUNEL-positive spermatocytes/ total spermatocytes per each stage of the seminiferous tubules . Significant differences were assessed by statistical analysis using non-parametric Mann Whitney test with the threshold of significance set at 0 . 05 . Spermatocyte spreads were prepared following the procedures described by Manterola et al . [24] . Briefly , seminiferous tubules were isolated , placed in a petri dish , and mechanically disaggregated using forceps . Then , 80 to 200 μl of 100 mM sucrose was slowly added and mixed with the cells . From this suspension , 14 μl was dropped onto a slide previously submerged in 1% paraformaldehyde ( PFA ) , pH 9 . 2 , and spread throughout the slide . The slides were then slowly dried in a humid chamber for 3h , washed with Photo-Flo 0 . 08% in distilled water , dried , and stored at −80°C until their use . The slides were then placed in PBS and incubated with the following primary antibodies: mouse anti-SYCP3 1:200 ( Abcam , Ab12452 ) ; rabbit anti-SYCP3 1:200 ( Abcam , Ab15093 ) ; rabbit anti-BRDT 1:100 [6]; rabbit anti-trimethyl histone H3 ( Lys 9 ) 1:300 ( EMD Millipore 07–442 ) ; rabbit anti-histone H3 ( acetyl K9 ) 1:200 ( Abcam , Ab4441 ) ; rabbit anti-monomethyl histone H3 ( Lys4 ) 1:200 ( EMD Millipore ABE1353 ) ; rabbit anti-trimethyl histone H3 ( Lys4 ) , clone CMA304 1:50 ( EMD Millipore 05–1339 ) ; mouse anti-RNA pol II CTD4H8 1:200 ( Upstate 05–623 ) ; mouse anti-γH2AX ( Ser139 ) 1:1000 clone JBW301 ( Upstate 05–636 ) ( EMD Millipore ) ; mouse anti-MLH1 1:100 Clone G168-15 ( BD pharmingen , 550838 ) ; rabbit anti-RAD51 1:200 ( Calbiochem , PC130 ) and human anti-centromere 1:300 ( Antibodies Incorporated , 15-235-0001 ) . After rinsing in PBST ( phosphate-buffered saline , 1% Tween-20 ) , the slides were incubated with the appropriate secondary antibodies diluted 1:200 in PBS as follows: Alexa 488-conjugated donkey anti-rabbit IgG ( H+L ) , Alexa 488-conjugated donkey anti-mouse IgG ( H+L ) , Alexa 594-conjugated donkey anti-mouse IgG ( H+L ) , Alexa 594-conjugated donkey anti-rabbit IgG ( H+L ) , Alexa 647-conjugated donkey anti-human IgG ( H+L ) . Slides were counter-stained with DAPI and mounted with Prolong ( Thermofisher ) . Chromosome spreads were examined using a Nikon eclipse E800 microscope equipped with epifluorescence optics , and the images were photographed with a high-definition cooled color camera head DS-Fi1c . All images were processed with Adobe Photoshop CS5 software . The different meiotic stages were identified by following previously described criteria [24] [28] . Briefly , we identified changes in chromosome morphology of autosomes and sex chromosomes , such as the appearance of excrescences on the AEs of sex chromosomes and the widening of SC attachment plates on the autosomes , as well as the synapsis behavior of autosomes and sex chromosomes , with special regard to the length of the pairing region between X and Y chromosomes . To measure the signal intensity of BRDT , we analyzed a total of 480 spermatocytes from three 3 month-old wild type mice , distributed in 10 leptonema , 10 zygonema , 20 early pachynema , 30 mid pachynema , late pachynema , early diplonema and mid/late diplonema spermatocytes respectively per each mouse . For measurement of the signal intensity of each histone modification H3K9me3 , H3K9ac , H3K4me1 as well as for RNA polymerase II ( pol II ) in the sex chromosomes , we analyzed an average of 450 spermatocytes from three 3 month-old wild type and Brdt-/- mice , From these , an average of 20 corresponded to early pachynema , 30 to mid pachynema , 40 to late pachynema , 30 to early diplonema and 30 to mid/late diplonema spermatocytes per each protein , per mouse and per each genotype . BRDT , H3K9me3 , H3K9ac , H3K4me1 and RNA polymerase II ( pol II ) intensities were quantified by selecting 5 areas of the immunofluorescent signal in the XY body and measuring the intensity of the signal using the measurement/mean value tool in ImageJ ( NIH ) . For BRDT and H3K4me1 signals , 5 random areas in the chromatin of autosome chromosomes were also quantified . Both immunofluorescent signals and background were quantified and the final intensities were calculated by subtracting the background immunofluorescent signal from the immunofluorescent signal from the protein . Results represent the mean±SD . It is important to note that in spread preparations , apoptotic cells exhibit a particular morphology that does not resemble a normal spermatocyte [56] and thus , we discarded those cells from our analysis . Statistical significance between mice was calculated by the one-way analysis of variance ( ANOVA ) , followed by Tuckey post test . A Z-test for two proportions was calculated to determine statistical differences between the signal intensity of BRDT in the chromatin of autosomes and in the chromatin of the sex chromosomes . The threshold of significance was set at 0 . 05 with a confidence interval of 95% . Statistically significant differences between wild type and Brdt-/- spermatocytes were determined using a non-parametric Mann Whitney test with the threshold of significance set at 0 . 05 . No statistically significant differences in the distribution of the results between animals of the same genotype were observed . The lengths of autosomes of WT and Brdt-/- early , mid , and late pachytene , and early and mid/late diplotene spermatocyte spreads were measured by using the measurement tool in ImageJ ( NIH ) , calibrated to micrometer ( μm ) by setting a known distance according to the microscope and digital camera . The position of the MLH1 foci along the autosome chromosomes of WT and Brdt-/- mid and late pachytene spermatocyte spreads was measured by determining the relative distance of each MLH1 focus in each SC using distance ( percentage of SC length ) from the centromere and distal telomere , according to the method and criteria described by Anderson et al . [35] . Ten mid and ten late pachytene WT and Brdt-/- spermatocyte spreads ( n = 3 animals per each genotype ) , immunolabeled with SYCP3 , MLH1 and centromere , were selected for measurement based on i ) the presence of unbroken and unstretched SC , and ii ) the presence of at least 19 MLH1 foci in the case of WT cells , and 10 MLH1 in Brdt-/- spermatocytes . Each SC was measured using the measurement tool in ImageJ ( NIH ) , and the length was divided into 10 equal ( 10% ) intervals . Using the same measurement tool , the position of each MLH1 foci was calculated by its distance to the centromere and to the distal telomere , respectively . The relative interference distance between two MLH1 foci on an SC was determined by following the method described by Anderson et al . [35] . Preparation of enriched populations WT and Brdt-/- pachytene spermatocytes was performed according to our laboratory's established protocols [14] by using seven WT and sixteen Brdt-/- 2–3 month-old mice . The purity of cell populations was by flow cytometric analysis on a Becton Dickinson FACScan Flow Cytometer . Pools of primary spermatocytes were obtained at a purity of >85% . RNA was isolated by using a Qiagen RNeasy Micro Kit and one microgram of total RNA was used to prepare cDNA with random hexamer primers and the SuperScriptIII First-Strand Synthesis System ( Invitrogen , #18080–051 ) . Quantitative real-time PCR ( qPCR ) was performed according to our standard protocol [57] . Primer sets are provided in S1 Table . The acidic ribosomal phosphoprotein P0 ( Arbp ) gene was used as an internal control for data normalization and the fold change between wild type and mutant samples calculated . All samples were analyzed in triplicate . Statistical differences of the expression of each gene between WT and Brdt-/- pachytene spermatocytes was performed using 2-way Anova with Bonferroni posttests . The transcriptomic analyses of the X chromosome were performed by using gene expression data from Brdt-/- testes performed by Gaucher et al [8] and obtained using the Illumina mouse WG-6 V2 . 0 gene expression array . The raw data from 17dpp and 20dpp WT and Brdt-/- mice was obtained from Gene Expression Omnibus , access number GSE39910 , platform GPL6887 . Within this page , we used the tool “Analyze with GEO2R” to establish the differential expression of genes by using the R package “Limma” , which uses a linear model to evaluate the differential expression and an empirical bayes method to moderate the standard errors of the estimated log-fold changes . The genes with an adjusted p-value of ≤0 . 05 and an absolute fold-change of <1 . 5 were considered as differentially expressed genes ( DEGs ) . All the dataset generated , including DEGs and genes with no differential expression ( no-change ) were then grouped per chromosome by using the genome database biomaRt , obtaining the chromosome localization and expression pattern ( DEG or no-change ) for each gene . From the obtained list , we used R to filter the genes that were only expressed in testes by using the transcript profile of developing mouse testes described by Scultz et al . [30] . Ggplot2 was used to generate the graphs of the gene-expression profile of the X chromosome at 17dpp and 20dpp testes . Nuclei were isolated from testes from twelve 17 day-old WT and Brdt-/- mice , which are enriched for pachytene and diplotene spermatocytes . As these testes are in the first spermatogenic wave , they contain no spermatids ( in the case of WT animals ) , and very few apoptotic cells ( in the case of Brdt-/- ) [8] . The MNase sensitivity assay and salt extraction analysis of chromatin were performed by following the procedure described by Kishi et al . [58] , with few modifications . Briefly , seminiferous tubules from 6 testes ( per each genotype ) were suspended in 5 ml of buffer A ( 0 . 32 M sucrose , 15 mM HEPES-NaOH ( pH 7 . 9 ) , 60 mM KCl , 2 mM EDTA , 0 . 5 mM EGTA , 0 . 5% bovine serum albumin ( wt/vol ) , 0 . 5 mM spermidine , 0 . 15 mM spermine and 0 . 5 mM dithiothreitol ) and ground using a Dounce tissue grinder . The cells were then filtered through a 40 μM mesh filter and the resulting suspension was layered on a cushion of 45 ml buffer A containing 30% of sucrose ( wt/vol ) and centrifuged at 3 , 000 g for 10 minutes at 4°C . Then , each sucrose layer was carefully removed and the nuclear pellet was suspended in 5 ml of buffer C ( 20 mM Tris-HCl ( pH 7 . 5 ) , 70 mM NaCl , 20 mM MgCl2 , 3 mM CaCl2 , 0 . 5 mM spermidine , 0 . 15 mM spermine and 0 . 5 mM dithiothreitol ) and centrifuged at 2 , 000 g for 5 minutes at 4°C . For MNase digestion , the nuclear pellet was suspended in 100 μl of buffer C and a minimum of 15 μg of chromatin was digested with 1 U/μl MNase ( N5386-50U , Sigma-Aldrich ) , followed by 3 minutes incubation at 37°C . The digestion was stopped by adding 18 μl of 0 . 5 M EDTA and 14 μl 0 . 1 M EGTA . The digested DNA fragments were then purified by using a QIAquick PCR purification kit ( Qiagen ) and subjected to a 2 . 5% agarose gel electrophoresis . All MNase assay experiments were repeated three times with four animals per genotype per experiment . Analysis of the relative intensity of the bands from three different chromatin digestions per each genotype was performed using the gel analyzer tool on Image J . Background was subtracted prior to the analysis of the bands . Statistical analysis was performed using 2-way Anova with Bonferroni posttests . Chromatin from pachytene spermatocytes was obtained following the procedure described by Getun et al . [39] with small modifications . Briefly , pachytene spermatocytes were re-suspended in 1ml lysis buffer without calcium ( 10 mM Tris pH 7 . 5 , 10 mM NaCl , 3 mM MgCl2 , 0 . 4% NP-40 , 0 . 5 mM Spermidine , 0 . 15 mM Spermine ) , incubated on ice for 10 minutes and centrifuged at 150 g for 10 minutes at 4°C . The pellet was then washed in 1ml lysis buffer with calcium ( 1 mM CaCl2 ) , centrifuged at 150 g for 10 minutes at 4°C and re-suspended in 50μl lysis buffer with calcium . To standardize the amount of MNase necessary to obtain only a mono-nucleosome fraction , chromatin was divided in 6 equal amounts and digested with increasing concentrations of MNase ( N5386-50U , Sigma-Aldrich ) ( 0 . 125 U/μl , 0 . 25 U/μl , 0 . 5 U/μl , 1 U/μl , 2 U/μl , 3 U/μl ) , followed by 10 minutes incubation at 37°C . The digestion was stopped by adding 18 μl of 0 . 5 M EDTA and 14 μl 0 . 1M EGTA . A single mononucleosome fraction was successfully obtained by using 1 U/μl MNase . Finally , DNA from the mononucleosome fractions was extracted and purified using a QIAquick PCR purification kit ( Qiagen ) . The concentration of the resulting DNA fragments was then quantified and used for tiled qPCR analysis . Tiled primer pairs covering the region comprising from 166 , 425 to 166 , 428 kb in the X chromosome , were designed using the PCRTiler software [59] . We obtained 50 pairs of primers with an overlap of ~50 kb that yield amplicons of ~100 kb . Primer sets are provided in S2 Table . Primer pairs were tested and those that produced multiples amplicons in the melting curves were discarded . Only primer pairs with similar amplification efficiency were used . qPCR was then performed using iTaq Universal SYBR Green Supermix ( Bio-Rad Laboratories ) on a CFX96 Touch Real-Time PCR Detection System ( Bio-Rad Laboratories ) . A standard two-step real-time PCR program was used with an annealing temperature of 61°C and 40 cycles of amplification . DNA protection profiles were calculated as the fold enrichment ( or protection capability ) by calculating the 2-ΔCt ( R value ) of MNase digested input DNA Ct subtracted to the Ct of undigested genomic DNA ( R = 2-ΔCt , ΔCt = Ct MNase treated DNA ( Input ) –Ct Undigested DNA ( Genomic ) ) , following the method described by Getun et al . [39] . | BRDT , a testis-specific member of the bromodomain and extra-terminal ( BET ) subfamily of epigenetic reader proteins , is essential for the generation of male gametes . In post-meiotic cells , BRDT is involved in chromatin organization and transcriptional regulation through its first bromodomain motif , as loss of the BD1 results in a truncated BRDT protein that fully interrupts the differentiation of the germ cells during the process of spermiogenesis . Complete loss of BRDT function results in an arrest during meiotic prophase with no cells progressing into post-meiotic stages . However , neither the specific role of BRDT in meiosis nor the pathways affected by its depletion are known . We investigated how BRDT controls meiosis by examining its subcellular localization during prophase I as well as the meiotic consequences observed with the loss of BRDT function . BRDT localizes throughout the chromatin of autosomes and sex chromosomes in a dynamic pattern during pachynema and diplonema . Loss of BRDT severely disrupts the epigenetic reprograming and silencing of transcription of the sex chromosomes , the global and regional chromatin configuration , and the formation and localization of crossovers in spermatocytes . Thus , BRDT regulates key meiotic processes that determine the genetic and epigenetic homeostasis of the male gamete . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"meiosis",
"spermatocytes",
"cell",
"cycle",
"and",
"cell",
"division",
"cell",
"processes",
"germ",
"cells",
"epigenetics",
"chromatin",
"sperm",
"sex",
"chromosomes",
"prophase",
"synapsis",
"animal",
"cells",
"chromosome",
"biology",
"gene",
"expression",
"autosomes",
"cell",
"biology",
"genetics",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"chromosomes"
] | 2018 | BRDT is an essential epigenetic regulator for proper chromatin organization, silencing of sex chromosomes and crossover formation in male meiosis |
Noncoding sequence contains pathogenic mutations . Yet , compared with mutations in protein-coding sequence , pathogenic regulatory mutations are notoriously difficult to recognize . Most fundamentally , we are not yet adept at recognizing the sequence stretches in the human genome that are most important in regulating the expression of genes . For this reason , it is difficult to apply to the regulatory regions the same kinds of analytical paradigms that are being successfully applied to identify mutations among protein-coding regions that influence risk . To determine whether dosage sensitive genes have distinct patterns among their noncoding sequence , we present two primary approaches that focus solely on a gene’s proximal noncoding regulatory sequence . The first approach is a regulatory sequence analogue of the recently introduced residual variation intolerance score ( RVIS ) , termed noncoding RVIS , or ncRVIS . The ncRVIS compares observed and predicted levels of standing variation in the regulatory sequence of human genes . The second approach , termed ncGERP , reflects the phylogenetic conservation of a gene’s regulatory sequence using GERP++ . We assess how well these two approaches correlate with four gene lists that use different ways to identify genes known or likely to cause disease through changes in expression: 1 ) genes that are known to cause disease through haploinsufficiency , 2 ) genes curated as dosage sensitive in ClinGen’s Genome Dosage Map , 3 ) genes judged likely to be under purifying selection for mutations that change expression levels because they are statistically depleted of loss-of-function variants in the general population , and 4 ) genes judged unlikely to cause disease based on the presence of copy number variants in the general population . We find that both noncoding scores are highly predictive of dosage sensitivity using any of these criteria . In a similar way to ncGERP , we assess two ensemble-based predictors of regional noncoding importance , ncCADD and ncGWAVA , and find both scores are significantly predictive of human dosage sensitive genes and appear to carry information beyond conservation , as assessed by ncGERP . These results highlight that the intolerance of noncoding sequence stretches in the human genome can provide a critical complementary tool to other genome annotation approaches to help identify the parts of the human genome increasingly likely to harbor mutations that influence risk of disease .
Despite strong evidence that regulatory regions can be affected by pathogenic mutations , such as in fragile-X syndrome , β-thalassemia , Charcot-Marie-Tooth neuropathy , breast cancer and others [1–5] , little has been done to quantify stretches of regulatory sequence in the context of both phylogenetic conservation and human-specific intolerance to variation , and then correlate it back to disease causing potential . While methods to assess phylogenetic conservation at a single site are established , such as GERP++ [6 , 7] , purely phylogenetic approaches are at a risk of ignoring human specific regulatory sequence [8 , 9] . Furthermore , while efforts have been made to create predictors that seek to identify variants in noncoding sequence that might influence expression or have higher chance of causing disease [10–13] , no framework has been introduced that focuses on standing variation in the human population to estimate the relative intolerance of a gene’s noncoding exome sequence to genetic variation . Since this regional-based approach proved effective for protein coding genes , it is natural to assess its application to noncoding exome sequence . To assess whether noncoding sequence can predict genes that cause human disease through gene dosage aberration , we derive two measures: a phylogenetic conservation based score and a score reflecting intolerance to standing variation in a human population . To permit an unambiguous comparison to gene lists , we concentrate on each gene’s proximal regulatory regions: 5’ UTR , 3’ UTR , and the 250bp upstream of the transcription start site; recognizing that these three regions are only a subset of the relevant regulatory sequence for protein-coding genes . We generate a GERP++ region-based conservation score to assess the overall conservation of each gene’s proximal noncoding sequence [6 , 7] . To capture regulatory function that might be human-specific we formulate a novel human population genetic approach ( ncRVIS ) . We then assess each gene’s proximal noncoding region for phylogenetic conservation and intolerance to genetic variation in the human population , and tie these scores back to genes known to cause disease due to a gene dosage aberration . An important clarification is that the current RVIS framework is a regional-based measure of intolerance to variation , and as such is complementary to traditional variant-level predictions . More recent ensemble-based predictors , such as CADD [12] and GWAVA [13] leverage multiple features including phylogenetic conservation to make predictions of functionality even for noncoding variants . To assess the levels of contribution from information beyond conservation , we adapted CADD and GWAVA into regionalized scores in a way analogous to ncGERP by taking the average CADD and subsequently the average GWAVA score across a gene’s noncoding proximal regulatory sequence as its ncCADD and ncGWAVA score , respectively ( Methods ) . Our results show that it is possible to use a combination of phylogenetic and human standing variation to identify regions of noncoding sequence that associate with gene-dosage sensitivity . Beyond the immediate noncoding flanking sequence of protein-coding genes , the framework introduced in this paper can be elaborated to include stretches of regulatory sequence beyond UTRs . Another important goal of this work is to illustrate that in addition to traditional phylogenetic signatures of important noncoding sequence , we can use signatures from human standing variation to help define boundaries of noncoding sequence that when considered as a unit might show an excess of mutations identified in cases compared to controls—similar to what is currently done in exome-sequencing studies where we assess excess mutations per each protein-coding gene . [14]
To evaluate whether a gene’s regulatory sequence can predict dosage sensitivity , we took four gene lists derived from independent sources . The first list contained OMIM disease-associated genes previously characterized as “haploinsufficient” [15] . The second list took a set of genes curated as dosage sensitive in ClinGen’s Genome Dosage Map ( http://www . ncbi . nlm . nih . gov/projects/dbvar/clingen/ ) . The third list—a novel list introduced here—relies solely on human polymorphism data from the 6503 whole exome sequences made available by the NHLBI Exome Sequencing Project ( ESP ) [16] to identify genes where , based on the sequence context and mutability , we observed fewer loss-of-function variants than we expected to observe . Finally , to look at the opposite end of the dosage sensitivity spectrum , we identified genes that are tolerant to copy number variations ( CNVs ) based on the CNV data from two large Database of Genomic Variants ( DGV ) studies [17–19] . We used pre-calculated hg19 GERP++ scores ( accessed January 2014 ) to calculate a single average GERP++ score across a gene’s noncoding sequence ( 3’ UTR , 5’ UTR and immediate promoter region ) . We refer to this score as the noncoding GERP ( ncGERP ) score . We then constructed a protein-coding conservation based score for each gene , pcGERP , by using the same methodology across the gene’s protein-coding sequence . As described in the relevant papers , GERP++ provides a score per nucleotide base , which has been shown to reflect a base’s conservation across the mammalian lineage [6 , 7] . A limitation of phylogenetic approaches is that they are unable to capture sequence with human specific function . To address this , we used the pattern of standing genetic variation in a human population . This approach is a noncoding formulation of the Residual Variance Intolerance Score ( RVIS ) , a regression framework we recently developed to score the protein-coding sequence of genes in terms of their tolerance to functional genetic variation . We showed in Petrovski et al ( 2013 ) that this approach provides significant information for which genes are likely to carry protein-coding pathogenic mutations [15] . To adapt this approach to noncoding sequence , however , several changes are needed ( Methods ) . First , instead of using the total number of observed variants to predict the expected number of common variants , we used the estimated mutation rate to reflect the mutability of the noncoding sequence . Second , since we cannot reliably distinguish functional and non-functional UTR variation we compared the prediction to all possible common noncoding variants . Finally , because most currently available exome kits do not provide sufficient coverage of UTRs , we relied on whole-genome sequence ( WGS ) data from 690 samples generated at the Duke Center for Human Genome Variation ( CHGV ( S1 Table ) . We first demonstrated that the RVIS itself when applied to protein-coding sequence of genes still has predictive utility when each of these adjustments are made , suggesting that a similar approach is possible for regulatory sequence ( S2 and S3 Tables ) . For comparisons to our previously published protein-coding RVIS , we also generated RVIS-CHGV , a score that is the exact formulation of the published RVIS [15] , but is dependent on the 690 CHGV whole-genome sequenced samples used to construct the ncRVIS score and similarly to ncRVIS , adopts the mutation rate of the effectively sequenced sites ( Methods ) . We found that the noncoding ncRVIS and protein-coding RVIS-CHGV scores are weakly correlated ( Pearson’s r2 correlation of 0 . 04 , S1A Fig ) . The ncRVIS ( Fig 1 ) , RVIS-CHGV and ncGERP scores and their corresponding genome-wide percentile scores can be found in S1 Data and at http://igm . cumc . columbia . edu/GenicIntolerance/ . To evaluate whether the ncGERP and ncRVIS scores correlate with known disease genes , we used the same gene lists as previously described [15] . We found that , using a logistic regression model , RVIS-CHGV , ncRVIS and ncGERP significantly predict OMIM haploinsufficient genes that have been linked through de novo mutations: p = 4 . 7x10-21 ( AUC = 0 . 75 ) , p = 2 . 4x10-7 ( AUC = 0 . 63 ) and p = 2 . 7x10-24 ( AUC = 0 . 78 ) , respectively . A joint model of the three scores achieved an AUC of 0 . 816 when predicting OMIM haploinsufficient genes that have been linked through de novo mutations ( Table 1 ) . However , based on the other OMIM gene sets it does not appear that the noncoding sequence of genes can currently distinguish the broader set of OMIM disease genes , indicating that the patterns within the noncoding sequence are likely to be for the most part specific to diseases linked to haploinsufficiency ( Table 1 ) . ClinGen’s dosage sensitivity map is another growing resource for genes that are curated by experts as being haploinsufficient or triplosensitive ( http://www . ncbi . nlm . nih . gov/projects/dbvar/clingen/ ) . As of the 1st of May 2015 , 263 genes had been annotated as having either “Some evidence for dosage pathogenicity” or “Sufficient evidence for dosage pathogenicity” ( S2 Data ) . We repeated the gene dosage sensitivity assessment using this better curated set of 263 genes . We again observed that both protein and noncoding RVIS and GERP scores were significantly associated ( p < 1x10-16 ) with genes curated to be dosage sensitive in ClinGen ( Fig 2 ) . ncGERP had the highest AUC ( 0 . 78 ) among the set of scores ( Table 1 and Fig 2 ) . Based on the curated ClinGen list this indicates that genes with highly conserved noncoding sequence relative to the rest of the genome are strongly correlated with gene dosage sensitivity . Additional noncoding scores were constructed for both CADD [12] and GWAVA [13] , by taking the average of the nucleotide-level score across a gene’s noncoding sequence ( Methods ) . These are designated ncCADD and ncGWAVA , respectively . These scores were assessed against ClinGen’s dosage sensitivity genes as well; both ncCADD ( p = 1 . 5x10-12; AUC = 0 . 63 ) and ncGWAVA ( p = 4 . 7x10-17; AUC = 0 . 66 ) were also found to be significantly predictive of ClinGen’s dosage sensitivity genes ( Fig 2 ) . We performed a joint logistic regression model to investigate performance in predicting ClinGen dosage sensitive genes from the genome-wide background using six features: the two RVIS and two GERP scores supplemented with two additional noncoding scores derived from ncCADD and ncGWAVA ( S4 Table , Fig 2 ) . There was modest improvement ( AUC = 0 . 83 ) compared to a joint model using three features: ncGERP , ncRVIS and RVIS-CHGV ( AUC = 0 . 82 , Table 1 ) . The haploinsufficiency gene lists in the previous section relied on known Mendelian gene-disease relationships . Another way to identify a list of genes sensitive to gene dosage is to use the absence , where expected , of protein-coding loss-of-function ( LoF ) variants in a large human population ( Methods ) . Such a population-based LoF deficient gene list highlights genes where changes in expression levels could be selected against , yet are independent of known gene-disease associations . Using the standing variation from the ESP6500SI reference population we identified 1 , 235 LoF deficient genes ( FDR < 1% ) and 1 , 762 LoF control genes that had both an ncRVIS and ncGERP score assigned ( Methods ) . We found that 2 . 3% of the 1 , 235 LoF deficient genes overlap with known OMIM “haploinsufficient” genes compared to one ( 0 . 06% ) of the 1 , 762 control genes ( Fisher’s Exact test , two-tailed p = 2 . 4x10-10 ) . Given that the construction of the LoF deficient gene list is independent of gene-disease databases , this list could include genes where haploinsufficiency might be incompatible with life ( non-viable ) , genes that are yet to be associated to disease through haploinsufficiency , or genes that cause disease through mechanisms other than haploinsufficiency . The overlaps of the 1 , 235 LoF deficient genes and the 1 , 762 control genes with the OMIM disease gene list ( minus haploinsufficiency genes ) were 26 . 6% and 13 . 2% , respectively ( Fisher’s Exact two-tail p = 3 . 1x10-20 ) . We find that the median ncRVIS of the collective 2 , 997 genes is 50 . 8% . By comparing the distribution of ncRVIS scores between the 1 , 235 LoF deficient genes and the 1 , 762 LoF control genes we demonstrated that LoF deficient genes have significantly more intolerant noncoding sequence ( median 37 . 9% vs . 58 . 1%; Mann-Whitney U test , p = 7 . 1x10-34 , Fig 3A and S2A Fig ) . We repeated the LoF deficient assessment with ncGERP , which showed that LoF deficient protein-coding genes preferentially have a more phylogenetically conserved regulatory sequence ( median ncGERP 23 . 4% vs . 64 . 5%; Mann-Whitney U test , p = 3 . 4x10-171 , S2B Fig . To understand whether information can be gained from combining ncRVIS and ncGERP scores , we used a multivariate logistic regression model , which showed that ncRVIS ( p = 5 . 4x10-6 ) maintains a significant signal for predicting LoF deficient genes . This supports the expectation that regulatory functions specific to humans may not always be captured by ncGERP , while ncRVIS is likely picking up such patterns of human-specific selection within the regulatory sequence of genes where regulated dosage is critical to normal function . An investigation of alternative noncoding scores showed that both ncCADD and ncGWAVA were also significantly associated with LoF deficient genes ( Fig 3B and S2 Fig ) . Taken together , these data advocate prioritizing coding LoF mutations and potentially the regulatory region mutations among LoF deficient genes that have conserved or intolerant noncoding sequence ( Fig 3B ) . This conclusion is corroborated by earlier results showing that ncRVIS and ncGERP are both significantly predictive of OMIM and ClinGen disease genes with a primary mechanism of haploinsufficiency . Individual CNVs distorting single or contiguous gene dosage have been linked to human diseases [20] . Inversely , genes that tolerate CNVs in the general population are unlikely to be dosage sensitive [21] . In this section we extended our assessment of ncRVIS and ncGERP to CNVs by asking whether a relationship exists between genes that have been shown to overlap ( ≥50% of the consensus coding sequence [CCDS] ) with a deletion/duplication based on two study populations from Database of Genome Variation ( DGV ) [18]: Conrad et al ( 2010 ) and the 1K Genomes Project ( 2012 ) [17 , 19] . These two studies amass 1 , 602 individuals with comprehensive CNV data across 14 , 714 assessable CCDS genes . Of these assessable genes , 861 genes were found to have at least one CNV overlap among the combined population of 1 , 602 samples . Genic tolerance to CNVs shows a clear relationship with the genes whose regulatory sequence also tolerates variation . We found that , on average , the 861 genes with a CNV overlap in these public databases have significantly higher ncRVIS ( p = 2 . 3x10-28; AUC = 0 . 61 ) and ncGERP percentile scores ( p = 9 . 2x10-31; AUC = 0 . 62 ) than the 13 , 853 genes without a reported CNV overlap in those data . Moreover , in a multivariate logistic regression model , RVIS ( p = 3 . 1x10-26 ) , ncRVIS ( p = 1 . 9x10-9 ) and ncGERP ( p = 8 . 9x10-12 ) each individually contribute to an improved overall prediction of genes that tolerate CNVs ( AUC = 0 . 68 ) . The current data indicates that genes tolerating CNVs in the general population are also more likely to tolerate variation in their noncoding regulatory sequence . With 13 , 853 genes reporting no CNV overlap in this CNV dataset , much larger populations of high-quality , genome-wide CNV data are required to appropriately assess the question of whether intolerance in the regulatory sequence of a gene can strongly predict intolerance to specifically CNV deletions . The correlation between RVIS-CHGV and ncRVIS is r2 = 0 . 04 ( S1A Fig ) . We included RVIS-CHGV and ncRVIS in a multivariate logistic regression model and found that the signals from RVIS-CHGV and ncRVIS provided significant independent information in predicting OMIM haploinsufficiency genes annotated as carrying de novo pathogenic mutations . This multivariate logistic regression achieves an AUC estimate of 0 . 77; higher than each of the RVIS-CHGV ( AUC = 0 . 75 ) and ncRVIS ( AUC = 0 . 63 ) models . Next , we generated two additional scores for each gene ( Fig 4A ) . The first was a combined genic intolerance assessment that considers the sum of the regulatory and protein-coding sequence by summing the values corresponding to a gene’s RVIS-CHGV and ncRVIS genome-wide percentiles , termed “RVIS-sum . ” Using the list of OMIM haploinsufficient genes , we found that 84% of genes are in the lower 50th percentile of RVIS-sum scores ( Fig 4B ) . The second score is meant to reflect the extent to which these two measures diverge , which we term “RVIS-diff . ” A positive RVIS-diff score indicates that the noncoding regulatory sequence of the gene is ranked as more intolerant than the protein-coding sequence of the same gene ( Fig 4A ) . To assess how ncRVIS may be useful in interpreting mutations among patients , we specifically evaluated ncRVIS in the context of loss-of-function ( LoF ) de novo mutations reported across cohorts of individuals ascertained for the presence[22–33] and absence[30–34] of neuropsychiatric disorders . Here , loss-of-function de novo mutations were defined as nonsense , canonical splice and protein-coding indels that occurred within CCDS sequence and were absent in the ESP6500SI database . Firstly , among controls , we identified 180 LoF de novo mutations and the median ncRVIS percentile score of the genes those mutations were found in was 45 . 8% . When we considered the 494 LoF DNMs identified across cohorts of simplex trios ascertained for various neuropsychiatric disorders , we found that the LoF DNMs preferentially occurred among noncoding intolerant genes , with the median ncRVIS being 36 . 2% . No single LoF de novo mutation was observed twice among controls . Among cases , a SCN1A splice-donor de novo mutation was found among two probands , both ascertained for an epileptic encephalopathy [28] . Taking the combination of ncRVIS and protein-coding RVIS , the RVIS-sum vector , we found that among controls the median RVIS-sum was 85 . 91 , while among neuro-ascertained cases it was 70 . 30 ( Mann-Whitney U 2-tail test p = 0 . 001 , S3A Fig ) . The significance remained even after excluding 19 loss-of-function de novo mutations among six previously known disease genes: CDKL5 , NRXN1 , SCN1A , SCN2A , STXBP1 and SYNGAP1 ( Mann-Whitney U 2-tail test p = 0 . 008 ) . A similar assessment is to use the information from a gene’s noncoding and protein-coding percentiles to calculate a single metric that reflects Euclidean distance from the most intolerant coordinate ( 0 , 0 ) . Genes close to the ( 0 , 0 ) coordinate are characterized as having both the most intolerant noncoding and protein-coding sequence . We found that loss-of-function de novo mutations among cases preferentially occurred among genes closer to ( 0 , 0 ) with a median Euclidean distance for case-ascertained LoF DNMs of 0 . 588 compared to 0 . 698 for control LoF DNMs ( Mann Whitney U test , p = 0 . 0035 ) . A logistic regression model regressing case/control LoF DNM assignment on the Euclidean distance achieved an AUC of 0 . 58 ( S3B Fig ) . We then combined the genic information from the Euclidean distance metric with the previously defined loss-of-function deficiency bioinformatics signature . We took only the LoF de novo mutations that fell in genes with a Euclidean distance ≤0 . 4 and also occurred in loss-of-function deficient genes with no more than a single LoF variant reported among the Exome Variant Server ( EVS ) [16] ( S5 Table ) . This identified nine observations among the controls—corresponding to 5 . 0% of LoF DNMs—and 70 observations among cases , corresponding to 14 . 2% of all LoF de novo mutations among cases ( Fisher’s Exact test two-tail p = 6 . 5x10-4; odds ratio of 3 . 2 ) ; a modest boost to what we got when we relied solely on the loss-of-function deficient bioinformatics signature ( Fisher’s Exact two-tail p = 9 . 5x10-4; odds ratio of 2 . 4 ) . The list of genes carrying one of the 70 case loss-of-function de novo mutations includes established genes: NRXN1 , SCN1A and SCN2A . The list also includes recently implicated genes: CHD2 [35] , CHD8 [36] , KMT2E [37] , MBD5 [35] , SETD5 [38] , and WDFY3 [39] . It is important to note that among the cases , the above six loss-of-function mediated pathogenic genes were of unknown significance when the de novo mutation data were first reported . This helps highlight the utility of this loss-of-function bioinformatics signature . The remaining case loss-of-function de novo mutations include some Mendelian disease genes with an existing neurological association , such as NIPBL , which is known to cause Cornelia de Lange syndrome [40] and KMT2A , which is known to cause Wiedemann-Steiner Syndrome [41] . The remaining genes with the same bioinformatics signature as the above established genes are: ANK2 , ARHGAP5 , ASH1L , BRD4 , BTAF1 , DLL1 , DNAJC6 , DOT1L , EPHB2 , FAM91A1 , GIGYF2 , INTS6 , ITGA5 , KIAA1429 , LARP4B , MED13 , MED13L , NCKAP1 , NOTCH1 , PHF3 , POGZ , RALGAPA1 , RALGAPB , RANBP2 , RB1CC1 , SPAG9 , STAG1 , UBN2 , UBR5 , ZC3H4 and ZNF292 ( S5 Table ) . It is unclear which of these genes could have their gene-disease association confirmed in the coming years; however , five of these candidates already have multiple LoF de novo mutation observations across neuropsychiatric ascertained patients: ANK2 , MED13L , NCKAP1 , POGZ and ZNF292 .
Developing methods to recognize functional mutations in the regulatory part of the human genome is widely recognized as one of the central challenges facing modern human genetics . The difficulty is well illustrated by the results of the ENCODE project . Considerable effort and progress has been made in identifying parts of the genome with clear regulatory potential based on experimentally confirmed transcription factor binding sites and related approaches . However , since much of the genome is currently assigned a possible regulatory role it is difficult to use only those data to prioritize mutations in the study of human disease . Here , we show that population genetic and phylogenetic approaches can help fill this gap by adding further information about the possible functional role of a noncoding stretch of sequence . Integrating these approaches with the sequence regions identified by ENCODE [42] and related studies may ultimately prove to be the most effective approach . There are many additional regulatory sequences that can be included using the framework described here . Examples include distal enhancers , noncoding RNAs and larger promoter regions . However , correctly and unambiguously associating distal regulatory elements to the genes they regulate requires highly curated data , which is not yet straightforward to acquire . Therefore , here we focus only on regulatory sequences that can be unambiguously associated with specific genes in order to test the ability of the noncoding exome sequence to predict genes that cause human disease via gene dosage aberrations . Using multiple resources , we show that dosage sensitive genes have distinct patterns of genetic variation in their proximal noncoding regulatory sequence . To the extent that more distant regulatory sequences may also carry variants that influence expression , we may expect a correlation between the intolerance patterns of a gene’s proximal and distal regulatory sequence . This possibility suggests that a sliding window of intolerance data throughout the human genome may provide a valuable new tool for identifying important regulatory sequence . Interpreting genome wide patterns of intolerance and relating those patterns to genes will not be a trivial task , but our results imply that genome wide patterns of intolerance have the potential to provide an important complement to other tools [42] used to identify important regulatory parts of the genome . ncRVIS is a ‘regional-based’ guide to patterns of standing variation in the proximal noncoding sequence of a gene in the human population ( Fig 1 ) . It leverages the collective information from the standing variation in a stretch of noncoding sequence to assess whether that stretch of noncoding sequence has more or less polymorphic variation than expected . This is distinct from variant based scores that look at individual variants . By identifying stretches of noncoding sequence with preferential depletion of standing variation we are hypothesizing that in many cases this is driven by purifying selection among the human population acting against variation in that noncoding region as a whole , rather than at an individual variant site . We and others have previously found that for the protein-coding sequence , RVIS and other estimates of human constraint are more indicative of disease causing genes than mammalian conservation [15 , 43] . However , in its current formulation , ncGERP outperforms ncRVIS in all current assessments . There can be a few explanations for this . Firstly , it is possible that the coding region is highly conserved throughout the genome to the point that there is limited allowance for big enough deviations between genic conservation in order to create an informative ranking . However , the noncoding regions may be more prone to allow such deviations . Secondly , the current ncRVIS formulation is based on a comparatively modest cohort ( n = 690 samples ) . There are two reasons we think ncRVIS remains important in light of the stronger signal observed from ncGERP . First , as we have shown throughout this work , the two scores are only weakly correlated ( r2 = 0 . 06 , S1K Fig ) and ncRVIS can add information beyond ncGERP . This is evidenced by the various dosage sensitive gene list assessments including the ClinGen assessment where in a joint logistic regression model of just ncRVIS and ncGERP , ncRVIS had a significant contribution ( p = 7 . 2x10-7 ) . This likely occurs , at least in part , for the interesting reason that there are genomic regions that have important functions only in humans . Evolutionary conservation will miss these regions , population genetic approaches will not . Second , the performance of ncGERP is close to its limit , as we already have a fairly good assessment of which sites are phylogenetically constrained , and which are not . ncRVIS , however , we anticipate will increase in predictive value as sample sizes of sequenced genomes grow , and thus a more extensive dataset of noncoding standing variation is available . Alternative noncoding predictors of dosage sensitive genes , which take the overall propensity for a gene’s proximal noncoding sequence to score as more ‘functional’ based on the average nucleotide-level CADD or GWAVA scores , suggest that nucleotide-level predictors of noncoding functionality do appear to detect additional signatures of regulatory function beyond conservation . We observe correlation between a gene’s ncGERP and ncCADD score ( r2 = 0 . 32 , S1V Fig ) , and to a lesser degree its ncGWAVA score ( r2 = 0 . 06 , S1AA Fig ) , as a result of their dependence on conservation-based signals in their construction . In a joint model , however , we found that both ncCADD and ncGWAVA provide signal independent of ncGERP and ncRVIS when predicting human dosage sensitive genes ( Fig 2 ) . This suggests that even though conservation is a major component of their predictive signal for ClinGen’s dosage sensitive genes , additional information not directly captured by conservation might be captured by these two ensemble predictors ( S4 Table ) . Currently , the basic paradigm to analyze protein-coding sequence is to use aggregate statistics that integrate the effect of different rare mutations affecting the same functional unit , often defined as the protein-coding sequence of a single human gene . This has proven effective in whole-exome sequence data because we know the protein-coding sequence boundaries we need to consider in order to effectively aggregate variants that affect the same functional unit [14] . In order to effectively interpret whole-genome regulatory sequence data , and find the noncoding regions that harbor risk-influencing mutations , we need to learn to recognize the functional noncoding stretches of sequences that affect gene expression . Current annotations lack specificity to define truly functional noncoding regions . Here , we have shown that a phylogenetic and population genetic framework can help define and prioritize the functional noncoding regions , and this is expected to improve when combined with information about sequences with regulatory potential from ENCODE [42] and related resources . Here , we also explore additional signals beyond conservation and human standing variation by assessing the dosage sensitivity predictive value of ncCADD and ncGWAVA scores , two nucleotide-level scoring frameworks that in addition to capturing signals of conservation , leverage other features and annotations from the noncoding sequence . Such an integrated framework will enable the definition of intolerant noncoding regulatory regions that have been under both strong evolutionary ( ncGERP ) and human population ( ncRVIS ) constraint . For these reasons , ncRVIS and related approaches are likely to play a key role in the development of a statistical genetic framework to support the interpretation of large scale whole genome sequence data that will soon emerge , for example through the recently announced National Human Genome Research Institute ( NHGRI ) call for genomics of common disease centers . In this context , it is essential to appreciate that the resolution of the ncRVIS approach depends critically on the total number of individuals that have been sequenced , and therefore its value is expected to continue to increase as whole-genome sequenced sample sizes increase .
Eleven data sources were used to develop and assess noncoding RVIS ( ncRVIS ) and noncoding GERP ( ncGERP ) . As exome sequencing kits only capture a fraction of the untranslated region ( UTR ) sequence of genes , we utilized human whole-genome sequenced samples from the Institute for Genomic Medicine , Columbia University database ( formerly Center for Human Genome Variation ( CHGV ) , Duke University ) to assess noncoding intolerance . We used Consensus Coding Sequence ( CCDS ) release 14 as the set of protein-coding genes of interest for scoring noncoding intolerance [44] . We used Ensembl 73 to define the UTR sequence of CCDS genes that did not overlap with CCDS protein coding regions of the same or overlapping genes [45] . We extracted gene-lists from OMIM database to reflect differing genetic models . We extracted a heavily curated list of haploinsufficient or triplosensitive genes from ClinGen’s Genome Dosage Map ( http://www . ncbi . nlm . nih . gov/projects/dbvar/clingen/ ) . For copy number variants ( CNVs ) , we identified a set of deletions and duplications reported across two published studies: The 1K Genomes Project and Conrad et al . ( 2010 ) [17 , 19] . We used the GERP++ database to derive noncoding and protein-coding regional GERP scores to compare to phylogenetic conservation at the genic level [6 , 7] . We also used two noncoding ensemble nucleotide-level predictors , CADD [12] and GWAVA [13] , to derive noncoding regional scores for each gene’s noncoding sequence as done for GERP++ . Finally , we relied on the ESP6500SI [16] database to extrapolate a loss-of-function ( LoF ) deficient gene list , based on observing less than expected LoF variants in a gene . To define the noncoding sequence for each gene , we relied on the Ensembl 73 noncoding annotation from that gene’s canonical transcript ( downloaded 19th September 2013 ) . We refer to noncoding exonic sequence of genes as the collection of 5'-UTR , 3'-UTR and an additional non-exonic 250bp upstream of transcription start site ( TSS ) . The 5’ and 3’ UTR are derived based on the canonical transcript annotation . The additional 250bp upstream of the TSS is defined as the 250 bases upstream of the initial exon junction , taking into consideration whether the transcript lies on the ( +/- ) strand . For three Ensembl genes ( PKD1L2 , SPIB , and UGT2A1 ) that had multiple canonical transcripts , we took the larger of the two canonical transcripts . Given the challenge of defining the upstream promoter region , we opted to choose a relatively small number of bases immediately adjacent to the TSS , and this was set at 250 for all genes . Defining different sized promoters per gene guided by the distribution of conservation ( e . g . , GERP++ scores ) or human polymorphism density would create a situation where we specifically define the promoter region of the gene we are assessing based on the more intolerant or more conserved set of bases upstream of the TSS . Evidently , this could create a bias towards intolerant or conserved promoters in our score , and therefore we prefer for this formulation to define the upstream promoter regions agnostic to the data we use for the assessments . The initial dataset was comprised of 56 , 715 noncoding units . These 56 , 715 units resulted in 19 , 563 unique Ensembl genes . We found that 18 , 507 genes had a 5’ UTR ( average = 260 bases , median = 182 bases ) . 18 , 638 genes had a 3’ UTR . And , by design , all 19 , 563 genes had a 250bp promoter region . Of the 19 , 563 genes , 18 , 148 ( 92 . 77% ) had all three noncoding units represented . 849 had only two units represented , whereas 566 were based solely on their promoter unit , with no UTR boundaries defined . The overall genomic noncoding sequence comprised of these 19 , 563 unique Ensembl genes is 34 , 065 , 650 bases . These reflect the number of sites prior to exclusion of inadequately covered sites and sites that overlap with protein-coding position among other genes , as discussed below . We found that whole-exome sequencing is inadequate for capturing the noncoding exonic sequence of protein-coding genes . To derive a noncoding RVIS , and to generate a comparative protein-coding RVIS based on the same subset of samples , we selected 690 internally sequenced ( CHGV ) control-approved whole genomes ( 78% Caucasian ancestry ) . For these 690 whole-genome sequenced samples , an average of 92 . 7% sites were covered , with at least 10-fold coverage across the 34 , 065 , 650 Ensembl defined noncoding sites . Similarly , relying on CCDS release 14 for the protein-coding sequence , we observed that these 690 whole-genome sequenced samples covered on average 94 . 6% of the 33 , 266 , 994 protein-coding sites in CCDS release 14 with at least 10-fold coverage . The set of phenotypes contributing to the whole-genome sequenced set of 690 cases is summarized in S1 Table . It is our experience that sites sequenced with consistently good coverage represent sites with more reliable alignment and variant calling than sites where coverage is sparse and inconsistently represented among a population . S4A Fig ( blue curve ) represents the number of our 690 whole-genome sequenced samples that had at least 10-fold coverage ( Y-axis ) versus the cumulative percentage of the 34 . 1Mbp Ensembl-defined UTR noncoding sites ( X-axis ) . The intersection between the blue curve and green line ( an illustrative cutoff ) indicates that at this point approximately 92% of samples have at least 10-fold read coverage at approximately 83% of the Ensembl noncoding sites . Together , the green threshold line and the blue sample-site coverage profile partition the space into four regions . Region II and region III represent the overall heterogeneity of coverage and the amount of noncoding sequence pruned from analysis , respectively . Shifting the green line left retains noncoding sequence ( smaller region III ) at a cost of increased coverage heterogeneity ( larger region II ) . Moving the threshold right reduces the noncoding sequence used in the analysis , but also reduces the noise from coverage heterogeneity . There are multiple ways to select a cutoff from these data; however , a balanced approach is to choose a cutoff that ensures region II and region III are as close as possible in terms of area . To evaluate the area for region II and region III , we first smooth the sample-coverage profile ( blue curve ) by fitting smooth spines , as illustrated by S4B Fig where the blue dots represent the original profile , while the red curve represents the smoothed splines . The smoothed curve traces the original data well . We use the smoothed curve to compute the areas for region II and region III ( through numeric integrations ) for a selection of evenly spaced cutoff values . The areas for region II and region III for different cutoff values is shown in S4C Fig , with red and blue curves representing region II and region III , respectively . We choose the balanced cutoff to be the point where the curves intersect , representing a balance between loss of data ( noncoding sequence sites ) and variability from coverage heterogeneity . The method yields an optimal value of 0 . 074 based on the noncoding sequence data . This suggests that removing the 7 . 4% most inconsistently covered noncoding sites corresponds to requiring noncoding sites to have at least 67% of samples with at least 10-fold read coverage . We selected 70% for the manuscript , corresponding to removing the 8% ‘noisiest’ noncoding sites with respect to inadequate coverage at the population level ( S4D Fig ) . We performed a sensitivity test varying the 70% threshold to a threshold of 60% ( r2 = 0 . 986 ) or 80% ( r2 = 0 . 971 ) and show that the final ncRVIS score is not highly sensitive to varying this threshold ( S5 Fig ) . Requiring ≥70% of the 690 samples have at least 10-fold coverage at a site prunes the noncoding sequence down to 31 , 355 , 520 ( 92 . 0% ) of the initial 34 , 065 , 650 Ensembl-defined UTR noncoding sites . For the CCDS release 14 protein-coding sites , we found that this pruning process retained 31 , 528 , 600 ( 94 . 8% ) of 33 , 266 , 994 CCDS sites . Although it is expected that some variants in protein-coding sequence will affect gene regulation and that these would be easily associated with the genes they fall in , we excluded all protein-coding regions in order not to confound the ncRVIS score with protein-coding sequence . Through this additional step , we ultimately retained 31 , 112 , 586 ( 91 . 3% ) of the 34 , 065 , 650 noncoding sites . We find that on average each of the 690 genomes has at least 10-fold coverage across 97 . 8% of the 31 . 1Mbps of noncoding sequence used to derive ncRVIS . Overall , the GC content of the 5’ UTR sequence is 61% in comparison to the GC content of the 3’ UTR sequence which is 42 . 5% . Combining the three noncoding components into a single genic noncoding unit resulted in 19 , 484 ( 99 . 6% ) of the 19 , 563 Ensembl genes retaining at least one noncoding component . The average length of noncoding sequence across the 19 , 484 Ensembl 73 genes was 1 , 597 ( median = 1 , 096 sites ) . Finally , we defined ncRVIS “assessable” genes as Ensembl genes not located on the Y chromosome , and with at least 70% of their noncoding sequence surviving the aforementioned filters . Through this , we retained 16 , 273 CCDS release 14 protein-coding genes that fulfilled the coverage requirements of having at least 10-fold coverage of at least 70% of the gene protein-coding sites across at least 70% of the CHGV whole-genome sequenced samples . The overlap between CHGV-derived RVIS and ncRVIS indicates that 15 , 471 genes were “assessable” for both coding and noncoding RVIS ( S1 Data ) . To accommodate the uncertainty surrounding the percentage of noncoding sequence sites that are neutral , we used an alternative metric to reflect mutability of a given sequence context in our ncRVIS and RVIS-CHGV adaptations . For the sites reflecting a genic unit ( noncoding or coding ) we use an in-house script developed by Dr Yujun Han . This script leverages the DNA tri-mer mutation rate matrix ( kindly provided by Drs . Shamil Sunyaev and Paz Polak of The Broad Institute of MIT and Harvard , Cambridge ) to generate a mutation rate for a given genic unit , which is calculated for each gene by summing the point mutation rates across the effectively captured sequence [28] . The mutation rate model provides an estimated rate of mutation per base . The rate is based solely on three bases: the interrogated base , the base immediately before , and the base immediately after the interrogated base . The model is based on human , chimpanzee and baboon genomic sequences [46] . The mutation rate model does not currently account for effects of larger sequence context or biological processes that affect mutation rate , such as background selection , distance to CpG islands , or replication timing . At the level of the gene , like others [43] , we find very high correlation ( r2 = 0 . 95 ) between gene coding length and mutation rate . While the high correlation suggests it is possible to use gene size as a proxy , we prefer leveraging the mutation rate to accommodate for some additional information that is likely lost when using gene size . The source code can be found in S3 Data . All sequencing was performed on the Illumina HiSeq2000 platform ( Illumina , San Diego , CA ) in the Genomic Analysis Facility in the Center for Human Genome Variation ( CHGV ) at Duke University . After sequencing , reads were aligned to Genome Reference Consortium Human Genome build 37 ( GRCh37 ) using the Burrows-Wheeler Alignment Tool ( BWA ) [47] and PCR duplicates were removed using Picard software ( http://picard . sourceforge . net ) . The reference sequence we used is identical to the 1000 Genomes Phase II reference and it consists of chromosomes 1–22 , X , Y , MT , unplaced and unlocalized contigs , the human herpesvirus 4 type 1 ( NC_007605 ) , and decoy sequences ( hs37d5 ) derived from HuRef , Human Bac and Fosmid clones and NA12878 . Variants were called using the Genome Analysis Toolkit[48] . SnpEff was used to annotate the variants[49] . To construct the ncRVIS score , we defined the minor allele frequency threshold dividing “common” and “rare” variants as ρ . To identify the number of variants with a MAF > ρ in the noncoding region of a gene , we use an in-house package , Analysis Tool for Annotated Variants ( ATAV ) . ATAV communicates with our in-house relational database that houses all the variant call ( and non-carrier ) information for all sites across each of the 690 whole-genome sequenced samples . Additional filtering consisted of excluding indel calls and requiring a minimum of 10-fold coverage to call a variant ( or be confident that a variant wasn’t present in a non-carrier sample ) . To increase confidence in called variants the following additional filters were applied: relying on GATK VQSLOD “pass” and “intermediate tranches , ” requiring a QUAL score of at least 30 , a QD ( quality by depth ) score of at least 2 , a genotyping quality ( GQ ) score of at least 20 , and a fisher strand bias ( FS ) score of less than 60 . For noncoding regions , we considered all common variants residing in the noncoding sequence as contributors to ( Y ) , the number of common variants . For the CHGV-based protein-coding RVIS score ( based upon the same 690 whole-genome sequenced samples as ncRVIS ) , we adopted the same criteria as in our earlier work introducing RVIS . That is , synonymous protein-coding variants did not contribute to the number of common ‘functional’ variants when deriving the CHGV protein-coding RVIS score . However , we did assess a secondary score ( RVIS-Yall ) for comparison purposes . RVIS-Yall considered all protein-coding variants as eligible to contribute to ( Y ) , including the putatively neutral , synonymous coding variants . We defined Y as the total number of common ( Minor Allele Frequency [MAF]>ρ ) SNVs in the noncoding sequence of a gene , and X as the effective mutation rate of the noncoding sequence of the gene , using the mutation matrix described previously . We then regressed Y on X and took the studentized residual as the noncoding Residual Variation Intolerance Score ( ncRVIS ) . The raw residual was divided by an estimate of its standard deviation to account for differences in variability that come with differing mutational burdens . The ncRVIS then provides a measure of the departure from the ( genome-wide ) average number of common variants found in the noncoding sequence of genes with a similar amount of noncoding mutational burden . When S = 0 , the gene has the average number of common noncoding variants given its total mutational burden; when S<0 , the gene has fewer common noncoding variation than predicted; when S>0 , it has more . Each ncRVIS is then translated to a corresponding percentile to reflect the relative position of that gene on the genome-wide spectrum of ncRVIS based on the relative intolerance of that gene’s noncoding sequence . S1 Data contains the X and Y estimates used to construct ncRVIS . The R code to reproduce ncRVIS relies on the MASS package [50]: studres ( glm ( Y~X ) ) . As we only had 690 whole-genome sequenced samples available , we chose to adopt a MAF threshold ρ of 1% for the noncoding RVIS and RVIS-CHGV . We had in our previous publication explored the behavior of the original RVIS for ρ of 0 . 01% and 1% , and found both of these to be strongly correlated with ρ = 0 . 1% ( Pearson's r = 0 . 849 and Pearson's r = 0 . 813 , respectively ) . The collection of genomes used to derive ncRVIS includes various sample ascertainments ( S1 Table ) . Given that we use the mutation rate matrix to define the underlying mutation rate ( X ) , and only consider variants with a MAF>1% when determining ( Y ) , we consider it highly unlikely that case-ascertained variants could be systematically influencing the current ncRVIS or RVIS-CHGV scores . We highlight F8 as the single gene that might require careful interpretation due to our collection of WGS samples that were ascertained for haemophilia . Under the residual variation intolerance framework , ncRVIS will not correlate with either the noncoding mutability or noncoding sequence length . To confirm this , we find that the correlation between ncRVIS and the corresponding mutability or size of the effective noncoding sequence to be r2 = 3 . 0x10-8 and r2 = 2 . 0x10-5 , respectively ( S1B and S1C Fig ) . We further confirmed that the ncRVIS is not strongly correlated to the corresponding genes ‘protein-coding’ sequence size or ‘protein-coding’ mutability: r2 = 0 . 0031 and r2 = 0 . 0026 , respectively ( S1D and S1E Fig ) . We do note , however , that there is high correlation between a gene’s noncoding sequence length ( number of bases ) and its derived mutability using the mutation rate matrix ( r2 = 0 . 9493 , S1F Fig ) . We first assessed the likely impact of using the estimated mutation rate instead of the observed variation by comparing two formulations of the original RVIS . To construct RVIS-mut we replaced the observed variation among the EVS population with the estimated mutation rate for that gene to represent ( X ) and kept the original Y variable from RVIS . Reassuringly , we find that RVIS-mut , using the estimated mutation rate , correlates highly ( Pearson’s r2 = 0 . 83 ) with that using the total number of variants in each gene ( RVIS ) ( S2 Table and S1G Fig ) . We next evaluated the effect of not being able to identify functional mutations by comparing RVIS to a third formulation ( RVIS-YALL ) . For RVIS-YALL we again use the effective mutation rate to represent X; however , we now permit all common protein-coding variants ( including synonymous variants ) for the Y variable . We find that RVIS-YALL remains highly correlated with the original RVIS ( Pearson’s r2 = 0 . 59 , S1H Fig ) ; more importantly , it remains predictive of genes causing Mendelian disease ( S2 Table ) . Finally , we show that a fourth formulation of RVIS , using an independent set of 690 whole-genome samples that were sequenced at the CHGV ( RVIS-CHGV ) , remains highly correlated with the original RVIS ( Pearson’s r2 of 0 . 63 , S1I Fig ) despite a decreased sample size , and continues to be significantly predictive of the disease gene lists , with the exception of genes causing recessive disease ( S2 Table ) . These comparisons suggest that , in principle , the ncRVIS formulation should work similarly to RVIS when regulatory sequences are subject to purifying selection . In our original RVIS paper we used omega ( ω ) as the phylogenetic approach to compare non-synonymous substitutions per non-synonymous site ( dN ) to the synonymous substitutions per synonymous site ( dS ) ( aka Ka/Ks , dN/dS ) . Given we are now interested in noncoding sequence , we have generated an alternative estimate to assess whether correlation exists between the ncRVIS and that of possible phylogenetic conservation at noncoding sites . For each gene we constructed two conservation vectors: one reflecting the noncoding sequence of a gene after excluding protein-coding overlapping sites ( ncGERP ) , and the other reflecting the protein-coding sequence of a gene ( pcGERP ) . Both conservation vectors were based on the average GERP++ score [6] of the qualifying chromosomal sites within the defined sequence . We found that ncGERP and pcGERP were moderately correlated to each other ( r2 = 0 . 30 ) . Compared to ncRVIS , both ncGERP and pcGERP had low correlation: r2 = 0 . 06 and r2 = 0 . 04 , respectively . Likewise , compared to previously described RVIS [15] , both ncGERP and pcGERP had relatively low correlation r2 = 0 . 06 and r2 = 0 . 15 , respectively . These five correlation tests were performed based on the 14 , 998 genes with the corresponding scatterplots available in S1J–S1N Fig . Interestingly , we found that pcGERP was inferior to ncGERP when comparing the 1 , 235 LoF deficient genes to the 1 , 762 LoF control genes , as described above ( median pcGERP 32 . 38% vs . 65 . 57%; Mann-Whitney U test , p = 5 . 6x10-141; in comparison to median ncGERP 23 . 39% vs . 64 . 49%; Mann-Whitney U test , p = 3 . 4x10-171 ) . While protein-coding genes are generally fairly phylogenetically conserved overall , there is variability inside the protein-coding genes in the phylogenetic conservation that correlates with whether a site causes disease or not . Overall , however , the majority of protein-coding genes are reasonably conserved across species . This leaves less scope for pcGERP variability among genes that can then be related to disease gene status ( Fig 5 ) . This is less true for the regulatory regions , where single-site variation is unlikely to systematically experience the same overall constraint as sites coding for structural components of the proteins . As a consequence of this , there is more scope for variability among the average ncGERP across the genome-wide spectrum of genes ( Fig 5 ) . Literature includes alternatives to GERP++ for quantifying the degree of importance ( sometimes referred to as functionality ) of noncoding sequence in the human genome . Unlike GERP++ , which is a direct measure of the phylogenetic conservation of a single site or a stretch of sequence , more recent alternatives include ensemble based predictors that leverage many features beyond conservation . Although we recognize that nucleotide-level scores were constructed specifically for variant-level assessments; we nonetheless investigate whether a regionalized version of these scores could add information to predicting dosage-sensitive gene lists as well or better than ncRVIS or ncGERP . To this end , we calculated noncoding regional scores based on two popular nucleotide-level scoring frameworks: CADD [12] and GWAVA [13] . Using the same coordinates as ncGERP , we took the average CADD and GWAVA scores across the defined noncoding regions as a gene’s noncoding score . To calculate regional noncoding CADD scores , referred to as ncCADD , we used the scaled C-scores from CADD version 1 . 0 [12] . In a regionalized form , ncCADD reflects the average CADD score for all possible single nucleotide substitutions across a gene’s defined noncoding sequence . For regionalized noncoding GWAVA , referred to here as ncGWAVA , we downloaded the required training data and scripts from ( ftp://ftp . sanger . ac . uk/pub/resources/software/gwava/ ) and followed instructions given by the developers to generate the site-specific scores for all noncoding exome sites . We were advised that for UTR sequence the TSS-distance matched training set would be optimal ( personal communication with Dr . Graham Ritchie ) . Using the TSS-distance matched training set we derived the GWAVA score for each noncoding nucleotide site in a gene’s defined noncoding exome sequence and took the average to be the gene’s ncGWAVA score . Neither CADD nor GWAVA were specifically developed to be interpreted as regional assessments . However , understanding the overall importance of a gene’s noncoding sequence as inferred from CADD or GWAVA could still be of interest . Both noncoding scores are provided in S1 Data . Scatterplots assessing correlations with other scores ( including ncGERP ) are available in S1S–S1AA Fig . To assess the possible contributions of each ncRVIS subunit , we generated an ncRVIS score for promoter regions , 5’ UTR regions , and 3’ UTR regions for the set of 10 , 726 genes that had “assessable” sequence across all three distinct noncoding subunits . To permit comparisons with the original RVIS score , we further restricted comparison to the 9 , 644 distinct genes that also had a published RVIS score ( Petrovski et al . 2013 [15] ) , an assessable ncRVIS score . We find that the highest correlation with the ncRVIS score comes from the 3’ UTR ncRVIS ( r2 = 0 . 79 ) , compared to promoter and 5’ UTR regions , which had r2 correlation of 0 . 25 and 0 . 20 , respectively ( S3 Table ) . To generate a loss-of-function ( LoF ) deficient gene list , we take the five distinct mutation rates provided per gene by Samocha et al . ( 2014 ) [43] and calculate the expected frequency of protein-coding loss-of-function variants for each assessable consensus coding sequence ( CCDS ) gene ( Ps ) . We achieve this by first summing the mutation rates corresponding to the three loss-of-function variant effect classes ( nonsense , splice and frameshift ) and dividing that by the total sum of the mutation rates of every possible mutation effect in the gene . We then use the resulting rate to determine the percentage of variants in a gene that we expect to result in a LoF effect , accommodating for the mutation rate . Based on the above , the average percentage of possible protein-coding mutations in a gene that are expected to result in a loss-of-function annotated variant ( whether it is subsequently selected against or not ) is ~9 . 22% of the sum of all possible protein-coding and canonical splice site mutation events . We then use the ESP6500SI database ( accessed 20th March 2013 ) to extract for each gene both the total number of observed unique variants ( n ) and specifically the number of observed unique loss-of-function variants reported in the CCDS of each gene ( x ) . This gives us our observed rate of LoF variants given all the protein-coding variation identified in the gene . Taking a gene’s expected percentage of unique loss-of-function variation under neutrality as calculated by ( Ps ) , a subset of 1 , 235 genes with ncRVIS and ncGERP scores were identified as being significantly deficient of loss-of-function variants using a one-sided binomial exact test with Benjamini & Hochberg false discovery rate multiple-testing correction ( FDR = 1% ) [51] ( S4 Data ) . As a comparative group , we identified a set of 1 , 762 ‘control’ genes where we observe greater than the expected number of loss-of-function variants . While this list of LoF control genes cannot be interpreted as significantly LoF tolerant , we consider the list a useful comparative group to the genes found to be significantly LoF deficient . It is clear that we are currently missing some true LoF intolerant genes due to insufficient resolution ( power ) from the EVS reference cohort . The result of this reduced power is that the majority of the exome is considered non-informative for LoF deficiency . Larger cohorts will enable better discrimination of truly LoF deficient genes . However , even though it is currently an incomplete list , the list of genes that are already significantly LoF deficient is already a valuable resource . Finally , to illustrate that this list is robust to false positives driven by how much of the gene has been effectively sequenced , we repeated the exact implementation only this time asking whether any genes were significantly deficient of synonymous ( presumed neutral ) variation . In comparison to the LoF assessment where we identified 1 , 235 genes with an FDR < 1% , genome-wide the lowest FDR among the synonymous assessment was an FDR of 61% , with no other gene achieving an FDR < 99 . 99% for the synonymous assessment . This further highlights the integrity of this approach to detect LoF deficient genes in the human genome . | Mutations in noncoding sequence can cause disease but are very difficult to recognize . Here , we present two approaches intended to help identify noncoding regions of the genome that may carry mutations influencing disease . The first approach is based on comparing observed and predicted levels of standing human variation in the noncoding exome sequence of a gene . The second approach is based on the phylogenetic conservation of a gene’s noncoding exome sequence using GERP++ . We find that both approaches can predict genes known to cause disease through changes in expression level , genes depleted of loss-of-function alleles in the general population , and genes permissive of copy number variants in the general population . We find that both scores aid in interpreting loss-of-function mutations and in defining regions of noncoding sequence that are more likely to harbor mutations that influence risk of disease . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [] | 2015 | The Intolerance of Regulatory Sequence to Genetic Variation Predicts Gene Dosage Sensitivity |
Plasmodium vivax is the most prevalent cause of human malaria in the world and can lead to severe disease with high potential for relapse . Its genetic and geographic diversities make it challenging to control . P . vivax is understudied and to achieve control of malaria in endemic areas , a rapid , accurate , and simple diagnostic tool is necessary . In this pilot study , we found that a colorimetric system using AuNPs and MSP10 DNA detection in urine can provide fast , easy , and inexpensive identification of P . vivax . The test exhibited promising sensitivity ( 84% ) , high specificity ( 97% ) , and only mild cross-reactivity with P . falciparum ( 21% ) . It is simple to use , with a visible color change that negates the need for a spectrometer , making it suitable for use in austere conditions . Using urine eliminates the need for finger-prick , increasing both the safety profile and patient acceptance of this model .
Malaria is the most common infectious disease in the tropics and subtropics [1] . Currently , P . vivax is endemic across Asia , the South Pacific , North Africa , Middle East , and South and Central America [2] , and has recently reappeared in regions where it had previously been eradicated , including North America and Europe [3] . Currently , an estimated 2 . 9 billion people live at risk of P . vivax infection [4] . Research in malaria has been primarily focused on P . falciparum , the most fatal form of malaria . However , P . vivax can also cause severe illness with serious complications and costs , especially in children , in whom it has a major impact on growth [5–7] . Relying on microscopic identification of malaria species jeopardizes malaria control due to its limitations [8 , 9] . The WHO recognizes a need for rapid , accurate , and easy diagnostic tools in order to control malaria and need for such a test is mounting in developing countries [10] . Currently available rapid diagnostic tests ( RDTs ) are controversial due to their sensitivity and specificity , and differentiate poorly between plasmodium species [11–13] . With the expanding use of nanotechnology in the biomedical arena , nanoparticles can play a role in low cost , innovative diagnostics [14] . Gold nanoparticles aggregate and change their color from red to purple-blue upon exposure to single stranded DNA in aqueous solution while double stranded DNA stabilize them to preserve their red color and thus present an opportunity to develop a fast and easily interpreted diagnostic test [15–20] . Merozoite Surface Protein 10 ( MSP10 ) is an immunogenic protein encoded by a single copy gene ( in P . falciparum , GenBan Accession PF3D7 0620400; in P . vivax , PVX_114145 ) which is expressed in the asexual blood stages of Plasmodium falciparum and P . vivax [21] . One of at least 10 epidermal growth factor domain-containing proteins , the role of MSP10 in the biology of Plasmodium parasites has yet to be determined . Plasmodium MSP10 proteins have been identified as being subject to positive selection for amino acid-changing polymorphisms at the population genomic level [22 , 23] . Population genomics studies identify signatures of global dispersal and drug resistance in Plasmodium vivax [24] . Blood based tests can discourage screening both because of the pain associated with finger-prick and because of social and cultural beliefs about blood sampling . Less painful , more culturally sensitive , and safer tools for malaria diagnosis should encourage participation in mass screening programs and improve public health [25] . Urine contains circulating P . vivax DNA in detectable quantities [26–28] and can therefore serve as a less invasive and more acceptable sample for malaria screening and diagnosis . Also , urine contains less interfering proteins and inhibitors than blood which allows easier DNA extraction [29] . Furthermore , urine provides lower risks to healthcare personnel , with reliable amounts of malaria DNA found in urine despite being substantially lower than blood samples [30] . Additionally , urine color is not expected to obscure color change of gold nanoparticles . In this pilot study , we tested the hypothesis that a colorimetric system using gold nanoparticles and MSP10 DNA detection in urine would be useful as a safe diagnostic and surveillance tool for P . vivax . Such a tool is needed for improving malaria control in the endemic setting .
Citrate reduced gold 15nm nanoparticles , and KCl were purchased from Sigma Aldrich ( St . Louis , MO , United States ) . PBS was obtained from Invitrogen ( Grand Island , NY , United States ) . NaCl and NaOH were acquired from Merck Millipore ( Kenilworth , NJ , United States ) . The two MSP10 oligonucleotides utilized in this study were a generous gift from Professor Mirko Zimic ( Universidad Peruana Cayetano Heredia , Lima , Peru ) and designed by Dr . Joseph Vinetz ( University of California San Diego , United States ) . Crafted to represent the C-Terminal segment of MSP10 , the first oligonucleotide has a sequence of 5´CACCATGGAACAGTTTATCCTGAAGAC3' . The other oligonucleotide was used as a representative of the N-terminal segment of MSP10 . It has a sequence of 5´AGCCATGGAACGTGCTAAGTGCAACA3’ . Archived urine samples positive for P . vivax and P . falciparum were collected from Iquitos in Peru and Ghana , respectively . Negative control urine samples were collected from volunteers who were blood smear negative in Iquitos , Peru and in Ghana , as well as in Lima , Peru , which is a non-endemic site . All urine samples were collected by clean catch procedures . Ghana urine samples were pelleted in the field and shipped on dry ice , pH was adjusted and samples were refrozen at -80°C as described earlier [31] . Peru urine samples were stored initially at -20°C prior to freezing at -80°C ( Table 1 ) . Peru’s urine samples were stored for 8 months while all Ghana samples were stored for more than one year . During the epidemiological surveys on the communities , the field microscopist reports whether a slide is positive or negative , and identifies the species , P . vivax and P . falciparum . They read 300 microscopy fields before the slide is reported as negative . In the laboratory , a second reader ( an experienced microscopist working for research projects for more than 15 years ) read the slide to report species and parasite density assuming a white blood cell count of 6 , 000/μl . The research microscopist read until 500 microscopy fields , before the slide is reported as negative . For quality control , 10% randomly selected slides ( positive and negative ) were reexamined by two blinded , expert microscopists at a reference laboratory in Loreto from Peru’s Ministry of Health . From the quality control examinations , the level of concordance varies between 98–100% for species and parasite density . Urine was thawed at 25°C . Once urine was at room temperature , dipsticks were carried out to determine urine pH and the presence of protein . Each urine sample was centrifuged at 15 , 000 rpm for 5 minutes to remove sediments and then filtered using a 0 . 2 mm membrane ( Minisart , Bohemia , NY , United States ) to remove possible confounding particulates . The urine samples were diluted 1:16 with PBS . Diluted samples’ pH was adjusted to reach ≈ 6 . 4 using pH meter , and HCl and NaOH solutions . 50 uL of each diluted urine sample was heated at 95°C for 30 seconds using a thermocycler . Samples were cooled at room temperature for 10 minutes and 10 uL of either C-Terminal or N-Terminal MSP10 oligonucleotides and 20 uL of 0 . 25 M NaCl were added . The sample was heated at 59°C for two minutes and allowed to cool to room temperature for ten minutes . Finally , 50 uL of citrate reduced AuNPs were added . Two minutes later , the system was read visually and by spectrophotometer . Urine and blood were collected for previous studies that were approved by institutional review boards of Universidad Peruana Caytano Heredia and University of Ghana , respectively . Written informed consents were obtained prior to storing samples as anonymous and unidentified .
Up to 84% of P . vivax positive samples stabilized the gold nanoparticles and maintained a red color while 97% of negative controls induced aggregation and allowed color change to purple-blue . This color change was distinctly distinguished by naked eye ( Fig 1 ) . Additionally , the color difference was well defined by spectrophotometer , with positive samples at wavelengths of 520 and negative samples exhibiting wavelengths of 610–630 ( Fig 2 ) . The colorimetric system was able to detect P . vivax with variable sensitivity . The sensitivity was dependent on which segment of MSP10 was used . The N-terminal segment distinguished 26 of 31 positive samples ( 84% ) while the C-terminal segment distinguished only 20 of 31 samples ( 65% ) . Both the N-terminal and C-terminal segments of MSP10 had an overall specificity of 97% in urine , with only one false positive out of 45 control samples . The false positive was from a laboratory control in Lima ( Fig 3 ) . Parasitemia level was determined by blood smears collected at the same time as the urine samples . Data were shared after running the colorimetric system on urine to minimize investigator bias . There was no correlation between parasitemia level in blood and false negativity . The lowest parasitemia level observed in blood smear that also had a positive colorimetric test was 12 parasite/uL . However , there were two false negative samples with average parasitemia level in blood of 2510 parasite/uL ( Table 2 ) . N-Terminal MSP10 oligonucleotide had a lower cross-reactivity ( 21% ) . C-Terminal MSP10 oligonucleotides showed high cross-reactivity with P . falciparum in urine utilizing colorimetric system ( 36% ) ( Table 3 ) . It took an average of 45 minutes from collecting urine to reading the test . Cost of the raw materials for each test is $0 . 20 .
To our knowledge , this test is the first RDT utilizing urine samples rather than blood and employing nanoparticles . The colorimetric assay using AuNPs and MSP10 oligonucleotides to detect P . vivax in urine holds potentials to provide a safe , simple , rapid , and cheap tool to diagnose one of the most common form of malaria . Innovative use of MSP10 as a marker for P . vivax has potential for global application in mass screening programs . | To control malaria , there is an urgent need for applying innovative diagnostics and new technologies . Nanoparticles can augment detection of malaria at lower parasite levels while providing fast and simple methodology . Novel use of MSP10 and gold nanoparticles to identify Plasmodium vivax’s DNA in urine can be utilized as screening tool with global application potentials . The proposed test could impact the control of the most common species of malaria in low resource settings as it could present a simple , fast , cheap and easy to interpret test . Furthermore , utilizing urine instead of blood eliminates the need for finger-prick which would increase safety profile and likely increase participation rate in mass screening programs . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"medicine",
"and",
"health",
"sciences",
"parasite",
"groups",
"body",
"fluids",
"plasmodium",
"engineering",
"and",
"technology",
"immunology",
"tropical",
"diseases",
"nucleotides",
"parasitic",
"diseases",
"parasitology",
"urine",
"parasitemia",
"apicomplexa",
"nanoparticles",
"nanotechnology",
"cross",
"reactivity",
"quantitative",
"parasitology",
"hematology",
"biochemistry",
"blood",
"anatomy",
"oligonucleotides",
"physiology",
"biology",
"and",
"life",
"sciences",
"malaria"
] | 2016 | Colorimetric Detection of Plasmodium vivax in Urine Using MSP10 Oligonucleotides and Gold Nanoparticles |
Pathogen access to host nutrients in infected tissues is fundamental for pathogen growth and virulence , disease progression , and infection control . However , our understanding of this crucial process is still rather limited because of experimental and conceptual challenges . Here , we used proteomics , microbial genetics , competitive infections , and computational approaches to obtain a comprehensive overview of Salmonella nutrition and growth in a mouse typhoid fever model . The data revealed that Salmonella accessed an unexpectedly diverse set of at least 31 different host nutrients in infected tissues but the individual nutrients were available in only scarce amounts . Salmonella adapted to this situation by expressing versatile catabolic pathways to simultaneously exploit multiple host nutrients . A genome-scale computational model of Salmonella in vivo metabolism based on these data was fully consistent with independent large-scale experimental data on Salmonella enzyme quantities , and correctly predicted 92% of 738 reported experimental mutant virulence phenotypes , suggesting that our analysis provided a comprehensive overview of host nutrient supply , Salmonella metabolism , and Salmonella growth during infection . Comparison of metabolic networks of other pathogens suggested that complex host/pathogen nutritional interfaces are a common feature underlying many infectious diseases .
Infectious diseases are a major worldwide threat to human health [1] . The situation is worsening because of rapidly rising antimicrobial resistance and insufficient development of new antibiotics . Most infectious diseases start with a few pathogenic organisms that invade host tissues , but disease symptoms develop only later when pathogens exploit host nutrients to grow to high tissue loads . Despite this crucial role of pathogen nutrition and growth , only a few host nutrients that are relevant for some pathogens have been identified [2] , [3] , [4] , [5] , [6] , [7] , [8] , [9] , [10] , [11] , [12] , [13] , [14] , [15] , and comprehensive quantitative data are lacking . The poor understanding of in vivo growth conditions can cause major antimicrobial drug development failures [16] , [17] , [18] , [19] and might compromise antibiotic treatment [20] . In this study , we investigated Salmonella nutrition and growth in a mouse infection model mimicking human enteric fever . Enteric fever is caused by ingestion of food or water contaminated with Salmonella enterica serovars Typhi and Paratyphi ( “typhoid/paratyphoid fever” ) [21] . Salmonella invade intestinal tissues and disseminate to inner organs including spleen , liver , kidney , bone marrow , and brain , where they proliferate and cause tissue damages that can result in strong inflammation and organ failure . Enteric fever causes tremendous morbidity and mortality worldwide . Current control strategies become increasingly inefficient as a result of increasing antimicrobial resistance [22] , [23] and emergence of Salmonella serovars that are not covered by currently available safe vaccines [24] , [25] . In mice , Salmonella enterica serovars that cause human enteric fever usually do not cause any disease [26] , in part because of expression of Toll-like receptor 11 in mice but not humans [27] . However , serovar Typhimurium , which can cause human diarrhea , causes in mice a systemic infection with pathology and disease progression similar to human typhoid fever . Some mouse strains carry a functional allele Slc11a1 ( also called NRAMP ) coding for a Fe2+/Mn2+/Zn2+transporter , and such mice can successfully control systemic salmonellosis [28] . However , widely used laboratory mouse strains ( i . e . , BALB/c , C57BL/6 ) carry a dysfunctional Slc11a1 allele which makes them highly susceptible to lethal systemic Salmonella infections . Salmonella infections in these genetically susceptible mice thus represent an excellent model for severe human typhoid ( and paratyphoid ) fever [26] . This disease model is particularly suitable for comprehensive experimental and computational analysis because of facile Salmonella genetics , availability of genome-scale in silico metabolic reconstructions [29] , [30] , [31] , extensive literature , and close similarities between Salmonella and the prime model organism E . coli . In this study , we used proteomics , mutant phenotyping , and computational approaches to investigate Salmonella nutrition and growth in this mouse typhoid fever model . Our data revealed an unexpectedly complex Salmonella nutritional landscape in infected host tissues , where many chemically diverse nutrients were available in scarce amounts . Salmonella adapted to this situation by simultaneously employing versatile nutrient utilization pathways .
To characterize Salmonella metabolic capabilities during infection , we sorted Salmonella from infected mouse spleen and determined copy numbers of 477 metabolic enzymes ( among 1182 identified proteins ) using the well-established proteomics iBAQ label-free quantification approach [32] with 30 isotope labeled AQUA [33] peptides as internal standards ( Table S1 ) . This analysis extended our previous qualitative detection of 178 Salmonella enzymes in the same disease model [34] as a consequence of improved sorting and proteomics technologies . The detected enzymes are known to catalyze 925 metabolic reactions , a remarkably high proportion of all known/inferred 2023 Salmonella metabolic reactions for which catalyzing enzymes have been annotated [31] . Interestingly , this included 102 transporters and enzymes involved in 77 reactions in 24 pathways for utilization of various carbohydrates , lipids , nucleosides , and amino acids ( Figure 1 ) . It is important to note that these enzyme numbers likely underrepresented the entire Salmonella in vivo proteome as limited material availability and mass spectrometry detection thresholds likely prevented identification of weakly expressed enzymes . These data suggested that during infection , Salmonella mobilized a large part of their diverse metabolic capabilities . In comparison , closely related E . coli requires only 293 reactions for optimal growth in a minimal in vitro medium [35] . However , even under such well-defined conditions , E . coli expresses more than 200 apparently not required enzymes suggesting that enzyme expression alone is not indicative of metabolic relevance [36] ( see below ) . In addition to the qualitative identification of expressed Salmonella enzymes and associated metabolic reactions , our proteome data also provided quantitative data on Salmonella metabolic capabilities . We combined enzyme copy numbers with available turnover numbers to calculate maximal reaction rates for 469 reactions ( Table S2 ) . As an example , we detected 20’000±1000 copies per Salmonella cell of glycerol kinase GlpK that catalyzes MgATP-dependent phosphorylation of glycerol to yield sn-glycerol 3-phosphate . The closely related E . coli ortholog ( 95% amino acid identity ) has vmax = 22 µmol min−1 mg−1 [37] equivalent to a turnover number of 21 s−1 . Based on these data , a single Salmonella cell would have the catalytic capacity to phosphorylate up to 420’000 glycerol molecules s−1 . Such results should be taken as approximate only since turnover numbers are usually determined for somewhat non-physiological in vitro conditions ( e . g . , glycerol kinase was assayed in low osmolarity buffer at 25°C ) . Moreover , these data were incomplete because of undetected enzymes with abundance below the proteomics detection threshold and missing kinetic data . Nevertheless , the data yielded an unprecedented large-scale overview of Salmonella catalytic capacities in an infected host tissue , and provided a unique quantitative basis for in-depth analysis of metabolic activities involved in Salmonella virulence ( Figure S1; an interactive map with detailed descriptions is available at http://www . biozentrum . unibas . ch/personal/bumann/steeb_et_al/index . html ) . As expected [38] , central carbon metabolism had particularly high catalytic power in contrast to biosynthesis pathways for minor biomass components such as vitamins . Many nutrient utilization pathways had also substantial catalytic power with especially high values for glycerol utilization ( Figure 1 ) . Together , these data suggested that Salmonella maintained versatile catabolic capabilities for diverse nutrients during infection . To determine the actual relevance of specific nutrients for supporting Salmonella host tissue colonization , we inactivated defined utilization pathways . We preferentially deleted transporters to prevent high-affinity nutrient uptake instead of inactivating degradation enzymes that could result in accumulation of toxic upstream metabolites such as phosphorylated carbohydrates , which can cause pleiotropic effects [39] , [40] . Some nutrients can permeate membranes without a dedicated transporter ( glycerol , short-chain fatty acids , myo-inositol , ethanolamine ) . In these cases , we inactivated enzymes that were unlikely to cause toxic intermediate accumulation based on available literature [41] , [42] , [43] , [44] . Utilization defects have previously been used in several studies , for example to determine the relevance of several carbohydrates for E . coli growth in the intestinal lumen [2] . As a potential caveat , an excess supply of alternative nutrients may mask specific utilization defects . Moreover , some mutations might cause polar effects on the expression of downstream genes . In most of our mutants , this would only affect genes coding for subunits in the same transporters or enzymes involved in the same degradation pathways as the inactivated gene ( Table S3 ) . However , it was still possible that some such polar effects influenced mutant phenotypes . In a complementary second set of Salmonella mutants , we inactivated well-characterized biosynthesis pathways for essential biomass components . The resulting Salmonella auxotrophs were unable to grow unless the missing biomass components were provided externally ( Table S4 ) . Any growth of such mutants in infected spleen was , therefore , indicative of host supply of the respective supplement . Similar approaches have previously been used in various infection models . To measure tissue colonization capabilities of the various mutants , we used competitive infections with mixtures of mutant and wildtype Salmonella ( Figure 2; Table S3 ) . Mutant fitness was measured as competitive indices ( CI = output ratio ( mutant/wildtype ) /input ratio ( mutant/wildtype ) ) . A CI value of 1 ( equivalent to log2 ( CI ) = 0 ) indicated that a mutant had equal colonization capabilities as wildtype Salmonella . Complementation of mutant alleles to verify mutation phenotypes was often difficult because most strains contained multiple mutations . However , we independently reconstructed the most attenuated mutant and confirmed the resulting colonization defect ( Figure 2 ) . In the statistical analysis we avoided the “multiple comparison problem” using the widely accepted Benjamini-Hochberg [45] “false discovery rate” ( FDR ) approach to identify the subset of attenuated mutants ( Table S3 ) . Interestingly , several Salmonella mutants with nutrient utilization defects had significantly diminished colonization capabilities ( Figure 2; for detailed interpretation see Table S5 ) . This suggested that there was no large excess of nutrients that would mask any utilization defects , and no single major nutrient that alone could support full Salmonella virulence . Instead , Salmonella colonization depended on utilization of glycerol , fatty acids , N-acetylglucosamine , gluconate , glucose , lactate , and arginine . Glucose was the only nutrient that had previously been identified to contribute to systemic Salmonella infection [11] . All seven identified nutrients can serve as a sole carbon source for Salmonella growth [46] and can be interconverted into each other . It was thus unlikely that any of these nutrients was required because it provided a unique chemical structure . Instead , the seven metabolites seemed to supply individual small nutritional contributions that only together enabled normal Salmonella in vivo growth ( see below ) . Other utilization mutants had non-significant colonization phenotypes suggesting limited contributions of the corresponding nutrients . Most of our infection experiments used BALB/c mice that carry a dysfunctional Slc11a1 allele ( see Introduction ) . Such mice are highly susceptible for systemic salmonellosis providing a useful model for severe human typhoid fever . For comparison , we also did some small-scale experiments in 129/Sv mice that carry a functional Slc11a1 allele and are therefore resistant to lethal salmonellosis . Competitive infections confirmed the importance of glycerol ( or glycerol-3-phosphate ) and N-acetyl-glucosamine for Salmonella growth ( Figure S2 ) suggesting similarities of Salmonella nutrition in susceptible and resistant mice . Additional evidence for nutrient availability came from the substantial colonization capabilities of most tested Salmonella auxotrophs ( Figure 2; Tables S3 , S5 ) . In particular , Salmonella readily accessed sufficient quantities of several ( pro- ) vitamins and all tested amino acids ( except proline ) . Similar colonization phenotypes were obtained for Salmonella mutants with utilization or biosynthesis defects in infected liver ( Table S3 ) indicating that similar nutrients supported Salmonella growth in two different host organs . Combination of our data with previously reported additional mutant virulence phenotypes indicated Salmonella access to a large set of at least 31 chemically diverse host nutrients in infected mouse spleen ( Table S5 ) . This analysis thus revealed a highly complex host/Salmonella nutritional interface , which is still likely incomplete because of limited mutant coverage and our inability to detect small colonization defects . Our data suggested that Salmonella exploited a wide range of diverse host metabolites . This was initially surprising since most microorganisms utilize only a single preferred nutrient such as glucose when exposed to nutrient mixtures [47] . Other nutrients and their utilization pathways remain irrelevant as long as this preferred nutrient is available . This was evidently not the case during infection , as glucose utilization only partially supported Salmonella growth in agreement with previous observations [11] . As one possible explanation , various nutrients including glucose might have been available in only limited amounts that together just supported Salmonella growth and tissue colonization . Indeed , colonization defects of Salmonella utilization mutants ( Figure 2 ) suggested that Salmonella virulence depended on simultaneous effective exploitation of several nutrients instead of relying on only one preferred nutrient . To further test the hypothesis of parallel utilization of different available nutrients , we used a cell culture infection model where Salmonella replicated intracellularly in macrophage-like RAW 264 . 7 cells mimicking conditions during systemic salmonellosis [48] . In this cell culture model , extracellular metabolites can reach intracellular Salmonella and contribute to their nutrition [49] , [50] , [51] , [52] . To test the impact of nutrient availability , we added external glucose or mannitol at 4 h post infection when Salmonella had already established their intracellular niche ( Figure 3A ) . Interestingly , both extracellular nutrients accelerated subsequent intracellular Salmonella growth ( Figure 3B ) . This growth promoting effect was dependent on specific Salmonella glucose/mannitol utilization capabilities , suggesting that external glucose and mannitol directly contributed to Salmonella growth , whereas nutrient-induced changes in the host cell had negligible impact ( e . g . , moderate changes in osmolarity ( 2 . 7 mOsm per added nutrient , some 1% of the total osmolarity ) , glucose metabolization by host cells ( mannitol cannot be metabolized by mammalian cells [53] ) , or modulation of host cell phagocytosis and oxidative bursting as observed at much higher mannitol doses [54] ) . These data indicated that intracellular Salmonella growth was limited by nutrient availability , and Salmonella exploited both a typically preferred ( glucose ) and a non-preferred carbon source ( mannitol ) when available thus supporting our nutrient limitation hypothesis . Taken together , both mutant colonization defects and cell culture experiments were consistent with Salmonella growth being dependent on diverse scarce nutrients during infection . In addition to these qualitative results on nutrient-limited Salmonella growth , we were interested to obtain quantitative nutrient supply rates as a basis for comprehensive understanding and computational modeling of Salmonella nutrition , metabolism and growth . Quantitative nutrient supply rates have not yet been reported for any infection model , but the severity of mutant colonization defects could provide some hints . As an example , the strong colonization defect of Salmonella glpFK gldA glpT ugpB defective for glycerol utilization , compared to Salmonella manX nagE defective for GlcNAc utilization , could suggest that more glycerol was available as compared to GlcNAc . This rationale has previously been used to assess the relative relevance of various carbohydrates for E . coli gut colonization [2] . However , direct calculation of the respective nutrient supply rates from such mutant colonization defects was hampered by the parallel utilization of many diverse nutrients . Moreover , nutrients such as glycerol and GlcNAc differ in their nutritional value per molecule . To quantitatively assess the availability of multiple host nutrients and their utilization by Salmonella , we therefore used a computational approach called Flux-Balance Analysis ( FBA ) [55] . This approach had been successfully applied to predict nutrient utilization and growth in a wide variety of organisms in excellent agreement with large-scale experimental data [56] . As a precondition for the application of FBA to Salmonella , we recently established together with more than 20 Salmonella experts an in silico reconstruction of the entire Salmonella metabolic network that contains all experimentally determined Salmonella metabolic activities , all enzymes with annotated metabolic activity encoded in the Salmonella genome , and their catalyzed reactions with all participating metabolites , stoichiometries , and information on reaction reversibility [31] . This consensus Salmonella metabolic reconstruction has been extensively documented and is continuously being updated by manual curation of newly available literature for Salmonella and closely related E . coli enzyme orthologs ( reconstruction STMv1 . 1 with 1279 Salmonella enzymes , 1824 metabolites , 2573 reactions; Tables S6 , S7 , S8; the reconstruction is available in SBML format at http://www . biozentrum . unibas . ch/personal/bumann/steeb_et_al/index . html ) and in the Supporting Information ( Model S1 ) . Flux-balance analysis can be used to determine if the metabolic network is capable of producing all components required for Salmonella biomass generation . Importantly , biomass requirements can differ between growth conditions [57] , [58] . To deduce Salmonella biomass requirements during infection , we analyzed published informative mutant virulence phenotypes and modified the biomass function accordingly ( for detailed descriptions see Table S8; for limitations in the in vivo biomass definition see Discussion ) . Flux-balance analysis revealed that the metabolic reconstruction could generate all included biomass components in the correct stoichiometry under observance of fundamental thermodynamic laws such as preservation of mass and charge ( “flux-balance” ) [31] . In addition to biomass generation , all cells have growth-unrelated demands for survival and these are commonly accounted for as “maintenance requirements” [59] . Such demands could be especially important in pathogens during infection when they need resources to resist host antimicrobial defense . To model Salmonella nutrition and growth in infected spleen , we provided the in silico reconstruction with all experimentally identified nutrients and used FBA to compute the resulting Salmonella biomass generation ( which we used as an approximation for growth throughout this study ) . We adjusted nutrient uptake rates to reproduce our experimental Salmonella mutant phenotypes ( for a detailed description of our approach , see Material and Methods and Figure S3 ) . This yielded supply rates for 31 organic nutrients ( Table S9 ) , as well as 13 inorganic ions ( Table S9 ) . To obtain consistent data we needed to assume enhanced maintenance requirements ( 145±20% of the value for axenic in vitro cultures ) . Such enhanced maintenance costs could reflect defense against hostile host environments ( see Discussion ) . We also explored scenarios of excess nutrient supply ( see Materials and Methods ) . The results revealed that the computation model could accommodate only modest nutrient excess up to 118% of the minimal nutrient supply values , and this would require improbably high maintenance costs for consistency with our experimental colonization data ( Figure 3C ) . These data provided additional in silico support for nutrient-limited Salmonella growth ( see above ) . It is important to note that our computational approach had several caveats ( see Discussion ) . On the other hand , the resulting model provided a first comprehensive quantitative approximation to the host nutritional landscape and its exploitation by Salmonella that could serve as a basis for subsequent improvements ( Figure 4; the model is available in SBML format at http://www . biozentrum . unibas . ch/personal/bumann/steeb_et_al/index . html ) and in the Supporting Information ( Model S1 ) . To assess how well the current computational model captured relevant aspects of Salmonella nutrition and growth during infection , we compared model predictions with large-scale experimental data sets on Salmonella mutant phenotypes , enzyme expression , and metabolic capabilities . To validate functional aspects of the computational model , we systematically predicted in vivo growth phenotypes for all 1279 model enzymes , and compared these predictions to reported experimental Salmonella mutant colonization phenotypes ( Table S10; Figure 5A; interactive maps for predicted and experimental mutant phenotypes are available at http://www . biozentrum . unibas . ch/personal/bumann/steeb_et_al/index . html ) . Inactivation of most enzymes had no impact on predicted growth rate . Only few , mostly biosynthetic , enzymes were essential for Salmonella virulence ( predicted mutant growth rate below 60% of wildtype ) , while some genes contributed to virulence ( predicted mutant growth between 60% and 98% of wildtype ) , and the vast majority of enzymes had non-detectable effects ( mutant growth rate higher than 98% of wildtype ) in agreement with our previous experimental data [34] . Detailed analysis of 738 single mutants with available experimental data revealed an overall prediction accuracy of 92% ( Figure 5A; Table S10 ) similar to accuracies achieved for the best computational models for E . coli in vitro cultures [59] . This analysis included 14 mutant phenotypes that we had used to deduce the biomass function ( Table S8 ) and 69 mutation phenotypes that we had used to deduce nutrient supply ( Table S5 ) . Consistency of model predictions for these mutants and additional mutants with linked phenotypes ( such as enzymes in the same pathways ) was , therefore , unsurprising . Moreover , gene selection for mutant testing in our and other labs was likely influenced by previous knowledge . Mutant phenotypes thus did not provide truly independent validation , but they demonstrated that the model yielded consistent quantitative explanations for the function of hundreds of Salmonella genes during infection . On the other hand , there were 61 discrepancies between computational predictions and experimental data that could help to identify remaining errors and knowledge gaps ( Figure 5B; for detailed analysis , see Table S10 and Discussion ) . We also compared model predictions with our protein identification data . Specifically , we used a recently developed approach called parsimonious enzyme usage FBA ( pFBA ) [36] to predict enzyme classes with differential functional relevance for efficient Salmonella biomass generation . These enzyme classes included , in order of decreasing relevance , ( i ) essential enzymes , ( ii ) enzymes required for optimal growth with minimal overall flux ( “optima” ) , ( iii ) enzymatically less efficient enzymes ( which could sustain optimal growth but would require more total flux ) , ( iv ) metabolically less efficient enzymes ( which could sustain only suboptimal growth ) , and ( v ) genes with no contribution to Salmonella growth ( zero flux in the associated reactions ) . Comparison with our proteome data for ex vivo sorted Salmonella revealed that there was a statistically highly significant relationship between relevance and the proportion and abundance of detected Salmonella enzymes in the various classes ( Figure 5C; Figure S4 ) , similar to what has been observed for computational models of well-characterized E . coli in vitro cultures [36] . On the other hand , we still detected only some 50% of the relevant enzymes ( classes “essential” and “optima” ) . Many non-detected enzymes were associated with rather low predicted reaction rates ( Figure S5 ) , suggesting that these enzymes might have been present in small quantities below our ex vivo proteome detection threshold . Incomplete proteome coverage of important enzymes has also been observed for E . coli in vitro cultures [36] . On the other hand , we detected several enzymes that were predicted to mediate no flux , again similar to observations for in vitro cultures [36] . Many such enzymes were involved in amino acid biosynthesis , nutrient utilization , gluconeogenesis , glycogen metabolism , and other pathways that all had experimentally non-detectable mutant phenotypes , consistent with their predicted non-functionality . It is possible , however , that these pathways were actually active , but accounted for minor contributions to Salmonella virulence that were undetectable with current in vivo methods . Alternatively , Salmonella might have prepared themselves for subsequent environments in their life cycle where these pathways would offer fitness benefits . Finally , Salmonella might be unable to optimally regulate its enzyme expression to shut down all dispensable enzymes ( as it is likely the case in E . coli in vitro cultures ) . Further research is required to test these and other hypotheses . We also compared our in vivo model with a model for Salmonella growth in minimal medium with glucose as sole source of carbon and energy . Interestingly , there was a large overlap between enzymes that were important for optimal growth of Salmonella under these two conditions . We detected 30 proteins that were predicted to be specifically required in vivo but not in vitro , providing some support for our in vivo model . On the other hand , we also detected 15 proteins that should be required only in the in vitro minimal medium but not in vivo . Interestingly , eleven of these 15 proteins were involved in amino acid biosynthesis suggesting that Salmonella maintained such biosynthetic capabilities in vivo despite access to host amino acids ( see above ) . It is possible that the amino acid supply was just marginally sufficient and Salmonella prepared itself for future amino acid starvation . Further work is required to clarify this issue . In addition to predicting enzyme relevance , the model also provided quantitative predictions for fluxes through hundreds of metabolic reactions . For some reactions , a large range of reaction rates was possible whereas others had more restricted rates ( Figure 5D ) as previously observed in other systems ( “flux variability” [60] ) . We determined the state with the lowest overall metabolic activity corresponding to economical use of costly enzymes . Such states have shown to correspond well with experimental flux data in other systems [36] , [61] . To determine the feasibility of these predicted reaction rates , we compared them to Salmonella catalytic capacities calculated from experimentally determined enzyme concentrations and turnover numbers ( see above; Table S2; Figure S1; http://www . biozentrum . unibas . ch/personal/bumann/steeb_et_al/index . html ) . Interestingly , 459 out of 469 analyzed reactions had feasible predicted rates ( Figure 5D ) . Three reactions had clearly infeasible reactions rates in the most economical computational state with lowest overall metabolic activity ( >3 fold above the corresponding catalytic capacities; these reactions are labeled in Figure 5D: 1 , formyltetrahydrofolate dehydrogenase; 2 , phosphoserine aminotransferase; 3 , glycerol dehydrogenase ) . However , all these reactions could be restrained to feasible rates without compromising predicted Salmonella growth or making other reactions infeasible . All seven other reactions had only moderate discrepancies between simulated and feasible rates , and four of them could again be restrained without compromising growth . The remaining three reactions had simulated reaction rates that remained slightly infeasible in all states ( simulated fluxes 1 . 2 to 2 . 5 fold too high ) . Interestingly , all three reactions were aminoacyl tRNA ligations ( for proline , alanine , and threonine ) . Possible causes for these discrepancies included inaccurate biomass assumptions for proline , alanine , and threonine protein content , experimental errors in protein quantification , and/or suboptimal assay conditions for tRNA ligase turnover number measurements . Moreover , the computational model disregards important processes outside metabolism such as macromolecular expression [58] , which could contribute to discrepancies between feasible and simulated reaction rates . Despite these three minor discrepancies , the overwhelming feasibility of reaction rates indicated that Salmonella had sufficient in vivo enzyme amounts and catalytic power to mediate nutrient utilization , metabolization , and biomass generation as predicted by the computational model . Although almost all reactions had entirely plausible reaction rates in the computational state with lowest overall metabolic activity , the entire flux solution space also included many reaction fluxes that exceeded plausible rates . In a next step , we prevented such implausible fluxes by setting upper/lower limits according to the maximum experimental enzyme capacities ( except for the problematic three tRNA ligations , see above ) . Interestingly , these large-scale constraints still allowed normal Salmonella in silico growth , but resulted in a dramatically reduced flux solution space ( Figure 5E , F ) . Specifically , the vast majority ( 80% ) of reactions had narrowly defined flux ranges ( relative flux variability below 10% ) , whereas in the initial unrestrained model only a small minority ( 16% ) had such narrowly defined reaction rates ( Figure 5F ) . This enzyme capacity-based model might thus provide a much better defined approximation to the actual in vivo flux state . Together , these data revealed that the model predicted ( i ) largely correct mutant virulence phenotypes , ( ii ) predicted enzyme relevance that correlated with experimental protein detection , and ( iii ) predicted reaction rates that were biologically plausible . This large-scale consistency with experimental data suggested that the computational model captured major aspects of Salmonella nutrition , metabolism , and growth in infected host tissues . To investigate if Salmonella conditions during mouse typhoid fever might be generally representative for pathogen nutrition in infected host tissues , we compared pathogen metabolic networks based on genome pathway annotations [62] . We analyzed 154 different mammalian pathogens ( Table S11 ) for presence of 254 nutrient utilization pathways and 118 biosynthetic pathways ( Figure 6A ) . Most pathogens shared the capability to utilize glycerol , fatty acids , various carbohydrates , nucleosides , and amino acids that could serve as N-sources ( such as arginine ) , suggesting a general preference for nutrients that Salmonella used in the mouse typhoid fever model . Additional genome comparisons for 316 non-pathogenic species revealed that they might also preferentially utilize similar nutrients ( Figure S6 ) . On the other hand , many pathogens have smaller genomes compared to related non-pathogenic species as a result of reductive genome evolution [63] resulting in loss of many pathways . To identify biosynthesis pathways that were commonly lost during this process , we determined a “biosynthesis depletion frequency” ( DF ) as follows . For each biomass component , we determined the fractions of pathogenic ( P ) and non-pathogenic ( NP ) species that encoded corresponding biosynthesis pathways in their genomes . The “biosynthesis depletion frequency” was calculated as the difference of the respective frequencies DF = NP−P . As an example , 89% of environmental bacteria but only 47% of mammalian pathogens had an apparently functional tyrosine biosynthesis pathway yielding a “depletion frequency” of 89%−47% = 42% . The results revealed that many pathogens lost biosynthesis pathways for several amino acids , nucleosides , and ( pro ) vitamins indicating that these pathogens - like Salmonella - might obtain such biomass components from their respective host environments ( Fig . 6B ) . Together , these genome comparisons suggested that many pathogens share with Salmonella access to a common large set of diverse host metabolites in infected mammalian tissues .
Host nutrients are essential for pathogen proliferation , disease progression , and efficacy of antimicrobial treatments . However , only few relevant nutrients have been identified and quantitative data on nutrient supply rates are lacking . In this study , we combined experimental enzyme abundance data , previously reported enzyme kinetic parameters , competitive infections with metabolic mutants , and computational modeling , to build and validate a comprehensive genome-scale model of Salmonella nutrition , metabolism , and growth in infected mouse tissues . Virulence phenotypes of nutrient utilization mutants and auxotrophic mutants revealed that Salmonella accessed a surprisingly large number of chemically diverse host nutrients including lipids , carbohydrates , amino acids , nucleosides , and various ( pro ) vitamins . Surprisingly , this included facile availability of all three aromatic amino acids based on full virulence of auxotrophic Salmonella pheA tyrA trpA . This was also consistent with common tryptophan auxotrophy of Salmonella enterica serovar Typhi clinical isolates from human typhoid fever patients [64] . On the other hand , aromatic amino acids were previously thought to be unavailable in infected mouse tissues based on strongly attenuating mutations in chorismate biosynthesis [65] , [66] , a precursor for aromatic amino acids . However , such mutants are not informative for aromatic amino acid availability since chorismate is also a precursor for ubiquinone which is essential for Salmonella virulence [34] . Similar conditions might exist for intracellular Listeria [67] . The large diversity of accessible host nutrients posed complex challenges to Salmonella metabolism but , in addition , also relieved Salmonella dependence on any particular nutrient and its corresponding utilization pathway , thus enabling Salmonella to maintain high virulence even when biosynthesis pathways for important biomass components such as amino acids were defective . This buffering capacity of the complex nutritional landscape significantly contributed to the remarkable robustness of Salmonella metabolism against internal perturbations during infection [34] . Proteome analysis of Salmonella purified from infected tissues revealed in vivo expression of enzymes involved in degradation of the major nutrients glycerol , fatty acids , and N-acetylglucosamine , glucose , lactate , and arginine suggesting that Salmonella allocated major enzyme resources to relevant pathways in agreement with earlier observations [68] . Exceptions included mannose-6-phosphate isomerase ( ManA ) and UDP-glucose 4-epimerase ( GalE ) that can participate in degradation of mannose and galactose , respectively . Both enzymes were present in concentrations that would sustain high reaction rates ( Figure 1 ) , yet neither mannose nor galactose had a detectable nutritional contribution during infection ( Figure 2; Table S5 ) . However , both enzymes can also operate in reverse direction for biosynthesis , and corresponding mutant phenotypes [69] , [70] support this as their dominant role in Salmonella virulence . Together , the proteome data revealed versatile Salmonella adaptation to a complex nutritional landscape . To deduce quantitative in vivo supply rates for the various nutrients , we used a genome-scale computational approach based on Salmonella mutant colonization phenotypes . Specifically , we updated a genome-scale reconstruction of the Salmonella metabolic network and established a modified in vivo biomass composition . We then determined which nutrient uptake rates would support Salmonella biomass production consistent with experimental colonization data for wildtype and mutant Salmonella . This approach yielded uptake rates for 31 organic and 13 inorganic nutrients . For consistency with the experimental wildtype Salmonella in vivo generation time , we had to increase the non-growth associated ATP maintenance requirements to some 145% of their original value for axenic in vitro cultures [31] . Increased maintenance costs might be expected for hostile host environments compared to axenic in vitro cultures , but accurate experimental validation of maintenance requirements is generally challenging [59] , [71] . It is important to note that our entire computational approach relied on several simplifying assumptions . ( i ) We disregarded nutrient utilization for purposes other than biomass generation or maintenance/virulence . ( ii ) We disregarded additional non-metabolic functions of the various mutated Salmonella genes ( “moonlighting functions” [72] ) . Such additional functions are possible although they have not yet been observed for any of the specific transporters/enzymes that we had inactivated . ( iii ) We assumed similar in vitro and in vivo biomass composition ( except for a few components for which informative mutant phenotypes had been reported ) disregarding well-documented effects of differential growth rates on biomass composition [58] , [73] , [74] . ( iv ) We deduced average nutrient supply rates but conditions might change during infection and could also differ between various Salmonella subpopulations . Some kinetic information could be obtained from competitive infection time series but this would require an extensive number of experimental animals . Because of all these caveats , the predicted nutrient supply rates and maintenance costs should be regarded only as rough estimates providing order-of-magnitude information as an approximation to the actual in vivo situation . On the other hand , the resulting model provided a first comprehensive quantitative approximation to the host nutritional landscape and its exploitation by Salmonella that could serve as a basis for subsequent improvements . To assess how well the current stage of this model reflected Salmonella nutrition and metabolism during infection , we extensively validated model predictions with large-scale experimental data . Interestingly , enzymes with predicted high relevance for optimal Salmonella growth were experimentally detected at higher rates compared to non-functional enzymes . Moreover , rate predictions for hundreds of reactions were consistent with experimentally determined enzyme levels . This indicated that the simulated metabolic flux distribution was fully feasible with the amounts of enzymes that are actually present in Salmonella during infection . The model predicted hundreds of mutant virulence phenotypes with an accuracy of over 90% thus indicating large-scale consistency with experimental data . The few remaining discrepancies may provide hints for further model improvements and targeted research to close knowledge gaps . Detailed examination suggested various typical limitations of our computational approach including ( i ) overestimated redundancies due to neglected regulation of isozyme/alternative pathway expression and/or differential substrate affinities ( e . g . , possibly poor expression of the sodium transporter NhaB which might fail to compensate for a nhaA defect in contrast to model predictions; low affinity zinc uptake through YgiE , which might be insufficient to compensate a defective ZnuABC zinc high-affinity transporter ) , ( ii ) incomplete biomass/maintenance functions that neglect signaling and detoxification needs ( e . g . , SpoT-dependent ppGpp homeostasis ) , ( iii ) inappropriate treatment of biomass components that contribute to virulence but are not absolutely essential ( e . g . , enterobacterial common antigen ) , ( iv ) neglect of continuous uptake of nutrients despite accumulation of toxic downstream intermediates ( e . g . , accumulation of GlcNAc-phosphate in absence of N-acetylglucosamine-6-phosphate deacetylase NagA [75] ) , and ( v ) knowledge gaps ( e . g . , bypass of dihydropteroate synthase FolP ) ( Figure 5B; for detailed analysis see Table S10 ) . Subsequent model versions might overcome some of these limitations to further improve prediction accuracy . In addition , a few experimental data might possibly be wrong based on inconsistencies between different studies . We also tested the predictive power of an enzyme capacity-restrained model that might more closely reproduce the wildtype flux state . Single gene deletion analysis of this model had very similar accuracy with three additional discrepancies ( too severe predicted growth defects for thrB , thrC , and aceA ) while resolving only one discrepancy ( detectable growth defect of zwf ) as compared to the unrestrained model . The enzyme capacities that we used as constraints in this model were based on protein profiles of wildtype Salmonella . Mutant Salmonella might have somewhat different protein profiles and enzyme capacities , and this might explain why the restrained model was not superior to the unrestrained model in predicting mutant colonization phenotypes . Taken together , the excellent agreement of model predictions and large-scale experiment data suggested that the model accurately captured major aspects of Salmonella nutrition , metabolism , and growth during infection in a comprehensive and quantitatively consistent way . Experimental mutant phenotypes and cell culture experiments suggested that despite Salmonella access to many host nutrients , these nutrients were available in only scarce amounts that individually would be insufficient to support full Salmonella virulence . This was also supported by modeling results that were incompatible with any substantial nutrient excess . Salmonella thus seemed to depend on simultaneous exploitation of several chemically diverse host nutrients through versatile utilization pathways . This apparent nutrient limitation inside infected host cells was initially surprising , since host cells contain numerous abundant metabolites that could provide rich carbon , nitrogen , and energy sources for Salmonella . However , intracellular Salmonella are separated from the nutrient-rich host cell cytosol by a vacuolar membrane that might restrict nutrient access . Further studies are required to better characterize this membrane and to test various hypotheses on host control of nutrient supply to Salmonella . This study extends previous work on metabolic host/pathogen interactions . In particular , combinations of transcriptome data , mutant phenotypes , and genome-scale computational metabolism models have been used to analyze metabolism and growth of Neisseria meningitidis in serum [76] , and Mycobacterium tuberculosis [57] , [77] and Listeria monocytogenes [78] in macrophages . One study even incorporated the metabolic networks of both Mycobacterium tuberculosis and its infected host macrophage cell in one integrated model that describes the entire host/pathogen metabolic interaction [57] . These studies identified several relevant host nutrients such as amino acids driving pathogen growth and provided the first genome-scale descriptions of pathogen metabolism as a basis for a system-level understanding of metabolic host/pathogen interactions . On the other hand , previous studies were limited to in vitro/cell culture conditions , included only a moderate number of host nutrients , and lacked quantitative data on nutrient supply rates and absolute enzyme levels . Our integrated experimental and computational approach addressed some of these limitations and yielded a comprehensive quantitative analysis of the highly complex nutritional in vivo landscape for Salmonella in infected host tissues . These data enabled us to generate a genome-scale model that accurately predicted enzyme requirements for Salmonella virulence in an important animal disease model . However , there still remain important issues that should be addressed in future studies . ( i ) We interpreted net Salmonella colonization phenotypes always as division rate differences ( similar to what has been done in most other studies ) . However , this is probably an oversimplification as some colonization defects might be caused by increased Salmonella killing by host antibacterial defenses , instead of differential Salmonella proliferation rates . In such cases , a simple metabolic interpretation in terms of diminished biomass production might be misleading . Future studies using methods such as Fluorescence Dilution [79] and direct detection of killed Salmonella [80] could provide suitable experimental data to address this issue . ( ii ) This and previous studies were based only on bulk measurements ( transcriptomics , proteomics , mutant colonization phenotypes ) that fail to account for any pathogen subpopulations . However , heterogeneous Salmonella subpopulations with different growth characteristics exist in vivo [79] , [81] . So far , nothing is known about possible metabolic differences among distinct subpopulations , and future studies should address this issue since subpopulations might play important roles in virulence , transmission , persistence , and treatment failures [82] . ( iii ) A complete picture should include host metabolic processes that provide nutrients for Salmonella . An impressive study on tuberculosis already revealed some aspects of the interplay between host and pathogen metabolic networks in Mycobacterium tuberculosis-infected macrophages [57] , and this approach might be extended to Salmonella as well . For Salmonella infections , analysis is complicated by the fact that the Salmonella-containing vacuole ( SCV ) communicates with late endosomes , from where it receives some incoming endocytosis cargo from the extracellular environment [51] thereby bypassing the metabolic network of the infected host cell . In addition , Salmonella might access some metabolites of the infected cell but additional experimental data will be needed to clarify the relative importance of the various nutrient supply routes . Another important aspect of metabolic host/Salmonella interactions is the question how Salmonella metabolism might influence host cell physiology . As an example , the capture by Salmonella of various host amino acids and nucleosides , as observed in this study , could modulate host cell functions that depend on these metabolites including antibacterial defense such as generation of nitric oxide [83] . Some indications for infection-induced changes in Salmonella-infected macrophages was already obtained in recent transcriptome and proteome studies [84] , [85] . Increasingly accurate modeling of all these aspects might ultimately provide a complete quantitative description of the host/Salmonella metabolic interactions that enable Salmonella growth and enteric fever disease progression . In addition to salmonellosis , the findings of this study also have some implications for infectious diseases in general . In particular , metabolic network comparisons suggested that many mammalian pathogens might share access to similar complex host nutrients that reflect general biochemical features of mammalian tissues . These results might provide a basis to establish in vitro culture conditions that closely mimic relevant in vivo conditions , helping to avoid drug development failures and to facilitate successful development of novel control strategies . On the other hand , the actual relevance of individual nutrients can vary . As an example , ethanolamine is an important nutrient for Salmonella in inflamed intestine [13] but not in our systemic infections . As another example , Mycobacterium tuberculosis access fatty acids and proline ( like Salmonella in mouse spleen ) , but glycerol is not a major nutrient , and lysine , tryptophan , and leucine are apparently available in insufficient amounts to meet mycobacterial biomass needs [19] , [86] , [87] , [88] , [89] . Interestingly , some of the commonly encountered nutrients are predominantly present as part of high molecular weight compounds such as glycans/glycoproteins ( GlcNAc ) , proteins ( most amino acids ) , or lipids ( glycerol , fatty acids ) suggesting that macromolecule hydrolysis might be an important aspect of pathogen nutrition in infected tissues . Indeed , many pathogens express hydrolases that degrade macromolecules such lipases , proteases , carbohydratases , etc . , as part of their virulence program . It might also be interesting to compare the common pathogen nutritional signature to the metabolism of commensal bacteria that inhabit body parts such as skin , genital mucosa , the oral cavity , or the intestine . Indeed , previous studies have already revealed commonalities among commensal gut bacteria such as the ability to digest complex carbohydrates [90] . Future studies might consider food components such as dietary plant sugars , host nutrients such as mucus , and waste products from other gut microbes . Moreover , such an analysis should also account for striking inter-individual differences in commensal microbial communities such as the recently described distinct enterotypes [91] . In conclusion , this study provided a comprehensive quantitative description of the Salmonella nutritional landscape during systemic salmonellosis and established a genome-scale model of Salmonella metabolism that explains major aspects of Salmonella infection biology . The results revealed an unexpectedly complex host/Salmonella nutritional interface that Salmonella exploited with versatile catabolic pathways . Similar complex host nutrients and versatile pathogen utilization pathways appear to be general features of many infectious diseases .
All animal experiments were approved by Kantonales Veterinäramt Basel-Stadt ( license 2239 ) and performed according to local guidelines ( Tierschutz-Verordnung , Basel-Stadt ) and the Swiss animal protection law ( Tierschutz-Gesetz ) . Salmonella mutants were constructed by lambda red-recombinase mediated allelic replacement [92] followed by general transduction using phage P22 int [93] . In multiple mutants , usage of the same resistance cassettes was enabled by FLP recombinase-mediated excision of the first cassette [92] . Strains were cultivated on Lennox LB medium containing 90 µg ml−1 streptomycin , 50 µg ml−1 kanamycin , 20 µg ml−1 chloramphenicol , and/or 100 µg ml−1 ampicillin . All auxotrophs required supplementation for growth as expected ( Table S4 ) . We infected female , 8 to 12 weeks old BALB/c mice intravenously with 500–2000 CFU Salmonella from late exponential LB cultures . For some experiments , we used female , 8–12 weeks old 129/Sv mice . Three to four days post-infection ( or five days for 129/Sv ) , mice were sacrificed and bacterial loads in spleen and liver were determined by plating of tissue homogenates treated with 0 . 3% Triton Tx-100 . In competitive infections , wildtype and mutant Salmonella carrying different antibiotic resistance markers were mixed before administration . Individual strain tissue loads were determined by replica plating on selective media and competitive indices ( CI = output ratio/input ratio ) were calculated . Statistical significance was analyzed using t-test of log-transformed CI values ( a parametric test was appropriate based on the normal distribution of such values [34] ) . Our experiments involved a large set of strains . To avoid the multiple comparison problem , we used the Benjamini-Hochberg false discovery rate ( FDR ) approach [45] . For Salmonella ex vivo purification , Salmonella sifB::gfp [94] were sorted from infected mouse spleen as described [34] using a FACSAria III sorter ( BD Biosciences ) . We used optical emission filters ( green fluorescence , 499–529 nm; orange fluorescence , 564–606 nm ) that optimally separated Salmonella GFP fluorescence from host cell autofluorescence . Proteome changes were minimized by preventing de novo synthesis with 170 µM chloramphenicol and delaying proteolysis by maintaining the samples at 0–4°C . Our previous results suggested that these conditions were effective to largely preserve the in vivo Salmonella proteome during sorting [34] . Preparation of tryptic peptides and analysis by LC-MS/MS was done essentially as described [95] with some modifications . Protein LoBind tubes and pipette tips ( Axygen ) were used throughout the procedure to minimize protein loss through adsorption . Frozen FACS sorted Salmonella pellets were resuspended in 15 µl lysis buffer ( 100 mM ammonium bicarbonate , 8 M urea , 0 . 1% RapiGest ) and sonicated for 2× 30 seconds . Released proteins were reduced and alkylated , and first digested for 4 hrs with sequencing grade LysC peptidase ( 10 ng/µl; Promega ) before overnight trypsin digestion . The detergent was cleaved by adding 2 M HCL and 5% TFA to final concentrations of 50 mM and 0 . 5% respectively , and incubating for 45 min at 37°C . Prior to centrifugation to remove the cleaved detergent ( 14 , 000×g , 10 min , 4°C ) , a mixture containing 32 heavy labeled reference peptides were added to the samples ( 5×10−5 fmoles per Salmonella for expected “high” abundance proteins , 5×10−6 fmoles per Salmonella for expected “low” abundance proteins; Table S12 ) . The recovered peptides were desalted on C18 reverse-phase spin columns ( Macrospin columns , Harvard apparatus ) , dried under vacuum and subjected to LC-MS/MS using an LTQ-Orbitrap-Velos instrument ( Thermo-Fischer Scientific ) . Between 5×105 and 2×106 Salmonella sorted from individual mice were analyzed in replicate LC-MS/MS runs . In order to increase the number of Salmonella protein identifications , MS-sequencing was partially focused on previously identified Salmonella peptides using the recently developed inclusion list driven workflow [95] . Peptides and proteins were database searched against a decoy database consisting of the SL1344 genome sequence ( ftp://ftp . sanger . ac . uk/pub/pathogens/Salmonella/ ) , GFP , 204 frequently observed contaminants , all mouse entries from SwissProt ( Version 57 . 12 ) , and all sequences in reversed order ( total 42502 entries ) using the Mascot search algorithm . The search criteria were set as follows: full tryptic specificity was required ( cleavage after lysine or arginine residues ) ; 2 missed cleavages were allowed; carbamidomethylation ( C ) was set as fixed modification; oxidation ( M ) as variable modification . The mass tolerance was set to 10 ppm for precursor ions and 0 . 5 Da for fragment ions . The false discovery rate was set to 1% for protein and peptide identifications . In addition to Salmonella proteins a substantial number of contaminating mouse proteins were identified in the samples as previously noted [34] . Absolute quantities were determined for those 18–20 “anchor” Salmonella proteins that were detected along with a corresponding labeled AQUA peptide ( Table S12 ) using the Trans-Proteomic Pipeline ( TPP , V4 . 4 . 0 ) . We then used the iBAQ method to establish absolute quantities of all remaining protein identifications , with a linear model error of between 47 and 60% . Comparison of samples from four independently infected mice revealed good reproducibility ( Table S1 ) . The data associated with this manuscript may be downloaded from ProteomeCommons . org Tranche using the following hash: HaSHrE4Paqa3Io3NARhJsV/7XeqYsNNvHYX3tt++xYcVYOf47nChKFB9E/PCD+j+xt5meJ1+4ytJIHVUeXx9Xb+ohBEAAAAAAAACZw = = Enzyme abundance was combined with reported turnover numbers for the respective Salmonella enzymes ( or closely related E . coli orthologs ) to calculate maximal feasible reaction rates ( Table S2 ) . Data were visualized using the pathway tools package [96] . Raw 264 . 7 macrophage-like cells were cultured in DMEM cell culture medium containing 10% serum and 0 . 5 g l−1 glucose . Cells were infected with Salmonella from stationary cultures at a multiplicity of infection of 30 for 30 min with an initial 5 min 1100×g centrifugation step . Medium was exchanged against DMEM containing 0 . 5 g l−1 glucose and 50 mg l−1 gentamycin . At 4 hours post infection , medium was exchanged with DMEM containing 0 . 5 g l−1 glucose , 1 g l−1 glucose , or 0 . 5 g l−1 glucose and 0 . 5 g l−1 mannitol . Cells were washed and lysed 10 h after infection , and aliquots were plated to determine CFU numbers . The consensus genome-scale metabolism reconstruction STMv1 [31] was updated to STMv1 . 1 based on recent literature ( Tables S6 , S7 ) . For in vivo modeling , we modified biomass requirements based on published mutant virulence phenotypes in infected host tissues . As an example , the high virulence of Salmonella mutants rfbH , rfbJ , rfbV , rfbF , rfbG [97] suggested that lipopolysaccharide with O-sidechains containing the carbohydrate abequose was not required in vivo . In total , these biomass modifications accounted for 14 mutant phenotypes ( for detailed descriptions of all modifications see Table S8 ) . We generated an in vivo model using Flux-Balance Analysis ( FBA ) with the COBRA toolbox [98] in a MATLAB environment . Nutrient uptake rates were adjusted to yield consistent results with experimental competitive indices of Salmonella mutants and reported phenotypes ( Tables S3 , S5 , S9 ) as well as the experimentally determined Salmonella wildtype in vivo generation time of 6 . 4 h [34] using the new MATLAB function nutrientUtilization ( ) ( Script S1 ) . Specifically , uptake of each nutrient was varied and glucose was added to achieve the wildtype growth rate ( in case of glucose as the nutrient of interest , we compensated with glycerol; in case of arginine that we modeled as a nitrogen source , we used ammonium for compensation ) . We then determined the nutrient uptake rate that matched the Competitive Index of a mutant that was unable to utilize this specific nutrient . After completing this procedure for each nutrient , we incorporated all first round nutrient uptake rates in an updated model . We then adjusted the maintenance costs to ensure a normal wildtype growth rate . We repeated this procedure a few times until values converged . We report these as final uptake rates in Table S9 . We determined simulation error margins by analyzing error propagation from the experimental data ( for examples , see Fig . S3 ) . We used the calculated median uptake rate for 14 amino acids to estimate uptake of amino acids alanine , asparagine , aspartate , glutamate , glycine , and serine , for which we lacked informative mutant data . Biomass requirements suggested uptake rates for additional 13 inorganic components ( Table S9 ) . We also explored the possibility of Salmonella access to excess nutrients using the new MATLAB function excess ( ) ( Script S2 ) . Specifically , we increased the growth rate to higher values than experimentally observed . For these scenarios , we determined uptake rates for the six major carbon/energy sources and adjusted maintenance costs as described above . To calculate the corresponding nutrient excess , we then compared the total nutrient uptake for these scenarios to what would be needed for normal growth at the experimentally determined rate . We predicted flux states with “minimal total flux” at maximal rates for biomass generation ( “objective function” ) using the respective options in the optimize ( ) function . We determined flux variability in alternative solutions using the fluxVariability ( ) function . This flux variability analysis was performed without assuming lowest overall metabolic activity to obtain the full range of possible flux states compatible with optimal Salmonella growth . We predicted biomass generation ( which we used as an approximation for growth throughout this study ) for all single gene deletions using the deleteModelGenes ( ) function . Genes were defined as essential if predicted mutant growth rates were below 60% of wildtype ( based on experimental growth data [34] for the avirulent aroA mutant [65] ) , contributing if mutants growth rates were between 60% and 98% , and non-detectable if mutants had growth rates higher than 98% of wildtype . We performed parsimonious FBA using the pFBA ( ) function of the COBRA toolbox . To validate these predictions , we examined reported experimental Salmonella colonization phenotypes and classified genes again as essential ( lethal dose 1000fold higher than wildtype , or CI after four days below 0 . 005 ) , contributing ( significant colonization defect below thresholds for essential genes ) , or non-detectable ( no significant difference to wildtype ) . We also used large-scale mutant phenotypes from two recent high-throughput studies [97] , [99] . In these cases , we converted the reported mutant phenotype scores to growth rates and estimated confidence intervals based on the data provided ( their Table S3 [97]; their Table S3 [99] ) and the Salmonella in vivo generation time of 6 . 4 h in susceptible mice [34] . In cases where conflicting data had been reported , we preferentially used data from studies with low infection dose . Metabolic Pathway predictions for 909 genomes were generated by the MetaCyc consortium [62] and kindly by provided Tomer Altman and Peter Karp on November 22 , 2010 . We identified 287 mammalian pathogens and 367 environmental organisms in this data set . We merged multiple strains belonging to the same species resulting in data for 154 pathogen species and 316 environmental species ( Table S11 ) . We then determined how many organisms in each group were capable to degrade a specific nutrient , or to synthesize a certain biomass component . | Infectious diseases are a major health problem worldwide . To cause disease , pathogens need to acquire host nutrients for growth in infected tissues and for the expression of virulence factors . In this study , we investigated Salmonella nutrition and growth in a well-characterized mouse model of human typhoid fever . We used a panel of Salmonella mutants with metabolic defects to assess the importance of various nutrient utilization pathways for Salmonella growth . We derived from these experimental data a computational model that predicts nutrient uptake rates , activity of metabolic pathways , and the effects of Salmonella enzyme defects on in vivo growth . The vast majority of these predictions were in close agreement with independent experimental data suggesting the model provided a consistent overview of Salmonella metabolism during infection . The data showed that Salmonella depend on a highly complex diet with many different host nutrients , but each of these nutrients is available in only scarce amounts . To grow and cause disease , Salmonella must simultaneously exploit these various nutrients with versatile degradation pathways . Similar complex pathogen diets might also drive many other infectious diseases . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"microbial",
"metabolism",
"microbial",
"pathogens",
"host-pathogen",
"interaction",
"biology",
"salmonella",
"microbiology",
"bacterial",
"pathogens"
] | 2013 | Parallel Exploitation of Diverse Host Nutrients Enhances Salmonella Virulence |
Open , collaborative research is a powerful paradigm that can immensely strengthen the scientific process by integrating broad and diverse expertise . However , traditional research and multi-author writing processes break down at scale . We present new software named Manubot , available at https://manubot . org , to address the challenges of open scholarly writing . Manubot adopts the contribution workflow used by many large-scale open source software projects to enable collaborative authoring of scholarly manuscripts . With Manubot , manuscripts are written in Markdown and stored in a Git repository to precisely track changes over time . By hosting manuscript repositories publicly , such as on GitHub , multiple authors can simultaneously propose and review changes . A cloud service automatically evaluates proposed changes to catch errors . Publication with Manubot is continuous: When a manuscript’s source changes , the rendered outputs are rebuilt and republished to a web page . Manubot automates bibliographic tasks by implementing citation by identifier , where users cite persistent identifiers ( e . g . DOIs , PubMed IDs , ISBNs , URLs ) , whose metadata is then retrieved and converted to a user-specified style . Manubot modernizes publishing to align with the ideals of open science by making it transparent , reproducible , immediate , versioned , collaborative , and free of charge .
There are many existing collaborative writing platforms ( Table 1 ) [6] . In general , platforms with “what you see is what you get” ( WYSIWYG ) editors , such as Microsoft Word or Google Docs , require the least technical expertise to use . On the flip side , WYSIWYG platforms can be difficult to customize and incorporate into automated computational workflows . Traditionally , LaTeX has been used for these needs , since documents are written in plain text and the system is open source and extensible . Rendering LaTeX documents requires specialized software , but webapps like Overleaf now enable collaborative authoring of LaTeX documents . Nonetheless , LaTeX-based systems are limited in that PDF ( or similar ) is the only fully supported output format . Alternatively , Authorea is a collaborative writing webapp whose primary output format is HTML . Authorea allows authors to write in Markdown , a limited subset of LaTeX , or their WYSIWYG HTML editor . Existing platforms work well for editing text and are widely used for scholarly writing . However , they often lack features that are important for open collaborative writing , such as versatile version control and multiple permission levels . For example , Manubot is the only platform listed in Table 1 that offers the ability to address thematically related changes together and enables multiple authors to iteratively refine proposed changes . Manubot’s collaborative writing workflow adopts standard software development strategies that enable any contributor to edit any part of the manuscript but enforce discussion and review of all proposed changes . The GitHub platform supports organizing and editing the manuscript . Manubot projects use GitHub issues for organization , opening a new issue for each discussion topic . For example , in a review manuscript like the Deep Review , this includes each primary paper under consideration . Within a paper’s issue , contributors summarize the research , discuss it ( sometimes with participation from the original authors ) , and assess its relevance to the review . In a primary research article , issues can instead track progress on specific figures or subsections of text being drafted . Issues serve as an open to-do list and a forum for debating the main messages of the manuscript . GitHub and the underlying Git version control system [7 , 8] also structure the writing process . The official version of the manuscript is forked by individual contributors , creating a copy they can freely modify . A contributor then adds and revises files , grouping these changes into commits . When the changes are ready to be reviewed , the series of commits are submitted as a pull request through GitHub , which notifies other authors of the pending changes . GitHub’s review interface allows anyone to comment on the changes , globally or at specific lines , asking questions or requesting modifications [9] . Conversations during review can reference other pull requests , issues , or authors , linking the relevant people and content ( Fig 1 ) . Reviewing batches of revisions that focus on a single theme is more efficient than independently discussing isolated comments and edits and helps maintain consistent content and tone across different authors and reviewers . Once all requested modifications are made , the manuscript maintainers , a subset of authors with elevated GitHub permissions , formally approve the pull request and merge the changes into the official version . The process of writing and revising material can be orchestrated through GitHub with a web browser ( as shown in S1 Video ) or through a local text editor . The Deep Review issue ( https://github . com/greenelab/deep-review/issues/575 ) and pull request ( https://github . com/greenelab/deep-review/pull/638 ) on protein-protein interactions demonstrate this process in practice . A new contributor identified a relevant research topic that was missing from the review manuscript with examples of how the literature would be summarized , critiqued , and integrated into the review . A maintainer confirmed that this was a desirable topic and referred to related open issues . The contributor made the pull request , and two maintainers and another participant made recommendations . After four rounds of reviews and pull request edits , a maintainer merged the changes . We found that this workflow was an effective compromise between fully unrestricted editing and a more heavily-structured approach that limited the authors or the sections they could edit . In addition , authors are associated with their commits , which makes it easy for contributors to receive credit for their work . Fig 2 and the GitHub contributors page ( https://github . com/greenelab/deep-review/graphs/contributors ) summarize all edits and commits from each author , providing aggregated information that is not available on most other collaborative writing platforms . Because the Manubot writing process tracks the complete history through Git commits , it enables detailed retrospective contribution analysis . These pull request and contribution tracking examples both come from Deep Review , the largest Manubot project to date , but illustrate the general principles of transparency and collaboration that are shared by all open Manubot manuscripts . GitHub issues can also be used for formal peer review by independent or journal-selected reviewers . A reviewer conducting open peer review can create issues using their own GitHub account , as one reviewer did for this manuscript ( https://github . com/greenelab/meta-review/issues/124 ) . Alternatively , a reviewer can post feedback with a pseudonymous GitHub account or have a trusted third party such as a journal editor post their comments anonymously . Authors can elect to respond to reviews in the GitHub issues or a public response letter ( https://github . com/greenelab/meta-review/blob/v3 . 0/content/response-to-reviewers . md ) , creating open peer review . Although we developed Manubot with collaborative writing in mind , it can also be helpful for individuals preparing scholarly documents . Authors may choose to make their changes directly to the master branch , forgoing pull requests and reviews . This workflow retains many of Manubot’s benefits , such as transparent history , automation , and allowing outside contributors to propose changes . In cases where outside contributions are unwanted , authors can disable pull requests on GitHub . It is also possible to use Manubot on a private GitHub repository . Private manuscripts require some additional customization to disable GitHub Pages and may require a paid continuous integration plan . See the existing manuscripts for examples of the range of contribution workflows and Manubot use cases . Manubot is a system for writing scholarly manuscripts via GitHub . For each manuscript , there is a corresponding Git repository . The master branch of the repository contains all of the necessary inputs to build the manuscript . Specifically , a content directory contains one or more Markdown files that define the body of the manuscript as well as a metadata file to set information such as the title , authors , keywords , and language . Figures can be hosted in the content/images subdirectory or elsewhere and specified by URL . Repositories contain scripts and other files that define how to build and deploy the manuscript . Many of these operations are delegated to the manubot Python package or other dependencies such as Pandoc , which converts between document formats , and Travis CI , which builds the manuscript in the cloud . Manubot pieces together many existing standards and technologies to encapsulate a manuscript in a repository and automatically generate outputs .
Manubot does not impose any restrictions on authorship . It allows authors to adhere to the author inclusion and ordering conventions of their field , which vary considerably across disciplines [53] . Some Manubot projects create a table in their GitHub repository to track contributors who did not commit text to the manuscript ( https://github . com/Benjamin-Lee/deep-rules/blob/cfb7f744573ca0532a19ca1a8e9473a555cf8eb2/contributors . md ) . This provides a transparent way to record contributions such as experimental research that generated data for the manuscript and discuss whether they meet that project’s authorship criteria . Contribution transparency helps prevent ghostwriting [54] and is especially important in collaborative writing [55] . Although we recommend authors provide their ORCID and GitHub username , Manubot also supports pseudonyms , pseudonymous GitHub usernames , and authors without an ORCID or GitHub account . To determine authorship for the Deep Review , we followed the International Committee of Medical Journal Editors ( ICMJE ) guidelines and used GitHub to track contributions . ICMJE recommends authors substantially contribute to , draft , approve , and agree to be accountable for the manuscript . We acknowledged other contributors who did not meet all four criteria , including contributors who provided text but did not review and approve the complete manuscript . Although these criteria provided a straightforward , equitable way to determine who would be an author , they did not produce a traditionally ordered author list . In biomedical journals , the convention is that the first and last authors made the most substantial contributions to the manuscript . This convention can be difficult to reconcile in a collaborative effort . Using Git , we could quantify the number of commits each author made or the number of sentences an author wrote or edited , but these metrics discount intellectual contributions such as discussing primary literature and reviewing pull requests . Therefore , we concluded that it is not possible to construct an objective system to compare and weight the different types of contributions and produce an ordered author list [56] . To address this issue , we generalized the concept of “co-first” authorship , in which two or more authors are denoted as making equal contributions to a paper . We defined four types of contributions [5] , from major to minor , and reviewed the GitHub discussions and commits to assign authors to these categories . A randomized algorithm then arbitrarily ordered authors within each contribution category , and we combined the category-specific author lists to produce a traditional ordering . The randomization procedure was shared with the authors in advance ( pre-registered ) and run in a deterministic manner . Given the same author contributions , it always produced the same ordered author list . We annotated the author list to indicate that author order was partly randomized and emphasize that the order did not indicate one author contributed more than another from the same category . The Deep Review author ordering procedure illustrates authorship possibilities when all contributions are publicly tracked and recorded that would be difficult with a traditional collaborative writing platform . Papers with hundreds or thousands of authors are on the rise , such as the article describing the experiments and data that led to the discovery of the Higgs Boson [57] ( 5000 authors ) and the report of the Drosophila genome [58] ( 1000 authors ) . Yet the number of people that participated in writing those papers , as opposed to generating and analyzing the data , is not always clear and is likely to be far below the number of authors [59 , 60] . Manubot provides the scientists involved in large collaborations the opportunity to actively participate , through a public forum , in the writing process .
The open scholarly writing Manubot enables has particular benefits for review articles , which present the state of the art in a scientific field [61] . Literature reviews are typically written in private by an invited team of colleagues . In contrast , broadly opening the process to anyone engaged in the topic—such that planning , organizing , writing , and editing occur collaboratively in a public forum where anyone is welcome to participate—can maximize a review’s value . Open drafting of reviews is especially helpful for capturing state-of-the-art knowledge about rapidly advancing research topics at the intersection of existing disciplines where contributors bring diverse opinions and expertise . Writing review articles in a public forum allows review authors to engage with the original researchers to clarify their methods and results and present them accurately , as exemplified at https://github . com/greenelab/deep-review/issues/213 . Additionally , discussing manuscripts in the open generates valuable pre-publication peer review of preprints [22] or post-publication peer review [16 , 62 , 63] . Because incentives to provide public peer review of existing literature [64] are lacking , open collaborative reviews—where authorship is open to anyone who makes a valid contribution—could help spur more post-publication peer review . The Deep Review was not the first scholarly manuscript written online via an open collaborative process . In 2013 , two dozen mathematicians created the 600-page Homotopy Type Theory book , writing collaboratively in LaTeX on GitHub [65 , 66] . Two technical books on cryptocurrency—Mastering Bitcoin ( https://github . com/bitcoinbook/bitcoinbook ) and Mastering Ethereum ( https://github . com/ethereumbook/ethereumbook ) —written on GitHub in AsciiDoc format have engaged hundreds of contributors . Both Homotopy Type Theory and Mastering Bitcoin continue to be maintained years after their initial publication . A 2017 perspective on the future of peer review was written collaboratively on Overleaf , with contributions from 32 authors [67] . While debate was raging over tightening the default threshold for statistical significance , nearly 150 scientists contributed to a Google Doc discussion that was condensed into a traditional journal commentary [68 , 69] . The greatest success to date of open collaborative writing is arguably Wikipedia , whose English version contains over 5 . 5 million articles . Wikipedia scaled encyclopedias far beyond any privately-written alternative . These examples illustrate how open collaborative writing can scale scholarly manuscripts where diverse opinion and expertise are paramount beyond what would otherwise be possible . Open writing also presents new opportunities for distributing scholarly communication . Though it is still valuable to have versioned drafts of a manuscript with digital identifiers , journal publication may not be the terminal endpoint for collaborative manuscripts . After releasing the first version of the Deep Review [10] , 14 new contributors updated the manuscript ( Fig 2 ) . Existing authors continue to discuss new literature , creating a living document . Manubot provides an ideal platform for perpetual reviews [70 , 71] . Concepts for the future of scholarly publishing extend beyond collaborative writing [72–74] . Pandoc Scholar [12] and Bookdown [75] , which has been used for open writing [76] , both enhance traditional Markdown to better support publishing . The knitcitations ( https://github . com/cboettig/knitcitations ) package enables citation by DOI or URL in R Markdown documents . Examples of continuous integration to automate manuscript generation include gh-publisher ( https://github . com/ewanmellor/gh-publisher ) and jekyll-travis ( https://github . com/mfenner/jekyll-travis ) , which was used to produce a continuously published webpage ( http://book . openingscience . org/ ) for the Opening Science book [77 , 78] . Binder [11] , Distill journal articles [79] , Idyll [80] , and Stencila [81 , 82] support manuscripts with interactive graphics and close integration with the underlying code . As an open source project , Manubot can be extended to adopt best practices from these other emerging platforms . Several other open science efforts are GitHub-based like our collaborative writing process . ReScience [83] as well as titles from Open Journals , such as the Journal of Open Source Software [52] , rely on GitHub for peer review and hosting . Distill uses GitHub for transparent peer review and post-publication peer review [84] . GitHub is increasingly used for resource curation [85] , and collaborative scholarly reviews combine literature curation with discussion and interpretation . There are potential limitations of our GitHub-based approach . Because the Deep Review pertained to a computational topic , most of the authors had computational backgrounds , including previous experience with version control workflows and GitHub . In other disciplines , collaborative writing via GitHub and Manubot could present a steeper barrier to entry and deter participants . In addition , Git carefully tracks all revisions to the manuscript text but not the surrounding conversations that take place through GitHub issues and pull requests . These discussions must be archived to ensure that important decisions about the manuscript are preserved and authors receive credit for intellectual contributions that are not directly reflected in the manuscript’s text . GitHub supports programmatic access to issues , pull requests , and reviews so tracking these conversations is feasible in the future . In the Deep Review , we established contributor guidelines ( https://github . com/greenelab/deep-review/blob/v1 . 0/CONTRIBUTING . md ) that discussed norms in the areas of text contribution , peer review , and authorship , which we identified in advance as potential areas of disagreement . Our contributor guidelines required verifiable participation for authorship: either directly attributable changes to the text or participation in the discussion on GitHub . These guidelines did not discuss broader community norms that may have improved inclusiveness . It is also important to consider how the move to an open contribution model affects under-represented minority members of the scientific community [19] . Recent work has identified clear social norms and processes as helpful to maintaining a collaborative culture [86] . Conferences and open source projects have used codes of conduct to establish these norms ( e . g . Contributor Covenant at https://www . contributor-covenant . org/ ) [87] . We would encourage the maintainers of similar projects to consider broader codes of conduct for project participants that build on social as well as academic norms . Science is undergoing a transition towards openness . The internet provides a global information commons , where scholarship can be publicly shared at a minimal cost . For example , open access publishing provides an economic model that encourages maximal dissemination and reuse of scholarly articles [18 , 88 , 89] . More broadly , open licensing solves legal barriers to content reuse , enabling any type of scholarly output to become part of the commons [90 , 91] . The opportunity to reuse data and code for new investigations , as well as a push for increased reproducibility , has begot a movement to make all research outputs public , unless there are bona fide privacy or security concerns [92–94] . New tools and services make it increasingly feasible to publicly share the unabridged methods of a study , especially for computational research , which consists solely of software and data . Greater openness in both research methods and publishing creates an opportunity to redefine peer review and the role journals play in communicating science [67] . At the extreme is real-time open science , whereby studies are performed entirely in the open from their inception [95] . Many such research projects have now been completed , benefiting from the associated early-stage peer review , additional opportunity for online collaboration , and increased visibility [50 , 96] . Manubot is an ideal authoring protocol for real-time open science , especially for projects that are already using an open source software workflow to manage their research . While Manubot does require technical expertise , the benefits are manyfold . Specifically , Manubot demonstrates a system for publishing that is transparent , reproducible , immediate , permissionless , versioned , automated , collaborative , open , linked , provenanced , decentralized , hackable , interactive , annotated , and free of charge . These attributes empower integrating Manubot with an ecosystem of other community-driven tools to make science as open and collaborative as possible . | Traditionally , scholarly manuscripts have been written in private by a predefined team of collaborators . But now the internet enables realtime open science , where project communication occurs online in a public venue and anyone is able to contribute . Dispersed teams of online contributors require new tools to jointly prepare manuscripts . Existing tools fail to scale beyond tens of authors and struggle to support iterative refinement of proposed changes . Therefore , we created a system called Manubot for writing manuscripts based on collaborative version control . Manubot adopts the workflow from open source software development , which has enabled hundreds of contributors to simultaneously develop complex codebases such as Python and Linux , and applies it to open collaborative writing . Manubot also addresses other shortcomings of current publishing tools . Specifically , all changes to a manuscript are tracked , enabling transparency and better attribution of credit . Manubot automates many tasks , including creating the bibliography and deploying the manuscript as a webpage . Manubot webpages preserve old versions and provide a simple yet interactive interface for reading . As such , Manubot is a suitable foundation for next-generation preprints . Manuscript readers have ample opportunity to not only provide public peer review but also to contribute improvements , before and after journal publication . | [
"Abstract",
"Introduction",
"Authorship",
"Discussion"
] | [
"computer",
"applications",
"open",
"science",
"peer",
"review",
"computer",
"and",
"information",
"sciences",
"data",
"management",
"web-based",
"applications",
"computer",
"networks",
"science",
"policy",
"internet",
"computer",
"software",
"citation",
"analysis",
"open",
"source",
"software",
"metadata",
"scientific",
"publishing",
"research",
"assessment",
"research",
"and",
"analysis",
"methods"
] | 2019 | Open collaborative writing with Manubot |
Japanese encephalitis virus ( JEV ) is the major cause of viral encephalitis in Southeast Asia . Vaccination of domestic pigs has been suggested as a “one health” strategy to reduce viral disease transmission to humans . The efficiency of two lentiviral TRIP/JEV vectors expressing the JEV envelope prM and E glycoproteins at eliciting protective humoral response was assessed in a mouse model and piglets . A gene encoding the envelope proteins prM and E from a genotype 3 JEV strain was inserted into a lentiviral TRIP vector . Two lentiviral vectors TRIP/JEV were generated , each expressing the prM signal peptide followed by the prM protein and the E glycoprotein , the latter being expressed either in its native form or lacking its two C-terminal transmembrane domains . In vitro transduction of cells with the TRIP/JEV vector expressing the native prM and E resulted in the efficient secretion of virus-like particles of Japanese encephalitis virus . Immunization of BALB/c mice with TRIP/JEV vectors resulted in the production of IgGs against Japanese encephalitis virus , and the injection of a second dose one month after the prime injection greatly boosted antibody titers . The TRIP/JEV vectors elicited neutralizing antibodies against JEV strains belonging to genotypes 1 , 3 , and 5 . Immunization of piglets with two doses of the lentiviral vector expressing JEV virus-like particles led to high titers of anti-JEV antibodies , that had efficient neutralizing activity regardless of the JEV genotype tested . Immunization of pigs with the lentiviral vector expressing JEV virus-like particles is particularly efficient to prime antigen-specific humoral immunity and trigger neutralizing antibody responses against JEV genotypes 1 , 3 , and 5 . The titers of neutralizing antibodies elicited by the TRIP/JEV vector are sufficient to confer protection in domestic pigs against different genotypes of JEV and this could be of a great utility in endemic regions where more than one genotype is circulating .
Mosquito-borne Japanese encephalitis virus is a member of the Flavivirus genus in the Flaviviridae family [1–4] . Flaviviruses contain a positive single-stranded RNA genome encoding a polyprotein that is processed into three structural proteins , the capsid ( C ) , the precursor of membrane ( prM ) and the envelope ( E ) , and seven non-structural proteins NS1 to NS5 [4] . Viral assembly occurs in the lumen of the endoplasmic reticulum: the nucleocapsids associate with prM-E heterodimers to form an immature JEV virion . The latter transits through the secretory pathway , where it is matured through cleavage of prM into the membrane ( M ) protein by furin in the trans-Golgi [4] . Additionally , JEV produces virus-like particles ( VLPs ) , which are assembled solely from prM and E proteins , and undergo the same maturation process as genuine viral particles [5] . These VLPs can be produced in the absence of any other viral component [5] . JEV is the etiologic agent of the most important viral encephalitis of medical interest in South Asia , with an incidence of 50 , 000 cases and about 10 , 000 deaths per year [1 , 3 , 6] . About 20 to 30% of the symptomatic human cases are fatal , while 30 to 50% of survivors can develop long-term neurologic sequelae . JEV is usually maintained in an enzootic cycle between Culex tritaeniorhynchus mosquitoes and amplifying vertebrate hosts , such as waterbirds and domestic pigs [1 , 3 , 7] . Horses and humans are thought to be dead-end hosts , since they do not develop a level of viremia sufficient to infect mosquitoes [7] . In the past decades , there has been an expansion of the geographic distribution of JEV in Asia and a possible introduction of JEV into Europe has been documented recently [6 , 8] . Phylogenetic studies based on the viral envelope protein sequences allow the division of JEV strains into genotypes G1 to G5 [1 , 3 , 9–15] . Initially , most of the circulating strains of JEV belonged to G3 and were at the origin of major epidemics in Southeast Asian countries . Recently a shift in prevalence from JEV G3 to G1 has been observed in several Asian countries , while some strains of JEV G5 have been occasionally isolated in China and South Korea [9–16] . We previously demonstrated that both integrative and non-integrative lentiviral vectors are promising vaccination vectors against arboviruses such as West Nile virus ( WNV ) , a neurotropic Flavivirus that belongs to the JEV serocomplex [17 , 18] . Immunization with a single minute dose of recombinant lentiviral TRIP vectors that express the soluble form of WNV E protein resulted to a robust protection against a lethal challenge with WNV in mice [17 , 18] . The currently used lentiviral delivery vectors , mostly derived from human deficiency virus-1 ( HIV-1 ) , allow in vivo stable transduction of dendritic cells . This allows a sustained antigen presentation through the endogenous pathway , which in turn elicits robust both humoral and cellular adaptative immunity [19–21] . Humoral immunity plays a pivotal role in protection against JEV infection [22–25] and consequently , the elicitation of a protective antibody response is critical in the development of safe JEV vaccines [25] . In the present study , we evaluated the ability of two lentiviral vectors TRIP/JEV expressing JEV G3 prM and E to induce a protective humoral immune response against JEV infection in a mouse model and in pigs . Both TRIP/JEV . prME vectors were efficient at producing broad anti-JEV neutralizing antibodies in a mouse model . Immunization of piglets with a TRIP/JE vector expressing JEV VLPs elicited high titers of specific neutralizing antibodies that could give a sufficient protection against different JEV genotypes .
Mosquito Aedes albopictus C6/36 cells were maintained at 28°C in Leibovitz medium ( L15 ) supplemented with 10% heat-inactivated fetal bovine serum ( FBS ) . Vero cells and were maintained at 37°C in Dulbecco’s modified Eagle medium ( DMEM ) supplemented with 5% FBS . BHK-21 , SK-N-SH , and HEK-293T cells were maintained at 37°C in DMEM supplemented with 10% FBS . Highly purified mouse anti-pan Flavivirus E monoclonal antibody ( mAb ) 4G2 was produced by RD Biotech ( Besançon , France ) . Mouse mAb anti-JEV NS5 was kindly provided by Y . Matsuura [26] . Antibodies against Calnexin and SNAP-tag were purchased from Enzo Life Sciences and New England Biolabs , respectively . Horseradish peroxidase ( HRP ) -conjugated goat anti-mouse IgG and anti-rabbit IgG antibodies were obtained from Bio-Rad Laboratories . HRP-conjugated goat anti-pig antibody was obtained from Bethyl Laboratories . Alexa Fluor 488-conjugated goat anti-mouse IgG antibody was obtained from Jackson ImmunoResearch . The JEV G1 strain G1 CNS769_Laos_2009 [27] was kindly provided by R . Charrel [27] . The JEV G3 strain Nakayama was obtained from the National Collection of Pathogenic Viruses ( NCPV , Salisbury , UK ) and passaged twice on Vero cells . The molecular clone of JEV G3 strain RP-9 was kindly provided by Y-L . Lin [28] and modified to produce pBR322 ( CMV ) -JEV-RP9 , as described previously [29] . To produce infectious virus , the molecular clone was transfected into HEK-293T cells using Lipofectamine 2000 ( Life Technologies ) . At three days post-transfection , viral supernatants were collected and used to infect C6/36 cells in order to grow final virus stocks for experiments . The chimera JEV S-g5/NS-g3 ( G5/G3 ) which express the structural proteins C , prM , and E from the JEV G5 strain XZ0934 fused to the nonstructural proteins of JEV G3 RP-9 has been already described [29] . The chimeric JEV S-g1/NS-g3 ( G1/G3 ) virus containing the structural protein region from the JEV G1 CNS769_Laos_2009 fused to the RP-9 nonstructural protein region was produced as follows . A silent mutation that created a unique restriction site ( AflII ) at position 2208–2213 ( residues 705 and 706 of the viral polyprotein ) was introduced directly in pBR322 ( CMV ) -JEV-RP9 through PCR mutagenesis using primers 5’-actgggaaaggctttcacgaccactcttaagggtgctcagagac-3’ and 5’-gtctctgagcacccttaagagtggtcgtgaaagcctttcccagt-3’ ( the Afl II site is underlined ) . The resulting pBR322 ( CMV ) -JEV-RP9 ( Afl II ) plasmid was used as template to generate the chimeric JEV . The fragment corresponding to nucleotides 114 to 2213 and flanked by the unique sites ApaI and AflII was substituted with the homologous fragments of JEV G1 strain ( nucleotides 115–2214 ) . Both chimeric viruses were produced through transfection of the cDNA infectious clone , as described for JEV G3 strain RP-9 . For the construction of recombinant lentiviral vectors expressing JEV proteins , modifications that optimize the expression of prM and E genes in mammalian cells were done on the original sequence of JEV strain RP-9 of G3 using a synthetic gene ( Genecust , Lux . ) . The mammalian codon-optimized sequence coding for prM signal peptide followed by prM and E glycoproteins was cloned into the BamH I and Xho I restriction sites of the pTRIPΔU3CMV plasmid , to generate pTRIPΔU3CMV/JEV . prME . The optimized sequence was further modified by mutagenesis PCR to generate pTRIPΔU3CMV/JEV . prMEΔTM that contains the genes encoding prM and E lacking its two transmembrane domains ( EΔTM ) . Lentiviral particles were produced by transient calcium co-transfection of HEK-293T cells as described previously [30] , but with the following modifications: at 24h hours post-transfection , the cell culture medium was replaced by serum-free DMEM . Supernatants were collected at 48 hours post-transfection , clarified by several rounds of low-speed centrifugation , and stored at -20°C . The recombinant lentiviral vectors were pseudotyped with VSV-G envelope protein of serotype Indiana ( IND ) or New Jersey ( NJ ) [31] . In the resulting vectors TRIP/JEV . prME and TRIP/JEV . prMEΔTM the CMV immediate early promoter ( CMVie ) drives the constitutive expression of recombinant JEV proteins . The TRIP/JEV vector stocks were titrated by real-time PCR on cell lysates from transduced HEK-293T cells and expressed as transduction unit ( TU ) per ml [17 , 18] . Titers of non-concentrated TRIP/JEV . prME vector bearing IND or NJ VSV . G envelope protein were 6 . 8 log10 TU . mL-1 and 6 . 2 log10 TU . mL-1 , respectively . Titers of lentiviral TRIP/JEV . prMEΔTM vector bearing IND or NJ VSV . G envelope protein were 7 . 1 log10 TU . mL-1 and 6 . 2 log10 TU . mL-1 , respectively . Lentiviral vector stocks were adjusted by dilution in sterile PBS and were inoculated in mice or pigs without further concentration . Human HEK-293T cells were transduced with recombinant lentiviral vectors . The supernatants were collected at 2 days post-transduction and clarified by centrifugation at 3 , 000g for 5 min at 4°C . The clarified supernatant was loaded over a 15% sucrose cushion in TNE buffer ( 10 mM Tris-HCl [pH 7 . 5] , 2 . 5 mM EDTA , 50 mM NaCl ) , and centrifuged at 100 , 000g for 2 . 5 h at 4°C . The supernatants were discarded , and the purified virus-like particles ( VLPs ) were suspended in 50 μl of TNE buffer . For titration of JEV infectivity , focus-forming assays ( FFA ) were performed on BHK-21 cells as previously described [29] . Virus titers were given in focus forming units per ml ( FFU . mL-1 ) . Large flasks of Vero cell monolayers were inoculated with JEV at low multiplicity of infection . The supernatant fluids of JEV-infected ( JEV antigen ) or mock-infected ( normal cell antigen or NCA ) cells were harvested and clarified . The supernatants were precipitated with 7% w/v PEG 6 , 000 ( Fluka ) , centrifuged , and the viral pellet was suspended in cold PBS supplemented with 0 . 1% ß-propiolactone in 0 . 1 M Sorensen buffer ( pH 9 . 0 ) for JEV inactivation . The working dilution of inactivated JEV antigen ( 1:200 ) was estimated based on an « in-house » indirect ELISA using well-characterized human positive JEV serum samples and already validated JEV antigen . The DES expression system ( Life Technologies ) was required for the production of recombinant viral antigens in Drosophila S2 cells . A synthetic gene coding for prM followed by EΔTM from JEV G3 strain SA-14 was cloned into the shuttle vector pMT/BiP/SNAP , a derived pMT/BiP/V5-HisA plasmid ( Life Technologies ) in which the SNAP-tag sequence ( Covalys BioSciences AG ) had been inserted in frame with the insect BiP signal peptide . The resulting plasmid pMT/BiP/JEV . prMEΔTM-SNAP encodes prM followed by EΔTM in fusion with the N-terminus of SNAP-tag . The recombinant domain III from the E protein ( EDIII ) of JEV G1 strain JaNAr0102/Japan/2002/Mosquito , JEV G3 strain GP05 , and JEV G5 strain 10–1827 were fused in frame to the C-terminus of SNAP-tag into the plasmid pMT/BiP/SNAP . The resulting plasmids pMT/BiP/JEV . prMEΔTM-SNAP and pMT/BiP/SNAP-JEV . EDIII were transfected into S2 cells to establish stable cell lines S2/JEV . prMEΔTM-SNAP and S2/SNAP-JEV . EDIII for G1 , G3 , and G5 as previously described [29] . The production and the purification of recombinant viral antigens from stable S2 cell lines were performed as previously described [28] . Western blot assay was essentially performed as previously described [29] . For immunofluorescence ( IF ) assay , cells were fixed with 3 . 2% paraformaldehyde in PBS and permeabilized with 0 . 1% Triton X-100 in PBS . JEV E protein was detected with the mAb 4G2 , followed by incubation with AlexaFluor488-conjugated secondary antibody . The cover slips were mounted with ProLong Gold Antifade Reagent with DAPI ( Life Technologies ) . The slides were examined using a fluorescent microscope ( Axioplan 2 Imaging , Zeiss ) . Groups of 6-week-old female BALB/c mice ( n = 6 ) were intraperitoneally ( i . p . ) inoculated with recombinant lentiviral vectors in 0 . 1 ml DPBS supplemented with 0 . 2% endotoxin-free serum albumin . Animals were bled by puncturing at the retro-orbital sinus level . A very low individual variability exists within each group of mice inoculated with recombinant lentiviral vectors justifying the use of pooled sera in subsequent experiments [18] . A control group of 3-week-old female BALB/c mice ( n = 6 ) was inoculated with 3 log10 FFU of JEV G3 ( strain RP-9 ) and immune sera was collected and pooled at 21 days post-inoculation . For passive seroprotection experiments , pooled immune sera were transferred i . p . into groups of 3-week-old C57BL/6 female mice ( n = 6 or 12 ) one day before a challenge with 10 FFU of JEV G3 ( strain RP-9 ) inoculated by the i . p . route . The challenged mice were monitored for signs of morbidity and mortality . Euthanasia was applied on animals showing the symptoms of viral encephalitis . Groups of 7-week-old specific pathogen free Swiss Land Race piglets from in-house breeding were housed in groups , and an adaptation time to the new environment of one week was given before starting the experiment . For immunization , the TRIP/JEV . prME lentiviral vector was diluted to a final volume of 0 . 5 ml with PBS ( Life Technologies ) . Immunization with the TRIP/GFP lentiviral vector was used as a negative control [18] . From a group of 5 piglets , four were vaccinated intramuscularly with various doses of the TRIP/JEV . prME vector and one was injected with the equivalent dose of control lentiviral vector TRIP/GFP . Immunized animals were bled before the first vaccination and then weekly until the end of the experiment . Four weeks after the first vaccination , all animals got a booster vaccination with the same dose of recombinant lentiviral vectors as at the first time point . For ethical reasons no lethal challenge was performed as protection in pigs . As a control , 3 animals were inoculated by the oronasal route with 7 log10TCID50 of live JEV Nakayama G3 . All pigs developed temporary fever and viremia and recovered completely after 4–6 days . The animal sera were examined weekly for anti-JEV antibody . Indirect ELISA measured the production of anti-JEV IgGs in immunized mice and piglets . A series of 96-well ELISA plates ( Nunc ) was coated with 0 . 1 ml of inactivated native JEV antigen or highly purified recombinant JEV antigens diluted in PBS at the concentration of 1 μg . mL-1 at 4°C overnight . NCA and SNAP served as negative control antigens . Indirect ELISA were performed as previously described ( 29 ) . The Immune Status Ratio ( ISR ) of each group of immunized mice is obtained by dividing the average of JEV antigen OD450 values by the average control antigen OD450 values . The end-point titers of anti-JEV antibodies in mouse sera were calculated as the reciprocal of the last dilution of serum having ISR value > 3 . 0 . Pig sera were tested as described for the mice , using HRP-conjugated goat anti-pig antibody as a secondary antibody . Pig sera obtained prior immunization were used as a negative control . Neutralizing ability of mouse and pig serum antibodies against JEV was determined by focus ( FRNT ) or plaque ( PRNT ) reduction neutralization tests on Vero cells , respectively . Mouse serum samples from each group were pooled . Pig sera were tested individually in triplicates starting at a 1:5 serum dilution . Pooled mouse or individual pig serum were two-fold serial diluted in DMEM supplemented with 2% FBS , with a starting dilution of 1:10 , and incubated for 2 h at 37°C with an equal volume of viral suspension containing 100 FFU of JEV . The end-point titer was calculated as the reciprocal of the highest serum dilution tested that reduced the number of FFU ( FRNT50 ) or PFU ( PRNT50 ) by 50% . A Log-rank ( Mantel-Cox ) test was used to compare survival data . Antibody levels between groups of immunized pigs were compared by Mann Whitney U test and the level of significance was set at 5% . GraphPad Prism ( GrapPad Software Inc . La Jolla , CA , USA ) was used for all statistical analysis . All mice were housed under pathogen-free conditions at the Institut Pasteur animal facility . The protocols and subsequent experiments were ethically approved by the CETEA ( Ethic Committee for Control of Experiments on Animals: http://cache . media . enseignementsup-recherche . gouv . fr/file/Encadrement_des_pratiques_de_recherche/58/1/Charte_nationale_portant_sur_l_ethique_de_l_experimentation_animale-version_anglaise_243581 . pdf ) at the Institut Pasteur ( C2A N°89/CETEA ) with the reference n°2013–0071 and declared to the French Ministère de l’Enseignement Supérieur et de la Recherche ( reference n° 000762 . 1 ) in accordance with the articles R . 214-24 et R . 214-125 du Code Rural et de la Pêche Maritime and R . 214-120 du décret n°2013–118 du 1er février 2013 in France . Experiments were conducted following the guidelines of the Office Laboratory of Animal Care at the Institut Pasteur . The protocols and subsequent experiments on pigs were ethically approved by the cantonal ethical committee of Bern ( number BE 118–13 ) and the FSVO ( Federal Food Safety and Veterinary Office , website: http://www . blv . admin . ch/index . html ? lang=en ) in Switzerland . Pig experiments were conducted following the guidelines of Swiss Animal Welfare Regulations ( Veterinary Service of LANAT ) . The Genbank accession numbers of JEV strains SA-14 , JaNAr0102/Japan/2002/Mosquito , GP05 and 10–187 are M55506 , AY377577 , FJ979830 , and JN587258 , respectively .
We have reported earlier that a single minute dose of a non-replicative lentiviral vector expressing the soluble form of WNV E glycoprotein induced a robust protective humoral response in a mouse model of WNV encephalitis [17 , 18] . To assess the potential of lentiviral vectors expressing JEV proteins at eliciting protective humoral response against JEV infection , a mammalian codon-optimized gene encoding prM and E from the JEV strain RP9 of G3 was inserted into the lentivirus TRIP vector ( Fig 1 ) . We generated two lentiviral vectors , expressing the prM signal peptide followed by the prM protein and the E glycoprotein , the latter being expressed either in the native form ( TRIP/JEV . prME ) or lacking its two C-terminal transmembrane domains ( TRIP/JEV . prMEΔTM ) . In these constructs , prM contributes to the folding , stability , and efficient secretion of the glycoprotein E . Lentiviral vectors which expressed JEV proteins were pseudotyped with VSV-G protein of the IND serotype . Non-replicative TRIP/JEV . prME and TRIP/JEV . prMEΔTM particles were produced from HEK-293T cells , achieving titers of about 7 log10 TU . mL-1 . The antigenicity of recombinant JEV proteins was assessed by transducing HEK-293T cells with the TRIP/JEV . prME or TRIP/JEV . prMEΔTM vectors ( Fig 2A ) . An empty vector served as a control . At 48 h post-transduction , the intracellular form of the E protein was detected using the anti-E MAb 4G2 by IF assay . A similar staining pattern for E was observed in TRIP/JEV-transduced cells expressing prME or prMEΔTM . Immunoblot assays using mouse anti-JEV antisera detected recombinant prM and E in lysates from HEK-293T cells transduced with TRIP/JEV vectors ( Fig 2B ) . We observed an efficient release of E protein to the supernatants of HEK-293T cells transduced with either TRIP/JEV vectors ( Fig 2C , top ) . The presence of prM was only detected in supernatants from cells transduced with the TRIP/JEV . prME vector ( Fig 2C , top ) . Because JEV prM and E have the capacity to self-assemble into VLPs , we assessed whether VLPs were secreted from HEK-293T cells transduced with TRIP/JEV vectors . The VLPs were detected by immunoblot assay using anti-E mAb 4G2 and anti-JEV immune serum ( Fig 2C , bottom ) . Extracellular JEV VLPs containing prM and E accumulated in the supernatant of HEK-293T cells transduced with TRIP/JEV . prME vector but not with TRIP/JEV . prMEΔTM vector . Thus , transduction of cells with TRIP/JEV . prME vector leads to efficient secretion of recombinant JEV VLPs . To evaluate humoral responses induced by the lentiviral TRIP/JEV vectors , adult BALB/c mice were inoculated with increasing doses of TRIP/JEV . prME or TRIP/JEV . prMEΔTM ( 3 to 5 log10 TU per animal ) by i . p . route . At 21 days post-immunization , sera were collected from each group of mice and pooled . Indirect ELISA was performed to detect anti-JEV IgGs using inactivated JEV particles as capture antigens ( Table 1 ) . NCA served as a control antigen . There was little to no antibody responses against JEV at TRIP/JEV vector doses lower than 5 log10 TU per animal . The dose of 5 log10 TU induced a significant production of anti-JEV specific antibodies with a mean titer reaching 1 , 600 for TRIP/JEV . prME and 400 for TRIP/JEV . prMEΔTM ( Table 1 ) . At the highest dose ( 6 log10 TU ) inoculated to mice , the mean titer of TRIP/JEV . prME antibody reached 10 , 000 . The latter dose was not further used due to the too large volume of non-concentrated TRIP/JEV vector inoculated to mice by i . p . route . We therefore decided to select the unique dose of 5 log10TU in subsequent mouse immunizations . To determine the time course of anti-JEV production , BALB/c mice that received 5 log10TU of TRIP/JEV . prME or TRIP/JEV . prMEΔTM were bled at 7 , 14 and 21 days post-immunization ( Table 1 ) . Anti-JEV antibodies were detectable at day 14 of immunization and reached significant titers at day 21 . To enhance the production of anti-JEV specific antibodies , groups of 12 mice previously immunized with 5 log10 TU of recombinant TRIP/JEV vectors received a booster dose of 5 log10 TU of homologous vectors bearing the VSV-G envelope protein of a different VSV strain ( NJ ) , 4 weeks after the first inoculation . Immune sera were collected 3 weeks after the boosting inoculation and ELISA measurements on pooled sera showed a 40-fold increase in anti-JEV antibody titers reaching the mean titers of 64 , 000 for TRIP/JEV . prME and 16 , 000 for TRIP/JEV . prMEΔTM . It is of interest that the level of anti-JEV antibody in mice twice immunized with 5 log10 TU of recombinant lentiviral vector expressing JEV VLPs was similar to that obtained following a challenge with 3 log10 FFU of live JEV G3 RP-9 . The reactivity of antibodies raised in mice after TRIP/JEV immunization was evaluated by immunoblotting using lysates of JEV-infected cells as viral antigens ( Fig 3 ) . Mice that received TRIP/JEV . prME displayed specific antibodies against prM and E whereas TRIP/JEV . prMEΔTM antisera contained only anti-E antibody . Thus , the co-expression of prM and the soluble form of E failed to stimulate the production of anti-prM antibody in a mouse model . Indirect ELISA tests were performed to determine the anti-E antibody levels in sera from mice immunized with TRIP/JEV . prME or TRIP/JEV . prMEΔTM . The soluble form of JEV G3 E fused to SNAP tag ( JEV . EΔTM-SNAP ) was used as a capture antigen of anti-JEV E IgGs . Given that flavivirus EDIII contains sub-type specific neutralizing epitopes , the recombinant SNAP-tagged EDIII proteins of G1 , G3 , and G5 were also used as capture viral antigens . Sequence alignment of the three EDIII identified a similarity of 96 . 3% between G3 and G1 and 92 . 5% between G3 and G5 ( Fig 4A ) . Purity and specificity of recombinant JEV . EΔTM-SNAP protein ( Fig 4B ) and SNAP-JEV . EDIII of G1 , G3 , and G5 protein ( Fig 4C ) were verified by immunoblotting using an antibody against SNAP-tag . Indirect ELISA showed that BALB/c mice that recovered from a lethal challenge with JEV G3 gave titers of anti-E antibody of 1 , 300 and anti-EDIII antibody from 1 , 000 to 4 , 000 ( Table 2 ) . BALB/c mice that received two doses of TRIP/JEV . prMEΔTM or TRIP/JEV . prME elicited anti-E antibody titers with a similar range of about 1 , 000 . Both lentiviral TRIP/JEV vectors were capable of inducing a similar level of anti-EDIII antibodies that are broadly reactive with different genotypes of JEV . It is of interest that immunization with TRIP/JEV vectors induced higher levels of anti-EDIII G5 antibody than live JEV G3 . A focus reduction neutralization test ( FRNT50 ) was performed to evaluate the ability of TRIP/JEV vectors to elicit a neutralizing antibody response against JEV G3 ( Table 3 ) . Immune sera obtained from BALB/c mice that recovered from a lethal challenge with JEV strain RP-9 had a FRNT50 of 150 . A weak FRNT50 titre of 10 was observed in mice inoculated with a single dose of 5 log10 TU of TRIP/JEV vector . A booster dose one month after the prime elicited JEV-neutralizing antibodies titers to 40 ( TRIP/JEV . prMEΔTM ) and 80 ( TRIP/JEV . prME ) ( Table 3 ) . Recent studies addressed the ability of already existing G3 derived vaccines at protecting against strains belonging to distinct genotypes [32–34] , and such testing should be systematically included when assessing newly designed JEV vaccines . We assessed the protective capacity of TRIP/JEV immunization against other circulating JEV genotypes , namely G1 and G5 . To investigate this issue , we decided to substitute the region encoding C , prM and E into the infectious cDNA clone of JEV G3 by the counterpart from JEV G1 or G5 ( Fig 5A ) . Since immunizations with the TRIP/JEV vectors are solely directed against JEV structural proteins , the contribution of non-structural proteins of JEV G1 and G5 was not explored . The growth of chimera JEV G1/3 or JEV G5/3 was comparable to that of parental JEV virus of G3 in cultured cell lines [29] . Immunoblot analysis showed that JEV G3 antisera recognized both prM and E from JEV G1 and G5 ( Fig 5B , left panel ) . However , JEV G3 antisera weakly reacted with prM from the chimera JEV G5/G3 . Essentially similar results were obtained when immunoblot assay was performed with anti-TRIP/JEV antisera ( Fig 5B ) . The immune sera from BALB/c mice inoculated with TRIP/JEV . prME or TRIP/JEV . prMEΔTM recognized the E protein from chimera JEV G1/G3 and G5/G3 . Only TRIP/JEV . prME antisera reacted with prM from chimera JEV G1/G3 and a to lesser extent , JEV G5/G3 . This lower reactivity was not due to a defect in JEV G5/3 protein accumulation since the reactivity of anti-JEV NS5 antibody was comparable amongst the different lysates of JEV-infected cells ( Fig 5B , lower right panel ) . We observed that TRIP/JEV . prMEΔTM was capable of inducing antibodies that can similarly react with the E protein from chimera JEV G1/3 and G5/3 , thus suggesting that soluble EΔTM exhibits a greater propensity to generate anti-E antibodies that recognize conserved epitopes regardless of the genotype ( Fig 5B ) . FRNT assays were performed to evaluate the ability of TRIP/JEV vectors to elicit a neutralizing antibody response against JEV G1/3 or G5/3 ( Table 3 ) . Pooled immune sera obtained from BALB/c mice infected with JEV G3 gave had FRNT50 values of 140 and 50 against chimeric JEV G1/3 and G5/3 , respectively . Immunized mice that received TRIP/JEV vectors developed neutralizing antibody that were also active against chimeric JEV G1/3 and G5/3 ( Table 3 ) . We noted that immunization with the TRIP/JEV . prME vector elicited slightly higher levels of neutralizing anti-JEV antibodies . The lower neutralization capability of TRIP/JEV-induced antibodies against chimera G5/3 correlated with the weaker reactivity of immune sera towards the prM and E proteins from JEV of G5 ( Fig 5B ) . These data show that both TRIP/JEV vectors were capable of stimulating the production of anti-JEV antibodies that neutralize JEV G1 and G3 , and to a lesser extent JEV G5 . We previously reported that inoculation of JEV G3 strain RP-9 to three-week-old C57BL/6 mice was lethal within one week [28] . Given that mouse susceptibility to JEV is age-dependent , we were unable to challenge animals following the long prime-boost vaccination period with TRIP/JEV vectors . Consequently , we decided to apply a protocol of passive transfer of TRIP/JEV antisera into three-week-old C57BL/6 mice . To address whether the humoral immunity elicited in mice after TRIP/JEV . prME or TRIP/JEV . prMEΔTM vaccination was protective in vivo , groups of 3-week-old C57BL/6 mice received i . p . inoculation of 10 μl of pooled , heat-inactivated immune sera collected from TRIP/JEV-inoculated mice two months after boosting . Pooled immune sera of BALB/c mice inoculated with JEV G3 served as a positive control . A group of mice inoculated with PBS was included . One day after the passive transfer of antisera , the mice were i . p . challenged with a lethal dose of JEV G3 . The animals were observed daily for clinical signs of illness and mortality over three weeks ( Fig 6 ) . Approximately 70% of the mice inoculated with PBS died within the 9–11 days post-challenge whereas administration of JEV immune sera induced a survival rate of 85% ( P = 0 . 05 ) . A survival rate of 50–60% was observed in mice after transfer of TRIP/JEV . prME or TRIP/JEV . prMEΔTM antisera . While inoculation with a single dose of TRIP/JEV antisera did not confer satisfactory levels of protection in our mouse model , we do note that a slight level of protection was obtained . Because lentiviral based-expression of JEV VLPs is particularly efficient at triggering neutralizing antibody responses , we assessed the capacity of TRIP/JEV . prME to stimulate a protective humoral response in pigs . Groups of four 7-week-old piglets were immunized intramuscularly with 6 ( low dose ) or 7 ( high dose ) log10 TU of TRIP/JEV . prME ( Fig 7 ) . As a control , two animals received a low or high dose of a recombinant lentiviral vector expressing reporter GFP . Indirect ELISA using recombinant EΔTM-SNAP protein as a viral antigen was used to assess the production of anti-JEV E antibodies in immunized pigs weekly ( Fig 7A ) . The monitoring of the antibody responses during the first 4 weeks after the prime inoculation revealed an efficient production of anti-JEV E antibodies . Comparison of the low and high dose immunization did not show statistically significant differences in anti-JEV E antibody production over this time period . The levels of anti-JEV E antibodies was enhanced after the boost performed on week 4 , and reached a plateau at least 1 . 5 month after the prime . When compared to the low dose , the high dose of TRIP/JEV . prME was more effective at eliciting a high level of specific antibody production ( P = 0 . 028 ) . As shown in the Fig 7B , the anti-JEV antibody titers induced 3 weeks after experimental infection of pigs with a single dose of live JEV were comparable to those stimulated in animals by a prime/boost immunization with 7 log10 TU of TRIP/JEV . prME lentiviral vector . The isotyping of anti-JEV E antibodies showed that TRIP/JEV . prME stimulated the production of both IgG1 and IgG2 by 2 weeks after the prime , and was followed by a decline at week 3 even at the high dose ( Fig 7C and 7D ) . The levels of both anti-JEV E IgG1 and IgG2 were similar to those observed in piglets challenged with JEV strain Nakayama at the week 3 of infection ( Fig 7E ) . In animals primed with TRIP/JEV . prME , the boost at week 4 enhanced preferentially the production of IgG2 by 10 weeks after the prime regardless of the inoculated dose . The individual serum samples obtained from animals immunized with the lentiviral TRIP/JEV . prME vector were also examined for neutralizing antibodies at 3 weeks after the prime and at 6 weeks after the boost ( Fig 8 ) . Immunized piglets that received a single dose of 6 to 7 log10 TU of TRIP/JEV . prME developed neutralizing antibody titers ranging from 10 to 30 against the homologous JEV G3 strain RP-9 and reached titers up to 160 after the boost ( Fig 8A ) . The higher dose of TRIP/JEV . prME induced a stronger anamnestic neutralizing antibody response . Examination of the piglet immune sera revealed that , regardless of the inoculated dose , TRIP/JEV . prME elicited neutralizing antibodies against the Nakayama strain of JEV G3 , the strain XZ0934 ( tested using the JEV G5/G3 chimera ) of JEV G5 and , to a lesser extent , the strain CNS769_Laos_2009 of JEV G1 ( Fig 8B ) . Importantly , the pattern of neutralizing activity of anti-TRIP/JEV . prME antibody was similar to that observed in immune sera collected from a group of piglets experimentally infected with the JEV strain Nakayama ( Fig 8C ) . These results showed that TRIP/JEV . prME is able to elicit high titers of neutralizing antibodies in piglets that received two inoculations with 7 log10 TU of lentiviral vector with an interval of one month . Additionally , we found that TRIP/JEV . prME is capable of stimulating the production of anti-JEV antibodies that neutralize JEV G1 and G5 .
Several vaccines against JEV are currently available to humans and for some animals such as horses and pigs: those are inactivated mouse brain-derived , inactivated cell culture-derived , live-attenuated and chimeric yellow fever virus-JEV vaccines [4 , 7 , 23 , 32–44] . However , some of them lack induction of long-term immunity and live-attenuated vaccine strains carry a possible risk of reversion to virulence [4] . Also the cost effectiveness of JEV vaccines is considered as a major obstacle [7] . DNA vaccines were shown to be as successful as commercial live-attenuated vaccines at eliciting anti-JEV immune responses in a mouse [25] and non-human primate models [44] . While those vaccines were designed to express the prM and E proteins , the fact that they were DNA vaccines required greater production costs . Lentiviral vectors represent a novel and attractive platform for gene-based immunization [20] . The ability of lentiviral vectors to efficiently transduce non-dividing dendritic cells allows a prolonged antigen presentation through the endogenous pathway , which in turns translates into the induction of strong , multi-epitopic and long lasting humoral as well as cellular immune responses . Consequently , an increasing number of pre-clinical studies show a great vaccine efficacy of lentiviral vectors in both infectious diseases and anti-tumor vaccination fields [19–21 , 31 , 45–47] . The purpose of our study was to evaluate the ability of two lentiviral TRIP-based vectors expressing the prM and E proteins from JEV , namely the TRIP/JEV . prME and TRIP/JEV . prMEΔTM vectors , at eliciting protective humoral immune response in mice and piglets . We showed that JEV VLPs accumulated in the supernatant of cells transduced with TRIP/JEV . prME but not TRIP/JEV . prMEΔTM . Thus TRIP/JEV . prME expresses recombinant prM and E that are secreted efficiently , which represents a supplementary asset in the ability at eliciting anti-JEV antibody response . Mice inoculated with a single low dose ( 5 log10TU ) of TRIP/JEV vectors developed JEV-specific IgGs and a booster dose one month after the prime resulted in a 40-fold increase in anti-JEV antibody titers . The reactivity of anti-JEV antibodies was documented by indirect ELISA and immunoblot assays . Mice immunized with TRIP/JEV . prME developed both anti-E and anti-prM antibodies whereas only anti-E antibodies were detected in sera of mice immunized with TRIP/JEV . prMEΔTM . The E protein acts as the main target for imparting protective immunity against JEV-related disease [23 , 32] , and its antigenic domain EDIII contains important epitopes that are recognized by neutralizing antibodies [48] . Further analysis of the recognition of JEV antigens by TRIP/JEV antisera showed that immunization with either TRIP/JEV . prME or TRIP/JEV . prMEΔTM generated comparable levels of antibodies against the E protein , as well as type-specific epitopes located in its antigenic EDIII domain . Immunization with TRIP/JEV vectors induced neutralizing antibodies against JEV belonging to genotypes , G1 , G3 , and G5 . Such observation is of particular importance at the period where it is evidenced that G1 JEV strains are replacing G3 strains [14] , from which most vaccines were designed . It is widely accepted that the humoral immune response is an essential component of protective immunity against JEV infection [23 , 24] . Consistent with the notion that VLPs are suitable as vaccine against arboviral disease including Japanese encephalitis [5 , 49] , TRIP/JEV . prME was the more efficient lentiviral vector in the production of neutralizing anti-JEV antibodies that conferred partial protection after their passive transfer in mice challenged with JEV . Inoculation of two doses of 7 log10 TU with a one-month of interval of TRIP/JEV . prME vector in piglets was highly efficient at eliciting high titers of anti-JEV neutralizing antibody that are potentially able to protect pigs from JEV infection . TRIP/JEV . prME was capable of stimulating the production of anti-JEV antibodies that neutralize JEV G3 and G5 , and , to a lesser extent , G1 . The potential impact of JEV genotype change on vaccine potency has been estimated and immune sera obtained from pigs injected with a G3 vaccine showed lower strain-specific cross-neutralizing antibody titers against JEV of G1 [36] . Such observation led to the development of new veterinary vaccines for pigs specifically directed against this particular genotype of JEV [43] . Although in our hands the TRIP/JEV . prME vector elicited neutralizing antibodies against a G1 virus in pigs , we did note that their levels were lower when compared to the other JEV genotypes tested . However , neutralizing antibodies titers against JEV of G1 could reach 1:40 , and thus could be sufficient to achieve protection in pigs . It could be nevertheless important to design alternate JEV antigen with a broader cross-reactivity against JEV strains of G1 . In this study , we demonstrated that immunization of pigs with a TRIP/JEV vector expressing JEV VLPs is particularly efficient at priming antigen-specific humoral immunity and triggers neutralizing antibody responses against the genotypes 1 , 3 , and 5 of JEV . The production of virus neutralizing antibodies is critical to protection against JEV infection in pigs [50] and a titer at least 1:10 is indicative of protective humoral immunity [51] . The titers of neutralizing antibodies elicited by the lentiviral TRIP/JEV . prME vector are sufficient to confer protection in domestic pigs against different genotypes of JEV and this could be of a great utility in endemic regions where more than one genotype circulates . | Japanese encephalitis virus is the etiologic agent of the most medically important viral encephalitis in South Asia with thousands of deaths per year . The virus is maintained in an enzootic cycle between Culex mosquitoes and amplifying vertebrate hosts , such as wild boars and pigs . Vaccination of domestic pigs has been suggested as a strategy to reduce viral disease transmission to humans , in line with the now-called “One Health” concept . Lentiviral gene transfer vectors represent a novel vaccination platform with an unprecedented ability to induce robust humoral immunity in various animal species . In our study , we demonstrated that immunization of pigs with a recombinant lentiviral vector expressing virus-like particles of Japanese encephalitis virus is particularly efficient at eliciting specific humoral immunity . The titers of neutralizing antibodies elicited by the lentiviral vector are sufficient to confer protection in domestic pigs against the different genotypes of Japanese encephalitis virus observed in Asia . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [] | 2015 | A Lentiviral Vector Expressing Japanese Encephalitis Virus-like Particles Elicits Broad Neutralizing Antibody Response in Pigs |
Somatic mutations drive the growth of tumor cells and are pivotal biomarkers for many cancer treatments . Genetic association analysis using somatic mutations is an effective approach to study the functional impact of somatic mutations . However , standard regression methods are not appropriate for somatic mutation association studies because somatic mutation calls often have non-ignorable false positive rate and/or false negative rate . While large scale association analysis using somatic mutations becomes feasible recently—thanks for the improvement of sequencing techniques and the reduction of sequencing cost—there is an urgent need for a new statistical method designed for somatic mutation association analysis . We propose such a method with computationally efficient software implementation: Somatic mutation Association test with Measurement Errors ( SAME ) . SAME accounts for somatic mutation calling uncertainty using a likelihood based approach . It can be used to assess the associations between continuous/dichotomous outcomes and individual mutations or gene-level mutations . Through simulation studies across a wide range of realistic scenarios , we show that SAME can significantly improve statistical power than the naive generalized linear model that ignores mutation calling uncertainty . Finally , using the data collected from The Cancer Genome Atlas ( TCGA ) project , we apply SAME to study the associations between somatic mutations and gene expression in 12 cancer types , as well as the associations between somatic mutations and colon cancer subtype defined by DNA methylation data . SAME recovered some interesting findings that were missed by the generalized linear model . In addition , we demonstrated that mutation-level and gene-level analyses are often more appropriate for oncogene and tumor-suppressor gene , respectively .
Somatic mutations play a central role in the development and progression of cancer . Associations between somatic mutations and molecular/clinical outcomes can provide important insights into cancer etiology or the mechanism of tumor growth , and potentially contribute to precision cancer therapy . Despite the functional importance of somatic mutations , few computational methods have been developed for association studies using somatic mutations . There are probably two reasons for this . First , since somatic mutation data are relatively new , most efforts were spent on bioinformatic challenges such as somatic mutation calling and functional annotations , e . g . , inference of driver mutations [1–3] , or estimation of cancer subtypes using somatic mutations [4 , 5] . Second , systematic studies of somatic mutations in large observational studies are not feasible until recently , thanks for the drop of sequencing cost and the improved capability to handle formalin-fixed paraffin-embedded ( FFPE ) tissue samples . While these challenges on sequencing tumor samples and calling mutations have been addressed , a limiting factor to harvest the rich information of somatic mutation associations is appropriate statistical methods for data analysis . A unique feature of somatic mutations , in contrast to germline mutations , is the difficulty to confidently call mutations from sequencing data . A major factor that contributes to this challenge is that a tumor sample is often a mixture of tumor cells and non-tumor cells ( e . g . , infiltrating immune cells ) and a somatic mutation may only occur in a subset of the tumor cells , known as intra-tumor heterogeneity [6] . Therefore the signals of a somatic mutation may be visible only in a small proportion of sequence reads , and it is challenging to separate such weak signals from sequencing errors or DNA damages caused by FFPE [7] . Another factor that limits mutation call availability/accuracy is low coverage of sequencing reads , particularly in whole genome sequencing data . Although many methods have been developed for somatic mutation calling [8–11] , there is no consensus on the best variant calling algorithm . The general recommendation is to take the intersection of the mutations called by a few methods , followed by additional filters [12 , 13] . Such a strategy reduces false positive rate , but at the price of inflated false negative rate . Therefore it is important to account for somatic mutation calling uncertainty in association studies . Such uncertainty of somatic mutation calling renders association methods for germline genetic variants inappropriate for somatic mutation associations . Generalized linear models are the most commonly used tools to assess germline genetic associations , for example , linear model for continuous traits and logistic regression for binary traits . Such methods do not account for mutation calling uncertainty . A few germline genetic association methods have been developed when the germline genomic features have inherent uncertainty , for example , for haplotype association [14 , 15] or for case-control associations with systematic difference between cases and controls [16] . However , these methods are designed for specific problems and are not applicable to somatic mutation association studies . A few earlier works have studied the associations between somatic mutations and gene expression using gene-level mutation [17] , by integrating gene-gene interaction networks [18] , or by a meta-analysis across multiple cancer types [19] . However , none of these works has considered the uncertainty of somatic mutation calling . In this paper , we propose a Somatic mutation Association test with Measurement Errors ( SAME ) , which accounts for somatic mutation calling uncertainty by modeling the true somatic mutation status as a latent variable and exploiting read count data to augment the mutation calls . We develop two versions of this test , one for mutation-level analysis using a single somatic mutation ( mSAME ) and the other one for gene-level analysis using multiple mutations within a gene ( gSAME ) . We have implemented SAME in an R package , and it is computationally efficient enough for genome-wide analysis . We evaluated the performance of SAME through extensive simulations and a real data application using the data from 12 cancer types of The Cancer Genome Atlas ( TCGA ) project . Our results demonstrated that SAME controls type I error and has improved statistical power compared to the competing methods that ignore somatic mutation calling uncertainty .
We first describe the mSAME test that works on a single somatic mutation . To simplify notations , we omit the index for somatic mutations in the following discussions . For a specific somatic mutation , we denote the mutation call and true mutation status in the i-th sample by Oi and Si , respectively , where 1 ≤ i ≤ n and n is sample size . Si equals to 1 if this mutation is present in the i-th sample , and 0 otherwise . The value of Oi depends on the read-depth information . Let the read-depth and the number of alternative reads of this mutation in the i-th sample be Di and Ai , respectively . A somatic mutation can be called only if there is enough coverage , i . e , Oi = 0 or 1 as mutation call indicator if Di ≥ d0 , and Oi is unobserved if Di < d0 , where d0 is a threshold used in the mutation calling process . Denote the outcome variable of the i-th sample by Yi and the set of additional covariates by xi . Let ρ0 = P ( Si = 0 ) and ρ1 = 1 − ρ0 = P ( Si = 1 ) , then the likelihood for the observed data can be written as L = ∏ i = 1 n ∑ j = 0 1 ρ j f Y , A , D , O ( Y i , A i , D i , O i | S i = j ) = ∏ i = 1 n ∑ j = 0 1 ρ j f Y ( Y i | S i = j ) f A , D , O ( A i , D i , O i | S i = j ) , ( 1 ) where fT denotes the density function for random variable T . We further assume that the conditional distribution of Yi given Si ( i . e . , fY ( Yi|Si = j ) in Eq ( 1 ) ) can be modeled by a generalized linear model with mean E ( Y i ) = g - 1 ( x i T α + S i β ) , ( 2 ) and a dispersion parameter ϕ , where g ( ⋅ ) is a link function , and α , β are the regression coefficients . We are interested in the association testing problem H0: β = 0 . For continuous outcomes , we can write fY ( Yi|Si ) as a normal density with the identity link function g ( t ) = t . For binary outcomes , we write fY ( Yi|Si ) as a Bernoulli density using the logit link function g ( t ) = log ( t/ ( 1 − t ) ) . For the distribution of read counts and observed mutation calls ( i . e . , fA , D , O ( Ai , Di , Oi|Si = j ) in Eq ( 1 ) ) , we use beta-binomial distributions to model allele-specific read counts Ai given Di , Oi and Si , and use a Bernoulli distribution to model Oi given Si . Beta-binomial distributions have been used to model allele-specific read counts from ChIP-seq [20] , RNA-seq [21] , DNA sequencing [22] , and somatic mutations [23 , 24] . The Bernoulli likelihood of observed somatic mutation calls given true somatic mutation status has been used to model somatic mutation calls from single cell DNA sequencing data [25 , 26] . These previous work have shown that these distributions are appropriate for real data . We have also compared the distributions of observed read counts versus expected ones from beta-binomial model fit and they agree very well ( Fig S1 in S1 Appendix ) . We denote the unknown parameters in the model by θ and the likelihood-ratio test statistic for the mSAME model is T = - 2 [ log L ( θ ^ 0 ; Y , A , D , O ) - log L ( θ ^ ; Y , A , D , O ) ] , ( 3 ) where θ ^ is the maximum likelihood estimator of θ in the whole parameter space , and θ ^ 0 is the maximum likelihood estimator of θ under H0: β = 0 . All the technical details for the likelihood function and parameter estimation can be found in Section 1 . 1-1 . 4 of S1 Appendix . Under H0 , the test statistic T asymptotically follows a Chi-square distribution with degree of freedom 1 , thus we can reject H0 if T > χ 1 2 ( 1 - ξ ) where χ 1 2 ( 1 - ξ ) is the ( 1 − ξ ) quantile of this Chi-square distribution . Next we discuss our gSAME model that aggregates the information of multiple somatic mutation loci within a gene ( or any arbitrarily defined unit ) for association testing . We start by defining some notations . Suppose that there are p mutation loci within a gene of interest , and we drop the index for gene for notational convenience . We use superscripts m and g to denote mutation-level and gene-level data , respectively . We denote the observed mutation calls for the i-th sample by O i m = { O i 1 m , ⋯ , O i p m } , the read-depth and the number of the alternative reads by D i m = { D i 1 m , ⋯ , D i p m } and A i m = { A i 1 m , ⋯ , A i p m } , respectively . Analogously , we denote the underlying true mutation status by S i m = { S i 1 m , ⋯ , S i p m } . We define the gene-level mutation status to be 1 if there is one or more mutations within this gene: S i g = { 1 if any S i j m = 1 , 0 if all S i j m = 0 . ( 4 ) The outcome variable Yi and the covariates xi are defined as before . In gene-level analysis , we model Yi as a function of S i g and xi . Then the likelihood function is L = ∏ i = 1 n ∑ j = 0 1 ρ j g f Y , A , D , O ( Y i , A i m , D i m , O i m | S i g = j ) = ∏ i = 1 n ∑ j = 0 1 ρ j g f Y ( Y i | S i g = j ) f A , D , O ( A i m , D i m , O i m | S i g = j ) , ( 5 ) where ρ 0 g = P ( S i g = 0 ) and ρ 1 g = 1 - ρ 0 g = P ( S i g = 1 ) . Since read count data ( i . e . , D i m and A i m ) and mutation calls ( O i m ) are collected for each mutated locus , their distributions are modeled given S i m . Then the remaining steps to complete this likelihood is to model S i m conditional on S i g . When S i g = 0 , it is clear that S i j m = 0 for all the p mutations . When S i g = 1 , S i m can have 2p − 1 possible values , which is computationally onerous to enumerate for large p . We notice that in practice , it is impossible to call a somatic mutation if the corresponding number of alternative reads equals to 0 . Hence to reduce computational burden , we assume that the j-th mutation may occur only if A i j m > 0 , otherwise we assign S i j m = 0 directly . Thus the number of the combinations is limited to 2 m i - 1 , where mi is the number of mutations with A i j m > 0 . Let θg be the unknown parameters in the gSAME model . The likelihood ratio test statistic of gSAME model for testing the effect of somatic mutation S i g is T = - 2 [ log L ( θ ^ 0 g ; Y , A m , D m , O m ) - log L ( θ ^ g ; Y , A m , D m , O m ) ] , ( 6 ) where θ ^ g is the estimator of θg in the whole parameter space , and θ ^ 0 g is the estimator of θg under H0 . All the technical details for the likelihood function and parameter estimation can be found in Section 1 . 5-1 . 6 of S1 Appendix .
We applied the proposed mSAME and gSAME methods as well as GLM to study the associations between somatic mutations and genome-wide gene expression in TCGA colon cancer patients . Briefly , we downloaded the bam files of exome-seq data for paired tumor-normal samples from NCI Genomic Data Commons ( GDC ) Data Portal . We called somatic mutations using the intersection of MuTect and Strelka , followed by the read-depth filter to keep those mutations with read-depth ≥20 in both tumor and paired-normal samples ( Section 3 . 1 in S3 Appendix ) . Colon cancer patients can be separated into two subtypes based on mutation load [27] . We classified a sample as hyper-mutated if it has more than 375 non-silent mutations and this cutoff is chosen to separate the two modes of the distribution of mutation load ( Fig S9 in S3 Appendix ) . Our analysis requires allele-specific read counts for each mutation across all samples . While collecting such information , we noticed that 24 samples have much smaller number of allele-specific read counts than the remaining samples and we removed them from our data analysis ( Section 3 . 3 in S3 Appendix ) . For gene expression data , we downloaded the . htseq . counts files from NCI GDC , which include the number of RNA-seq reads mapped to 60 , 483 genomic features . Most of these features are non-coding RNAs or pseudo genes that have zero or very small number of RNA-seq read counts across most tumor samples . We selected 17 , 986 genes that have at least 20 reads in more than 25% of the samples for the down-stream analysis . Let Tij be the read count for the j-th gene in the i-th sample . We correct for read-depth variation across samples using Tij/di , where di is the 75 percentile of gene expression within the i-th sample , a robust measurement of read-depth [28] . Then we quantified gene expression by log ( Tij/di ) , to make variation of gene expression similar across orders of expression levels [29] . We further regressed out copy number effect from gene expression data ( Section 3 . 4 in S3 Appendix ) . Since copy number measurement may be missing for some genes ( usually the genes around the beginning/end of a chromosome or around a centromere ) , we removed genes with missing copy number information , and ended up with 16 , 339 genes for the following analysis . Taking the intersection of the samples with somatic mutations and gene expression , we obtained 386 samples . We further included age , gender , and hyper-mutation status as covariates . We also removed those potential germline mutations by checking the read-depth data in the paired normal samples ( Section 3 . 5 in S3 Appendix ) . In the following analysis , we only studied non-silent mutations because silent mutations in exonic regions are most likely to be passenger mutations that do not have functional impact . To illustrate somatic mutation association analysis using dichotomous outcomes , we applied both mSAME and gSAME to identify somatic mutations associated with colon cancer subtypes defined by DNA methylation data . One of the most well known subtype of colon cancer is the hypermutation subtype [4 , 27] . By definition , it is associated with many somatic mutations and thus we used it as a covariate in all the analysis of this paper . Here we consider another subtype , defined by clustering analysis of genome-wide DNA methylation data [32] ( Fig S13 in S3 Appendix ) . See Section 3 . 8 in S3 Appendix for details of methylation data processing . We used this clustering results to classify the cancer patients into two groups and treated it as a binary outcome . Then we associated this subtype indicator with somatic mutations . Similar to the eQTL analysis , we performed mutation-level association analysis using mSAME and GLM ( logistic regression ) on 37 mutations that are present in at least 5 samples . At the significance level 0 . 05/37 ≈ 0 . 00135 , mSAME and GLM both detected one significant mutation of BRAF V600E , where mSAME yields a smaller p-value than GLM ( Table 2 ) . We also performed gene-level analysis by gSAME and GLM for the 180 gene-level mutations used in eQTL analysis . Both methods discovered two significant gene-level mutations: BRAF and KMT2C , using the p-value threshold 0 . 05/180 = 0 . 00028 . KMT2C is known as a tumor suppressor gene [33] . Our results suggest that the mutations of KMT2C are associated with DNA methylation , which is consistent with its role as histone methyltransferases because DNA methylation and histone methylation often work together to establish epigenetic landscape for gene expression regulation . Following the workflow of the eQTL analysis for TCGA Colon Adenocarcinoma ( COAD ) samples , we conducted eQTL analysis using somatic mutations for 11 other TCGA cancer types , including Bladder Urothelial Carcinoma ( BLCA ) , Brain Lower Grade Glioma ( LGG ) , Glioblastoma multiforme ( GBM ) , Head and Neck squamous cell carcinoma ( HNSC ) , Kidney renal clear cell carcinoma ( KIRC ) , Liver Hepatocellular Carcinoma ( LIHC ) , Lung adenocarcinoma ( LUAD ) , Lung squamous cell carcinoma ( LUSC ) , Ovarian serous cystadenocarcinoma ( OV ) , Skin Cutaneous Melanoma ( SKCM ) , and Stomach adenocarcinoma ( STAD ) . These cancer types are chosen due to their relatively large sample sizes and relatively higher rates of somatic mutations . We dowloaded the gene expression and somatic mutation data for association analysis from NCI GDC Data Portal , using the workflow of “HTSeq—Counts” for gene expression data , and the workflow of “MuTect2 Variant Aggregation and Masking” for somatic mutation data . For mutation-level association analysis , we selected the mutations that occur in at least 5 samples . For gene-level analysis , we selected the gene-level mutations that occur in at least 5% of the samples . For each mutated locus , we need read count data ( read depth and the number of alternative reads ) for all samples , regardless of mutation call status . For COAD analysis , we downloaded all the bam files to local server and then collected these counts . However , this approach is not feasible for pan-cancer study across 11 cancer types because downloading and storing all the bam files requires too many resources . Instead , we obtained the read-count data using the cloud service provided by The Seven Bridges Cancer Genomics Cloud [34] ( Section 3 . 9 of S3 Appendix ) . In all association studies , we included age and gender ( except for gender-specific cancer PRAD and OV ) as covariates . For LGG , we further adjusted for cancer subtype defined based on the IDH1 or IDH2 mutation and chromosome 1p and 19q co-deletion [35] . We recorded significant findings using genome-wide Bonferroni correction , and summarized the number of the significant findings by GLM or SAME in Table S9 in S3 Appendix ( Section 3 . 10 of S3 Appendix ) . The complete lists of the results are provided as supplementary text files . Examining the number of significant eQTLs for each mutation or each gene across cancer types shows no apparent pattern: most mutation-level or gene-level eQTLs are not shared across cancer types . However , one exception is gene-level TP53 mutation ( Fig 3 ) , which is among the significant eQTLs in 7 out of the 12 cancer types . This is partly due to the fact that TP53 is mutated with relatively high frequency across cancer types and it is a transcription factor that can directly regulate gene expression . When we relax the p-value cutoff to use transcriptome-wide significance ( i . e . , p-value cutoff = 0 . 05/# of genes ) , gene-level TP53 eQTLs were identified in 9 cancer types . In addition , several other gene-level eQTLs are shared among multiple cancer types ( Fig S14 in S3 Appendix ) . Overall the pattern of mutation/gene eQTLs shared across cancer types are similar between SAME and GLM ( Fig S15-S16 in S3 Appendix ) , though in general mSAME/gSAME identify more eQTLs than GLM . Next we examine the eGenes ( genes whose expression are associated with an eQTL ) associated with TP53 gene level mutation across cancer types . Since we focus on one mutation , we select the eGenes identified by transcriptome-wide significance . At this significance level , TP53 has no eGene by either gSAME or GLM in three cancer types: KIRC , LUSC and HNSC , and thus the following results only involve the remaining nine cancer types . We are interested in similarities of TP53 eGenes across cancer types . Towards this end , we examine the 50 eGenes identified in at least 3 cancer types by either gSAME ( 35 eGenes ) or GLM ( 46 eGenes ) , with an intersection of 31 genes identified by both methods ( Fig 4 ) . The difference of gSAME and GLM results are most due to potential mutation calling errors in TP53 ( Fig S17 in S3 Appendix ) . The protein product of TP53 , p53 , is a very well studied tumor suppressor and is involved in different biological processes such as cell cycle arrest , DNA repair , and apoptosis [36] . Many target genes of p53 have been reported [37] , and these target genes can be used to evaluate the relevance of the eGenes identified from our study . About 29% ( 10 out of 35 ) of the eGenes identified by gSAME and 20% ( 9 out of 46 ) identified by GLM are among 343 high confidence p53 target genes [37] ( Fig 4 ) . The only difference is gene CDKN1A ( encoding protein p21 ) where gSAME and GLM identified it as an eGene for three and two cancer types , respectively . CDKN1A is one of the most important targets of p53 and is requested for p53-mediated cell cycle arrest [37] . Visualization of the mutation status of these 50 genes show an interesting pattern: three cancer types , LUAD , LIHC and SKCM are clustered together since many genes are eGenes only in these three cancer types ( Fig 4 ) . The relatively larger number of eGenes in these cancer types can not be explained by the mutation frequency of TP53 ( Fig S18 in S3 Appendix ) or genome-wide somatic mutation load ( Fig S19 in S3 Appendix ) . None of these eGenes are among the 343 high confidence p53 target genes , suggesting that they may be indirectly regulated by p53 . Gene ontology analysis shows that these eGenes are enriched with genes involved in cell cycle related biological processes such as chromosome segregation . Therefore our results suggest that somatic mutation of TP53 may have similar functional roles in cell cycle control in LUAD , LIHC and SKCM .
Understanding the associations between somatic mutations and cancer-related traits is of fundamental importance for precision cancer therapy . In this paper , we present a statistically powerful and computationally efficient approach for association analysis of somatic mutations while accounting for measurement errors of somatic mutations . By modeling the calling uncertainty of the somatic mutations and including the read-depth data into our statistical model , the proposed SAME method can significantly improve the statistical power for the association analysis . The SAME method can accommodate both continuous and dichotomous outcomes , and it is applicable to both mutation-level and gene-level association testing . While we have demonstrated SAME using the publicly available exome-seq data , it will provide larger degree of power gain for whole genome sequencing studies where read depth are typically lower . One practical question of using our method is that how to choose between mutation-level ( mSAME ) versus gene-level ( gSAME ) analysis . Our eQTL analysis results suggest that mSAME may be more suitable for recurrent mutations in oncogene ( e . g . , the BRAF V600E mutation ) . This is because an oncogene is often activated by some specific “gain of function” mutations that drive tumor growth , and such driver mutations are often recurrent across patients . Other rare mutations in the same gene may be passenger mutations , even if they are non-silent ones . For example , BRAF harbors 10 non-silent mutations in TCGA colon cancer dataset . Except for the V600E mutation , the remaining 9 mutations only occur in one or two samples , and thus are likely passenger mutations . When collapsing both driver and passenger mutations to create a gene-level mutation , the mutation pattern may be “diluted” by those passenger mutations , and thus gene-level associations may yield less significant results than mutation-level associations . This is indeed the pattern observed when we compare the eQTL results for BRAF V600E mutation versus BRAF gene level mutations ( Fig 5 ) . In contrast , gSAME may be more suitable for tumor suppressor gene ( e . g . , TP53 ) . The function of a tumor suppressor gene may be perturbed by multiple “loss of function” mutations and thus there is no evolutionary pressure to select a specific one . Since all the loss of function mutations have similar functional consequence , gene-level association can have much larger power than mutation-level analysis . For example , TP53 has 68 individual mutations in TCGA colon cancer dataset , among which only 6 mutations occur at more than 2% of the samples and are significant eQTLs with transcriptome-wide multiple testing correction . For each gene expression trait , we take the minimum mutation-level p-value across these 6 mutations and compare it with gene-level p-value . In most cases , the gene level analysis yields stronger associations than mutation-level analysis ( Fig 5 ) . We have carefully implemented mSAME/gSAME to maximize computational efficiency , so that it is computationally feasible for genome-wide eQTL mapping . However , it still takes about 1-5 seconds per association testing . In contract , GLM is computationally much more efficient , taking about 0 . 01-0 . 02 seconds per association testing . Therefore , when there is limited mutation calling error ( e . g . , with high quality samples and high sequencing coverage ) one strategy to balance computational time and accuracy is to use GLM for a quick initial scan , and then apply mSAME/gSAME for a subset of associations at a relatively liberal p-value cutoff . In addition , gSAME will become computationally more inefficient for larger analysis units , such as several genes within a pathway . Further development is needed in such situations . In fact , simply collapsing individual mutations may not be a good strategy for pathway level association analysis and better strategies to summarize pathway level somatic mutations warrant further studies . Somatic mutation association is a new field with great potential to deliver key findings for precision cancer therapy . Accounting for somatic mutation calling uncertainty and low read-depth is an initial step to develop more rigorous and powerful association methods . We expect that more methods will be developed to exploit other types of information , such as intra-tumor heterogeneity or pathway level analysis where mutation information across genes is aggregated . | Cancer is a genetic disease that is driven by the accumulation of somatic mutations . Association studies using somatic mutations is a powerful approach to identify the potential impact of somatic mutations on molecular or clinical features . One challenge for such tasks is the non-ignorable somatic mutation calling errors . We have developed a statistical method to address this challenge and applied our method to study the gene expression traits associated with somatic mutations in 12 cancer types . Our results show that some somatic mutations affect gene expression in several cancer types . In particular , we show that the associations between gene expression traits and TP53 gene level mutation reveal some similarities across a few cancer types . | [
"Abstract",
"Introduction",
"Materials",
"and",
"methods",
"Results",
"Discussion"
] | [
"medicine",
"and",
"health",
"sciences",
"cancers",
"and",
"neoplasms",
"simulation",
"and",
"modeling",
"oncology",
"mutation",
"molecular",
"biology",
"techniques",
"mutagenesis",
"and",
"gene",
"deletion",
"techniques",
"epigenetics",
"dna",
"information",
"technology",
"data",
"processing",
"chromatin",
"dna",
"methylation",
"research",
"and",
"analysis",
"methods",
"computer",
"and",
"information",
"sciences",
"chromosome",
"biology",
"gene",
"expression",
"chromatin",
"modification",
"mutational",
"analysis",
"dna",
"modification",
"molecular",
"biology",
"somatic",
"mutation",
"biochemistry",
"colorectal",
"cancer",
"cell",
"biology",
"nucleic",
"acids",
"genetics",
"biology",
"and",
"life",
"sciences"
] | 2018 | Association analysis using somatic mutations |
Common causes of acute febrile illness in tropical countries have similar symptoms , which often mimic those of dengue . Accurate clinical diagnosis can be difficult without laboratory confirmation and disease burden is generally under-reported . Accurate , population-based , laboratory-confirmed incidence data on dengue and other causes of acute fever in dengue-endemic Asian countries are needed . This prospective , multicenter , active fever surveillance , cohort study was conducted in selected centers in Indonesia , Malaysia , Philippines , Thailand and Vietnam to determine the incidence density of acute febrile episodes ( ≥38°C for ≥2 days ) in 1 , 500 healthy children aged 2–14 years , followed for a mean 237 days . Causes of fever were assessed by testing acute and convalescent sera from febrile participants for dengue , chikungunya , hepatitis A , influenza A , leptospirosis , rickettsia , and Salmonella Typhi . Overall , 289 participants had acute fever , an incidence density of 33 . 6 per 100 person-years ( 95% CI: 30 . 0; 37 . 8 ) ; 57% were IgM-positive for at least one of these diseases . The most common causes of fever by IgM ELISA were chikungunya ( in 35 . 0% of in febrile participants ) and S . Typhi ( in 29 . 4% ) . The overall incidence density of dengue per 100 person-years was 3 . 4 by nonstructural protein 1 ( NS1 ) antigen positivity ( 95% CI: 2 . 4; 4 . 8 ) and 7 . 3 ( 95% CI: 5 . 7; 9 . 2 ) by serology . Dengue was diagnosed in 11 . 4% ( 95% CI: 8 . 0; 15 . 7 ) and 23 . 9% ( 95% CI: 19 . 1; 29 . 2 ) of febrile participants by NS1 positivity and serology , respectively . Of the febrile episodes not clinically diagnosed as dengue , 5 . 3% were dengue-positive by NS1 antigen testing and 16 . 0% were dengue-positive by serology . During the study period , the most common identified causes of pediatric acute febrile illness among the seven tested for were chikungunya , S . Typhi and dengue . Not all dengue cases were clinically diagnosed; laboratory confirmation is essential to refine disease burden estimates .
Undifferentiated febrile illnesses are common in children living in tropical areas of Asia . Common causes include dengue , malaria , leptospirosis , influenza A , Salmonella Typhi , rickettsia , Japanese encephalitis and chikungunya , [1]–[8] . The symptoms and differential diagnoses of these diseases are similar , often mimicking those of dengue and making accurate clinical diagnosis difficult without laboratory confirmation [1] , [9] . Reliable laboratory-confirmed diagnoses of acute febrile illness require a positive bacteriological/virological test such as culture results and PCR; serological confirmation of pathogen-specific antibodies ( immunoglobulin ( Ig ) M or a four-fold rise in IgG ) can also support such assessments . Dengue is caused by four serotypes ( DEN1–4 ) of the genus Flavivirus [10] . Transmitted by Aedes mosquitoes , it is one of the most widespread of the arthropod-borne viral diseases . It is a major public health concern because of the huge burden it exerts on populations , health systems and economies [11] . Asian-Pacific countries have more than 70% of the worldwide disease burden [12] , and in Indonesia and Thailand , dengue is one of the leading causes of hospitalization and death among children [13] . Although dengue prevention currently relies on mosquito control , vaccine candidates are under development [5] , [14] , [15] and the World Health Organization ( WHO ) has included dengue among its targets for the control of neglected tropical diseases during 2015–2020 [11] . However , the burden of dengue is generally under-reported in many Asian countries , because national surveillance systems ( where they exist ) are passive and/or based largely on clinical diagnosis without laboratory confirmation [16]–[19] . Thus , there is a need for accurate , population-based , laboratory-confirmed data on the incidence of dengue in high-risk populations . Determining the local etiology of acute febrile illness and the operational suitability of field sites in endemic regions is also important for the success of large-scale clinical trials of dengue vaccines . This prospective cohort study in children was therefore carried out in five dengue-endemic countries: Indonesia , Malaysia , Philippines , Thailand and Vietnam . Active surveillance for febrile illness was carried out in the cohort population to determine the incidence and proportion of acute febrile episodes that were caused by dengue , as well as by chikungunya , hepatitis A and influenza A viruses , leptospirosis , rickettsia , and S . Typhi . These non-dengue diseases were chosen because they are the more frequently reported causes of febrile illness caused by a single pathogenic organism at the study sites and because their differential diagnoses mimic that for dengue [9] .
The study protocol was approved by the site-specific Independent Ethics Committee or Institutional Review Board ( IRB ) ; namely the Committee of Medical Research Ethics , Faculty of Medicine , University of Indonesia; the Health Research Ethics Committee , Faculty of Medicine , University Padjadjaran/Dr Hasan Sadikin Hospital; Faculty of Medicine Udayana University/Sanglah Hospital Ethics Committee , Bali , Indonesia; the Medical Research & Ethics Committee , Ministry of Health Malaysia; Research Institute for Tropical Medicine IRB , Philippines; Chong Hua Hospital IRB and the Vicente Sotto Memorial Medical Center Ethics Committee , Cebu , Philippines; the Walter Reed Army Institute of Research IRB , US ( Kamphaeng Phet Hospital ) and the Philippines ( Cebu ) ; the Ethical Review Committee for Research in Human Subjects , Ministry of Public Health , Thailand; the Ethics Committee , Faculty of Tropical Medicine , Mahidol University; and the Biomedical Research Ethics Committee , Ministry of Health , Vietnam . The study was conducted in accordance with the Seoul revision of the Declaration of Helsinki as adopted by the concerned regulatory authorities , using Good Clinical Practice and International Conference on Harmonization guidelines . Before any procedure associated with the study was performed , parents/guardians provided written informed consent on behalf of all child participants . In addition , written consent was also obtained through separate assent forms from participants according to local Ethics Committee regulation requirements or according to the study sponsor's standard operating procedures in countries that do not have local requirements for assent forms ( Indonesia , Malaysia , Thailand: 7–14 years ; Philippines: 12–14 years; Vietnam: 8–11 years ) . Furthermore , participants aged 12–14 years in Vietnam also provided signed informed consent on the same consent form as their parents/guardians . The study was conducted at 10 main sites in five Asian countries . These included district , city and provincial government hospitals and institutions in highly dengue-endemic areas , and associated health centers ( satellite sites ) . This prospective , multicenter , active surveillance cohort study conducted in Indonesia , Malaysia , Philippines , Thailand and Vietnam involved 150 participants from each main site , who were aged 2–14 years on the day of enrollment and recruited between June and September 2010 . The study was conducted from June 2010 to July 2011 . Participants were recruited from the community , schools , health centers and/or private health clinics , depending on each study site's setting . Because one objective of this study was site preparation for a subsequent Phase III study of a dengue vaccine , eligibility criteria were established: namely , that participants had to be in good health with no history of chronic illness or immunodeficiency; able to attend scheduled visits and comply with study procedures; and had not received any vaccine in the 4 weeks preceding the day of enrollment ( except for pandemic influenza vaccination , which could be received >2 weeks before enrollment ) , nor were planning to receive any vaccine in the 4 weeks following enrollment . The active surveillance system was designed to detect all acute febrile episodes in the cohort . Participants' guardians were given a thermometer and shown how to measure axillary temperature . All participants or their guardians were contacted weekly to monitor the occurrence of acute febrile episodes . In the event of an episode , participants were asked to go to their designated healthcare center . To determine whether participants had to be contacted and followed up , school registers were monitored for absenteeism . All participants made two visits to the study site: an enrollment visit and a termination visit . Additional visits were required if acute febrile episodes occurred: an acute visit and a convalescent visit , at which blood samples were obtained . Causes of fever were assessed by testing acute and convalescent sera for dengue , chikungunya , hepatitis A , influenza A , leptospirosis , rickettsia , and S . Typhi using the same standardized commercial kits at all sites . Malaria was not tested for because , based on the investigators' experience , malaria was not a significant cause of pediatric acute febrile illness at these study sites . The study objectives were to identify acute febrile episodes among the cohort , and then to determine some of the specific causes of the acute fever in these febrile participants using a preset list of laboratory tests . A secondary objective was to evaluate operational infrastructure at these study sites in preparation for a Phase III study of a dengue vaccine [20] . The primary outcome measures were the proportion and incidences of acute febrile episodes , and which of the seven diseased tested for were their most common causes , based on the following case definitions . For every acute febrile episode , blood samples for acute sera were taken from the participants at the study site within 5 days after fever onset , and convalescent paired samples were obtained 7–14 days after acute sample collection for serological dengue tests , complete blood count ( including platelet count and hematocrit ) , and to assess for other causes of febrile illness . Dengue NS1 tests were performed only on acute sera . Clinical study information gathered at each study site was electronically reported by the study investigator or an authorized designee using an electronic case report form . The same commercial kits were used at each study site . The Platelia Dengue NS1 Ag kit ( Bio-Rad , USA ) was used according to the manufacturer's instructions to detect dengue NS1 antigen in acute serum samples by ELISA . The Dengue Virus IgM Capture DxSelect ELISA kit and the Dengue Virus IgG Capture DxSelect ELISA kit ( Focus Diagnostics , USA ) were used according to the manufacturer's instructions to detect dengue-specific IgM or IgG , respectively , in both acute and convalescent samples . Other causes of fever were assessed using commercial kits to detect leptospirosis ( Leptospirosis Indirect Hemagglutination ( IHA ) Test; Focus Diagnostics , USA ) ; rickettsia ( Rickettsia IFA IgG/IgM; Focus Diagnostics , USA ) ; hepatitis A ( Anti-HAV IgM ELISA; DIAsource ImmunoAssays S . A . , Belgium ) ; S . Typhi ( Salmonella Typhi IgM ELISA; Calbiotech Inc , USA ) ; chikungunya ( NovaLisa Chikungunya IgM μ-capture ELISA; NovaTec Immundiagnostica GmbH , Germany ) and influenza A ( NovaLisa Influenza Virus A IgM-ELISA; NovaTec Immundiagnostica GmbH , Germany ) in both acute and convalescent sera . These kits were provided to all the sites for reasons of availability , ease of use and consistency , even though they were not all gold standard tests . The sensitivity and specificity of these tests , as determined by the manufacturers , are shown in Table S1 . The sample size of 150 participants per site was not hypothesis-driven , and was based on an estimated proportion of acute febrile episodes in these locations of 24% , in accordance with the investigators' experiences . The incidence and proportion of acute febrile episodes and their causes were described for the study cohort by country and for all countries combined . The Clopper-Pearson method was used to calculate the 95% confidence interval ( CI ) for the proportions of acute febrile illness and dengue [22] . The incidence density of acute febrile illness and of the causes was calculated as: ( 2 ) The Rothman-Greenland method [23] was used to calculate the CI of incidence density . Statistical analyses were performed using SAS 9 . 1 software ( SAS Institute Inc . ) . Missing data were not imputed .
The study cohort included 1 , 500 eligible participants , of which 1 , 487 ( 99 . 1% ) participants ( 446 [99 . 1%] , 299 [99 . 7%] , 297 [99 . 0%] , 299 [99 . 7%] and 146 [97 . 3%] in Indonesia , Malaysia , Philippines , Thailand and Vietnam , respectively ) completed the study . Demographic characteristics of the participants are shown in Table 1 . Of the 13 participants ( 0 . 9% ) who did not complete the study , 12 withdrew voluntarily and one was withdrawn due to noncompliance with the protocol . The overall study duration was 294 days ( 9 . 8 months ) : 285 , 294 , 233 , 244 and 292 days in Indonesia , Malaysia , Philippines , Thailand and Vietnam , respectively . Participants were followed up for a mean of 237 days ( 7 . 9 months ) , ranging at the different study sites from 211 days ( at Cebu , Philippines ) to 277 days ( at My Tho , Vietnam ) . All the acute febrile participants presented to their identified healthcare facility for an acute visit . Overall , 96 . 5% presented within 5 days after fever onset ( one participant [0 . 4%] presented initially to a non-study site , so the acute sample was taken outside the 5-day timeframe ) and 96 . 9% returned within the designated period to have their convalescent blood sample drawn . The incidence density of acute fever overall was 33 . 6 ( 95% CI: 30 . 0; 37 . 8 ) per 100 person-years of follow-up , ranging from 20 . 8 in Malaysia to 40 . 5 in Indonesia ( Table 2 ) . Overall , 19 . 3% ( 289/1 , 500 ) of the cohort experienced at least one acute febrile episode during the study period ( Table 2 ) . A total of 374 acute febrile episodes occurred in 289 participants – 60 of these participants had two or more acute febrile episodes during the study period . Of these 60 participants , 20 reported three or more acute febrile episodes , three had four febrile episodes and one participant had five febrile episodes . A clinical diagnosis was reported for 98 . 9% of febrile episodes ( 370/374 ) . The five most frequently made clinical diagnoses using the Medical Dictionary for Regulatory Activities ( MedDRA ) preferred terms were: pharyngitis including nasopharyngitis: 124/374 acute febrile episodes ( 33% ) ; upper and lower respiratory tract infections including upper respiratory tract infections , pneumonia , bronchitis: 72/374 ( 19% ) ; tonsillitis including pharyngotonsillitis: 39/374 ( 10 . 5% ) ; viral infection excluding dengue: 37/374 ( 10% ) ; dengue: 34/374 ( 9 . 1% ) ; and gastroenteritis including diarrhea: 8/374 ( 2% ) . Because of the length of time it took to process the laboratory tests , clinical diagnoses were often made independently of the reported laboratory results within the study context , and acute febrile participants were managed according to local standard practice . A laboratory test result ( i . e . laboratory diagnosis ) for dengue , chikungunya , hepatitis A , influenza A , leptospirosis , rickettsia , and/or S . Typhi was reported for 95 . 7% ( 358/374 ) of febrile episodes ( 97 . 8% in Indonesia , 86 . 4% in Malaysia , 96 . 6% in the Philippines , 95 . 7% in Thailand and 97 . 1% in Vietnam ) . Overall , 57% of participants tested positive by IgM for one of these seven etiological agents . The overall incidence density of laboratory-confirmed dengue by NS1 antigen was 3 . 4 ( 95% CI: 2 . 4; 4 . 8 ) per 100 person-years , and of probable dengue by serology was 7 . 3 ( 95% CI: 5 . 7; 9 . 2 ) per 100 person-years ( Table 3 ) . The mean duration of fever at the time of blood sampling for NS1 testing was 2 days ( median 2 . 4 days ) . Of the 289 febrile participants , 11 . 4% ( 95% CI: 8 . 0; 15 . 7 ) had laboratory-confirmed dengue , while 23 . 9% ( 95% CI: 19 . 1; 29 . 2 ) had probable dengue ( Figure 1 ) . Discrepancies between clinical diagnosis of dengue and laboratory test findings were observed . As mentioned previously , a clinical diagnosis of dengue was made in 34 out of 374 febrile episodes ( 9 . 1% ) . Sixteen per cent ( 60/374 ) of acute febrile episodes were not clinically diagnosed as dengue but were supported by serology ( i . e . probable dengue ) , and 5 . 3% ( 20/374 ) were not clinically diagnosed but were supported by virological testing ( i . e . laboratory-confirmed dengue ) . By contrast , 5 . 9% of acute febrile episodes ( 22/374 ) were clinically diagnosed as dengue , but were not supported by the laboratory tests for NS1 and/or or IgM/IgG . Of the prespecified panel of non-dengue diseases for which sera from acute febrile participants were tested , only chikungunya and typhoid fever were laboratory-diagnosed by IgM positivity at incidence densities that were higher than that of dengue ( 10 . 8 [95% CI: 8 . 9; 13 . 1] and 9 . 1 [95% CI: 7 . 3; 11 . 2] per 100 person-years , respectively; Table 4 ) . Chikungunya virus IgM antibodies were detected in 35 . 0% of febrile participants , and typhoid fever IgM antibodies in 29 . 4% ( Figure 2 ) . Influenza , rickettsia and hepatitis A were less common causes of febrile illness than dengue . Three cases of leptospirosis were detected by hemagglutination ( one in the Philippines and two in Thailand ) . However , this testing method does not allow one to distinguish between IgM and IgG antibodies , and therefore it could not be confirmed whether leptospirosis was the cause of these acute febrile episodes , or whether IgG antibodies remained from a previous infection . Amongst the 218 febrile participants who tested negative for dengue , there were 82 laboratory-diagnosed cases of chikungunya ( i . e . 28 . 4% of all febrile participants ) , 65 cases of typhoid fever ( 22 . 5% ) , 21 cases of influenza A ( 7 . 3% ) , 10 cases of rickettsia ( 3 . 5% ) and 4 cases of hepatitis A ( 1 . 4% ) . Among the 71 febrile participants who tested positive for dengue by NS1 antigen and/or serology , there were 17 laboratory-diagnosed cases each of chikungunya and typhoid fever ( each 5 . 9% of all febrile participants ) , 13 cases of influenza A ( 4 . 5% ) and 7 cases of rickettsia ( 2 . 4% ) .
This is the first prospective , multinational , active surveillance study with a focus on acute febrile illness to include these five dengue-endemic Asian countries . The overall incidence density of acute febrile illness was 33 . 6 per 100 person-years , with 19 . 3% of the 1 , 500 children experiencing at least one episode . This proportion is close to the estimated rate of 24% that was anticipated at the time of the study design , based on unpublished national and regional reports . Overall , 57% of participants with acute febrile illness tested positive by IgM for one of the seven etiological agents included in the predetermined panel of commercial tests . The incidence density of dengue was 3 . 4 per 100 person-years according to NS1 antigen positivity ( 11 . 4% of febrile participants ) and 7 . 3 by serology ( 23 . 9% of febrile participants ) , which confirms the high dengue endemicity in these countries . Where these findings differed from those reported previously for other dengue surveillance studies in this region , it was most likely because dengue incidences can vary from year to year , even at the same site [2] , [5] , [7] , [17] , [24] , [25] . For example , in an active surveillance study of acute febrile illness among school children in Ratchaburi , Thailand from 2006 to 2009 , dengue ( confirmed by IgM/IgG ELISA ) caused 6 . 74% of all acute febrile illnesses during the total study period , and the incidence ranged from 1 . 77% in 2006 to 5 . 74% in 2008 [5] . Another active surveillance study among children in Ratchaburi and Kamphaeng Phet reported dengue incidences of 23–25/1 , 000 in 2006–2007 using IgM/IgG and/or RT-PCR or virus isolation [17] . Incidence rates of dengue ranged from 16 . 9 to 38 . 6 per 1 , 000 person-years following active surveillance of 2–15-year-olds in Long Xuyen , Vietnam from 2003 to 2007 [25] . In the Philippines , a surveillance study conducted in San Pablo from 2007 to 2009 showed that 11% of acute febrile illnesses in infants were caused by dengue [26] . The proportion of dengue cases confirmed by serology was greater than those confirmed by NS1 antigen . Several explanations could account for this . Sensitivity of NS1 testing is directly related to the viral load , and thus the time since the start of viral replication: the mean duration of fever at the time of sampling was 2 . 4 days . Although the NS1 antigen test used in this study to test for laboratory-confirmed dengue has high specificity [18] , [27] , DENV-2 infections have been associated with significantly lower plasma NS1 levels relative to DENV-1 or DENV-3 infections [28] . In addition , a lower sensitivity for this test has been reported for secondary infections and , in dengue-endemic areas ( such as those where this study was conducted ) , a higher proportion of patients have secondary infections [27] . Differences in primary versus secondary dengue infections and non-dengue Flavivirus prevalences may have influenced the serology findings as follows . Firstly , IgM antibodies for dengue can remain elevated for 2–3 months after infection [29] , and positive IgM results could have been recorded for samples where an infection occurred 2–3 months before acute sample collection . Secondly , serological kinetics differ between primary and secondary dengue infections [30] . In secondary dengue infections , IgM levels are substantially lower than in primary infections [30] , while IgG antibody titers are detectable even in the acute phase , and rise rapidly and cross-react broadly with other Flaviviruses [29] . The differences in findings between methods may thus be due in part to variabilities in dengue epidemiology , seasonality and serotype prevalence between the study countries [31] . However , this effect could not be assessed directly because of this study's limitations that neither seroprevalence ( i . e . dengue serotypes ) nor primary versus secondary infections were determined . That chikungunya was the most commonly detected infection of those selected for evaluation in this study , occurring in a relatively high proportion ( 35 . 0% ) of acute febrile participants with an incidence density of 10 . 8 per 100 person-years , was rather unexpected . However , even allowing for the possibility that chikungunya-specific IgM levels can remain elevated for up to 60 days [32] , these findings are consistent with reports that chikungunya has re-emerged as an important infection in Asia [8] , [33]–[35] . The high incidence density of chikungunya observed in the Philippines ( 21 . 3 ) and Vietnam ( 18 . 5 per 100 person-years ) , which contributed to the high incidence observed overall , are consistent with local news and internet reports that chikungunya cases increased before or during the study period in these countries ( unpublished ) . Furthermore , a surveillance study of chikungunya in children aged <15 years , conducted in southern Vietnam in 2010 in a region that included our study site , showed that 0–33 . 3% of sera samples were positive for chikungunya at different sites [36] . Two chikungunya outbreaks were also reported in the region during the study period: one in China from September to October 2010 [37] , and 1 , 500 cases of chikungunya were reported in Cambodia between May 2011 and June 2012 [38] . Although studies that track the incidence of chikungunya during outbreaks have been carried out in southern Thailand [8] , [35] , [39] and Indonesia [40] , this is the first report to describe the incidence of chikungunya in these five countries following active surveillance . Typhoid fever was the second most commonly detected infection overall , being identified in 29 . 4% of febrile participants at an incidence density of 9 . 1 per 100 person years . This finding must be viewed in the context of the commercial test's limited sensitivity and specificity ( discussed further below ) . Nevertheless , our findings suggest that implementing routine tests for chikungunya ( namely , antigen detection ) and typhoid fever ( culture ) in these countries would increase the accuracy of diagnosis of undifferentiated acute febrile illness in children . Influenza vaccination is mainly used in the private market in these countries , hence the rates of influenza vaccination were most likely low in the study population ( which was defined as healthy children at the time of enrolment ) , although these data were not collected from participants . Although the commercial kits used for the pre-specified panel of non-dengue diseases were not the gold standard tests for some of these infectious agents , they were used because standardization was essential in this multicountry study and they provided ease of use . To eliminate confounding factors where positive serology results could have been indicative of previous infections or cross-reactivity , we reported only the IgM data for non-dengue causes of acute infections because IgG data ( and thus IgM/IgG ratios ) were not available for all the causative infectious agents . Nevertheless , we acknowledge the limitations of using these test results to calculate the incidences of these causative agents without correlating the findings with additional more stringent tests . The chikungunya test that was used has ≥95% sensitivity and specificity and , according to the manufacturer's specifications , does not cross-react with dengue antibodies ( Table S1 ) . However , test performances based on manufacturers' data may be higher than those based on published data , which generally include many more samples per study . Hence , it is possible that testing in the absence of correlation with clinical diagnoses could have led to an overestimation of the true disease incidence of chikungunya , although internet-based news bulletins such as the Program for Monitoring Emerging Diseases ( ProMED-mail , a programme of the International Society for Infectious Diseases ) confirm that chikungunya cases are increasing in these countries . The limitations of rapid testing for S . Typhi using currently available commercial tests have been well debated in the literature [3] , [41]–[43] . Blood culture from blood or bone marrow and microbiological characterization are the gold standard of enteric fever diagnosis , yet even these methods are only positive in 40%–60% of presumptive cases [41] . Despite the increased sensitivity of bone marrow culture , obtaining bone marrow by standard methods is technically challenging , invasive and not generally performed . In view of the importance of this pathogen in the highly endemic region of Southeast Asia [3] , [44] , we opted to use a commercialized test in order to try to capture as much information on S . Typhi infection as possible . The sensitivity and specificity of the S . Typhi IgM ELISA test we used were calculated from manufacturer's product information to be 86% and 96% , respectively ( Table S1 ) . In addition , a larger proportion of S . Typhi-positive participants than those positive for the other agents were IgM-positive for other causative organisms . Our findings should thus be viewed in the context of these factors and are likely to be an overestimation of the true incidence of S . Typhi as a causative agent of acute febrile disease . Discrepancies between clinical diagnoses and laboratory confirmation of dengue infection were observed . Dengue infections were clinically underdiagnosed: 16 . 0% of acute febrile episodes that were not clinically diagnosed as dengue were later supported by positive dengue serology , and 5 . 3% were confirmed by NS1 antigen testing . Clinical misdiagnoses were also made: 5 . 6% of all acute febrile episodes were clinically diagnosed as dengue but were not laboratory-confirmed . These findings confirm previous reports from this region [4] , [16] that clinical diagnosis of dengue has a limited predictive value and that laboratory analysis supports a more accurate assessment in differentiating causes of fever . Such discrepancies impact the accuracy of disease burden estimates . A study limitation is that the disease incidences that we report here are based on laboratory test findings alone , and may represent a misestimation of the true incidence of some of these etiological causes of febrile illness , in the context of the specificity and sensitivity of each of these respective tests . Clinical diagnosis was not taken into account when calculating disease incidence , except for dengue . Nevertheless , a strength of this study was its prospective cohort design involving intensive active surveillance to capture cases that might not otherwise have been detected on the basis of symptoms alone . Acting to ‘correct’ these incidences by excluding participants with atypical presentation would have compromised this study strength . The limited sample size ( 150 children per center ) and the relatively short duration of less than 1 year is another study limitation . Nevertheless , given that a Phase III efficacy study of a dengue vaccine commenced at these sites immediately following this study ( ClinicalTrials . gov NCT01373821 ) , these data provided valuable information about the baseline incidence of acute febrile illness and dengue . The study also met its secondary objective of showing that all these sites were capable of capturing and following up acute febrile episodes within a specific timeframe among a well-defined cohort , which lends additional validity to the data presented here . In conclusion , active fever surveillance showed that of the seven diseases for which we tested , the most common causes of pediatric acute febrile illness in these countries were chikungunya , S . Typhi and dengue . Clinical diagnosis was not sufficient to detect all dengue cases , and laboratory confirmation is essential to refine disease burden estimates of dengue and other common causes of acute febrile illness in children . These findings are of relevance to researchers planning clinical studies of vaccines against these infectious agents in Southeast Asia . | Acute febrile episodes are common in children living in tropical countries . Diagnosis can be challenging because symptoms of the more common infectious causes are similar and often mimic those of dengue . Asia Pacific has over 70% of the worldwide dengue disease burden , although dengue incidence is generally underestimated because most surveillance systems are passive or based on clinical diagnosis without laboratory confirmation . Understanding the local etiology of febrile illness and the incidence of dengue is important when planning large-scale vaccine trials . This prospective , active fever surveillance , cohort study was carried out in children in five dengue-endemic Asian countries – Indonesia , Malaysia , Philippines , Thailand and Vietnam – during 2010–2011 . Acute febrile episodes occurred in 289 ( 19 . 3% ) of the cohort of 1 , 500 children . Among the diseases for which antibodies were tested using commercial kits , the top three causes of acute fever were chikungunya , Salmonella Typhi and dengue , followed by influenza A , rickettsia and hepatitis A . Dengue was confirmed in 11 . 4% of the febrile children by viral protein detection and in 23 . 9% by serology . Clinical diagnosis was not sufficient to detect all dengue cases . These findings are of relevance to those planning clinical studies of vaccines against these infectious agents in Southeast Asia . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"medicine",
"bacterial",
"diseases",
"disease",
"mapping",
"infectious",
"diseases",
"survey",
"methods",
"clinical",
"epidemiology",
"tropical",
"diseases",
"(non-neglected)",
"infectious",
"disease",
"epidemiology",
"epidemiology",
"dengue",
"fever",
"flavivirus",
"influenza",
"neglected",
"tropical",
"diseases",
"viral",
"diseases",
"salmonella"
] | 2013 | Dengue and Other Common Causes of Acute Febrile Illness in Asia: An Active Surveillance Study in Children |
The Ashkenazi Jewish ( AJ ) population is important in genetics due to its high rate of Mendelian disorders . AJ appeared in Europe in the 10th century , and their ancestry is thought to comprise European ( EU ) and Middle-Eastern ( ME ) components . However , both the time and place of admixture are subject to debate . Here , we attempt to characterize the AJ admixture history using a careful application of new and existing methods on a large AJ sample . Our main approach was based on local ancestry inference , in which we first classified each AJ genomic segment as EU or ME , and then compared allele frequencies along the EU segments to those of different EU populations . The contribution of each EU source was also estimated using GLOBETROTTER and haplotype sharing . The time of admixture was inferred based on multiple statistics , including ME segment lengths , the total EU ancestry per chromosome , and the correlation of ancestries along the chromosome . The major source of EU ancestry in AJ was found to be Southern Europe ( ≈60–80% of EU ancestry ) , with the rest being likely Eastern European . The inferred admixture time was ≈30 generations ago , but multiple lines of evidence suggest that it represents an average over two or more events , pre- and post-dating the founder event experienced by AJ in late medieval times . The time of the pre-bottleneck admixture event , which was likely Southern European , was estimated to ≈25–50 generations ago .
Ashkenazi Jews ( AJ ) , numbering approximately 10 million worldwide [1] , are individuals of Jewish ancestry with a recent origin in Eastern Europe [2] . The first individuals to identify as Ashkenazi appeared in Northern France and the Rhineland ( Germany ) around the 10th century [3] . Three centuries later , Ashkenazi communities emerged in Poland , but the source ( s ) of migration are not completely clear . The Ashkenazi communities in Poland have grown rapidly , reaching , by the 20th century , millions in size and a wide geographic spread across Europe [2] . Due to the relative scarcity of relevant historical records , the ethnic origins of present-day Ashkenazi Jews are debated [2] , and in such a setting , genetic data provides crucial information . A number of recent studies have shown that Ashkenazi individuals have genetic ancestry intermediate between European ( EU ) and Middle-Eastern ( ME ) sources [4–8] , consistent with the long-held theory of a Levantine origin followed by partial assimilation in Europe . The estimated amount of accumulated EU gene flow varied across studies , with the most recent ones , employing genome-wide data , converging to a contribution of around 50% of the AJ ancestry [4 , 7 , 9] . Despite these advances , little is known about the identity of the European admixing population ( s ) and the time of the admixture events [2 , 10] . Speculations abound , due to the wide geographic dispersion of the Jewish populations since medieval times , but with very few historical records to support any claim [2] . Further complicating the picture is an Ashkenazi-specific founder event that has taken place less than a millennium ago , as manifested by elevated frequencies of disease mutations [11 , 12] , reduced genetic diversity [13 , 14] , and an abundance of long tracts of identity-by-descent [9 , 15 , 16] . Results from our recent study [9] were not decisive regarding the relative times of the European admixture and the founder event , calling for a more in-depth investigation . A number of previous population genetic studies have attempted , sometimes implicitly , to “localize” the Ashkenazi genomes to a single geographic region or source population [4–6 , 17] . However , such approaches may be confounded by the mixed EU and ME Ashkenazi ancestry , which necessarily implies the existence of multiple sources . Here , we overcome this obstacle , following studies in other populations [18 , 19] , by performing a preliminary step of local ancestry inference ( LAI ) , in which each locus in each Ashkenazi genome is assigned as either EU or ME . Following LAI , the source population of the European and Middle-Eastern “sub-genomes” can be independently localized . We begin our analysis by testing the ability of available LAI software to correctly infer ancestries for simulated EU/ME genomes . Proceeding with RFMix , we apply LAI to Ashkenazi SNP array data , and use a maximum likelihood approach to localize , separately , the EU and ME sources . We correct bias introduced by the method using simulations , and show that it is robust to potential errors in LAI . We also employ other methods based on allele frequency divergence between Ashkenazi Jews and other populations , although they turn out to be less informative . To estimate the time of admixture , we first use the lengths of EU and ME tracts and the decay in ancestry correlation along the genome . We further introduce a new method for dating admixture times based the genome-wide EU or ME ancestry proportions . We again remove bias from all methods using simulations . We integrate these results with an analysis of identity-by-descent ( IBD ) sharing both within AJ and between AJ and other populations . Finally , we compare our estimates to those produced by the GLOBETROTTER suite [20–22] . Our results suggest that the European gene flow was predominantly Southern European ( ≈60–80% ) , with the remaining contribution either from Western or ( more likely ) Eastern Europe . The time of admixture , under a model of a single event , was estimated at ≈30 generations ago . However , this admixture time is likely the average of at least two distinct events . We propose that admixture with Southern Europeans pre-dated the late medieval founder event , whereas the admixture event in Eastern Europe was more recent .
SNP arrays for Ashkenazi Jewish individuals were available from the schizophrenia study reported by Lencz et al . , 2013 [23] ( see also [24] ) . SNP arrays for European and Middle-Eastern populations were collected from several sources ( Table 1 ) . All genotypes were uniformly cleaned , merged , and phased ( Methods ) , resulting in 2540 AJ , 543 Europeans , and 293 Middle-Easterners genotyped at 252 , 358 SNPs . Note that while there are additional studies in these populations , we restricted ourselves to ( publicly available ) Illumina array data to guarantee a sufficient number of remaining SNPs after merging all datasets . We divided the European genomes into four regions: Iberia , North-Western Europe ( henceforth Western Europe ) , Eastern Europe , and Southern Europe ( Italy and Greece ) . The Middle-Eastern genomes were divided into three regions: Levant , Southern Middle-East , and Druze . See Table 1 for further details and S1 Fig for a PCA plot [25] supporting the partition into the indicated regions . A number of recent studies have shown that sharing of identical-by-descent ( IBD ) segments is abundant in the AJ population , and is likely due to a severe bottleneck around 30 generations ago [4 , 7 , 9 , 15 , 16] . An open question is the relative timing of the bottleneck and the European gene flow , with our current and past [9] point estimates dating admixture at around or slightly earlier than the bottleneck . Given that most IBD segments but the very long ones ( e . g . , of length >7cM ) coalesce around the time of the bottleneck , we contrast three hypotheses . If admixture completely predated the bottleneck , then IBD segments should have the same EU/ME ancestry proportions as observed genome-wide . If European admixture completely post-dated the bottleneck , then IBD segments should show exclusive ME ancestry . If , on the other hand , European gene flow occurred both before and after the bottleneck , then IBD segments should show an elevated ( though not exclusive ) ME ancestry compared to the rest of the genome [41–43] . Further , IBD segments of different lengths shared between AJ and other populations could shed light on the geographic origin of each admixture event . We detected long ( >3cM ) IBD segments using Germline [44] and Haploscore [45] ( Methods ) . For segments shared within AJ individuals , we then computed the total amount of genetic material in IBD segments associated with each pair of diploid ancestries , namely , the fraction of SNPs in IBD segments where each of the two individuals sharing the segment has either homozygous EU ancestry , homozygous ME ancestry , or heterozygous ancestry . Clearly , errors in IBD segment detection and local ancestry inference could severely bias the conclusions of such an analysis . Fortunately , we could naturally account for these errors using the observed amount of genetic material in IBD segments shared between individuals labeled homozygous ME and homozygous EU , since the proportion of such segments is a direct measure of the noise level ( Methods and S1 Text section 4 ) . Our results demonstrate an over-representation of Middle-Eastern IBD segments , consistent with two waves of gene flow . Specifically , we estimated the European fraction of the AJ ancestry at the bottleneck as 42% , less than the 53% observed genome-wide ( Methods ) . The contribution of post-bottleneck European gene flow required to explain these figures is 19% of the AJ ancestry ( Methods ) . Considering only segments of length between [3 , 7]cM ( as longer segments may descend from ancestors even more recent than the bottleneck ) slightly increased the inferred magnitude of post-bottleneck gene flow to 22% , or 23% when considering only segments between [3 , 4]cM . Given a history of multiple admixture events , a natural question is the geographic source of each event . According to the documented AJ migration history , we speculated that the Southern-European gene flow was pre-bottleneck and that the Western/Eastern European contribution came later . Indeed , we note that the estimated proportion of ≈20% post-bottleneck replacement is close to our above estimate of ≈16% EU gene flow from sources other than Southern-EU as well as to TreeMix’s and Globetrotter’s results below ( and perhaps also with our previous estimate of ≈15% EU ancestry based on AJ and Western European ( CEU ) data alone [46] ) . To test this hypothesis , we considered the European ancestry of IBD segments longer than 15cM , which are highly unlikely to predate the bottleneck . The proportion of AJ chromosomes with all regions masked but the >15cM IBD segments inferred by our geographic localization pipeline to be most likely Southern European decreased by 14 . 8% points compared to the genome-wide results . In contrast , the proportion of AJ chromosomes inferred to be most likely Eastern and Western European increased by 10 . 2 and 4 . 5% points , respectively . As a control , when we considered AJ individuals reduced to IBD segments of any length , there was no noticeable change . We also considered IBD segments shared between AJ and other populations ( Fig 5 ) , and observed that the number of segments shared between AJ and Eastern Europeans was ≈6-fold higher than shared between AJ and Southern Europeans ( consistent with [5] ) , with this ratio increasing to ≈60-fold for segments of more recent origin ( length >7cM ) . Further , the number of segments shared with Eastern Europeans was ≈2-fold higher than with Western Europeans or the people of Iberia ( P = 5∙10−3 for the difference , using permutations of the EU regional labels ) , pointing to Eastern Europe as the predominant source of recent gene flow . We have so far provided multiple estimates for the ancestry proportions from each source and the time of admixture events . We now attempt to bring these estimates together into a single model and provide bounds on the model’s parameters . The results of all analyses ( at least once examined in the light of simulations ) point to Southern Europe as the European source with the largest contribution . At the same time , relatively large contributions from Western and/or Eastern Europe were also detected , with some analyses ( IBD within AJ and between AJ and other sources , and GLOBETROTTER ) showing stronger support for an Eastern European source . Based on historical plausibility , these admixture events must have happened at different times , implying multiple events . The inferred admixture time , when modeled as a single event , was between 24–37 generations ago across the methods we examined ( corrected mean segment length and ancestry proportions , Alder , and GLOBETROTTER ) , very close to the time of the AJ bottleneck , previously estimated to ≈25–35 generations ago [9 , 16] . Therefore , it is plausible to argue that one admixture event occurred before or early during the bottleneck , while the other event happened after the bottleneck , with the IBD analysis suggesting that the more recent admixture was with Eastern Europeans . Based on these arguments , we propose that a minimal model for the AJ admixture history should include substantial pre-bottleneck admixture with Southern Europeans , followed by post-bottleneck admixture on a smaller scale with Western or ( more likely ) Eastern Europeans . The estimates for the total European ancestry in AJ range from ≈49% using our previous whole-genome sequencing analysis [9] , to ≈53% using the LAI analysis here , and ≈67% using the calibrated Globetrotter analysis . The proportion of Western/Eastern European ancestry was estimated between ≈15% ( Globetrotter and the LAI-based localization method ) , and , if identified as the source of the post-bottleneck admixture , 23% ( the IBD analysis ) . Therefore , the proportion of the Southern European ( presumably pre-bottleneck ) ancestry in AJ is between ≈26% to ≈52% , corresponding to [34 , 61]% ancestry at the time of the early admixture . Given these bounds , along with the admixture time estimate based on a single event ( 24–37 generations ago ) , we derived a constraint on the admixture times of the pre- and post-bottleneck events ( Methods ) . We further assumed that post-bottleneck admixture happened at most 20 generations ago , when the effective population size has already recovered from the bottleneck ( since our estimate of the post-bottleneck admixture proportions relied on the part of the genome not shared IBD; see the IBD analysis above and Methods ) . Finally , we assumed that post-bottleneck admixture happened no more recently than 10 generations ago , since no mass admixture events are known in the past 2–3 centuries of AJ history [52] . The results ( Fig 6 ) show that given these constraints , the pre-bottleneck admixture time is between 24–49 generations ago . Our proposed model is shown in Fig 7 .
The ethnic origins of Ashkenazi Jews have fascinated researchers for over a century [53 , 54] . The availability of dense genotypes for hundreds of AJ individuals , along with the development of new analysis tools , demonstrated genetic relatedness between AJ and other Jewish groups , and suggested Europe and the Middle-East as putative ancestral sources [4–8 , 24] . Here we attempted , for the first time , to create a detailed portrait of the admixture events experienced by AJ during their dwelling in Europe . To this end , we used previously generated genome-wide array data for AJ , European , and Middle-Eastern populations ( Table 1 ) , as well as a variety of current and newly developed population genetics methods . Before discussing the historical implications of our results , we point out two general lessons that emerge from the analysis . The first is that AJ genetics defies simple demographic theories . Hypotheses such as a wholly Khazar , Turkish , or Middle-Eastern origin have been disqualified [4–7 , 17 , 55] , but even a model of a single Middle-Eastern and European admixture event cannot account for all of our observations . The actual admixture history might have been highly complex , including multiple geographic sources and admixture events . Moreover , due to the genetic similarity and complex history of the European populations involved ( particularly in Southern Europe [51] ) , the multiple paths of AJ migration across Europe [10] , and the strong genetic drift experienced by AJ in the late Middle Ages [9 , 16] , there seems to be a limit on the resolution to which the AJ admixture history can be reconstructed . The second lesson is the importance of evaluating the results of off-the-shelf tools using simulations when studying closely related populations . When simulating relatively old ( ≈1k years ago ) Middle-Eastern and European admixture ( particularly Southern European ) , we found many tools to be of limited utility ( see , e . g . , the section on Alder , f-statistics , and TreeMix and S1 Text sections 1 and 2 on LAMP and PCAMask ) . Further , while we eventually were able to extract useful information off RFMix’s local ancestries , the raw results were not very accurate: the accuracy per SNP was only ≈70% , the mean segment length was more than twice than expected , and the variance of the ancestry proportion per chromosome was overestimated . When jointly analyzing LAI and IBD sharing , the inferred proportion of IBD segments that were either not IBD or had a random ancestry assignment was as high as ≈35% ( ( 1-λ ) in Methods ) , although fortunately , we were able to account for that in our model . We note , though , that problems of this nature are not expected for recent admixture events between more diverged populations . Our model of the AJ admixture history is presented in Fig 7 . Under our model , admixture in Europe first happened in Southern Europe , and was followed by a founder event and a minor admixture event ( likely ) in Eastern Europe . Admixture in Southern Europe possibly occurred in Italy , given the continued presence of Jews there and the proposed Italian source of the early Rhineland Ashkenazi communities [3] . What is perhaps surprising is the timing of the Southern European admixture to ≈24–49 generations ago , since Jews are known to have resided in Italy already since antiquity . This result would imply no gene flow between Jews and local Italian populations almost until the turn of the millennium , either due to endogamy , or because the group that eventually gave rise to contemporary Ashkenazi Jews did not reside in Southern Europe until that time . More detailed and/or alternative interpretations are left for future studies . Recent admixture in Northern Europe ( Western or Eastern ) is consistent with the presence of Ashkenazi Jews in the Rhineland since the 10th century and in Poland since the 13th century . Evidence from the IBD analysis suggests that Eastern European admixture is more likely; however , the results are not decisive . An open question in AJ history is the source of migration to Poland in late Medieval times; various speculations have been proposed , including Western and Central Europe [2 , 10] . The uncertainty on whether gene flow from Western Europeans did or did not occur leaves this question open . The historical model we proposed is based on careful weighting of various methods and simulations , and we attempted to account for known confounders . However , it is possible that some remain . One concern is the effect of the narrow AJ bottleneck ( effective size ≈300 around 30 generations ago [9 , 16] ) on local ancestry inference and on methods such as TreeMix and f-statistics . We did not explicitly model the AJ bottleneck in our simulations , though a bottleneck may have been artificially introduced since the number of independent haplotypes from each region used to generate the admixed genomes was very small . However , as we discuss in Methods , this is not expected to affect local ancestry inference , since each admixed chromosome was considered independently . Another general concern is that while we considered multiple methods , significant weight was given to the LAI approach; however , this may be justified as the LAI-based summary statistics were more thoroughly matched to simulations . Another caveat is that our estimation of the two-wave admixture model is based on heuristic arguments ( the multiple European sources and the differential ancestry at IBD segments ) , and similarly for the admixture dates . The IBD analysis itself relies on a number of assumptions , most importantly that the error in LAI and in IBD detection is independent of the ancestry and that most of the moderately long IBD segments descend from a common ancestor living close to the time of the bottleneck ( see S1 Text section 4 and S7 Fig ) . A general concern when studying past admixture events is that the true ancestral populations are not represented in the reference panels . Here , while our AJ sample is extensive , our reference panels , assembled from publicly available datasets , are necessarily incomplete . Specifically , sampling is relatively sparse in North-Western and Central Europe ( and particularly , Germany is missing ) , and sample sizes in Eastern Europe are small ( 10–20 individuals per population ) . In addition , we did not consider samples from the Caucasus ( however , this is not expected to significantly affect the results [5] ) . We also neglected any sub-Saharan African ancestry , even though Southern European and Middle-Eastern populations ( including Jews ) are known to harbor low levels ( ≈5–10% ) of such ancestry [49 , 56] . Generally , bias will be introduced if the original source population has become extinct , has experienced strong genetic drift , or has absorbed migration since the time of admixture . Additionally , a reference population currently representing one geographic region might have migrated there recently . We note , however , that as we do not attempt to identify the precise identity of the ancestral source , but rather its very broad geographic region , some of the above mentioned concerns are not expected to significantly affect our results . Additionally , as we show in S1 Text section 3 , our pipeline is reasonably robust to the case when the true source is absent from the reference panel . We note , though , that there may be other aspects of the real data that we are unaware of and did not model in our simulation framework that may introduce additional biases . Finally , we stress that our results are based on the working hypothesis that Ashkenazi Jews are the result of admixture between primarily Middle-Eastern and European ancestors , based on previous literature [4–8] and supported by the strong localization signal of the ME source to the Levant . Strong deviations from this assumption may lead to inaccuracies in our historical model . The admixture history of Ashkenazi Jews thus remains a challenging and partly open question . To make further progress , the natural next step is to use sequencing data . Whole-genomes are now available for several European populations ( e . g . , [57] ) as well as for Ashkenazi Jews [9] and some Middle-Eastern groups [58] . The accuracy of LAI is expected to increase for sequencing data , as also noted for other analysis tools ( e . g . , [59] ) . Additionally , whole-genomes will make it possible to run analyses based on the joint allele frequency spectrum of AJ and other populations . In parallel , denser sampling of relevant European and Middle-Eastern populations ( mostly from Central and Eastern Europe ) will be required in order to refine the geographic source ( s ) of gene flow . Beyond data acquisition , we identify three major methodological avenues for future research into AJ admixture . First , any improvement in the accuracy of local ancestry inference will translate into improved power to resolve admixture events . Second , methods will have to be developed for the inference of continuous and multi-wave admixture histories ( e . g . , [35] ) under LAI uncertainty . At the same time , inference limits will have to be established for events temporally or geographically near , as we began to develop here ( S1 Text section 6; see also [40] ) . Finally , one may use the signal in the lengths of IBD segments shared between AJ and other populations and within AJ to construct an admixture model ( e . g . , as in [60] ) , provided that we can reliably detect shorter segments than is currently possible .
We merged the genotypes from all sources ( Table 1 , lifting over to hg19 whenever necessary ) , and removed cryptic relatives by detecting IBD segments ( using Germline [44] ) and removing one of each pair of individuals sharing more than 300cM . Individuals with a non-Ashkenazi genetic ancestry ( defined to share less than 15cM , on average , with other AJ ) were also removed . Other standard QC measures ( carried out in Plink [61] ) included removal of SNPs or individuals with a high no-call rate and eliminating SNPs with an ambiguous strand assignment . The genotypes of all individuals ( of all ancestries ) were jointly phased using Shapeit [62] . For the geographical localization analysis , we thinned the SNPs to eliminate LD using Plink ( LD was measured in the entire dataset ) . RFMix was run using the TrioPhased option ( see S1 Text section 1 ) and the generation parameter set to 30 . Other parameters were set to default values . In each analysis involving the AJ individuals , we used a random subset of 200–500 individuals ( out of overall 2540 ) in order to reduce running times . We did not use the expectation maximization ( EM ) option of RFMix , as simulations of ME/EU admixture demonstrated that inference accuracy was not improved by running the highly time-consuming EM step . Additionally , the EM step makes an iterative use of the admixed ( Ashkenazi ) genomes themselves in order to supplement the reference panels , thereby potentially introducing biases due to the excessive haplotype sharing in AJ . The final assigned local ancestries were the maximum-a-posteriori estimates . We verified that setting RFMix’s admixture time parameter to 50 generations did not change the inferred ancestry proportions . For each admixed individual , we assumed that admixture ( from all source populations ) occurred at a single generation . The admixture parameters are the ancestry proportions contributed by each source and the admixture time G ( generations ago ) . We generated haploid admixed chromosomes sequentially . The ancestry of each chromosomal segment was randomly selected , using the weight of each source ( i . e . , its ancestry proportions ) . We then randomly selected a chromosome from the chosen source population , and drew the segment length ( in cM ) from an exponential distribution with rate G/100 . A haploid set of 22 chromosomes was then created for each individual . Diploid individuals were constructed by randomly pairing two sets of haploid chromosomes . Once generated , we did not evolve the simulated genomes . We ran Alder [47] with default parameters ( including an automatic detection of the minimal length cutoff ) , and with two reference populations . f4 statistics were computed using the implementation in the TreeMix package [50] . The TreeMix analysis itself was run with default parameters , except a block size ( -k ) of 500 ( corresponding to ≈5MB , beyond the extent of typical LD ) . On both simulations and real AJ data , GLOBETROTTER was run with default settings , as given in the example distributed with the program . For completeness , when generating only ancestry profiles ( the proportion of ancestry contributed to the target population by each reference population ) , we set prop . ind = 1 and num . mixing . iterations = 0 . When jointly inferring admixture events and proportions , we used , following the “Brahui-Yoruba” sample provided with GLOBETROTTER , boostrap . num = 20 , props . cutoff = 0 . 001 , and num . mixing . iterations = 5 . In both modes , in the initial step of chromosome “painting” ( CHROMOPAINTER ) , the AJ genomes were only allowed to be painted by donor/surrogate populations ( Southern/Western/Eastern EU and Levant ) . To reduce the computational burden when running GLOBETROTTER on the real data , we used a random subset of 200 AJ individuals . A number of methods exist for the estimation of historical admixture times . Johnson et al . [18] fitted the number of ancestry switches; Pugach et al . [64] used a wavelet transform of the local ancestry along the genome; and Pool and Nielsen [65] , as well as Gravel [35] , fitted the distribution of segment lengths . However , these methods require an accurate identification of the boundaries of admixture segments , which is not always possible , especially for computationally phased data . Reich and colleagues [47–49] fitted the decay of admixture linkage disequilibrium ( LD ) with genetic distance ( see main text ) , but their method can be confounded by background LD . Hellenthal et al . [21] recently proposed a promising approach based on the probability of two fixed loci to have given ancestries . Admixture parameters can also be inferred using more general demographic inference methods , e . g . , based on the allele frequency spectrum [66 , 67] or IBD sharing [60]; however , to use these methods one must specify and infer a model for the entire history . Recently , Rosenberg and colleagues [39 , 68] , Liang and Nielsen [69] , and Gravel [35] , derived analytical results for the moments of the ancestry proportion , namely the fraction of the chromosome that descends from each ancestral source . These ancestry proportions can be estimated ( e . g . , [63] ) and matched to the theoretical moments for admixture time inference ( e . g . , [36 , 37] ) . However , these methods do not make use of the information available in the entire distribution , and we therefore sought to derive it . Our method assumes a simple admixture model , where the admixed population under investigation formed t generations ago as a result of merging of populations A and B , and where the proportion of ancestry contributed by A and B was q and 1 − q , respectively . We assume that lineages change along the chromosome due to recombination at rate t per Morgan . Ignoring genetic drift and constraints imposed by the underlying biparental pedigree ( which is justified for admixture times around 10–100 generations ago and for typical human effective population sizes [40] ) , we assume that following recombination , the new source population is selected at random . Therefore , a recombination event will lead to a change of ancestry from A to B with probability 1 − q and from B to A with probability q . The lengths ( in Morgans ) of the chromosomal segments with A ancestry will therefore be exponentially distributed with rate ( 1 − q ) t , and similarly for the B segments ( rate qt ) [35] . We neglect the first generation after admixture when A and B segments do not yet mix [35] . Given a chromosome of length L ( Morgans ) , the ancestry along the chromosome can be modeled as a two-state process with states A and B , and with the distribution of segment lengths in each state given above . We are interested in the distribution of x , the fraction of the chromosome in state A . Adopting a result of Stam [38] , the desired distribution is given by f ( x;L ) = ( 1−q ) e−qhδ ( x ) +qe− ( 1−q ) hδ ( 1−x ) +q ( 1−q ) he−h[ ( 1−q ) x+q ( 1−x ) ]{[qx+ ( 1−q ) ( 1−x ) ]I1 ( 2hα ) α+2I0 ( 2hα ) } , ( 4 ) Where h ≡ tL , α≡q ( 1−q ) x ( 1−x ) , and I0 and I1 are the modified Bessel functions of the first kind of order 0 and 1 , respectively . Note the delta functions at x = 0 and x = 1 , corresponding to the probability of the entire chromosome to have B-only or A-only ancestry , respectively . The mean ancestry proportion satisfies ⟨x⟩ = q , as expected . The variance is given by Var[x]=2q ( 1−q ) h2 ( e−h+h−1 ) in agreement with Eq ( A16 ) in [35] . In practice , for unrelated individuals , phase switch errors are abundant , and hence it is difficult to accurately determine the ancestry proportion per chromosome . However , it is still possible to determine the diploid ancestry proportion , y = ( x1 + x2 ) /2 . Given that homologous chromosomes have independent histories ( unless the population is extremely small [70] ) , its distribution , fd ( y; L ) , can be computed from Eq ( 4 ) by convolution . Suppose we are now given the diploid ancestry proportions yij for individuals i = 1 , … , n and chromosomes j = 1 , … , 22 ( where each chromosome has length Lj ) . Assuming that chromosomes are independent both within and between individuals , the likelihood of the data is given by likelihood=∏i=1n∏j=122f ( yij;Lj ) We can then maximize the likelihood using a simple grid search over q and t . Simulation results with perfect knowledge of segment boundaries demonstrated that the method can correctly infer both q and t ( S2 Fig ) , although the variance of the estimate increases with t . We also considered a more complex historical model with an additional admixture event . Under this model , populations A and B had merged t1 generations ago , contributing proportions q and 1 − q to the admixed population . Then , t2 ( < t1 ) generations ago , migrants from population A have replaced a fraction μ of the gene pool of the admixed population . No other events then take place until the present . Using the Markov process representation of the admixture process of Gravel [35] , and using techniques of renewal theory , we were able to derive the distribution of the lengths of the A and B segments , which depend , in a complex way , on ( t1 , t2 , q , μ ) . We then obtained an implicit expression for the distribution of the ancestry proportions . ( More specifically , we obtained the Laplace transform of that distribution with respect to the chromosome length . ) Mathematical details are given in S1 Text section 6 . We observed that the distribution of ancestry proportions generated from the double admixture model can be fitted , for some parameter combination , with the pulse model ( S1 Text section 6 ) , and therefore , we did not use these theoretical results for inference . Nonetheless , these results are useful for understanding the range of double admixture models that will be mapped into the same pulse admixture event . Specifically , under double admixture , the distribution of B segments lengths is exponential with rate r = t1 – ( 1 − q ) ( t1 –μt2 ) , and the proportion of B ancestry is M = ( 1 − q ) ( 1 − μ ) . Since for pulse admixture , T generations ago , r = ( 1 − M ) T , the inferred time T under a pulse model satisfies T ( q+μ−qμ ) =t1− ( 1−q ) ( t1−μt2 ) . ( 5 ) Given T , Eq ( 5 ) then imposes a constraint on the parameters of the model , in particular if q and μ can be independently estimated , as in our case . | The Ashkenazi Jewish population has resided in Europe for much of its 1000-year existence . However , its ethnic and geographic origins are controversial , due to the scarcity of reliable historical records . Previous genetic studies have found links to Middle-Eastern and European ancestries , but the admixture history has not been studied in detail yet , partly due to technical difficulties in disentangling signals from multiple admixture events . Here , we present an in-depth analysis of the sources of European gene flow and the time of admixture events by using multiple new and existing methods and extensive simulations . Our results suggest a model of at least two events of European admixture . One event slightly pre-dated a late medieval founder event and was likely from a Southern European source . Another event post-dated the founder event and likely occurred in Eastern Europe . These results , as well as the methods introduced , will be highly valuable for geneticists and other researchers interested in Ashkenazi Jewish origins . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"population",
"genetics",
"geographical",
"locations",
"jews",
"genetic",
"mapping",
"ethnicities",
"simulation",
"and",
"modeling",
"population",
"biology",
"research",
"and",
"analysis",
"methods",
"europe",
"geography",
"chromosome",
"biology",
"historical",
"geography",
"people",
"and",
"places",
"haplotypes",
"cell",
"biology",
"heredity",
"earth",
"sciences",
"genetics",
"biology",
"and",
"life",
"sciences",
"population",
"groupings",
"gene",
"flow",
"genomics",
"evolutionary",
"biology",
"chromosomes"
] | 2017 | The time and place of European admixture in Ashkenazi Jewish history |
Spatially controlled release of sister chromatid cohesion during progression through the meiotic divisions is of paramount importance for error-free chromosome segregation during meiosis . Cohesion is mediated by the cohesin protein complex and cleavage of one of its subunits by the endoprotease separase removes cohesin first from chromosome arms during exit from meiosis I and later from the pericentromeric region during exit from meiosis II . At the onset of the meiotic divisions , cohesin has also been proposed to be present within the centromeric region for the unification of sister centromeres into a single functional entity , allowing bipolar orientation of paired homologs within the meiosis I spindle . Separase-mediated removal of centromeric cohesin during exit from meiosis I might explain sister centromere individualization which is essential for subsequent biorientation of sister centromeres during meiosis II . To characterize a potential involvement of separase in sister centromere individualization before meiosis II , we have studied meiosis in Drosophila melanogaster males where homologs are not paired in the canonical manner . Meiosis does not include meiotic recombination and synaptonemal complex formation in these males . Instead , an alternative homolog conjunction system keeps homologous chromosomes in pairs . Using independent strategies for spermatocyte-specific depletion of separase complex subunits in combination with time-lapse imaging , we demonstrate that separase is required for the inactivation of this alternative conjunction at anaphase I onset . Mutations that abolish alternative homolog conjunction therefore result in random segregation of univalents during meiosis I also after separase depletion . Interestingly , these univalents become bioriented during meiosis II , suggesting that sister centromere individualization before meiosis II does not require separase .
After their production during S phase , sister chromatids remain paired . This sister chromatid cohesion is crucial for proper bipolar chromosome orientation within mitotic spindles during early M phase . Sister chromatid cohesion is maintained primarily by cohesin , a protein complex composed of an SMC1/3 heterodimer and accessory subunits including an α-kleisin protein [1 , 2] . However , during late metaphase after biorientation of all chromosomes within the spindle , cohesion between sister chromatids needs to be released for chromosome segregation during anaphase . This cohesion release depends on separase , an endoprotease which specifically cleaves α-kleisin just before the metaphase to anaphase transition [1–5] . Loss of cohesin or separase function results in chromosome segregation errors due to premature separation of sister chromatids or failure of their separation , respectively . In comparison to mitosis , chromosome segregation during meiosis is more elaborate [6] . After pre-meiotic S phase , homologous chromosomes pair up and form bivalents . Maintenance of homologous chromosome pairs usually depends on chiasmata generated by meiotic recombination . Importantly , the two sister kinetochores in each chromosome are also united into a functional unit and co-oriented during the first meiotic division [7–12] . This allows bipolar orientation of bivalents in the meiosis I spindle . Separase-mediated release of cohesin from chromosome arms during late metaphase I permits terminalization of chiasmata and chromosome segregation during anaphase I [13–17] . Importantly , however , release of pericentromeric cohesin is prevented during meiosis I [18–24] . This keeps sister kinetochores paired . Moreover , since sister kinetochores regain functional individuality , they become bioriented within meiosis II spindles and segregate to opposite spindle poles after separase-mediated destruction of pericentromeric cohesin during late metaphase II [25–28] . Faithful chromosome segregation during meiosis therefore relies on the unique functional unification of sister kinetochores during meiosis I , in combination with temporally controlled release of arm and pericentromeric cohesion during meiosis I and II , respectively . Interestingly , the success of meiosis depends apparently not only on the two chromosomal cohesin populations in arm and pericentromeric regions , but also on yet another , functionally distinct cohesin pool acting within the centromere . Centromeric cohesin was proposed to be present and required exclusively before meiosis I in fission yeast for the functional unification and co-orientation of the two sister kinetochores [9 , 26 , 29 , 30] . Analyses in mouse oocytes [31 , 32] and plants [33] have provided further support for a functional unification of sister kinetochores by meiosis I-specific centromeric cohesin . The molecular mechanisms that establish and inactivate centromeric cohesin before meiosis I and II , respectively , are poorly understood . Spo13 , Moa1 and Meikin , which appear to provide a similar function required for sister kinetochore co-orientation during meiosis I in budding yeast , fission yeast and vertebrates , respectively , might be involved in the generation of centromeric cohesion [11] . Moreover , destruction of centromeric cohesin during exit from meiosis I by separase seems to be an obvious , highly probable mechanism for sister kinetochore individualization before meiosis II . Beyond unknown aspects of canonical meiosis , additional issues remain to be clarified in the context of derived meiotic variants . In Drosophila males , pairing and physical linkage of homologous chromosomes involves neither meiotic recombination nor synaptonemal complex formation [34] . Several components of an alternative homolog conjunction system have been identified genetically . Mutations in modifier of mdg4 in meiosis ( mnm ) and stromalin in meiosis/SA-2 ( snm ) result in random segregation of homologs in meiosis I [35] . Cytological analyses have revealed homolog conjunction defects in the mutants [35] . The proteins MNM and SNM accumulate exclusively in spermatocytes . At the start of meiosis , they are recruited prominently to a spot on the X-Y chromosome bivalent . Although the two sex chromosomes are strongly heteromorphic in Drosophila , they both contain a locus with an rDNA gene cluster and 240 bp repeats within the intergenic sequences . These repeats were shown to be required and sufficient for X-Y pairing [36] . Cytology has revealed co-localization of these repeats with MNM and SNM [35 , 37] . In addition , weak MNM/SNM spots were also observed on autosomes . The molecular details of MNM and SNM recruitment to meiotic chromosomes remain to be analyzed . MNM contains a unique C-terminal FLYWCH Zn-finger domain and an N-terminal BTB/POZ domain that is also present on many other mod ( mdg4 ) isoforms [35] . SNM is a distant relative of the stromalins which are known to be cohesin subunits [35] . Beyond mnm and snm , homolog conjunction in Drosophila males also requires teflon ( tef ) which codes for a protein with three C2H2-type zinc fingers and unknown localization during meiosis [38] . tef expression does not appear to be meiosis-specific and it is only required for autosome but not X-Y conjunction [38] . The fact that homologous chromosomes in Drosophila males are not linked by chiasmata and arm cohesion raises the question of how homologs are separated during meiosis I . In principle , separase activity might inactivate the alternative homolog conjunction system before anaphase I and thus cause homolog separation as in canonical meiosis . However , separase appears to target primarily cohesin which does not appear to be involved in alternative homolog conjunction . Hence , a separase-independent mechanism remains a possibility as well . Therefore , we evaluated the role of separase during Drosophila male meiosis . Moreover , the possibility of separase-independent chromosome segregation during Drosophila male meiosis I appeared to offer opportunities for confirmation that separase is actually required for sister kinetochore individualization during exit from meiosis I , since in this case normal chromosome segregation during meiosis I would be predicted to be followed by a failure of sister kinetochore biorientation during meiosis II in the absence of separase function . The apparent absence of a Drosophila rec8 homolog [39 , 40] provided yet another reason for our interest in meiotic separase functions . A meiosis-specific Rec8 alpha-kleisin has been shown to be absolutely required for protection of pericentromeric cohesion from separase cleavage during meiosis I in a wide range of species [15 , 21 , 23 , 24 , 26 , 33 , 41] . Its putative absence in Drosophila further emphasizes the non-canonical nature of its meiosis , making meiotic separase functions highly unpredictable in this species where potentially not just chromosome separation during meiosis I but also during meiosis II might be achieved in a separase-independent fashion . To evaluate meiotic functions of separase , we developed approaches for separase inhibition specifically during male meiosis . Thereby we were able to demonstrate that the inactivation of the alternative homolog conjunction system before anaphase I onset is entirely dependent on separase . Moreover , the fact that alternative conjunction in Drosophila males can be eliminated by specific mutations that do not affect sister chromatid cohesion also allowed us to address whether sister kinetochore individualization after metaphase I depends on separase . Interestingly , we find that sister kinetochore biorientation during meiosis II appears to be independent of separase .
Drosophila Separase ( SSE ) does not have a large N-terminal regulatory region as typically observed in other species [42] . However , it forms a complex with the product of the three rows ( thr ) gene which appears to have resulted from a separase gene split during Drosophila evolution [43 , 44] . Moreover , the product of the pimples ( pim ) gene also needs to accumulate during interphase and join the SSE-THR complex for eventual separase function during mitosis [42 , 45] . Therefore , embryos lacking pim function zygotically also display a separase loss-of-function phenotype . This mutant phenotype does not reveal that PIM actually also has an additional , separase-inhibitory role . However , PIM is known to be the Drosophila securin homolog that prevents premature separase activity until late metaphase [46 , 47] . While PIM is required initially for separase complex formation , it has to be degraded again eventually during late metaphase via activation of the ubiquitin ligase APC/C . The meiotic function of Sse , thr and pim cannot be studied in zygotic null mutants . They do not develop to the developmental stages where progression through meiosis starts because of sister chromatid separation failure during the earlier mitotic divisions [42 , 45 , 48] . To investigate meiotic functions , we applied transgenic RNAi expressed specifically in spermatocytes using the bam-GAL4-VP16 ( bG ) driver . Knock down of thr was found to be most effective , causing male sterility ( S1 Table ) . To characterize the effects of THR depletion at a cellular level , we analyzed testis squash preparations after staining with a DNA stain and anti-tubulin . Anti-tubulin labeling facilitates identification of meiotic cells and discrimination of meiosis I and II also when chromosome segregation is abnormal . Cells during the meiotic divisions have spindles that change in a characteristic manner during progression through these divisions . During late anaphase and telophase for example , formation of a prominent central spindle occurs . Anti-tubulin labeling also revealed the overall cell size which is halved by cytokinesis first during meiosis I and once again during meiosis II . Conversely , cell number per cyst increases and also provided information whether cells are in meiosis I or II . Preparations with testes from bG males with and without UASt-thrRNAi transgene displayed normal prometaphase I figures ( Fig 1A ) . Up to four DNA masses could be resolved corresponding to the bivalents with the sex chromosomes ( XY ) and the three autosomes ( 2nd , 3rd , and small 4th chromosome ) . In the control preparations , telophase I figures and the figures from prometaphase II and telophase II were normal as expected ( Fig 1A ) . Telophase cells contained two round daughter nuclei of comparable size close to the two spindle poles . In contrast , after THR depletion 86% of the telophase I figures were clearly abnormal ( Fig 1A and 1B ) . Bi-lobed DNA staining with a connecting chromosome bridge was observed ( 67% ) , as well as cases with a single DNA mass that was no longer bi-lobed ( 19% ) . Subsequent meiotic stages were also abnormal after THR depletion ( Fig 1A ) . Prometaphase II cells contained highly variable amounts of chromatin . Moreover , chromosome separation failure was again apparent during telophase II in 100% of the cysts ( Fig 1C ) . Time-lapse analysis of progression through the meiotic divisions without and with THR depletion ( S1 Movie and S2 Movie ) fully confirmed the findings from squash preparations . THR depletion did not affect meiosis I up to anaphase onset . However , chromosome separation during anaphase I did not succeed . Subsequent cytokinesis was irregular as well , producing a pair with a nucleate and an anucleate cell in some cases , or cutting through the undivided mass of chromatin in other cases . As a result , meiosis II spindles were often abnormal . But even in cases with normal meiosis II spindles , chromosome separation during anaphase II never occurred normally . Time-lapse imaging also revealed that THR depletion did not severely affect the dynamics of progression through meiosis I . However , the number of our movies that start before nuclear envelope breakdown in meiosis I is low ( only two cysts from independent preparations ) because time-lapse imaging was usually started only after finding a cyst within the testis preparation which already had initiated meiosis I . Starting time-lapse imaging earlier at a stage where cysts are still in the long premeiotic G2 phase allows analysis of progression through meiosis only in very rare fortuitous cases because cyst viability deteriorates in most of the cases before entry into meiosis . In the two completely tracked THR depleted meiotic cysts , the duration of prometaphase I , metaphase I , and anaphase I was 12/18 , 12/16 , and 11/12 minutes , respectively . The average duration of these meiosis I phases in controls was 12 . 3 ± 1 . 8 , 13 . 3 ± 3 . 7 , and 10 . 7 ± 1 . 2 minutes respectively ( ± s . d . ; n = 6 cysts from independent preparations ) . Additional movies allowing determination of the duration of metaphase I ( n = 2 ) and anaphase I ( n = 4 ) after THR depletion provided further support for our conclusion that the temporal dynamics of congression of bivalents and onset of anaphase during meiosis I were not severely altered by THR depletion , although subtle effects cannot be excluded . As a result of the chromosome separation defects caused by THR depletion during meiosis I , meiosis II was severely affected in various variable ways and therefore , analysis of temporal dynamics during interkinesis and meiosis II after THR depletion was not attempted . As bG-directed expression starts already during the final mitotic division cycles that generate the cysts of 16 interconnected spermatocytes , the meiotic defects described above might arise in principle as secondary consequences from earlier division abnormalities . However , several observations indicated that the cysts entering into meiosis after bG-directed THR depletion were normal . Cysts still comprised 16 cells as expected . Moreover , the number of centromeres detected during meiosis indicated that all spermatocytes were euploid . FISH with X and Y probes , as well as MNM/SNM immunolocalization also confirmed euploidy ( see below ) . However , testes of bG males with UASt-thrRNAi were smaller than those of control males . Therefore , some germline stem cell or gonial cyst depletion might occur as a result of leaky expression during earlier stages . To rule out off-target effects , we introduced into the males with bG and UASt-thrRNAi also an UASt-thrRr transgene predicted to be RNAi-resistant as a result of silent mutations . While the RNAi-resistant transgene was unable to restore a completely normal meiosis , it reduced the frequency of telophase I abnormalities fourfold ( Fig 1B ) . Moreover , the residual chromosome bridges during telophase I were less massive and less stable when the RNAi-resistant transgene was present ( S1 Fig ) . For further confirmation that the abnormalities resulting after THR depletion reflect a loss of separase function , we depleted other separase complex subunits . Analogous bG directed expression of UASt-SseRNAi and UASt-pimRNAi resulted in no or milder abnormalities , respectively . Abnormalities after PIM depletion were also first observed during telophase I , as in case of THR depletion . Chromosome bridges were apparent in 56% of the telophase I figures ( Fig 1B ) and all cysts during telophase II also displayed chromosome bridges . To achieve SSE depletion , we applied an alternative method , deGradFP which allows regulated proteolytic degradation of GFP fusion proteins [49] . Therefore , we first generated a line in which the lethality associated with hemizygosity for an Sse null mutation ( Sse13m/Df ( 3L ) SseA ) was prevented by a transgene driving expression of EGFP-SSE under control of the Sse cis-regulatory region ( gEGFP-Sse ) . The resulting flies were also fertile ( S1 Table ) . However , introduction of an additional transgene directing NSlmb-vhh-GFP4 expression by the bam regulatory region led to complete sterility ( S1 Table ) . NSlmb-vhh-GFP4 has been shown to result in polyubiquitination and proteasomal degradation of GFP fusion proteins [49] . Analyses of testis squash preparations revealed that SSE depletion by deGradFP resulted in abnormalities that were indistinguishable from those caused by thr-RNAi ( Figs 1B , 1C and S2 ) . In summary , spermatocyte-specific depletion of the three separase complex subunits , THR , PIM or SSE , with different strategies ( transgenic RNAi and deGradFP ) was found to cause the same specific defect during meiosis . Importantly , defects start already during anaphase I . Although the degree of interference was lower in case of PIM , presumably as a result of incomplete depletion , chromosome segregation during anaphase I was affected in each case . We conclude that separase function is required for chromosome segregation during both meiotic divisions in Drosophila males . To determine whether separase is required for the removal of alternative homolog conjunction proteins before chromosome segregation during meiosis I in Drosophila males , we analyzed the effects of THR depletion on the subcellular localization of MNM and SNM with the help of a fully functional mnm-EGFP transgene [35] and anti-SNM antibodies [35] . As previously described [35] , co-localized MNM-EGFP and anti-SNM signals were detected primarily in a highly prominent spot on the XY bivalent during prometaphase and metaphase of meiosis I ( Fig 2A ) . In contrast , no MNM-EGFP/anti-SNM spot was detected during telophase I ( n = 47 ) ( Fig 2A ) and later meiotic stages in control testis where meiosis I is normal [35] . However , after THR depletion , MNM-EGFP/anti-SNM spots were not only present early during meiosis I , but also during telophase I ( n = 209 spermatocytes ) ( Fig 2A ) and all subsequent stages up to the postmeiotic stages of sperm tail elongation ( S3 Fig ) . In telophase I cells with chromosome bridges , the MNM-EGFP/anti-SNM spots were usually associated with the bridge ( 94% ) . Telophase I cells without obvious chromosome bridges ( 6% ) also had always an MNM-EGFP/anti-SNM spot in one of the two daughter nuclei . For confirmation , we also performed time-lapse imaging with testis expressing MNM-EGFP and a red fluorescent histone ( Fig 2B , S3 Movie and S4 Movie ) . In control spermatocytes , the MNM-EGFP spot was observed to disappear during early anaphase I within 2–3 minutes ( n = 18 spermatocytes ) . In contrast , after THR depletion chromosome separation failure was accompanied by persistence of the MNM-EGFP spot during exit from meiosis I ( n = 76 spermatocytes ) . By anti-SNM staining of testis squash preparations , we evaluated whether PIM depletion by RNAi or SSE depletion by deGradFP also resulted in SNM perdurance beyond anaphase I ( S4 Fig ) . These stainings clearly revealed anti-SNM signals during telophase I and meiosis II in spermatocytes depleted for these other separase complex subunits . We conclude that separase function is required for the release of alternative homolog conjunction proteins from chromosomes during progression through meiosis I . The observed failure of MNM and SNM release from chromosomes at anaphase I onset after THR depletion is likely responsible for the associated failure of homolog separation during meiosis I . Accordingly , when alternative homolog conjunction fails to be established , THR depletion is no longer expected to cause chromosome bridging during telophase I . To evaluate this prediction , we depleted THR in mnm and snm mutants . However , we first confirmed that mutations in mnm and snm do not cause chromosome bridges during telophase I independent of THR depletion . Consistent with earlier reports [35] , instead of four bivalents as in wild type , we observed up to eight chromatin masses during prometaphase I in mnm and snm mutants ( Figs 2C and S5 ) , indicating the known defect in homolog conjunction . In addition , daughter nuclei were often unequal in size and DNA content ( Figs 2C and S5 ) , reflecting the known random distribution of chromosomes during meiosis I [35] . Chromosome bridges were very rare in telophase I ( Fig 3B ) . The few bridge-like structures in telophase I that were observed in mnm and snm mutants presumably represent univalents lagging in the division plane rather than stretched bivalents [35] . The low frequency of chromosome bridges during meiosis I in mnm and snm mutants also confirms that co-orientation of sister kinetochores during meiosis I does not depend on homolog conjunction during male meiosis [35] . THR depletion in mnm mutants did not induce any phenotypic change during meiosis I ( Figs 2C and S5 ) . Importantly , it did not cause chromosome bridges during telophase I ( Figs 2C and 3B ) , in striking contrast to the effect of THR depletion in spermatocytes with functional alternative homolog conjunction ( Fig 1 ) . Moreover , THR depletion did also not result in chromosome bridges during telophase I in snm mutants ( Figs 3B and S5 ) . In addition , SSE depletion by deGradFP did also no longer cause chromosome bridges during telophase I when performed in mnm or snm mutants ( Figs 3B and S5 ) . In summary , the fact that mnm and snm mutations , which abolish the alternative homolog conjunction in male meiosis , suppress chromosome bridge formation during telophase I in spermatocytes progressing through meiosis in the absence of separase function indicates that separase is required for the resolution of homolog conjunction during meiosis I . Our conclusion that absence of separase function during Drosophila male meiosis I results in a specific failure of homolog separation is further supported by analyses with an additional mutant . tef mutants were chosen because alternative homolog conjunction is also defective in these mutants but not completely . While autosomes fail to pair into bivalents , formation of the X-Y bivalent is not affected [38] . THR depletion in tef mutants , therefore , is predicted to result in a chromosome bridge during telophase I that always represents a stretched XY bivalent , if resolution of homolog conjunction during male meiosis I requires separase function . As in case of nmn and snm mutants , chromosome bridges during telophase I in tef mutants with undisturbed separase function ( Figs 2C and 3B ) were infrequent and presumably represent occasional lagging autosomes . THR depletion in tef mutants resulted in an increased frequency of chromosome bridges during telophase I ( Figs 2C and 3B ) . However , these bridges were fewer and less massive compared to those induced by THR depletion in spermatocytes with entirely normal homolog conjunction ( Figs 1 , 2A and 2B ) , consistent with the notion that only the XY bivalent forms bridges in tef mutants , while all bivalents form bridges in case of normal homolog conjunction . To demonstrate directly that the bridges resulting from THR depletion in tef mutants represent stretched XY bivalents , we performed fluorescence in situ hybridization ( FISH ) with a red fluorescent probe for the X and a green fluorescent probe for the Y chromosome ( Fig 2D ) . After THR depletion in tef mutants , the red and green signals were invariably observed on the bridge ( Fig 2D ) . Moreover , anti-SNM staining also resulted in a dot on the non-segregating DNA mass resulting after THR depletion in tef mutants ( Fig 2D ) . FISH was also applied to analyze the segregation of the X and Y chromosome after THR and SSE depletion in spermatocytes with normal homolog conjunction , and also in mnm and snm mutants after SSE depletion ( S6 Fig ) . These analyses confirmed that disjoining of X and Y was inhibited in the absence of separase function and that the suppression of chromosome bridges resulting from a lack of separase function by inactivation of the alternative homolog conjunction system was paralleled by random segregation of X and Y during meiosis I . Finally , live imaging of progression through meiosis in spermatocytes with green fluorescent centromeres and red fluorescent chromosomes indicated that THR depletion causes chromosome separation failure directly and not indirectly via impairment of spindle or kinetochore function or premature exit from meiosis I ( S7 Fig and S5 and S6 Movies ) . While mutations inactivating the alternative homolog conjunction system during Drosophila male meiosis very effectively suppressed chromosome bridges during telophase I after THR depletion ( Figs 2C and 3B ) , they completely failed to do so during telophase II ( Fig 3A and 3C ) . These telophase chromosome bridges presumably reflect a failure to separate sister centromeres during meiosis II in the absence of separase function . To confirm this notion , we performed experiments with mutations in sisters on the loose ( solo ) . SOLO has no sequence homology to known proteins [50 , 51] . However , analysis of the solo mutant phenotype revealed that it provides a function analogous to that of Rec8 in other eukaryotes . Rec8 is a meiosis-specific α-kleisin . Pericentric cohesin with Rec8 instead of the non-meiosis-specific Rad21 α-kleisin is protected from separase-dependent cleavage during meiosis I but no longer during meiosis II [6] . The Drosophila genome does not contain an obvious Rec8 homolog . But SOLO expression is also meiosis-specific [50 , 51] . It interacts physically and functionally with the SMC1 core cohesin subunit and it is present at meiotic centromeres until anaphase II . Mutations in solo result in premature separation of sister chromatids . The chromosome bridges observed during telophase II after THR depletion in mnm mutants are therefore predicted to be abolished when the spermatocytes also lack solo function , if the bridges reflect stretched sister chromatids . Indeed , chromosome bridges during telophase II were not only absent in solo single mutants but also after THR depletion in solo mnm double mutants ( Fig 3A and 3C ) . Chromosome bridges during telophase II were also missing after THR depletion in solo snm double mutants ( Figs 3C and S8 ) . These results demonstrate that Separase is required for resolution of sister chromatids in meiosis II . We point out that mutations in solo , in contrast to mutations in mnm and snm , did not abolish the THR depletion-induced chromosome bridges during meiosis I ( Figs 3B and S9 ) , indicating that the alternative homolog conjunction system functions independent of sister chromatid cohesion . Consistent with this conclusion , earlier data reported as unpublished [50] indicated that MNM and SNM are present on bivalents in solo mutants where the large majority of bivalents also remain intact through metaphase I . Anti-SNM staining revealed that the prominent SNM dot on the XY bivalent cannot be detected any longer in telophase I and subsequent stages in solo mutants ( S10 Fig ) , indicating that the inactivation of the alternative homolog conjunction during meiosis I occurs normally . However , THR depletion in solo mutants effectively prevented the disappearance of the SNM dot during exit from meiosis I ( S10 Fig ) . Intense SNM dots were still present during meiosis II and during the early postmeiotic stages ( S10 Fig ) . This perdurance of the alternative homolog conjunction presumably also explains the high frequency of chromosome bridges that was observed after THR depletion in solo mutants during meiosis II ( Fig 3C ) . The fact that separase is no longer required for chromosome separation during meiosis I in mutants with a non-functional alternative homolog conjunction system facilitates the analysis of a putative separase requirement for sister centromere individualization during progression through meiosis I . Since centromeric cohesin is thought to unite sister centromeres into a functional unit before meiosis I [12] , destruction of centromeric cohesin by separase during meiosis I might enable sister centromere individualization and biorientation during meiosis II . After THR depletion in mnm mutants , univalents are randomly distributed into the two daughter cells . If separase function during meiosis I is required for sister centromere individualization , biorientation of these univalents in the meiosis II spindle is expected to fail and co-orientation of sister kinetochores as in meiosis I will also occur during meiosis II . To monitor sister centromere behavior we performed live imaging with spermatocytes expressing green fluorescent CID/Cenp-A as well as red fluorescent histone His2Av . Splitting of sister centromeres as well as sister kinetochore biorientation in meiosis II was clearly observed not only in control spermatocytes but also after THR depletion in mnm and snm mutants ( Fig 4A and S7 and S9 Movies ) . In fact , sister centromere splitting and biorientation could also be clearly detected after THR depletion in spermatocytes with functional homolog conjunction although meiosis II was often highly irregular as a result of the meiosis I chromosome separation failure ( Fig 4A and S8 Movie ) . Importantly , sister centromere individualization and biorientation during meiosis II after THR depletion was detectable only transiently during metaphase II . During progression into metaphase II , the single centromere dots present in a secondary spermatocyte were split into a pair of dots along the spindle axis . However , the resolved sister kinetochores did not move apart towards opposite poles during anaphase II in the absence of separase function as a result of the failure in releasing sister chromatid cohesion . While the sister chromatid separation failure indicated effective THR depletion , the presence of residual separase function still sufficient for sister centromere individualization during meiosis I cannot be ruled out definitively . To address this possibility , we carefully compared the effects of THR depletion in spermatocytes with either two or only one functional thr gene copy quantitatively . A reduction of the thr gene copy number is predicted to increase the efficiency of THR depletion and hence sister centromere splitting and kinetochore biorientation might be more compromised , if separase function during meiosis I is crucial for these processes . Therefore , we also performed THR depletion in spermatocytes heterozygous for either a P-element insertion within thr ( thrk07805b ) or a deficiency that deletes thr ( Df ( 2R ) BSC338 ) . These THR depletion experiments were again performed in a background with mutations in mnm or snm , where meiosis II is not accompanied by spindle irregularities resulting from chromosome separation failure during meiosis I . We quantified the fraction of secondary spermatocytes that displayed centromere splitting during metaphase II ( Fig 4B ) . Moreover , we also determined the inter-sister kinetochore distance during metaphase II ( Fig 4C ) because partial centromere individualization during meiosis I might result in a reduced inter-sister kinetochore distance during metaphase II . These measurements did not reveal a difference between metaphase II spermatocytes in controls and after THR depletion in spermatocytes with only one functional thr gene copy . Finally , by estimating the duration of the different phases ( prometa- , meta- , ana- , telophase ) in the time-lapse movies , we assessed whether the temporal dynamics of progression through meiosis II was affected by THR depletion in either mnm or snm mutants that had only one functional thr+ gene copy ( Fig 4D ) . Compared to controls , the average time interval between nuclear envelope breakdown and anaphase onset was found to be slightly extended in the two THR depletion cases ( 28 . 4 ± 6 . 2 and 26 . 1 ± 4 . 2 versus 22 . 5 ± 2 . 5 min; n = 8 , 7 and 5 , respectively ) but not in a statistically significant manner . Similarly , prometaphase appeared to be extended in the two THR depletion cases ( Fig 4D ) although variablity in these two cases was considerable . Therefore , it is conceivable that separase makes a contribution to chromosome biorientation during meiosis II but it does not appear to be essential .
We demonstrate that the alternative system used for conjunction of homologous chromosomes before meiosis I that is used in Drosophila males instead of the canonical combination of chiasmata and sister chromatid cohesion , needs to be inactivated by separase during the transition from metaphase to anaphase for normal chromosome segregation during meiosis I . Moreover , we provide evidence that the individualization of sister kinetochores , which have been proposed to become united for co-orientation during meiosis I by centromeric cohesin , does not require separase function during exit from meiosis I . Our work also demonstrates that sister separation during meiosis II depends on separase . As separase is required for development to the stages where male meiosis occurs , spermatocyte-specific depletion of separase complex subunits had to be developed for our analysis of separase function during male meiosis . In case of the THR subunit , transgenic RNAi expressed in early spermatocytes was found to be highly effective . Live imaging after THR depletion revealed a complete failure of homolog separation during meiosis I in all cells of all analyzed spermatocyte cysts . Similarly , cytological characterization of fixed testis squash preparations revealed chromosome bridges in late meiosis I figures ( anaphase I , telophase I ) , consistent with the live imaging results . The 10% telophase I figures that did not display chromosome bridges in the fixed samples , at least in part , represent cases where the bridges did not persist long enough . We point out that chromosome bridges resulting from a failure of homolog separation during Drosophila male meiosis I are unlike those with continuous DNA throughout , resulting in mitosis after incomplete chromosome replication or in canonical meiosis after recombination defects . During Drosophila male meiosis I , homologs are thought to be conjoined by proteinaceous links that in exceptional cases might be severed eventually by spindle forces or other processes activated during exit from meiosis I . In principle , the few exceptional telophase I figures without chromosome bridges might also indicate residual THR function . To address the possibility of incomplete knockdown , we have carefully compared the effects of THR depletion in spermatocytes with either two or one functional copy of the endogenous thr+ gene . Since the effects of THR depletion were found to be entirely independent of the thr+ gene copy number , residual THR function does not appear to be present , although we cannot rule it out completely . RNAi can have off-target effects . Several of our findings indicate that the consequences of THR depletion reported here do not reflect such off-target effects . Expression of the RNAi-resistant UASt-thrRr transgene resulted in significant although partial suppression of the THR depletion effects . Suppression simply as a result of Gal4 titration away from the UAS-V20thrshmiR9 transgene by the UASt-thrRr transgene can be ruled out , because experiments with two UASt-thrRr copies did not produce stronger suppression . We do not understand why UASt-thrRr expression does not result in complete suppression , but suspect problems caused by an inappropriately early temporal window of bG-mediated UASt-thrRr transgene expression , insufficient mRNA stability and translational control because of missing untranslated regions . Translational control is particularly pervasive in spermatocytes and is known to occur in case of Cyclin B and Twine/Cdc25 phosphatase , two well-studied cases of meiotic M phase regulators [52 , 53] . We speculate that co-translation of all the separase complex subunits late during the four day spermatocyte growth phase might be required for the production of functional separase complexes for meiosis . UASt-thrRr transcripts are not present in late spermatocytes after expression using bG . Alternative GAL4 transgenes effectively driving expression in late spermatocytes do not exist . The fact that SSE depletion in spermatocytes by deGradFP results in the same defects as THR depletion by RNAi provides further evidence against RNAi off-target effects . In case of SSE depletion by deGradFP , we were unable to achieve suppression by expression of an UASp-Sse transgene , presumably also for the reasons discussed above . These technical difficulties have precluded meaningful experiments addressing whether the effects of SSE depletion are suppressible by expression of an SSE variant predicted to be a catalytically inactive protease . Our live imaging has revealed that the MNM-EGFP dot , which reflects XY chromosome conjunction , disappears very rapidly during the first 2–3 minutes of anaphase I in a THR-dependent manner . These observations and the corroborating analyses with anti-SNM on fixed samples strongly suggest that homolog separation during male meiosis I requires SSE protease activity , but they do not exclude a non-proteolytic role leading to MNM/SNM re-distribution throughout the cell that might not be detectable with our tools . A direct analysis of MNM/SNM protein levels during progression through meiosis I by immunoblotting would require the isolation of sufficient amounts of precisely staged meiotic cysts . The low abundance of meiosis I cysts has prevented such analyses so far . While the MNM-EGFP dot disappears during the cell cycle phase where separase is predicted to be active as a protease , a slightly earlier disappearance would have been expected . We suspect that proteolytic inactivation of the very high amounts of MNM/SNM present on the dot on the XY bivalent might be slower than the removal of the far less concentrated material from the autosome and that the XY bivalent therefore might be the last to separate during anaphase I . Assuming that homolog separation during male meiosis I depends on protease activity of separase raises the question what the critical substrates might be . Drosophila melanogaster MNM and SNM contain sequences conforming to the consensus of separase cleavage sites [54] , but they are poorly conserved within Drosophilid orthologs , and we have been unable to detect MNM and SNM cleavage during M phase after expression in mitotically proliferating cells ( preliminary observations ) . As the molecular basis of homolog conjunction by MNM and SNM is far from being clear , it remains readily possible that they function together with additional unknown protein partners that might be cleaved by separase . In principle , it appears conceivable that MNM and SNM co-operate with cohesin to bring about alternative homolog conjunction during male meiosis . The fact that SNM is a distant member of the SA/Stromalin family of cohesin subunits would appear to support this notion . Accordingly , separase might target α-kleisin as its critical substrate also during male meiosis I . Several observations argue against this . SNM as well as MNM do not co-localize with the SMC1 cohesin subunit [35] and therefore appear to function independent of cohesin . Moreover , the mitotic Rad21 α-kleisin does not appear to be involved during the meiotic divisions , at least in case of female meiosis [40] . The meiosis-specific α-kleisin family protein encoded in the Drosophila genome , C ( 2 ) M , is not required for male meiosis and does not function in a Rec8-like manner during female meiosis where it is required for normal synaptonemal complex formation but not sister chromatid cohesion [39 , 55] . Genetically several genes have been identified that based on their mutant phenotype appear to provide a Rec8-like function during Drosophila meiosis [50 , 51 , 56–59] . Their protein products ORD , SOLO , and SUNN have no sequence similarity to α-kleisins . But they are mutually dependent on each other for their predominant localization around centromeres where they are co-occurring with SMC1 and SMC3 . All evidence therefore suggests that these proteins might function in a Drosophila-specific variant cohesin complex providing sister chromatid cohesion during meiosis . While it is readily possible that one of these proteins is a separase substrate during meiotic divisions , this would not explain how separase brings about homolog separation during male meiosis I for reasons also provided by our experiments . We demonstrate that homolog conjunction by MNM and SNM during male meiosis does not depend on solo function , consistent with previous work [50] . This conclusion is suggested by our observations that THR depletion in solo mutants results in chromosome bridges during meiosis I which are no longer observed when THR is depleted in solo mnm or solo snm double mutants . In the absence of solo function , therefore , MNM/SNM establish chromosome conjunction that results in chromosome bridges during meiosis I in the absence of separase function . At present , a critical target that needs to be cleaved by separase for homolog separation during male meiosis I is not known . Similarly , the critical separase target for sister chromatid separation during meiosis II is also unknown . The possibility to by-pass the separase requirement for chromosome separation during meiosis I in Drosophila males by mutational inactivation of the alternative homolog conjunction system , in combination with our ability to monitor progression through both meiotic divisions by time-lapse imaging , has also allowed us to address the role of separase for sister centromere individualization during exit from meiosis I . A series of extremely elegant experiments in fission yeast has provided strong evidence suggesting that the co-orientation of sister centromeres that is established specifically during meiosis I for biorientation of bivalents in meiosis I spindles depends on the presence of centromeric sister chromatid cohesion mediated by meiotic Rec8 cohesin complexes [9 , 26 , 29 , 30] . Centromeric cohesion that keeps sister centromeres in close proximity might result in the assembly of a single kinetochore on each homolog at the onset of meiosis I . Importantly , to allow sister kinetochore biorientation within meiosis II spindles , centromeric cohesion would have to be resolved at some stage after co-orientation during meiosis I has been achieved . Our observations suggest that separase might not be required for sister centromere individualization during Drosophila male meiosis . We demonstrate that sister kinetochore biorientation is successful during meiosis II after THR depletion , while sister chromatid separation is completely abolished . Separase might be dispensable for the removal of centromeric cohesin during meiosis I because of an alternative cohesin removal mechanism . In principle , Wapl can open cohesin rings without separase although only when the rings have not yet been locked by SMC3 acetylation [60–64] . Alternatively , sister centromere co-orientation in Drosophila male meiosis might not involve centromeric cohesin . Finally , we acknowledge that our evidence cannot definitely rule out the possibility that some residual THR escaping depletion might still be sufficient for normal sister kinetochore biorientation but not for sister chromatid separation during meiosis II . Moreover , we also point out that our light microscopic analyses cannot resolve the aspects of sister centromere individualization during Drosophila male meiosis I that have been observed by serial sectioning and electron microscopic analysis [7] . At the ultrastructural level , a hemispherical kinetochore , where the two sister kinetochores cannot be resolved , is detected in early prometaphase I spermatocytes . By anaphase I , however , two closely associated but clearly distinct sister kinetochores in a side-by-side configuration were usually observed . For lack of spatial resolution , we cannot exclude that extent or dynamics of this sister kinetochore resolution process during meiosis I is abnormal after THR depletion . Similarly , our data can also not exclude that a possible separase contribution to sister kinetochore biorientation during meiosis II might eventually be compensated during prometaphase II in THR depleted spermatocytes by spindle forces for example . Our limited data on temporal dynamics of chromosome congression during meiosis II after THR depletion in mnm and snm mutants is consistent with the notion that separase makes some contribution to efficient chromosome biorientation during meiosis II but cannot prove it . The analysis of the role of separase for sister kinetochore biorientation during meiosis certainly deserves further attention , including studies in other organisms .
The following lines with previously characterized mutations or transgenes were used: Sse13m and Df ( 3L ) SseA [42] , thrk07805b [65] , Df ( 2R ) BSC338 [66] , P{ry+ , hsp70-mnm-EGFP} , mnmZ3-3298 , mnmZ3-5578 , snmZ3-0317 , and snmZ3-2138 [35] , soloZ2-0198 and soloZ2-0338 ( Yan et al . , 2010 ) , tefZ2-4169 and tefZ2-3455 [67] , P{w+ , bamP-GAL4-VP16}III [68] , P{w+ , His2Av-mRFP}II . 2 and P{w+ , gcid-EGFP-cid}II . 1 [69] , P{w+ , pUbi-EGFP-alphaTub84B}II [70] , and P{w+ , pUbi-EYFP-asl} [71] . Lines for transgenic RNA interference , UAS-V20thrshmiR9 , UAS-V20thrshmiR10 and UAS-W20thrshmiR45 , were generated by integrating the pVALIUM20 and pWALIUM20 constructs ( see below ) into the attP2 landing site . For production of P{w+ , UASt-thrRr}attP40 , allowing expression of a thr cDNA with silent mutations in the regions targeted by thrshmiR9 , thrshmiR10 and thrshmiR45 , a pUASt-attB construct ( see below ) was integrated into the attP40 landing site . For pim-RNAi , we used y w1118; P{w+ , KK106514}VIE-260B ( v100534 ) . PBac{3xP3-ECFP , gEGFP-Sse}III . 1 was generated by germline transformation with a PiggyBac construct . P{w+ , bamP-NSlmb-vhh-GFP4}II . 1 was isolated after P-element-mediated germline transformation with a pCaSpeR4 construct . For all experiments , flies were cultured at 25°C . Detailed genotypes of the flies analyzed are provided in the supplemental material ( S1 Table ) . For the production of transgenic lines allowing GAL4-dependent expression of short hairpin microRNAs ( shmiRs ) , we generated constructs using the vectors pVALIUM20 and pWALIUM20 [72] . Inserts were generated by annealing the following oligonucleotides: 5'-ctagcagt-CCCTTGGAAGCTACAAGTCAA-tagttatattcaagcata-TTGACTTGTAGCTTCCAAGGG-gcg-3' and 5'-aattcgc-CCCTTGGAAGCTACAAGTCAA-tatgcttgaatataacta-TTGACTTGTAGCTTCCAAGGG-actg-3' for thrshmir9 , 5'-ctagcagt-AACGCTTCTAGTTCAACTAAA-tagttatattcaagcata-TTTAGTTGAACTAGAAGCGTT-gcg-3' and 5'-aattcgc-AACGCTTCTAGTTCAACTAAA-tatgcttgaatataacta-TTTAGTTGAACTAGAAGCGTT-actg-3' ) for thrshmir10 , 5’-ctagcagt-AAGAAGTAGATCATTCTTCAA-tagttatattcaagcata-TTGAAGAATGATCTACTTCTT-gcg-3’ and 5’-aattcgc-AAGAAGTAGATCATTCTTCAA-tatgcttgaatataacta-TTGAAGAATGATCTACTTCTT-actg-3’ for thrshmir45 . Capital letters indicate the regions corresponding to sequences within the thr coding sequence . For the production of a transgenic line allowing expression of EGFP-Sse under control of the Sse cis-regulatory region , we generated pBac{3xP3-ECFP-gEGFP-Sse} using a modified version of the previously described gSse transgene construct [42] . During construction , an Sse cDNA fragment replacing the Sse genomic region containing the first two introns was introduced as well as the EGFP coding sequence fused at the N-terminus . For the construction of pCaSpeR4-bamP-NSlmb-vhh-GFP4 , the fragment coding for NSlmb-vhh-GFP4 was amplified from pUASt-NSlmb-vhh-GFP4 [49] using primers SCH1 ( 5’-GACTACCGGTATGATGAAAATGGAGACTGAC-3’ ) and SCH2 ( 5’-GACTGCGGCCGCTTAGCTGGAGACGGTGAC-3’ ) . After digestion with AgeI and NotI , the insert was used to replace the GAL4-VP16 coding region released by the same restriction enzymes in pCaSpeR4-bamP-GAL4-VP16 [68] . A modified thr cDNA was inserted into pUASt-attB [73] for the production of pUASt-attB-thrRr . The thr cDNA [48] was modified by replacing the region containing shmiR target sequences with a synthetic variant ( GenScript , Piscataway , NJ 08854 , USA ) . The fertility of 5–10 single males per genotype was assayed in parallel . Each single male was allowed to mate with three w virgin females for two days . Flies were transferred into a fresh vial and discarded after two more days . Flies eclosing from this vial were counted . Testis squash preparations were done as described [74] . For immunolabeling , mouse monoclonal anti-α-tubulin DM1A ( Sigma ) was used at 1:10’000 and affinity-purified rabbit polyclonal anti-SNM [35] at 1:250 . Secondary antibodies were Alexa488- or Alexa568-conjugated goat antibodies against mouse or rabbit IgG diluted 1:1000 . Dissection of testis , fixation with 4% PFA , permeabilization with PBST-DOC and anti- α-tubulin staining were done as described ( protocol 3 . 2 . 2 , steps 1–14 ) [74] . Cy5-conjugated goat anti-mouse IgG diluted 1:1000 was used as secondary antibody . Ethanol incubations and dehydration with a formamide series were also done as described ( immuno-FISH protocol 3 . 2 , steps 10–26 ) [75] . An oligonucleotide ( 5'-TTTTCCAAATTTCGGTCATCAAATAATCAT-3' ) with Atto-565 on 5’ and 3’ end ( Integrated DNA Technologies , B-3001 Leuven Belgium ) was used for detection of the X-specific 359 bp repeats at a concentration of ≈ 1 ng/μl in hybridization Buffer . An oligonucleotide ( 5'-AATACAATACAATACAATACAATACAATAC-3' ) with Alexa-488 fluorophore at the 3’ end ( Sigma-Aldrich , 8107 Buchs , Switzerland ) was used for detection of the Y-specific AATAC repeats at a concentration of ≈ 2 ng/μl in hybridization buffer . The denaturation step was performed at 98°C for 6 min , and hybridization over night at 16°C . Slides were washed twice in 50% formamide , 2x SSCT at 16°C for 1 hour each . Thereafter , additional washes were performed at room temperature in 25% formamide , 2x SSCT for 10 min and three times in 2x SSCT for 10 min each . DNA was stained with Hoechst 33258 ( 1 μg/ml ) for 10 min and slides were washed twice in PBS for 5 min . Slides were mounted in 70% Glycerol , 50 mM Tris-HCl pH 8 . 5 , 10 mg/ml propyl gallate , 0 . 5 mg/ml phenylendiamine . Image stacks with 250 nm spacing between focal planes were acquired with a 63×/1 . 4 oil immersion objective on a Zeiss Cell Observer HS microscope . If not stated differently , the images displayed in the figures represent maximum intensity projections . The data used for statistical analyses of a particular genotype was obtained from multiple slides and each slide was prepared with about 14 dissected testes . Testes from pupal or adult males were dissected in Schneider’s Drosophila Medium ( Invitrogen , #21720 ) , 10% fetal bovine serum ( Invitrogen ) , 1% penicillin/streptomycin ( Invitrogen , #15140 ) . The dissected testes were transferred into 40 μl of medium in a 35 mm glass bottom dish ( MatTek Corporation , #P35G-1 . 5-14-C ) , and opened with fine tungsten needles to release the cysts . To reduce movements within the sample , methylcellulose ( Sigma ) was added . A wet filter paper was placed inside along the dish wall before sealing the lid with parafilm . Time-lapse imaging was performed with a spinning disc confocal microscope ( Visitron ) with a 60×/1 . 4 oil immersion objective . Image stacks with 24–45 focal planes spaced by 0 . 5–1 μm were acquired with a time interval of 30–60 sec . Precise numbers are specified in the legends for each supplementary movie . The distance between sister centromeres during metaphase II was measured in 3D using Imaris software ( Bitplane ) . To exclude that the measurements actually represent early anaphase time points , the last three time points before anaphase onset were excluded from consideration . | Sexual reproduction depends on meiosis , a special cell division that occurs in two steps , meiosis I and II . Meiosis is distinct in males and females that produce two very different forms of compatible gametes , the sperm and egg , respectively . In the fly Drosophila melanogaster , sex-specific differences are pronounced . While pairing of homologous chromosomes into bivalents before the first meiotic division proceeds in a canonical manner in females , males use an alternative system . After production of bivalents and their biorientation in the meiosis I spindle , they need to be split up again in both sexes so that chromosomes can be segregated to opposite spindle poles during anaphase I . Here we demonstrate that separase , a protease that separates canonical bivalents by cleavage of a cohesin subunit , is also required during male meiosis I , even though alternative homolog conjunction does not depend on cohesin . Moreover , our results suggest that the separation of sister centromeres into functionally distinct units that has to occur between the first and second meiotic division does not depend on separase . Centromeric cohesin might therefore either not enforce the crucial sister centromere coorientation during meiosis I , or be removed in a separase-independent manner in preparation for meiosis II . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"meiosis",
"invertebrates",
"homologous",
"chromosomes",
"spermatocytes",
"chromosome",
"structure",
"and",
"function",
"anaphase",
"centromeres",
"metaphase",
"cell",
"cycle",
"and",
"cell",
"division",
"cell",
"processes",
"animals",
"germ",
"cells",
"animal",
"models",
"drosophila",
"melanogaster",
"model",
"organisms",
"drosophila",
"sperm",
"telophase",
"research",
"and",
"analysis",
"methods",
"chromosome",
"biology",
"animal",
"cells",
"insects",
"arthropoda",
"cell",
"biology",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"organisms",
"chromosomes"
] | 2016 | Separase Is Required for Homolog and Sister Disjunction during Drosophila melanogaster Male Meiosis, but Not for Biorientation of Sister Centromeres |
Realistic , individual-based models based on detailed census data are increasingly used to study disease transmission . Whether the rich structure of such models improves predictions is debated . This is studied here for the spread of varicella , a childhood disease , in a realistic population of children where infection occurs in the household , at school , or in the community at large . A methodology is first presented for simulating households with births and aging . Transmission probabilities were fitted for schools and community , which reproduced the overall cumulative incidence of varicella over the age range of 0–11 years old . Moreover , the individual-based model structure allowed us to reproduce several observed features of VZV epidemiology which were not included as hypotheses in the model: the age at varicella in first-born children was older than in other children , in accordance with observation; the same was true for children residing in rural areas . Model predicted incidence was comparable to observed incidence over time . These results show that models based on detailed census data on a small scale provide valid small scale prediction . By simulating several scenarios , we evaluate how varicella epidemiology is shaped by policies , such as age at first school enrolment , and school eviction . This supports the use of such models for investigating outcomes of public health measures .
Varicella is endemic in most Western countries where vaccination has not been implemented [1] , [2] . More than 90% people are infected with varicella-zoster-virus ( VZV ) before 12 years of age , but the age-specific seroprevalence of VZV shows large variability between countries [3] . For example , the median age at infection ranges between 2 years in The Netherlands and 6 years in Italy [3] . The explanation of such differences is likely to be found in factors shaping the possibilities of infection in children , such as household structure [4] , schools and social behaviours . For example , the median age at varicella infection is smaller in countries where more children attend pre-school ( Figure 1 ) . It has also been shown that summer holidays are synchronous with large troughs in varicella incidence in France [5] , as children have fewer contacts during these periods . Other characteristics may also shape the epidemiology of the disease , albeit the effect may be more subtle . For example , the age at varicella infection was found to decrease in successively born children of the same household , and an increase in risk followed after school enrolment of the first-born child [6]; the risk of varicella was smaller in less densely populated areas , in agreement with other studies [7] . The effectiveness of interventions , for example vaccination or school exclusion , could be changed by such differences . Using models describing the population and its structure in sufficient detail is the only way to capture this heterogeneity and improve predictions . Indeed , computational models of disease spread have increasingly tried to more accurately describe places where population mix and infection occur , using detailed demographic data ( see for example [8]–[12] ) . As for now , these models have largely been motivated by studying pandemic influenza or bioterrorist attacks , and none incorporated aging of the population or perpetuation of the disease over several years , because the typical timescale of the epidemics lasted only a few months . Including a realistic demographic process in such models has been described as a challenge[13] , and only rarely considered in practice[14] . Indeed , census data typically provide a cross-sectional view of the population , yielding the current distribution of household sizes and age of members . Using such data , it is possible to simulate populations in which household structure and age structure conform with the census ( see [15] ) . However , including births and aging in such a population leads to some difficulties . For example , households where only one child is reported at the time of census contain a mix of single-child households where no other child will be borne , and households in which the younger siblings have not yet been born . To properly account for the demographic process , new births must occur in the latter; but this changes the overall size distribution of households , in turn producing characteristics of the simulated population that may no longer compare to the observed population . Here , the simulation of a realistic population of children over several years is described , using detailed census data . The spread of varicella is explored in this population , where infection can be transmitted in several locations ( household , school and municipality ) . Model predictions are compared with data formerly obtained from the Corsican children population in 2008 [6] , and the impact of some socio-behavioural changes are considered .
Creating households directly from census data , as described above and in [15] , does not easily allow maintaining the population structure as aging is introduced into the simulation . For instance , no simple rule allows deciding in which households new children should be borne . For example , Ajelli allocated newborns to new or existing households on the basis of probabilities calculated from current household size and age of children [14] , requiring computations at almost each new birth . To circumvent this difficulty , we estimated first the final household size ( FHS ) distribution from census data , i . e . the number of children in households where no other children will be borne , as well as the age difference ( AD ) distribution of the time lag between successive siblings in the same household . An estimate of the FHS distribution ( see Figure 2A ) was obtained by computing the total number of children in households where the oldest child was between 13 and 17 years old , assuming that no additional births were possible in these households . The observed AD was fitted by a gamma distribution ( see Figure 2B ) and was assumed independent of the birth order within the household . The Corsican population ( 300 . 000 inhabitants , comprising 35 . 000 children aged under 12 y . o . ) was split over municipalities ( n = 360 , mean area: 24 km2 ) , according to current number of households comprising of at least one children less than 12 years old . Schools ( n = 268 ) were created using data from the French ministry of national education at the corresponding locations . The school capacity ( number of children during the school year ) , and type of school ( schools for children with ages from 3 to 7 y . o , for children with ages from 8 to 11 y . o , for all children with ages from 3 to 11 y . o ) was also taken into account . Information on commuting and data from the ministry of education were used to allocate children to schools ( Figure 2E ) : The probability of going to school in municipality j when residing in municipality i was estimated as where nij were the observed commuting flows for children . Allocation to a school in the municipality was proportional to the expected size of each school , and according to type of school . The unit of simulation time was the day . The natural history of varicella was described by an M/S/E/I/R compartmental model [18]–[21] . All children were born susceptible to varicella infection . After birth , they entered compartment M with decreasing protection from infection until 6 months of age [22] , then compartment S where they were susceptible to infection . In case of infection , the child was first latent ( stage E , infected not infectious ) , then infectious and asymptomatic ( stage Ia ) followed by infectious and symptomatic when the skin rash finally appears ( stage Is ) . Eventually , all children recovered ( stage R ) . In stage M , protection by maternal antibodies was 95% until month 1 , then 75% until week 9 , 50% until week 14 , 25% until week 20 , and 5% until month 6 [22] . The duration of the latent period ( stage E ) was sampled in a discretized normal distribution with mean = 14 days and standard deviation 2 . 5 days [23] , the infectious asymptomatic period in a discretized gamma distribution with mean = 1 . 7 days and standard deviation 0 . 2 days [1] , [2] and the subsequent symptomatic period in a discretized normal distribution with mean = 5 days and standard deviation 1 day [1] , [2] . From model simulations , we calculated the following quantities ( averaged over 80 successive years ) : The parameters , , and required to run a simulation are not precisely known for varicella . The household secondary attack rate ( HSAR ) , which approximates the pairwise probability of transmission in the household , is approximately 70% for varicella [6] , [26] . As this is little changed by household size ( 69% in households with 2 children , 76% in households with 3 children [6] ) , we assumed constant pairwise transmission irrespective of household size . Solving the theoretical pairwise transmission probability in the household for this value ( where πi is the probability distribution of the infectious period duration distribution ) led to ph = 20% . The other terms were determined using maximum likelihood . More precisely , we computed the likelihood of the model based on the cumulative incidence according to age as , where is the model predicted cumulated incidence by age a , and na the number of children who have had varicella before age a out of a total of Na in a sample of Corsican children [6] . Exploration of the likelihood was done using Latin hypercube sampling [27] . Parameter ph was fixed at 0 . 2 , pi was sampled in the range 0–10−4 and the two other were sampled between 0 and 1 . The Akaike Information Criterion was used to compare models [28] . A realistic age-structured model has been implemented as described in [29] . In this simpler model , the Corsican youth population was divided in 14 age groups ( 0–0 . 5 years , 0 . 5–1 year , 1–2 years , 2–3 years , … 11–12 years , 12 years and above ) . The 0–0 . 5 year olds were assumed to be immune to infection . We used the POLYMOD data ( from the UK ) for the contact matrix . Finally , the model was fit to the Corsican cumulated incidence data by maximum likelihood[29] .
Using the RAS model , it was possible to simulate cumulative incidence data close of what was observed in Corsica . However , the fit of this model , as measured by the likelihood , was worse than that derived in the best fitting individual-based models ( Figure 3 & Table 1 ) . In the individual-based model , when the probabilities of being infected with varicella in the municipality and in the school were set to 0 ( model HE ) , it was not possible to obtain a good fit with the cumulated incidence profile with age . In the best fitting combination , even when large external transmission was allowed , the cumulated incidence ( CI ) at age 12 was 48% , short of the 90% observed ( Figure 3 ) . In this model , 65% of infections occurred outside households ( see Table 1 ) . We then included transmission in the municipality , but not in schools ( model HME ) . The fit improved , with cumulative incidence of varicella reaching 90% by 12 years of age . The percentage of infections occurring outside households and the municipality was 8% ( Table 1 ) . However , in this model , the cumulated incidence increased too quickly with age as compared to the observed data . Indeed , the CI was 32% at 2 years old and 54% at 4 years old , compared with 21% and 48% . The model with transmission in schools , but not in the municipality ( model HSE ) , provided a better global fit as judged by the AIC . However , this time , the CI increased too slowly in children aged less than 3 years old , reaching 18% compared with 20% in the real data . Moreover , the CI at 12 years old was 92% , more than the observed 89% . Finally , the model allowing transmission in both schools and municipalities ( model HSME ) provided the best fit ( i . e . smallest AIC , see Table 1 ) . The simulated cumulated incidence of varicella was almost undistinguishable from the observed data . The range of parameter values leading to small differences in AIC ( <2 ) was narrow , between 0 . 114 and 0 . 122 for pm , between 0 . 128 and 0 . 134 for ps and between 1 . 9×10−5 and 2 . 1×10−5 for pi . The proportion of infections due to external exposure was reduced at 5 . 5% ( Table 1 ) . Using the best fitting model ( HSME ) , the cumulative incidence of varicella was predicted in two special cases: in first-born and in other children; and in rural and urban settings . Figure 4 show that the model predicting cumulative incidence with age matched the characteristics of the observed data . Indeed , the age difference at varicella observed among siblings of the same family was present ( Figure 4A ) , with an increase in incidence after 3 years of age in first-born children; the difference between rural and urban settings was also present with varicella occurring at a later age in children living in more rural settings . When all levels of mixing were not allowed , the model failed to reproduce these quantities . For example , when the municipality was not included , although the simulated overall cumulated incidence was close to that observed in Corsican children , the cumulated incidence in first-born children was largely underestimated at three years old: 8% at 3 years old ( not shown ) , compared with 22% in the observed data . The average weekly incidence agreed with those reported by the Sentinelles network system [30] . The simulated incidence time series showed a reduction in incidence that was associated with holidays , with a pronounced trough during the summer period ( Figure 4C ) . However , the observed data was not entirely reproduced , as the model predicting incidence was the highest after summer holidays and decreased during the school year , when the observed data suggested that it was the reverse . In France , infected children are excluded from school as soon as varicella symptoms are recognized . We explored how this public health measure impacted the spread of varicella . Simulations were run assuming that infected children remained present in the community and school until they were cured , with the same level of infectivity as during the asymptomatic stage . As a consequence , age at varicella decreased in the whole population , with median age at infection shifting from 4 . 7 to 2 . 6 years old ( Figure 5A ) . This would also lead to almost all children being infected with varicella before age 12 , compared with approximately 90% in the field data [6] . Next , we explored how age at schooling may explain the variability in varicella age-seroprevalence between countries where vaccination is not implemented . We simulated scenarios in which age at first school enrolment was set at 2 , 4 , 5 and 6 years old . The effect of this change is presented in Figure 5B . The median age at infection increased by about 8 months for each one-year increase in age at school enrolment .
In this paper , we have shown that including detailed population structure in models of varicella transmission allowed us to reproduce the cumulated incidence of varicella according to age . A better fit was obtained than with a realistic age structured model , even using mixing matrices based on real contacts . Importantly , specific features observed in varicella epidemiology were reproduced owing to the detailed structure of the model . This included , for example , differences in age at varicella according to birth rank and place of residence , indicating that the rich structure built into epidemiologic models using census data leads to improved models regarding disease spread . Modelling varicella , or other childhood diseases , requires simulating populations over several years . Indeed , seroprevalence studies show that most infections occur during the first 12 years of life [3] , [16] , [17] . It is therefore required to simulate realistic children populations over several years . Here we showed that it could be achieved by knowing only the number of households in each municipality , the household final size distribution and the age difference distribution between successive siblings . Throughout the entire simulation , the simulated population agreed with census data regarding age of children , household sizes , and movement flows ( see Figure 2 ) . To determine the household final size distribution , we focused on households where the oldest child was between 13 and 17 . A sensitivity analysis indicated that the bottom threshold could be chosen in the range of 10–16 with little change in the estimated HFS distribution . The average final size of the household was 1 . 9 , in agreement with the total fertility rate in women who have had at least one child [31] . Other demographic processes , such as change in household structure due to divorce or remarriage , were not modelled; neither were the opening or closing of schools , which happens depending on the number of children in small municipalities . Varicella provided an excellent case study , since it is a common childhood disease in France ( as universal vaccination is not recommended ) and surveillance data is available , however the model could however easily be applied to other childhood infectious diseases . The varicella natural history description was standard [18] , with an additional split of the infectious period according to the presence of symptoms . This allowed to model the prodromic infectious phase , often reported for varicella , where infectivity increases in the few days before a rash is present [1] , [2] . As seen in the simulations , these 1 or 2 days were important for transmission: indeed , the models not including school and municipality transmission failed to reproduce the incidence data . Truly asymptomatic varicella was not modelled as it is rare [32] and presumably less infectious [33] . One issue in the present modelling was how to initialise the population regarding susceptibility to the disease . Indeed , the susceptibility of siblings or schoolmates is not independent , since the disease is transmissible . We used two approaches: ( 1 ) start with an entirely susceptible population or ( 2 ) randomly assign a susceptibility status according to the observed cumulative incidence with age [6] . In either case , after discarding the initial first 20 years of simulation the results were not sensitive to the actual method of initialisation , although the simulated cumulated incidence rate with age was more quickly stabilized with the second method . A second issue was to quantify and average over stochastic variation . Averaging over at least 80 years of simulation was required in this respect . The choice of Corsica to build the model was motivated by the availability of epidemiological data for comparison , nevertheless changing the input census data would make it possible to use the model in other settings . A child contact network was described as household , school and municipality . The detailed description of these places structured the possibilities of interaction according to age and space , in place of mixing matrices used in other models [29] , [34] , [35] . In Table 2 , we report how changing the number of contacts at school or in the municipality changed our results . There was altogether little effect on the overall fit . Unexpectedly , the proportion of varicella cases due to school contacts decreased with increasing contacts at school . However , more contacts in school led to increased introduction of cases in households and the municipalities ( spatial diffusion ) , where transmission could start , and lead to more cases . The school size ( median school size = 63 children ) further limited its overall influence . Other typical mixing places for young children may include day care , but this is almost inexistent in Corsica ( less than 200 children overall ) and was not considered here . How transmission depends on household size is a current matter of debate , and our model assumed a constant pairwise transmission in household , as suggested by [6] . We found that this hypothesis was in good agreement with the data: In households comprising 2 susceptible children before the introduction of the virus , the observed distribution of the number of cases was 1 case: 31% , 2 cases: 69% , the simulated distribution was 1 case: 27% , 2 cases: 73% ( p . value = 0 . 93 ) ; whereas in households comprising 3 susceptible children before the introduction of the virus , the observed distribution of the number of cases was 1 case: 12% , 2 cases: 23% , 3 cases: 65% , the simulated distribution was 1 case: 8% , 2 cases: 24% , 2 cases: 68% ( p . value = 0 . 99 ) . The basic reproduction number , corresponding to the number of secondary cases caused by one case in a totally susceptible population , ( R0 ) was approximately 4 ( average over 500 simulations with 1 initial random case ) . This value was in the low end among European countries [3] . There was large variation depending on whether the initial children attended school ( 4 . 4 secondary cases ) or not ( 1 . 3 secondary case ) . Using the RAS model , this estimate was close to 14 . This shows that the reproduction number is indeed very dependent on how contact are calculated , as previously noted by other authors for varicella [29] , [36] . As seen in Figure 2C , the school commuting network is essentially based on geography , most children going to school in their own or a neighbouring municipality . The difference between contacts at school and in the municipality is therefore mostly due to the age distribution of contacts . Even if the main sources of mixing were included , it was necessary to include an external force of infection . When this was not allowed , the disease rapidly went extinct especially during the summer holidays . The persistence of a disease requires sufficient a community size , as is well-known in the case of measles [37] , which may not be met in Corsica . Introducing an external source of infection ( pi ) was necessary to perpetuate varicella . Indeed , persons previously infected with the varicella virus may start shedding the virus again , for example during zona episodes , and originate transmission chains . Indeed , Zona incidence is approximately constant over time in Corsica ( see http://www . sentiweb . org ) , and together with case importation , this supports introducing an external force of infection for varicella in small populations [38] . The model required estimating only 3 parameters corresponding to daily transmission probabilities , and provided a very good fit to the data . In the best fitting case , the daily probability of pairwise transmission in the household was approximately 17% , and it was 13% in the school and 12% in the municipality . Interestingly , these probabilities are consistent with the rate of varicella transmission ( 0 . 00133/minute ) derived from models based on time use data [34] . Indeed , a daily probability of 17% is obtained for a duration of contact of 140 minutes ( calculated as ) , when the average daily cumulated time of contact between 2 children aged less than 10 years old is indeed between 100 and 200 minutes [34] . The model did not reproduce all features of observed incidence with time . Overall , the observed incidence of varicella showed an increasing trend during the school year , with the highest point being before the summer holidays ( see Figure 4C ) , but the model predicted incidence , while highly variable , that often showed the opposite trend . To properly reproduce this feature , transmission in schools , in municipalities or the “external transmission” should be lower in winter and larger in the spring: Such seasonality could be the result of impaired transmission in cold weather [39] , or of fluctuations in the incidence of zona [40] . In conclusion , detailed simulation of realistic children populations over several years may improve the study of childhood disease transmission . Further comparisons with compartmental models using realistic mixing matrices are necessary to identify the best approaches to help public health decisions . | Individual-based models of disease transmission have increasingly included detailed demographic data to more accurately describe places where population mix and infection occur . These models may help to understand in more detail heterogeneities in transmission and improve public health decisions . Here , the spread of varicella , a childhood disease , is studied in such a model where spatial and population structures are explicitly modelled . The model focuses on children , organized in households , schools and municipalities , in agreement with census data . The detailed structure of the population used in the model allows for reproducing several observed differences in the epidemiology of varicella , for example , variation in age according to birth rank and place of residence . These results support using detailed models with the eventual aim of improving decisions in public health . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"medicine",
"infectious",
"diseases",
"infectious",
"disease",
"modeling"
] | 2011 | Modelling the Effects of Population Structure on Childhood Disease: The Case of Varicella |
China has seen a massive resurgence of rabies cases in the last 15 years with more than 25 , 000 human fatalities . Initial cases were reported in the southwest but are now reported in almost every province . There have been several phylogenetic investigations into the origin and spread of the virus within China but few reports investigating the impact of the epidemic on neighboring countries . We therefore collected nucleoprotein sequences from China and South East Asia and investigated their phylogenetic and phylogeographic relationship . Our results indicate that within South East Asia , isolates mainly cluster according to their geographic origin . We found evidence of sporadic exchange of strains between neighboring countries , but it appears that the major strain responsible for the current Chinese epidemic has not been exported . This suggests that national geographical boundaries and border controls are effective at halting the spread of rabies from China into adjacent regions . We further investigated the geographic structure of Chinese sequences and found that the current epidemic is dominated by variant strains that were likely present at low levels in previous domestic epidemics . We also identified epidemiological linkages between high incidence provinces consistent with observations based on surveillance data from human rabies cases .
Rabies is a fatal zoonotic disease , posing a severe public health problem with more than 55 , 000 human rabies deaths occurring annually . 99% of all fatalities occur in developing countries [1] , [2] and Asia accounts for 80% of the worldwide total [3] . After India , China reports the second highest number of human cases , with more than 117500 recorded deaths since 1950 and three major epidemics ( 1956–1957 , 1980–1990 and 1997 to the present day ) [3] , [4] . In the majority of cases in Asia , the domestic dog acts as the main reservoir for rabies transmission , with 85%–95% of human rabies cases ascribed to dog bites [5] , [6] which in turn is a consequence of poor dog population control . Rabies virus ( RABV ) belongs to the genus Lyssavirus , family Rhabdoviridae . Previous studies indicate that globally there are six distinct clades: Africa2 , Africa3 , Indian subcontinent , Arctic related , Cosmopolitan and Asian , with the last four lineages circulating in the Asia region [7] , [8] . The Indian subcontinent clade is confined to Sri Lanka and southern India , while the arctic-related clade is widely distributed , spanning from far east Siberia to western Asia including India , Pakistan and Iraq [8]–[10] . The Asia clade is disseminated widely throughout Southeast Asian countries including China , Vietnam , Thailand , Cambodia , Philippines , Myanmar and Laos [7] , [8] . In recent years , several studies involving phylogenetic analysis of RABVs have provided some insight into the evolutionary diversity of the rabies virus within China and the association with the strains in neighboring countries . In particular , previous findings indicate that RABVs from China are closely related to those from neighboring countries , possibly sharing a common ancestor [7] , [11]–[14] . However , despite the severity of the problem , there has been no extensive investigation of the impact of the current rabies epidemic in China on surrounding regions or , conversely , the influence of these regions on the epidemic . Therefore , to investigate this question further we conducted a comprehensive phylogeographic analysis to explore the phylodynamics of rabies isolates from China and neighboring countries .
From 2003 to 2010 , we collected dog brain samples from provinces and municipalities in China where rabies was endemic or emerging , with regions selected as described previously [15] . Specifically , there were two stages to the surveillance program . In the first stage of the program the goal was to examine the infection rate in the general dog population in high incidence regions . In this stage , samples were collected from local meat markets . As dogs are brought to the meat market from the surrounding area ( i . e . of the order of a few square kilometres ) and as there is no transportation of dogs to other markets this represented a random sample of the dog population for this region . In the second stage of the surveillance program , isolates were primarily collected from suspected rabid animals ( wildlife or domestic ) or human related cases . All samples were tested for RABV using direct fluorescent assay ( DFA ) as described previously [6] . Total RNA was extracted by Trizol reagent ( Invitrogen , Burlington , ON ) according to the manufacturer's instructions . Based on this , 84 samples tested positive for rabies virus . Complete RABV N ( nucleoprotein ) gene sequences were determined using RT-PCR and sequencing reactions as described elsewhere [8] , [16] . RABV sequences were collected from both China and neighboring Asian countries and , based on the N sequence , six different datasets were created that provided a compromise between number of sequences , alignment length and range of isolation date and geography . For the Asian phylogenetic analysis , dataset 1 comprised 110 sequences , spanning the full 1350 bp of the gene , dataset 2 consisted of 177 N sequences spanning nucleotides 1032–1350 and dataset 3 consisted of 312 sequences spanning nucleotides 64–399 , two highly variable regions of the gene . Two additional datasets were created to investigate the relationship between isolates from China and countries close to its southern border . Dataset 4 comprised sequences spanning nucleotides 40–399 and dataset 5 contained sequences spanning nucleotides 1033–1329 . Finally , to explore the phylogenetic diversity of RABVs in China , we retrieved all Chinese full length N sequences from Genbank . After combining with our newly acquired sequences and removing identical sequences from the same province , we composed a sixth dataset of 232 complete China N gene sequences ( nucleotides 1–1353 ) . A complete list of the new sequences and their background information , together with additional sequences retrieved from Genbank is given in Table S1 and the composition of the datasets are summarized in Table S2 . A map of all geographic regions incorporated in the study and the geographical location of all isolates in datasets 4 and 5 ( generated using the Google Maps API ) is shown in Figure 1 . For each of the datasets 1–3 , a maximum clade credibility ( MCC ) rooted tree was constructed using the Bayesian Markov Chain Monte Carlo ( MCMC ) methods implemented in the BEAST package ( v1 . 6 . 2 ) [17] . A relaxed ( uncorrelated lognormal ) clock model , a general time-reversible nucleotide substitution model with rate heterogeneity and an invariable sites ( GTR+I+Γ4 ) model of substitution determined by jModelTest [18] , and a constant coalescent model were used to conduct the analysis . For each dataset , the MCMC analysis was run for 50 million generations to ensure sufficient mixing . Convergence of parameters estimates was checked using TRACER ( http://beast . bio . ed . ac . uk/ ) and was indicated by an effective sample size ( ESS ) >200 . From this approach we derived the phylogenies of each dataset . Posterior probability values were presented as indicators of nodal support . To assess the geographic structure of RABV in Asia in a more quantitative manner , we examined the posterior distribution of genealogies within the trees produced in the previous step using the Bayesian Tip-Significance testing ( BaTS ) software tool [19] . For datasets 1 and 2 , sequences were assigned uppercase letters to define their state according to their geographic location ( See table 1 and table 2 for details ) . To determine the strength of geographical association with sampling locations across the entire tree , BaTS calculates the association index ( AI ) [20] and the parsimony score ( PS ) [21] , as well as the maximum monophyletic clade size ( MC ) [19] to assess the correlation for specific locations . The PS score takes a score between 1 and n , where n is the number of tips in the tree; a PS of 1 corresponds to complete phylogeny-trait association ( a measure of the extent to which neighboring taxa in a phylogenetic tree share a character of interest , in this case the geographical location ) . The AI is a sum across all internal nodes and is defined byWhere fi is the frequency of the most common trait ( here the geographical location ) among the tips subtended by node i , and mi is the number of tips subtended by i , Low AI values correspond to strong phylogeny-trait associations . The monophyletic clade ( MC ) statistic provides a measure of the phylogeny-trait correlation for each trait and is defined byWhere mi is the number of tips subtended by node i and Ii = 1 if all tips under i have trait x and 0 otherwise . For this statistic , higher MC values indicate stronger phylogeny-trait associations . As BaTS performs the association test from the credible set of trees generated by BEAST in the previous step , it can also estimate the uncertainty associated with the predictions . The BaTS analysis indicated there was clustering of geographical states according to region and strong phylogenetic trait association , suggesting the possibility of the occurrence of translocation events . To test how the rabies virus was dispersed across the geographic region of Asia , each isolate was assigned the following lowercase letters to define their state according to their country of origin ( a: Afghanistan; b: Cambodia; c: China; d: India; e: Indonesia; f: Japan; g: Kazakhstan; h: Laos; i: Mongolia; j: Myanmar; k: Nepal; l: Pakistan; m: Philippines; n: Russia; o: South Korea; p: Sri Lanka; q: Thailand; r: Vietnam ) and RABV translocation events were traced through phylogenies derived from the Asian datasets utilizing the program MigraPhyla [22] , [23] using both accelerated transformation of character states ( ACCTRAN ) and delayed transformation of character states ( DELTRAN ) parsimony optimization methods . To estimate the reliability of the predicted translocation events , a Monte Carlo test of 10 , 000 trials was used to randomly distribute the same localities across the tree tips and these ‘random’ trees were then examined for translocation events . The P value for a translocation event between two locations was estimated based on the number of times the translocation event was observed in the original tree compared to the number of times the events occurred in the ‘randomized’ trees . To correct for multiple tests and the sparsity of the generated translocation matrix , a sparse false discovery rate ( sFDR ) correction was applied to test the significance of the estimated P values . The sFDR cutoff was set by P value rank× ( 0 . 05/total of migration events ) >P value . Translocation results were visualized using the Circos software package [24] . To investigate the relationship between isolates from China and from regions close to the South China border , two ML trees based on datasets 4 and 5 respectively were constructed . The datasets were comprised of sequences from countries adjacent or close to the South China border ( Figure 1 ) . Sequences from India , Bhutan and Bangladesh were not included as these countries border Tibet and Sichuan which only began to record rabies cases in 2011 . Two datasets were used because different surveillance programs in the various countries sequenced different regions of the N gene and it was not possible to generate a single comprehensive dataset representative of the entire geographic region . The datasets are summarized in Table S1 and S2 . Using the same nucleotide substitution model as datasets 1 to 3 , we used dataset 6 to reconstruct a MCC tree using BEAST v . 1 . 6 . 2 [17] . As two major clades of Chinese isolates were identified in both the Asian and China analyses and accounted for most sequences in dataset 6 , we selected these for further investigation . To determine the viral dispersion among provinces in China , a non-reversible discrete phylogeography model was applied to each of these two lineages , with the sampling provinces of these Chinese isolates acting as the discrete states [25] . As the geographic origin of RABV remains unclear , we used a Bayesian stochastic search variable selection ( BSSVS ) method which employed a Bayes factor test to identify the best supported migration pathways between geographic locations ( i . e . provinces ) that were epidemiologically linked [25] . The SPREAD program [26] was used to produce an animation of the results in the keyhole markup language ( KML ) to illustrate the epidemiological links , which can be viewed by Google Earth ( http://earth . google . com ) .
Consistent with previous studies [7] , [8] , phylogenetic analysis of datasets 1 , 2 and 3 revealed six distinct clusters in Asia: Indian subcontinent; Cosmopolitan; Arctic-related; Southeast Asia ( SEA ) SEA1; SEA2; and SEA3 , all of which are supported with strong a posteriori probability values ( Figure 2a , 2b and Figure S1 ) . The geographic composition of these clades is also consistent with previous results . The Indian subcontinent cluster only contains isolates from India and Sri Lanka [8] . The Cosmopolitan cluster comprises isolates from a much broader region of Asia including Russia , Kazakhstan , Mongolia , China and India . Interestingly , an unpublished strain isolated from dog in Pantnagar in Uttarakhand Northern India ( HQ829841 ) was grouped with isolates from China rather than with those from India , but lack of background information makes it difficult to determine the significance of this result . The Arctic-related cluster is comprised of strains circulating in Russia , Mongolia , South Korea , China , India , Nepal as well as Afghanistan and Pakistan , and other publications also report strains from Middle Eastern countries such as Iran and Iraq placed within this clade [9] , [11] . The SEA1 cluster is confined to strains from China and Indonesia and there is clear subdivision according to geographic origin . The SEA2 cluster includes isolates from China and Philippines and is similarly split into two subgroups according to country of origin . The SEA3 cluster contains isolates from southwestern China and the biogeographical region referred to as the Indochina peninsula or approximately equivalent to Mainland Southeast Asia , similar to the Asian 2 group reported in another study [7] . Overall , our results agree with previous studies , but the structure within each clade offers new insight from a geographical and phylogenetic perspective . Despite the obvious correlation between China and neighboring countries , the distinct grouping of Chinese isolates suggest that the Chinese strains in the Cosmopolitan , SEA1 and SEA2 clusters , which contain the majority of the Chinese isolates , have evolved independently from their counterparts from neighboring countries , regardless of the collection date of isolates . In our phylogeny-geographic origin association analysis , we grouped countries according to their geographic proximity and examined their dispersion within the predicted trees by calculating the PS and AI indices . The results for dataset 1 and dataset 2 are summarized in Table 1 and Table 2 respectively . For each dataset , a measure of the overall tree structure is provided by the AI and PS statistics; these can be interpreted by comparison with the associated null value , which is the corresponding statistic calculated from a null distribution of trees randomly selected from the posterior sample of trees generated by BEAST . In both cases , the AI and PS statistics for the estimated trees are much less than the null values at P = 0 , indicating strong support for the presence of geographic structure and suggesting the isolates are mostly clustered according to their geographic origin . A measure of the phylogeny-trait association for each location is provided by the MC statistic and is calculated for each location in both datasets ( Table 1 and Table 2 ) . The MC statistic is positively correlated with the strength of phylogeny-trait association and values greater than the null value indicate strong association . All of the defined geographic regions show significant support ( P value<0 . 001 ) for population subdivision with the exception of region D ( Japan ) which indicates gene flow from other regions . Region C only contains isolates from China , and the large MC value with significant statistical support indicates a preponderance of in situ evolution within China . The results of the MigraPhyla translocation analysis of the Asian datasets are summarized in Figure 3 . After applying a sparse false discovery rate ( sFDR ) correction , the remaining translocation events inferred from the Asian datasets phylogenies indicate that China and Russia play an important role in transmitting RABVs across the Asian region . The following significant translocation pathways were identified ( Figure 3a/dataset 1: Russia to Mongolia , South Korea , China , Japan , Afghanistan , India and Nepal; Kazakhstan to Mongolia and Russia; Afghanistan to Pakistan; and Thailand to Cambodia and Viet Nam; Figure 3b/dataset 2: Russia to South Korea , Kazakhstan and Mongolia; India to Afghanistan; Afghanistan to Pakistan; and Thailand to Cambodia and Viet Nam ) . Among these significant translocation events , dispersal mainly occurred among geographically adjoining countries in all three datasets ( Figure 3 and Figure S2 ) , with the exception of Russian isolates which , according to dataset 1 , were predicted to have spread extensively to distant regions ( Figure 3a ) . However , many countries in dataset 1 are represented by only a few isolates which may have biased the result; this conclusion is supported by the results for datasets 2 and 3 ( Figure 3b and Figure S2 ) which , with the exception of South Korea , only retain translocation events from Russia to adjacent countries . However , these larger datasets still predict translocation events from China to every neighboring country ( although these migration events do not have high statistical support ) but relatively few predictions of translocation events in the opposite direction . i . e . , China has an impact on rabies in neighboring countries , but cases imported into the country have negligible effect on the epidemic in China , which appears to be driven by internal events . To investigate the possibility that the absence of statistically significant translocation events between China and neighboring countries was simply a consequence of bias towards China isolates in the datasets , we generated two additional datasets , 4 and 5 , comprising sequences from countries adjacent , or close to , the South China border . Surveillance data indicates that the border provinces of Guangxi , Guizhou , Guangdong and Henan in the Southwest represent the majority of early cases , so the majority of Chinese isolates were selected from these regions [27] . The ML trees for these two datasets are shown in Figure 4a and 4b respectively . The country of origin of the sequences are represented by the height and colour of the bars on the outside of the tree . In both trees the sequences are grouped into four major clades SEA1/China I , SEA2/China II , SEA3/China VI and Cosmopolitan/China III , consistent with their classification in the trees in Figure 2 and Figure S1 . If the national borders failed to halt the spread of rabies , we would expect to find a close evolutionary relationship between China isolates from the current epidemic and isolates from other countries in South East Asia ( Philippines , Laos , Myanmar , Thailand , Cambodia and Viet Nam ) . Furthermore , surveillance studies indicate that SEA1 is the dominant strain in the current Chinese epidemic and SEA2 is associated with the previous epidemic that occurred during the 1970s and 1980s [27] , [28] ( and Table S3 and unpublished data ) . Thus , if there was any spillover from the current epidemic into neighboring countries , we would expect to find some isolates from other countries placed in the SEA1 clade . After removing duplicate entries , the two datasets represent a total of 550 unique isolates . The majority of these isolates are placed in SEA3 , with the remainder dispersed between SEA2 and Cosmopolitan . A single group of 11 Viet Nam sequences isolated in the north of the country between 2007 and 2009 from both Human and Dog are located in the SEA1 ( Figure 4a insert ) . However , their branch point from the China sequences , which include isolates dating back to 1969 , indicates they are from a distinct lineage that is not associated with the current China epidemic . Interestingly , there is a single Viet Nam isolate placed in the middle of the Chinese China I sequences . Upon further investigation , it was found that this sequence was isolated from a human subject in Lang Son city in Lang Son province , which is the most important border crossing between Viet Nam and China , although no further information is available regarding the subject . Given this is a single Viet Nam isolate within the China branch , that all the Vietnamese dog isolates are in a separate branch , and considering the volume of cross border traffic between Lang Son and Pingxiang city ( ? ? ? ) in Guizhou province on the Chinese side , it seems probable that this infection event occurred within China . The above analyses indicate that SEA1 , the dominant variant rabies strain in China , has not spilled over into neighboring countries . However , to further explore the diversity of the rabies virus in China , we conducted a comprehensive phylogenetic analysis using all available Chinese RABV N sequences ( dataset 6 ) . Bayesian coalescent analysis of RABVs from China identified six distinct lineages ( China I–VI ) with high posterior value support ( Figure 5 ) , which is in accordance with previous studies using complete G and N sequences [12] . The China_VI lineage includes a few isolates that originated from Guangxi and Yunnan provinces in southeastern China and which are closely related to RABVs from countries in the Indochina peninsula/Mainland Southeast Asia ( corresponding to the SEA3 cluster in the Asian analysis results ( Figure 2a and 2b ) . The China V lineage only contains three isolates that were collected around 20 years ago; this lineage is probably associated with an earlier epidemic ( unpublished data - manuscript in preparation ) and died out due to a population bottleneck or remains present at low levels and cannot be easily sampled . The China IV lineage consists of samples that are only found in Inner Mongolia and are closely related to the Arctic_related clade [9] , [11] ( Figure 2 ) . The China III lineage has isolates collected across the country and corresponds to the Cosmopolitan clade [29] . The lack of diversity in this clade is highlighted by identification of four isolates in this clade possessing 99 . 4% nucleotide similarity despite being collected from four distant provinces ( Guizhou , Hunan , Henan and Jiangsu ) in the same year . The China I and China II lineages are representative of most RABV isolates prevalent in China over the last decade and correspond to the two Chinese subgroups samples in the SEA1 and SEA2 clusters in Figure 2 . Both of these lineages can be further divided into five clearly defined sublineages with varying support . In several sublineages ( China IIe , China Id and China Ie ) , recently acquired isolates shared a common ancestor with basal old strains from late 1980s or early 1990s , suggesting that those current prevalent RABV strains might evolved from earlier epidemic strains . Interestingly , 100% nucleotide identity was observed in eight new collected isolates ( AH12/Anhui/2005 , CGZ0518/Guizhou/2005 , CGX0516/Guangxi/2005 , CHN0813/Hunan/2008 , CJS0621/Jiangsu/2006 , CJX0902/Jiangxi/2009 , CYN0924/Yunnan/2009 , CZJ0804/Zhejiang/2008 ) from eight provinces spanning from 2005 to 2009 in the China_IIb sublineage . In our Bayesian coalescent approach of China RABVs , the mean rate of nucleotide substitution was estimated to be 5 . 23×10−4 substitutions per site per year ( 95% HPD values , 3 . 94×10−4–6 . 68×10−4 ) , which agrees with previous estimates [7] , [30] . Estimates of the Time to the Most Recent Common Ancestor ( TMRCA ) indicates that current strains diverged around 1711 CE ( 95% HPD values , 1399–1869 ) , concordant with previous estimation using N gene , but slightly earlier than the estimate for the G gene [7] , [13] , [31] . The divergence time of China I and China II lineages were determined to be 1907 and 1934 , respectively , i . e . , these two lineages evolved independently of external RABV strains over a long time span undergoing localized evolution . The phylogeographical analyses of China I and China II lineages identified several provinces that appear to be epidemiologically linked . The transmission pathways for these two clades with Bayes factor greater than 3 are shown in Figure 6a and 6b respectively . Notably , China I contains many more linkages than China II , which suggests that this lineage plays the dominant role in the spread of rabies in China . Figure 6a also indicates that east China appears to be not only epidemiologically related to adjoining provinces but also to distant provinces , and seems to act as an epidemic hub for transmission of rabies virus to other regions , which is consistent with results from our previous analysis [15] . Other long distance transmissions of rabies virus can also be identified as well as translocation events between neighboring provinces . For example , Shaanxi province has previously experienced very low rabies incidence but cases have begun to increase in recent years . Figure 6a indicates a strong epidemiological linkage from Shaanxi to Sichuan and from Sichuan to Yunnan . This is consistent with surveillance data for human rabies cases which show dissemination of the virus from southwest China to neighboring provinces and into regions such as Shaanxi in the northern part of the county that have previously been incident free for several years [32] . For both clades , rather than a random dispersion of epidemiological linkages , there appears to be a general trend of vertical transmission ( Shandong-Guangdong , Hebei-Fujian , Shandong-Zhejiang ) and horizontal transmission ( Yunnan-Shanghai , Guizhou-Shanghai , Hunan-Shanghai ) which is also consistent with human rabies surveillance data which highlights a flow of cases from high incidence regions in the south of the country to medium and low incidence regions [32] .
Rabies remains a serious public health problem throughout Asia . Nevertheless , the current goal is to eliminate rabies in China by 2020 ( a target set at the ASEAN plus 3 rabies conference ) . Thus , effective and feasible long term programs for prevention and control are essential . Nevertheless , the situation can vary among countries or regions due to local problems or specific conditions [33] , [34] and understanding these differences may aid the development of effective control measures . In this study we performed a detailed phylogenetic analysis of RABVs in Asia using a comprehensive dataset selected from all currently available samples , as well as new samples collected as part of a national surveillance program , with a view to obtaining a better understanding of the role of different countries in the distribution of Asia rabies . There are already many published reports on the distribution of rabies within China and from the broader perspective of Asia . [6] , [14] , [35] , [36] . These studies have investigated the phylogenetic relationship between Chinese strains and strains from other Asian countries and have demonstrated several isolates share a close phylogenetic relationship . However , the degree of exchange between neighboring countries and the relevance to the current rabies epidemic in China remains unclear . In our analysis we have examined the relationship between China and its Asian neighbors in far greater depth by investigating the geographical structure of the estimated phylogenies to try and interpret the contribution of specific regions to the observed epidemic in China , and conversely , the impact of the rabies epidemic in China on neighboring regions . One of the limitations of previous studies is the restricted number of samples that have been isolated from many Asian countries . In this work , we attempted to overcome this problem by constructing multiple datasets based on different regions of the N gene which allowed us to incorporate a broader range of isolates; the results for each dataset were consistent with those obtained with full gene sequences , indicating our results were robust . Phylogenetic analysis indicates that geographic structure is the defining feature of the tree and that RABVs are strongly clustered according to their geographic origins . The Chinese isolates could be classified into two types: type A strains comprise isolates that were mixed in with strains from neighboring countries , indicating they shared a close evolutionary relationship; type B strains , although placed in clades with other Asian strains , formed distinct subclades that only consisted of Chinese sequences . For the type A isolates , the majority of Chinese isolates ( from Guangxi and Yunnan provinces in Southwest of China ) belonged to the SEA3 clade and clustered with isolates from countries in the Indochina peninsula/mainland Southeast Asia region , suggesting this might be a convergent region for RABV panmixis due to frequent commerce including animal trade [12] , [37] . Additional Chinese RABV isolates from other clades were also clustered with isolates from other countries but their small number suggests these represented sporadic events . The majority of isolates were of type B strains . One major lineage of currently circulating Chinese rabies strains shared a common ancestor with those from Philippines in SEA2 clade , while a second Chinese lineage appears closely related to strains from Indonesia in SEA 1 clade . The estimated date of the TMRCA of these strains of 1907 and 1934 respectively , coincides with of historical emigration from China to Southeast Asian Countries [38] , suggesting some association might exist between emigration and the transmission of RABVs between these countries [6] , [35] . The presence of distinct clades implies that , after adapting to local hosts and environment , the Chinese RABV strains evolved separately , i . e . without gene flow in our out of the country , to become the predominant strains associated with the current rabies epidemic in China . Translocation analysis of RABVs between China and other Asian countries illustrated that gene flow of RABV principally occurred amongst geographically adjacent countries . However , although several translocation events from China to other countries were predicted , all of them lacked strong statistical support . While the translocation analysis should be interpreted with caution due to possible sampling bias ( due to the disproportionately high number of China isolates ) and the analytical method ( translocation events are based on best estimates of ancestor states which are not necessarily unique ) , our results are further supported by the classification of isolates from countries bordering south China . Thailand and Viet Nam in particular have comprehensive surveillance programs , but out of 550 isolates from the sampled countries , only 11 sequences ( from Viet Nam ) were grouped with Chinese sequences from the dominant variant strain , and these were in a separate and distinct branch . The identification of a single Viet Nam isolate within the Chinese sequences of China I clade is a cause for concern , although it does seem probable that the subject was infected within China . Nevertheless , it would be prudent to closely monitor the rabies situation in Lang Son city in Viet Nam , if similar cases were found in the canine population , then this would be evidence of spillover . Nevertheless , in spite of the scale of the epidemic within China , it appears that , currently , few cross border translocation events occur . Although China occupies a large geographic area bordering many countries , with the exception of arctic-related strains introduced from Russia to Inner Mongolia , there has been no major influx of rabies cases from outside China . Conversely , despite the large number of rabies cases currently experienced in China we identified relatively few translocation events . More importantly , these events originated from clades that are not significantly associated with the current epidemic . Having established the current epidemic is evolving independently of neighboring countries , we also investigated the dispersion and genetic variation of the virus within China . The phylogeographic analysis of the Chinese RABV isolates dataset identified six lineages existing in China with an isolation date ranging from 1969 to 2010 , spanning the current and previous epidemics [39] . Two major lineages ( China I and China II ) account for most of current rabies epidemic . Consistent with the Asian phylogeography analyses , these two lineages are highly localized , experiencing infrequent gene flow from outside mainland of China . The China III lineage corresponds to the Cosmopolitan clade , which had been predicted to a consequence of global colonization from Europe between the 15th to 19th century [40] . However , from the current dataset , it appears that this lineage only exists at a relatively low level and is associated with occasional events rather than significantly contributing towards the current epidemic . These three lineages can generally be classified as type B strains as described above . The remaining three lineages ( China IV to VI ) are more representative of type A strains and also , based on the total number and date of isolates , appear to contribute little to the current epidemic . Previous studies have demonstrated the role of humans in dispersion of rabies in Africa [41] , [42] . In particular , estimates of viral gene flow in localities in Algeria and Morocco in the Talbi study were 2 to 4 times higher than corresponding estimates in wildlife [42] . Although this highlights the importance of anthropogenic influences , it is difficult to make a direct comparison between our results and the Talbi study . Although the samples were collected over a 20 year period in Algeria and Morocco , there are marked socio-economic differences in the geographic regions . Firstly , the populations of Tunisia and Algeria are 10 and 35 million respectively , compared to 1 . 3 billion in China . Secondly , between 1995 and 2011 , the estimated Gross Domestic Product ( GDP ) of Tunisia and Algeria increased from $18billion to $45billion and from $40billion to $180 billion respectively . Over the same time period , the GDP of China increased from $730 billion to $7 . 3 trillion . It is the rapid economical expansion in China over the last twenty years that has probably had the most significant impact on the spread of rabies and , ironically , on its control as more funds have become available for vaccination programs , education and subsidies for post exposure treatment . Prior to the plan for economic reform plan instigated by Deng XiaoPing , travel was more restricted and more commonly at the local level . Long distance travel was generally by train or bus and large scale transportation of goods only began to increase as the industrial infrastructure expanded . As the economy grew and relocation was more straightforward the population became more mobile . This likely facilitated the spread of RABV as people moved from villages to towns and cities , or between cities , transporting their dogs as part of the relocation process . Long distance relocation may explain our identification of identical N gene sequences from eight difference and geographically distant provinces . This is also supported by recent reports of rabies cases in Beijing where infected dogs were brought to the capital by migrant workers . On the other hand , dog meat markets likely aid the establishment and dissemination of RABV within a local region , as large numbers of dogs are able to roam freely . However , it is improbable they are associated with long range dissemination of the virus as , in general , there is no transportation of dogs occurs over long distances . News reports in the foreign press featuring truckloads of animals in cages are related to large scale operations that are only located in the major cities . Within each facility , dogs are kept within compounds or cages and moved to market in a matter of days , thus they are unable to contribute to the spread of the disease and only contribute sporadic cases . In 1985 , a national rabies control and prevention program was implemented , and by 1996 rabies cases had decreased to 159 [39] , [43] . However , after this point the number of cases rapidly increased and a new epidemic emerged in the country . Available data showed that at least three distinct RABV lineages survived the control program and successfully reemerged , suggesting the presence of multiple reservoirs to allow RABVs' persistence over an extended period . In China , domestic dogs served as the main reservoir for RABVs with wildlife such as ferret badger also identified as a reservoir but playing an underdetermined role in the epidemic [44] , [45] . Bats act as an additional potential reservoir of RABV [46] although their role has yet to be investigated in the current epidemic . All of these factors further complicate the task of rabies control and long-term support coordinated at the national level is key to the success of such efforts . Based on surveillance data and epidemiological surveys from the past decade , new regulations on rabies control have been drafted by the Ministry of Agriculture and Health in China . These new regulations place emphasis on rabies control at the source ( such as vaccination of domestic animals , especially in rural area ) and have already proved to be effective , as seen from the reduction in rabies cases in high incidence provinces in recent years [32] . Also , trial dog vaccination programs implemented in certain high incident regions in southwest provinces in China have also proved effective in controlling rabies . In the next phase of the program , vaccination will be extended to additional regions to incorporate more of the dog population with the aim of building up a vaccination barrier to combat rabies spread . The rapid dispersal of rabies cases across the country indicates there are efficient transmission routes to facilitate dissemination of the virus . Evidence of the role of RABV transmission via human intervention and translocations has been well documented [8] , [42] , [47] and our predicted horizontal and vertical epidemiological linkages between provinces are consistent with the observed dispersion of the virus according to human rabies surveillance data [32] . Nevertheless , although there have been many reports regarding the recent spread of RABV across China , an detailed investigation of the impact of the epidemic in the context of Southeast Asia has yet to be considered . Our observation of a dominant variant strain that is unique to China is significant in that it suggests that neighboring countries have not been seriously impacted by the epidemic . In spite of the increasing trade between China and other countries in South East Asia , it further suggests that current border controls remain effective at restricting the passage of infected animals . The filtering of rabies cases at national borders shows that it is possible to limit the spread of the virus if suitable barriers exist and these findings may provide guidance for further determining effective measures for rabies control within China and to meet the goal of eliminating of rabies in China by 2020 . | Rabies as a fatal zoonotic disease continues to be a public threat to global public health . After India , China reports the second highest number of human cases , with more than 117 , 500 deaths and three major epidemics since 1950 . China remains in the middle of the third epidemic . In this work we investigate the impact of China on rabies in South East ( SE ) Asia . We collected nucleoprotein sequences from samples isolated throughout SE Asia and investigated their phylogenetic and geographic relationships . Our results indicate that clear geographic patterns exist within rabies virus in SE Asia , with isolates mainly clustered according to their geographic origin . While we found evidence of the sporadic exchange of strains between neighboring countries , the major strain responsible for the current Chinese epidemic does not appear to spread to neighboring countries . Our findings suggest that national geographical boundaries and border controls act as effective barriers to halt the spread of rabies from China into adjacent regions . We further investigated the geographic structure of Chinese sequences and found the current epidemic is dominated by variant strains that likely evolved from previous domestic epidemics . Our study provides valuable insight for rabies control and prevention in China and SE Asia . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"medicine",
"biogeography",
"public",
"health",
"and",
"epidemiology",
"spatial",
"epidemiology",
"microbiology",
"rabies",
"phylogenetics",
"neglected",
"tropical",
"diseases",
"population",
"biology",
"infectious",
"diseases",
"epidemiology",
"biology",
"evolutionary",
"systematics",
"viral",
"evolution",
"ecology",
"virology",
"viral",
"diseases",
"evolutionary",
"biology"
] | 2013 | National Borders Effectively Halt the Spread of Rabies: The Current Rabies Epidemic in China Is Dislocated from Cases in Neighboring Countries |
Epithelial cells are characterized by apical-basal polarity . Intrinsic factors underlying apical-basal polarity are crucial for tissue homeostasis and have often been identified to be tumor suppressors . Patterning and differentiation of epithelia are key processes of epithelial morphogenesis and are frequently regulated by highly conserved extrinsic factors . However , due to the complexity of morphogenesis , the mechanisms of precise interpretation of signal transduction as well as spatiotemporal control of extrinsic cues during dynamic morphogenesis remain poorly understood . Wing posterior crossvein ( PCV ) formation in Drosophila serves as a unique model to address how epithelial morphogenesis is regulated by secreted growth factors . Decapentaplegic ( Dpp ) , a conserved bone morphogenetic protein ( BMP ) -type ligand , is directionally trafficked from longitudinal veins ( LVs ) into the PCV region for patterning and differentiation . Our data reveal that the basolateral determinant Scribbled ( Scrib ) is required for PCV formation through optimizing BMP signaling . Scrib regulates BMP-type I receptor Thickveins ( Tkv ) localization at the basolateral region of PCV cells and subsequently facilitates Tkv internalization to Rab5 endosomes , where Tkv is active . BMP signaling also up-regulates scrib transcription in the pupal wing to form a positive feedback loop . Our data reveal a unique mechanism in which intrinsic polarity genes and extrinsic cues are coupled to promote robust morphogenesis .
Epithelial cells are characteristically polarized with apical-basal plasma membrane polarity , which is crucial for tissue homeostasis [1 , 2] . Many of intrinsic factors underlying apical-basal polarity have been identified to be tumour suppressors [1 , 2] . Multiple distinct but interacting groups of proteins , including the partitioning defective ( PAR ) proteins , the Crumb ( Crb ) complex and the Scribbled ( Scrib ) complex , play crucial roles in regulating epithelial polarity [1] . Among these , the Scrib complex , composed of Scrib , Discs large 1 ( DLG1 ) and Lethal giant larvae ( LGL ) [3] , has previously been identified as a basolateral determinant of epithelial cells by instructing epithelial-specific cytoskeletal rearrangements , the distribution of apical proteins , and polarized trafficking machinery [3–5] . Strong loss-of-function ( LOF ) of the Scrib complex causes tumor-like growths [4 , 5] . Decapentaplegic ( Dpp ) , a conserved Drosophila bone morphogenetic protein ( BMP ) -type ligand , plays crucial roles in various developmental contexts [6] . The Dpp ligands bind to the type I receptor Thickveins ( Tkv ) and type II receptor Punt . This stimulates Tkv to phosphorylate the transcriptional factor Mad . Upon phosphorylation , Mad binds co-Smad Medea and then translocates to the nucleus for transcriptional regulation of target genes . During pupal stages , Dpp is produced in the longitudinal veins ( LVs ) of the pupal wing and trafficked into the posterior crossvein ( PCV ) region from around 18 hr after pupariation ( AP ) to establish long-range signaling [7] . Prior to dpp transcription in the PCV region at around 26 hr AP [8] , long-range BMP/Dpp signaling appears to instruct PCV patterning and differentiation , since either loss- or gain-of-function of Dpp trafficking is sufficient for causing crossveinless or ectopic crossvein phenotypes [9 , 10] . Dpp trafficking is supported by the BMP binding proteins short-gastrulation ( Sog ) and crossveinless ( Cv ) to direct Dpp ligand localization in the PCV region [7] . Since the Dpp trafficking mechanism appears to limit ligand available in the PCV region [7 , 9] , systems that optimize the signaling response to limiting external ligands must be crucial . Thus PCV formation serves as a unique model to address molecular mechanisms underlying the optimization of BMP signaling [6 , 11–15] . Epithelial morphogenesis is one of the key processes in animal development and is frequently regulated by highly conserved signaling molecules . To promote robust morphogenesis , ensuring precise interpretation of signal transduction as well as spatiotemporal distribution of extracellular ligands are important . Here , using PCV formation in Drosophila as a model , we demonstrate that the basolateral determinant Scrib complex is required for PCV formation through optimizing BMP signaling . Scrib regulates Tkv localization by facilitating its meeting extracellular ligands , and subsequently is involved in Tkv internalization to Rab5 endosomes . Our data suggest a mechanism that optimizes the signaling response to limiting external ligands for patterning and differentiation of epithelia .
In a screen for novel components involved in PCV formation , we identified the components of the Scrib complex as candidates . To investigate how Scrib regulates PCV formation , we analyzed the Scrib complex in the Drosophila pupal wing using a conditional knockdown approach [16] . We observed that knockdown of scrib , dlg1 or lgl in the pupal wing blade disrupted PCV formation ( Fig 1A–1D and S1A Fig ) . Notably , knocking down Scrib complex components in this manner did not disrupt cell polarity , as shown by the apical marker aPKC and the basolateral marker DLG1 ( Fig 1E and 1F ) . Thus a partial reduction in Scrib complex gene expression is sufficient to disrupt PCV morphogenesis but not to disrupt apical-basal polarity , indicating that loss of the Scrib complex disrupts PCV formation by another mechanism . To understand whether the PCV-less phenotypes resulted from reduced BMP signaling , phosphorylated Mad ( pMad ) staining ( a readout of BMP signaling [17] ) was used to examine the effects of scrib knock down in the PCV region . Knockdown of scrib in the wing caused a loss of BMP signaling in the PCV region ( Fig 1G and 1H ) . scrib or dlg1 mutant clone analysis further suggested that Scrib and DLG1 are required for BMP signaling in the PCV region in a cell-autonomous manner ( Fig 1I and S1B Fig ) . Loss of BMP signal in scrib clones in the PCV region appears not be caused by perturbed ligand trafficking . GFP-Dpp ligands expressed in scrib clones are rather diffusible ( S1C–S1F Fig ) . These results support that Scrib is required for optimizing BMP signal in the PCV region but not for facilitating BMP ligand trafficking . Interestingly , Scrib proteins are enriched in the vein primordial cells at 24 hr AP , where BMP signaling is positive ( Fig 2A ) . These observations indicate that the Scrib complex may be up-regulated by BMP signaling and may constitute a feedback loop . To test this idea , we assessed pMad signaling and Scrib protein staining in pupal wings at different time points . At 18 h AP , the pMad signal is widely observed in the prospective LV regions , but faint in the PCV region , in line with previous reports that PCV morphogenesis initiates around 18h AP ( Fig 2A ) [7 , 8] . pMad staining becomes refined in LVs , and intensifies in the PCV around 20–24 hr AP ( Fig 2A–2E ) . Scrib is uniformly distributed throughout wing tissues at 18 h AP , when BMP signaling is low . Scrib then gradually accumulates in the vein progenitor cells while BMP signaling becomes refined at 20–24 hr AP ( Fig 2A–2E ) . Based on these data , we postulate that the BMP/Dpp signal may regulate scrib transcription . To study whether transcriptional levels of scrib and dlg1 are also susceptible to BMP/Dpp signal activity , we performed quantitative PCR ( qPCR ) and found that scrib and dlg1 are transcriptionally regulated by BMP/DPP signaling ( Fig 2F ) . Our data also indicate that scrib and dlg1 transcription are up-regulated by BMP signal in future intervein cells as well as wing vein progenitor cells , when the constitutively active form of Tkv ( caTkv ) are ectopically induced ( Fig 2G and 2H ) . To understand whether Scrib levels are up-regulated as vein material or specifically up-regulated by BMP signal , we compared Scrib and DE-cadherin ( DE-Cad ) , a key molecule for epithelial morphogenesis [18] , by producing ectopic clones of caTkv . Interestingly , up-regulation of Scrib but not DE-Cad become visible already at 20 h AP ( S1G Fig ) . DE-Cad then becomes up-regulated at 24 h AP ( S1H Fig ) , suggesting that both Scrib and DE-Cad are positively regulated by BMP signaling but in a distinct manner . Together , our data suggest that feedback through Scrib and DLG1 is crucial for long-range BMP/Dpp signaling during PCV development . How does Scrib affect long-range BMP/Dpp signaling in the PCV region ? During the course of Dpp trafficking into the PCV region , Dpp ligands appear to move through the basal side of the pupal wing epithelia [9] . Since the Scrib complex has been proposed to regulate the apical-basal protein distribution of epithelial cells [4 , 19] , we hypothesize that Scrib may affect BMP receptor localization . To test this hypothesis , we investigated the distribution of the BMP type I receptor Tkv in the PCV region . In wild-type pupal wings , Tkv is located at the basolateral region of PCV cells , however Tkv localization was observed more apically in scrib mutants ( Fig 3A–3C and S2A and S2B Fig ) . To investigate whether Tkv localization is regulated by Scrib or modulated as a secondary affect of changing cell polarity , we used an RNAi approach . Our data reveal that Tkv is enriched apically but reduced basally in scrib RNAi wings ( S2C and S2D Fig ) . These results suggest that Scrib ensures enhanced basal Tkv localization where BMP ligand trafficking takes place . We also found that Tkv was frequently observed in puncta at the basal side of the PCV region ( Fig 3D and 3E ) . We wondered whether the observed amounts of Tkv reflect internalized receptors . At the basal side of the PCV cells , Tkv often co-localizes with Scrib and DLG1 at the Rab5-positive early endosomes ( Fig 3D and S2E Fig ) . A key process in canonical BMP signal transduction is production of pMad by activated receptors and its accumulation in the nucleus . In wild type cells , substantial amounts of pMad co-localized with Tkv and Scrib as puncta in early endosomes ( Fig 3E ) . Since pMad subsequently accumulates in the nucleus , endosome-localized pMad appears to be transient . Our data also reveal that the number of co-localization of Tkv with Rab5 is decreased in scrib RNAi wings due to less number of Tkv puncta ( Fig 3F–3H and S2F Fig ) . Taken together , these data suggest that Scrib-dependent Tkv localization at Rab5 endosomes is key for producing robust BMP signaling in the PCV region . If Tkv localization at Rab5 endosomes is key for BMP signaling in the PCV region , ablation of receptor internalization may reduce the BMP signal . Speculating that Scrib facilitates Tkv internalization through a clathrin-dependent mechanism , we investigated whether clathrin-dependent endocytosis affects BMP signaling in the PCV region . Adaptor complex-2 ( AP-2 ) , which functions together with clathrin to initiate endocytosis in the plasma membrane , is composed of four subunits: α , β , μ and σ [20] . Due to its importance in various cellular functions , complete inhibition of endocytosis causes severe tissue damage . Therefore , the relevance of endocytosis was analyzed by investigating genetic interactions and weak loss-of-function analysis . First , we analyzed genetic interactions of dpp alleles with AP-2α ( ada ) , and found that ada interacted genetically with dpp in the PCV region ( Fig 4A–4D ) . Ectopic PCV phenotypes are occasionally observed when Dpp signaling in the PCV region is partially disturbed [21] . This might be caused by incomplete maintenance of positive feedback mechanisms , resulting in ectopic ligand trafficking . Next we found that RNAi of AP2 complex subunits shows crossveinless phenotypes ( Fig 4E–4G and S3A and S3B Fig ) . The PCV-less phenotypes were confirmed to be due to loss of BMP signaling in the PCV region , since the pMad signal was largely ablated in the prospective PCV region of wings expressing AP-2μ RNAi ( Fig 4H ) . Importantly , knocking down AP-2μ in this condition did not disrupt epithelial polarity , as shown by the apical localization of aPKC and the basolateral localization of DLG1 ( Fig 4I ) . We then investigated whether the early endosome protein Rab5 is required for BMP signaling in the PCV region . Ectopic expression of a dominant negative form of Rab5 significantly reduced BMP signaling in the PCV region ( Fig 4J ) . These results suggest that clathrin-mediated internalization of the BMP receptor to Rab5 endosomes is a crucial step for BMP signaling in the PCV region . How then does Scrib play a role in Tkv internalization and signaling ? First , we addressed whether physical interactions between Scrib and Tkv are key for Scrib function . Scrib belongs to the LAP protein family , containing Leucine-rich repeats ( LRR ) , PDZ and C-terminal ( CT ) domains ( Fig 5A ) . To investigate whether Tkv and Scrib are physically associated , we performed co-immunoprecipitation ( co-IP ) analysis . GFP-tagged Tkv and MYC-tagged full-length or truncated Scrib were co-transfected into Drosophila S2 cells . Our data suggest that full-length Scrib weakly associates with Tkv ( S4A Fig ) . In contrast , when different domains of Scrib were co-expressed , LRR regions showed strong interactions with Tkv ( Fig 5B ) , suggesting that the LRR domain is sufficient for interaction with Tkv . Previous studies suggest that the LRR domain is mainly responsible for cell polarity maintenance and proliferation control , while the PDZ domain mediates physical interactions with a variety of proteins [22] . A Scrib fragment including the LRR and PDZ domains , but not the CT domain , showed strong interactions with Tkv ( Fig 5A and 5C ) . Since the CT domain is capable of binding to LRR ( S4B Fig ) , this domain may function as a regulator of Scrib-Tkv interactions . Next , we examined whether the conformational change conferred by activation affects the association of Tkv with Scrib . Although caTkv associates with full-length Scrib less efficiently ( S4A Fig ) , LRR interacts with caTkv and pMad ( Fig 5D ) , suggesting that Tkv complexed with Scrib is able to phosphorylate Mad . We then tested whether LRR is sufficient for maintaining BMP signaling in the PCV region in vivo . BMP signaling in scrib mutant clones in the PCV region was sufficiently restored by LRR as well as by full-length Scrib , but not by Scrib without an LRR domain ( Fig 5E and 5F ) . Consistently , pMad signaling in the PCV region remains intact in mutant cells of scrib5 allele that lacks the PDZ3/4 and CT domains ( Fig 5G ) [22] . We further confirmed that the LRR , but not the PDZ domain substantially restores cell polarity in scrib mutant cells of the pupal wings ( S4C Fig ) . Therefore , restoration of BMP signaling in scrib mutant clones LRR might be mediated through biochemical interaction of Scrib , Tkv and pMad in endosomes or through restoration of tissue polarity . However , our RNAi data in which Scrib regulates BMP signaling in the PCV region in a distinct manner from regulation of tissue polarity ( Fig 1 ) sufficiently support that the former scenario is more likely . Taken together , these results indicate that the LRR domain not only sustains epithelial polarity , but also interacts with the BMP receptor to maintain BMP signaling . How is internalized Tkv regulated to effect BMP signaling ? One simple hypothesis is that Tkv preferentially localizes to Rab5 endosomes to optimize BMP signaling . In that case , Tkv may associate with Rab5 as well as with Scrib to form a complex in the early endosome . To test this hypothesis , Tkv , Rab5 and LRR were co-expressed in S2 cells . We found that Tkv , Rab5 and LRR form a protein complex ( Fig 6A ) . When caTkv and Rab5 are co-expressed , they not only interact with each other , but also with pMad ( Fig 6B and 6C and S5A Fig ) , suggesting that Tkv in the Rab5 endosome is capable of phosphorylating Mad . Tkv is able to associate with various forms of Rab5 , including wild-type , constitutively active ( Q88L ) and dominant negative ( S43N ) ( S5B Fig ) [23] . Furthermore , the BMP type-II receptor Punt was also detected in the complex with Rab5 and LRR ( S5C Fig ) . These results are consistent with our hypothesis that BMP receptors are preferentially localized in the Rab5 endosome to optimize BMP signaling . We further found that the constitutively active form of Rab5 partially rescued loss of BMP signaling in scrib mutants ( Fig 6D and 6E ) , suggesting that the Rab5 endosomes play significant roles in BMP signaling during PCV formation .
This study shows that the Scrib complex , a basolateral determinant , is a novel feedback component that optimizes BMP signaling in the PCV region of the Drosophila pupal wing . During PCV development , limited amounts of Dpp ligands are provided by the Dpp trafficking mechanism [7 , 9] . Furthermore , amounts of receptors appear to be limited since tkv transcription is down-regulated in the cells in which the BMP signal is positive [7 , 24] , a mechanism that serves to facilitate ligand diffusion and sustain long-range signaling in the larval wing imaginal disc [25 , 26] . To provide robust signal under conditions in which both ligands and receptors are limiting , additional molecular mechanisms are needed . Previous studies suggest that two molecules play such roles . Crossveinless-2 ( Cv-2 ) , which is highly expressed in the PCV region , serves to promote BMP signaling through facilitating receptor-ligand binding [12 , 27] . Additionally , the RhoGAP protein Crossveinless-c ( Cv-c ) provides an optimal extracellular environment to maintain ligand trafficking from LVs into PCV through down-regulating integrin distribution at the basal side of epithelia [9] . Importantly , both cv-2 and cv-c are transcriptionally regulated by BMP signaling to form a feedback or feed-forward loop for PCV formation . Scrib , a third component , sustains BMP signaling in the PCV region in a different manner . First , to utilize Tkv efficiently , Scrib regulates Tkv localization at the basolateral region in the PCV cells , where ligand trafficking takes place . Regulation of receptor localization could be a means of spatiotemporal regulation of signaling molecules during the dynamic process of morphogenesis . Second , to optimize the signal transduction after receptor-ligand binding , Scrib facilitates Tkv localization in the Rab5 endosomes . Localization of internalized Tkv is abundant at Rab5 endosomes in the PCV region of wild-type , but not scrib RNAi cells . While the physical interaction between Scrib , Tkv and Rab5 in the pupal wing remains to be addressed , our data in S2 cells suggest that physical interactions between these proteins are key for preferential localization of Tkv at the Rab5 endosomes . Recently , Scrib has been implicated in regulating NMDA receptor localization through an internalization-recycling pathway to sustain neural activity [28] . Therefore , Scrib may be involved in receptor trafficking in a context-specific manner , the molecular mechanisms of which , however , remain to be characterized . Third , BMP/Dpp signaling up-regulates scrib transcription in the pupal wing . Moreover , knockdown of scrib leads to loss of BMP signaling in PCV region but not loss of apical-basal polarity ( Fig 1 ) . These facts suggest that upregulation of Scrib is key for optimizing BMP signaling by forming a positive feedback loop . Previous studies indicate that cell competition takes place between scrib clones and the surrounding wild-type tissues in the larval wing imaginal disc [29] . Moreover , cell competition has been documented between loss-of-Dpp signal and the surrounding wild-type tissues [30] . We presume that the mechanisms proposed in this study are independent of cell competition for the following reasons . First , scrib RNAi and AP-2μ RNAi data reveal that loss of BMP signal in the PCV region is produced without affecting cell polarity ( Figs 1 and 3 ) . Thus , cell competition is unlikely to occur in this context . Second , BMP signal is intact in scrib mutant clones of the wing imaginal disc ( S6 Fig ) , suggesting that cell competition caused by scrib clones is not a direct cause of loss of BMP signaling in scrib mutant cells . Previous studies established that receptor trafficking plays crucial roles in signal transduction of conserved growth factors , including BMP signaling . Several co-factors have been identified as regulators of BMP receptor trafficking [31–38] . Some of them down-regulate BMP signaling [31–35 , 38] , while others help maintain it [36 , 37] . We propose that the Scrib-Rab5 system is a flexible module for receptor trafficking and can be utilized for optimizing a signal . During larval wing imaginal disc development , BMP ligands are trafficked through extracellular spaces to form a morphogen gradient . Although previous studies indicate that regulation of receptor trafficking impacts BMP signaling in wing imaginal discs [23 , 39] , BMP signaling persists in scrib or dlg1 mutant cells in wing discs ( S6 Fig ) . Wing disc cells interpret signaling intensities of a morphogen gradient . In this developmental context , an optimizing mechanism might not be beneficial to the system . In contrast , cells in the PCV region use the system to ensure robust BMP signaling . Taken together , our data reveal that a feedback loop through BMP and Scrib promotes Rab5 endosome-based BMP/Dpp signaling during PCV morphogenesis . Since the components ( BMP signaling , the Scrib complex , and Rab5 endosomes ) discussed in this work are highly conserved , similar mechanisms may be widely utilized throughout Animalia .
UAS-scrib RNAi ( #35748 ) , scrib2 ( #41755 ) , UAS-dlg1 RNAi ( #35772 ) , dlg1B FRT19A/FM7 ( #57087 ) , UAS-lgl RNAi ( #35773 ) , UAS-AP-2μ RNAi ( #28040 ) , UAS-AP-2α RNAi ( #32866 ) , UAS-AP-2σ RNAi ( #27322 ) , UAS-tkvQ253D ( #36536 ) , UAS-myrRFP ( #7118 ) , UASp-Rab5Q88L-YFP ( #9774 ) , UASp-Rab5S43N-YFP ( #9771 ) , UAS-scrib ( #59079 ) , UAS-scrib . LRR . LASPD ( #59081 ) and UAS-scrib . DeltaLRR . GFP ( #59084 ) , nubbin-Gal4 ( nub-GAL4 ) ( #25754 ) and tub-GAL80ts ( #7018 ) were obtained from the Bloomington Drosophila Stock Center . Scrib-GFP ( #CA07683 ) was obtained from Fly Trap projects [40 , 41] . Tkv-YFP ( #115298 ) was obtained from the Kyoto Drosophila Genetic Resource Center . dpphr4 , dppshv-Gal4 , UAS-tkv-HA , UAS-GFP-dpp were described previously [7 , 9] . UAS-myr-deleteLRR and Scrib5 were obtained from D . Bilder , ada3 from M . Gonzalez-Gaitan [42] . Fly stocks were maintained at 25°C unless otherwise mentioned . To induce MARCM clones [43] , larvae were heat-shocked for two hours at 37°C at 96 hours after egg laying ( AEL ) . Fig 1A , 1E and 1G: w/+; nub-Gal4/+; Tub-Gal80ts/+; B , F and H: w/+; nub-Gal4/+; Tub-Gal80ts/scrib RNAi; C: w/+; nub-Gal4/+; Tub-Gal80ts/ dlg1 RNAi; I: hsFlp/+; tub-Gal4 , UAS-mCD8-GFP/+; FRT82B tub-Gal80/FRT82B scrib2 Fig 2A–2E: w;; Scrib-GFP; F: yw as a control; w+; UAS-tkv . Q253D/+; dppshv-Gal4/+; G , H: hsFlp/+; tub-Gal4 , UAS-mCD8-GFP/ UAS-tkv . Q253D; FRT82B , tub-Gal80/FRT82B Fig 3B and 3C: hsFlp/+; UAS-myr-mRFP/Tkv-YFP; tub-Gal4 FRT82B tub-Gal80/FRT82B scrib2; D-E: w; Tkv-YFP/+; F: w/+; nub-Gal4/Tkv-YFP; Tub-Gal80ts/+; w/+; nub-Gal4/Tkv-YFP; Tub-Gal80ts/scrib RNAi Fig 4A: yw; ada3/+; B: yw; dpphr4/+; C: yw; dpphr4/ ada3; E and H: w/+; nub-Gal4/+; Tub-Gal80ts/ +; F , H and I: w/+; nub-Gal4/+; Tub-Gal80ts/ AP-2μ RNAi; J: hsFlp/+; tub-Gal4 , UAS-mCD8-GFP/ UASp-RAB5WT or S43N-YFP; FRT82B , tub-Gal80/FRT82B Fig 5E and 5F: hsFlp/+; tub-Gal4 , UAS-mCD8-GFP/UAS-Scrib . fl-GFP ( or UAS-scrib . LRR . LASPD or UAS-scrib . DeltaLRR . GFP ) ; FRT82B tub-Gal80/FRT82B scrib2; G: hsFlp/+; tub-Gal4 , UAS-mCD8-GFP/+; FRT82B tub-Gal80/FRT82B scrib5 Fig 6D: hsFlp/+; tub-Gal4 , UAS-mCD8-GFP/ UASp-RAB5Q88L-YFP; FRT82B tub-Gal80/FRT82B scrib2 S1A Fig: w/+; nub-Gal4/+; Tub-Gal80ts/lgl RNAi; B: FRT19A dlg1B/ hsFlp , tub-GAL80 , FRT19A; tub-GAL4 , UAS-mCD8-GFP/+; C and E: hsFlp/+; UAS-myr-mRFP/UAS-GFP-Dpp; tub-Gal4 FRT82B tub-Gal80/FRT82B; D and F: hsFlp/+; UAS-myr-mRFP/UAS-GFP-Dpp; tub-Gal4 FRT82B tub-Gal80/FRT82B scrib2; G , H: hsFlp/+; tub-Gal4 , UAS-mCD8-GFP/ UAS-tkv . Q253D; FRT82B , tub-Gal80/FRT82B S2B Fig: hsFlp/+; UAS-myr-mRFP/Tkv-YFP; tub-Gal4 FRT82B tub-Gal80/FRT82B scrib2; C: w/+; nub-Gal4/Tkv-YFP; Tub-Gal80ts/+; D: w/+; nub-Gal4/Tkv-YFP; Tub-Gal80ts/scrib RNAi; E: w; Tkv-YFP/+; F: w/+; nub-Gal4/Tkv-YFP; Tub-Gal80ts/+; w/+; nub-Gal4/Tkv-YFP; Tub-Gal80ts/scrib RNAi S3A Fig: w/+; nub-Gal4/+; Tub-Gal80ts/ AP-2α RNAi; B: w/+; nub-Gal4/+; Tub-Gal80ts/ AP-2σ RNAi S4C Fig: hsFlp/+; UAS-myr-mRFP/+; tub-Gal4 FRT82B tub-Gal80/FRT82B scrib2; hsFlp/+; UAS-myr-mRFP/UAS-Scrib . fl-GFP; tub-Gal4 FRT82B tub-Gal80/FRT82B scrib2; hsFlp/+; UAS-myr-mRFP/UAS-Scrib . LRR . LASPD; tub-Gal4 FRT82B tub-Gal80/FRT82B scrib2; hsFlp/UAS-myr-deleteLRR; UAS-myr-mRFP/+; tub-Gal4 FRT82B tub-Gal80/FRT82B scrib2 S6 Fig For scrib MARCM: hsFlp/+; tub-Gal4 , UAS-mCD8-GFP/+; FRT82B tub-Gal80/FRT82B scrib2 For dlg1 MARCM: FRT19A dlg1B/ hsFlp , tub-GAL80 , FRT19A; tub-GAL4 , UAS-mCD8-GFP/+ Pupal wings were fixed in 3 . 7% formaldehyde ( Sigma-Aldrich ) at 4°C overnight . Wing imaginal discs were fixed in 3 . 7% formaldehyde at room temperature ( RT ) for 20 minutes . All immunostaining and in situ hybridizations were performed as described previously [7 , 9] . The primary antibodies used are as follows: mouse anti-DLG1 , rat anti-DE-Cadherin and mouse anti-GFP ( for immunohistochemistry; all at 1:50 ) were obtained from Developmental Studies Hybridoma Bank , rabbit anti-phospho-SMAD1/5 ( 1: 200 for IF , 1:2000 for Western blotting ) from Cell Signaling Technology ( CST ) , rabbit anti-Rab5 ( 1:600 ) and rabbit anti-RFP ( 1:5000 for Western blotting ) from Abcam , mouse anti-RFP ( 1:5000 for Western blotting ) from Chromotek , mouse anti-GFP ( 1: 5000 for Western blotting ) from Millipore , mouse anti-β-tubulin ( 1:5000 ) from Sigma-Aldrich , rabbit anti-MYC ( 1:500 ) , goat anti-Scrib ( 1:100 ) , rabbit anti-aPKC ( 1:100 ) and mouse anti-LGL ( 1:200 ) from Santa Cruz Biotechnology , and rabbit anti-Scrib ( 1:2000 ) from C . Doe . Secondary antibodies were as follows: goat anti-mouse IgG Alexa 488 , goat anti-mouse IgG Alexa 568 , goat anti-mouse IgG Alexa 647 , goat anti-rabbit IgG Alexa 568 , goat anti-rabbit IgG Alexa 647 , goat anti-rat IgG Alexa 488 and goat anti-mouse IgG Cy5 , all from Molecular Probes ( 1:200 ) . GFP-booster ( 1:200 , ChromoTek ) was used to enhance the YFP signal in Fig 3F and S2C Fig . Fluorescent images were obtained with a Zeiss LSM700 upright confocal microscope . Images of in situ hybridization and adult wings were obtained with a Nikon Eclipse 90i microscope . All confocal immunofluorescent images were processed and analyzed with ImageJ ( NIH ) . The images were the composite of a stack with projection of max intensity unless specified in detailed . Mad-FLAG for cell culture was described previously [44] . Tkv-GFP and caTkv-GFP in UAS plasmids were obtained from G . Marquez . A Punt-GFP construct under UAS control was obtained from M . O’Connor . Full-length ( FL ) and truncated scrib cDNAs were obtained by PCR using scrib cDNA ( #42064 , Addgene ) as a template . FL and truncated scrib cDNAs with one copy of MYC tag at the C-terminus were cloned into BglII and KpnI sites of pUAST-attB [45] . Protein sequences of truncated scrib cDNAs are as follows: LRR , 1–692; PDZ1/2 , 531–1105; PDZ3/4 , 928–1400; CT , 1338–1757; LRR+PDZ1/2 , 1–1105; ∆CT , 1–1400 . Rab5 cDNAs from the BDGP Gold cDNA collection ( Drosophila Genomics Resource Center; DGRC ) were cloned into pENTER-D-TOPO entry vector ( Invitrogen ) and then subcloned into the destination vector pTWR ( DGRC ) . Dominant-negative ( S43N ) and constitutively-activated ( Q88L ) forms of Rab5 were generated using an overlap-PCR strategy . nub-GAL4 was used to drive the RNAi in the wing blade in combination with the temperature-sensitive GAL80 [16] . scrib , dlg1 or lgl RNAi flies were cultured at 25 , 27 or 25°C , respectively . AP-2 subunit RNAi flies were maintained at 18°C for 5 days and then cultured at 27°C . Drosophila S2 cells were used for producing recombinant proteins as previously described [44] . Cells were transfected with HiFugene transfection reagent ( Promega ) according to the manufacturer’s protocol . S2 cells were transfected with the plasmids expressing indicated cDNA and tub-GAL4 . Three days after transfection , cells were collected and cell lysates were subjected to immunoprecipitation using the GFP-Nanotrap A Kit ( Chromotek ) according to the manufacturer's instructions . S2 cells were subjected to the IP lysis buffer ( 25 mM Tris-HCl pH 7 . 4 , 150 mM NaCl , 1% NP-40 , 1 mM EDTA , 5% glycerol ) on ice for 30 min . The supernatants obtained were subjected to the WB as an input and subsequent IP . Western blotting was conducted as previously described [44] . All biochemical data shown are representative of no less than three independent assays . Pupal wings were dissected without fixation , and subjected to RNA purification using TRIzol reagent ( Invitrogen ) . RNA template was first converted to cDNA using Maxima First Strand cDNA Synthesis Kit ( Thermo Scientific ) . The cDNA template was then subjected to qPCR carried out with the StepOnePlus Real-Time PCR Systems Kit ( Applied Biosystems ) . The following primers were used: gapdh F , CGAAGATCGGAATTAACGGA; gapdh R , ACCGTGAGTCGAGTCGAATT; βTub F , GAACCCTGCTGATTTCCAAGAT; βTub R , ATATCGTAGAGAGCCTCGTTGT; scrib F , CTGGCATATTCATATCGCACATT; scrib R , TCATCACCTGGCTTCAACA; dlg1 F , CACCGAGGATATAACCAGAGAAC; dlg1 R , CAGGATGAAGGACACATAGATACC; lgl F , TGAGTCAATCCGCCAACTTCCA; lgl R , TTCACTGTAAGACCAACGCTCTGT; vvl F , CTGCACATACACCATCACAT; vvl R , GGAGAACACATTGCCATAGA . All experiments were carried out independently at least three times . Error bars indicate s . e . m . Statistical significance was calculated by the two paired t-test method . | Epithelial morphogenesis is one of the key processes in animal development . Evolutionarily conserved growth factors frequently instruct patterning and differentiation in morphogenesis . However , little is known about how extracellular cues and epithelial morphogenesis are mutually coordinated in vivo . Wing posterior crossvein ( PCV ) development in Drosophila provides an excellent system for understanding how bone morphogenetic protein ( BMP ) signaling regulates patterning and differentiation of epithelia . We find that the apical-basal polarity gene Scribbled ( Scrib ) is required for PCV formation by optimizing BMP signaling in the PCV region as follows . First , Scrib regulates BMP type-I receptor Thickveins ( Tkv ) localization basally . Second , Scrib facilitates Tkv internalization to the Rab5 endosomes to optimize signal transduction after receptor-ligand binding . Third , BMP signaling up-regulates scrib transcription in the pupal wing to form a positive feedback loop . These results suggest that coupling between epithelial polarity genes and conserved growth factors play crucial roles in patterning and differentiation of epithelia . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"cell",
"physiology",
"invertebrates",
"rna",
"interference",
"vesicles",
"cloning",
"animals",
"cell",
"polarity",
"animal",
"models",
"organisms",
"developmental",
"biology",
"drosophila",
"melanogaster",
"model",
"organisms",
"immunoprecipitation",
"epigenetics",
"molecular",
"biology",
"techniques",
"cellular",
"structures",
"and",
"organelles",
"morphogenesis",
"drosophila",
"endosomes",
"research",
"and",
"analysis",
"methods",
"genetic",
"interference",
"gene",
"expression",
"molecular",
"biology",
"insects",
"precipitation",
"techniques",
"arthropoda",
"biochemistry",
"signal",
"transduction",
"rna",
"cell",
"biology",
"nucleic",
"acids",
"genetics",
"biology",
"and",
"life",
"sciences",
"cell",
"signaling",
"bmp",
"signaling"
] | 2016 | Scribbled Optimizes BMP Signaling through Its Receptor Internalization to the Rab5 Endosome and Promote Robust Epithelial Morphogenesis |
Th17 cells are a subset of CD4+ T cells known to play a central role in the pathogenesis of many autoimmune diseases , as well as in the defense against some extracellular bacteria and fungi . However , Th17 cells are not believed to have a significant function against intracellular infections . In contrast to this paradigm , we have discovered that Th17 cells provide robust protection against Trypanosoma cruzi , the intracellular protozoan parasite that causes Chagas disease . Th17 cells confer significantly stronger protection against T . cruzi-related mortality than even Th1 cells , traditionally thought to be the CD4+ T cell subset most important for immunity to T . cruzi and other intracellular microorganisms . Mechanistically , Th17 cells can directly protect infected cells through the IL-17A-dependent induction of NADPH oxidase , involved in the phagocyte respiratory burst response , and provide indirect help through IL-21-dependent activation of CD8+ T cells . The discovery of these novel Th17 cell-mediated direct protective and indirect helper effects important for intracellular immunity highlights the diversity of Th17 cell roles , and increases understanding of protective T . cruzi immunity , aiding the development of therapeutics and vaccines for Chagas disease .
Chronic infection with the protozoan parasite Trypanosoma cruzi results in Chagas disease , a Neglected Tropical Disease currently affecting 8–11 million people worldwide and 300 , 000 people in the United States [1] . Humans usually acquire T . cruzi infection via reduviid insect vectors , but infections also sometimes occur through vertical transmission , the ingestion of contaminated food products , and the receipt of infected biological donations . The disease can cause significant cardiac and gastrointestinal morbidity , but drug therapies are typically non-curative and poorly tolerated [1] . The significant global burden of Chagas disease , coupled with the inefficacy of available treatments , indicates a pressing need to develop novel therapeutics , including preventative and/or therapeutic vaccines to induce protective T . cruzi immunity in at risk individuals . The development of effective vaccines requires a more detailed understanding of the protective host immune responses against T . cruzi infection . Because T . cruzi promiscuously infects both cells expressing MHC class II and cells expressing only MHC class I , CD8+ T cells are critical for protection against infection of all host targets cells . However , CD4+ T cells also are needed for optimal protection [2] . CD4+ Th1 cells have been shown to provide both systemic and mucosal protection against T . cruzi infection , consistent with the well-established framework of Th1 cells being the CD4+ T cell subset most important against intracellular pathogens [3 , 4] . Less is known about the role of other CD4+ T cell subsets during T . cruzi infection . More recently , studies have found a protective role for IL-17A , the major cytokine produced by CD4+ Th17 cells , raising questions about the possibility of a protective role for Th17 cells [5–7] . However , multiple subtypes of IL-17-producing cells exist , including αβ T cells , γδ T cells , innate lymphoid cells , and even B cells in T . cruzi infection [8 , 9] . In addition , Th17 cells have been shown to be involved in autoimmunity [10–12] . Thus , it remained unclear from these studies of global IL-17A deficiency whether Th17 cells specifically play a protective or pathologic role in T . cruzi immunity . Although Th17 cells are now known to protect against certain extracellular bacteria and fungi [13–16] , they are not thought to have a significant function in intracellular immunity . To investigate the specific effects of different CD4+ T cell subsets in T . cruzi infection , we generated T cell receptor transgenic ( TS-CD4-Tg ) mice with CD4+ T cell receptors that are specific for p7 , an immunodominant epitope encoded by the T . cruzi trans-sialidase , a highly conserved virulence factor and potential vaccine target ( S1 and S2 Figs ) [17–20] . Using CD4+ T cells from these mice , we were able to generate parasite-specific Th1 and Th17 cells in vitro . These cells were used for studies in vivo using adoptive transfer experiments , and in vitro using CD8+ T cell activation and macrophage infection assays . Through these experiments , we found that Th17 cells confer robust protection against T . cruzi infection , even surpassing that provided by Th1 cells , through both direct and indirect protective effects .
To study the roles of Th1 cells versus Th17 cells in vivo , in vitro-differentiated TS-CD4-Tg Th1 or Th17 cells were adoptively transferred with naïve polyclonal CD8+ T cells into RAG KO mice , which were then challenged subcutaneously with T . cruzi ( S3 Fig , Fig 1A and 1B ) . We confirmed the persistence of these transferred cells ( S4 Fig ) . As expected , control RAG KO mice ( no T cell transfer ) did not survive the challenge , and nearly all mice receiving CD8+ T cells alone also succumbed to infection . Consistent with previous studies supporting a protective function for Th1 cells against T . cruzi , mice receiving co-transfer of Th1 cells and CD8+ T cells exhibited lower blood parasite burdens and improved long-term survival . However , surprisingly , mice receiving co-transfer of Th17 cells and CD8+ T cells experienced the strongest protection , as indicated by the lowest blood parasite burdens and highest survival rates ( 100% long term ) of any group . This represented significantly improved protection even compared to mice receiving Th1 cells ( p< . 001 comparing Th17 and CD8+ T cells to Th1 and CD8+ T cells , Fig 1A and 1B ) . Because our Th17 cell cultures also contained undifferentiated cells not expressing the classical Th17 markers RORγt or IL-17 ( S3C Fig ) , we sought to confirm that the impressive protection was truly a Th17 cell-mediated effect , and not attributable to another cell type present in the Th17 cell cultures . To do so , we used cell purification kits to obtain >90% purity of IL-17+ cells from our Th17 cell cultures ( S5A Fig ) . Co-transfer of these cells with polyclonal CD8+ T cells into RAG KO mice resulted in sustained Th17 cell responses and similar activation of the CD8+ T cells ( S5B and S5C Fig ) . Even more importantly , the transfer of highly purified IL-17+ T cells resulted in protection from T . cruzi ( S5D and S5E Fig ) , confirming that bona fide Th17 cells provide protection against this parasite . We also tested polyclonal T . cruzi-specific Th17 cells , which were generated by stimulating wild-type CD4+ T cells with dendritic cells pulsed with whole parasite lysate antigens under Th17-skewing conditions ( S6A Fig ) . We found that polyclonal Th17 cells , which better model a Th17 cell response that could arise in vivo , also result in significantly lower parasitemia and 100% long-term survival when given with CD8+ T cells ( S6B–S6D Fig ) . Furthermore , Th17 cells have been reported to maintain plasticity and revert to Th1 phenotypes in vivo [21–23] , raising another important consideration . Indeed , although Th17 cells persisted after transfer , as indicated by sustained production of IL-17 , a small percentage of in vitro-biased Th17 cells expressed IFN-γ ( S3C Fig ) , and some Th1 cells emerged from Th17 cells in vivo ( S4A and S4C Fig ) . Therefore , we asked whether the IFN-γ expressed by a small subset of Th17 cells , or the trans-differentiation of Th17 cells into Th1 cells after transfer , was responsible for the protective effects . To answer this question , we generated T-bet KO TS-CD4-Tg mice lacking T-bet , the transcription factor that drives Th1 differentiation and IFN-γ production ( T-bet KO TS-CD4-Tg ) . Using CD4+ T cells from these mice , we made T . cruzi-specific T-bet KO Th17 cells , which are unable to adopt the Th1 phenotype . We gave RAG KO mice T-bet KO TS-CD4-Tg Th17 cells and polyclonal CD8+ T cells , and infected them with T . cruzi . CD4+ T cells recovered from recipient mice 9 and 101 days after infection confirmed that the transferred T-bet KO Th17 cells were veritable Th17 cells , as they did not produce IFN-γ in vivo , but maintained high levels of IL-17A expression ( S7A and S7B Fig ) . Mice receiving WT or T-bet KO Th17 cells were similarly protected , with significantly decreased parasite burdens compared to control and 100% long-term survival ( Fig 1C and 1D ) . The persistence of Th17 cells and their protective effects could be detected more than three months after initial infection ( S7C Fig ) , providing further evidence that a stable Th17 cell response to T . cruzi is possible , and that there is no requirement for Th1 cells or any other IFN-γ+ CD4+ T cells for robust T . cruzi immunity . To investigate mechanisms of direct protection against T . cruzi infection by Th17 cells , we infected murine peritoneal exudate macrophages ( PEM ) with T . cruzi trypomastigotes in vitro and added Th1 or Th17 cells for two days . In the absence of T cells , high-level parasite infection occurred . The addition of either Th1 cells or purified IFN-γ alone significantly reduced numbers of cells that became infected after two days . Th1 cells are known to prime macrophage activation for the killing of intracellular microorganisms through the secretion of IFN-γ . Consistent with this , a neutralizing antibody directed against IFN-γ reversed the effects of Th1-mediated protection ( Fig 2A–2C ) [24] . The addition of Th17 cells to the cultures also reduced the parasite burden among infected macrophages , indicating that Th17 cells are able to directly protect cells against T . cruzi infection . This effect could be recapitulated by replacing Th17 cells with purified IL-17 cytokine , or abrogated by adding an IL-17A-neutralizing antibody , indicating that the protection is mediated by the IL-17 produced by Th17 cells . Although Th17-derived IL-17A is known to recruit neutrophils and enhance inflammation in extracellular infections , a direct intracellular protective effect has not been previously reported . We confirmed previous studies showing that Th1 cells , through the production of IFN-γ , induce macrophage iNOS to produce nitric oxide [25] , the mechanism underlying the killing of intracellular pathogens . However , NO was not induced by Th17 cells ( Fig 2D ) , and although the addition of the iNOS inhibitor N6- ( 1-Iminoethyl ) -L-lysine ( L-NIL ) reversed Th1-mediated protection , it had no effect on Th17-mediated protection ( Fig 2E ) . These data indicate that the direct protection provided by Th17 cells operates independently of NO generation . Similar to NO-mediated protection , ROS can also inhibit the growth of intracellular pathogens in neutrophils and macrophages [26] . IL-17A has been reported to induce production of ROS in endothelial cells , but it was unknown whether the cytokine could induce ROS in other cells [27] . We hypothesized that Th17 cells could also induce ROS production in macrophages , and that the induction of ROS was necessary for IL-17-mediated direct protection . To test our hypothesis , we infected bone marrow-derived macrophages ( BMDMs ) generated from WT and gp91phox KO mice . The latter are genetically deficient in a critical subunit of the NADPH oxidase , resulting in a defective phagocyte respiratory burst response . Following infection of these cells , we treated them with IFN-γ or IL-17A for 48 hours . The gp91phox deficiency did not affect the induction of NO ( Fig 2F ) . In both infected WT and gp91phox KO BMDM , IFN-γ treatment induced NO production , but IL-17A did not ( Fig 2F ) . In WT BMDM , both IFN-γ and IL-17A protected against infection and intracellular growth of T . cruzi . Deficiency of the gp91phox subunit of NADPH oxidase had no effect on IFN-γ-mediated protection , but reversed IL-17A-mediated protection , indicating that functional NADPH oxidase is required for this mechanism , and suggesting the involvement of ROS generated during the respiratory burst ( Fig 2G ) . Given our findings concerning IL-17-mediated protection in vitro , we asked whether IL-17A was also responsible for the protection observed in vivo . We first neutralized IL-17A in RAG KO mice reconstituted with Th17 cells and polyclonal CD8+ T cells by injection of a monoclonal anti-IL-17A antibody previously shown to reverse IL-17A function in vivo [28 , 29] . Neutralizing IL-17A had no effect on parasitemia levels or survival percentages in these mice ( Fig 3A and 3B ) . We also overexpressed IL-17A in RAG KO mice receiving polyclonal CD8+ T cells alone using a recombinant IL-17A-producing adenovirus ( IL-17A AdV ) . Although mice injected with IL-17A AdV had markedly higher serum levels of IL-17A ( 28 ng/ml ± 16 in IL-17 AdV sera versus <0 . 02ng/ml in control AdV sera ) , overexpression of IL-17A alone was not enough to replace Th17 cells , and did not protect mice also receiving CD8+ T cells ( Fig 3C and 3D ) . Overall , these results indicate that IL-17 is dispensable for the major in vivo protective effects induced by Th17 cells , despite being the major Th17 cytokine implicated in previous reports on Th17-based protection and pathology . Therefore , in T . cruzi infection , Th17 cells must function through an IL-17-independent mechanism to protect . To discriminate between direct protective versus indirect T helper effects , we studied protection provided by Th17 cells transferred alone or with polyclonal CD8+ T cells . CD8+ T cells were needed for optimal Th17-mediated protection ( Fig 4A and 4B ) , indicating Th17 helper effects for CD8+ T cells were most important in vivo . However , the mechanism of Th17 helper effects on CD8+ T cells could be due to Th17 cells directly instructing these cells , or Th17 cells indirectly affecting these cells via the activation of antigen-presenting cells . To distinguish between these potential mechanisms , naïve polyclonal CD8+ T cells were stimulated with suboptimal α-CD3 in vitro . Th17 cells and/or immature dendritic cells ( DCs ) were added to these cultures to provide potential co-stimulatory help . In the absence of Th17 cells , CD8+ T cells exhibited progressive degrees of proliferation with sub-optimal α-CD3 stimulation and increasing doses of DC ( Fig 4C ) , as detected by dilution of the cell staining dye CFSE . However , nearly 100% of CD8+ T cells proliferated maximally when Th17 cells were introduced , even in the absence of DC . These data indicate that Th17 cells directly and strongly enhance the activation of naïve CD8+ T cells . In addition to increasing the proliferation of CD8+ T cells , Th17 cells also induced these cells to increase the expression of surface CD44 ( a marker of activated T cells ) , the effector cytokine IFN-γ , and the chemokine MIP-1α , indicating an activated state of CD8+ T cells ( Fig 4D ) . These helper effects on CD8+ T cells were also detectable using supernatants ( SN ) from activated Th17 cells , implicating soluble factors ( Fig 4D ) . However , consistent with our previous in vivo results , neutralization of IL-17A did not prevent the Th17-mediated activation of CD8+ T cells in vitro ( Fig 4D ) . By co-culturing sub-optimally stimulated CD8+ T cells with individual purified Th17 cytokines , we discovered that IL-21 alone recapitulated this potent help , inducing a state of CD8+ T cell activation similar to that seen with the addition of Th17 cells or Th17 cell SNs ( Fig 4E ) . We confirmed that our Th17 cells produced significant amounts of IL-21 , while the Th1 cells did not ( Fig 4F , S3F Fig ) . In vitro assays demonstrated that IL-21R KO CD8+ T cells are unable to be activated by Th17 cell SNs in the same manner as WT CD8+ T cells ( Fig 5A ) , confirming the IL-21 signaling requirement for Th17-mediated protection . To test these results in vivo , we transferred Th17 cells into RAG KO mice with either WT or IL-21R KO CD8+ T cells , and challenged the mice with T . cruzi . Recovered spleen cells from mice receiving IL-21R KO CD8+ T cells with Th17 cells had impaired responses to T . cruzi TS antigen/peptide stimulation ( Fig 5B ) . Furthermore , only mice receiving Th17 cells with WT CD8+ T cells were able to control infection . Although mice given WT CD8+ T cells reduced parasitemia after day 15 , parasitemia levels steadily increased in mice receiving IL-21R KO CD8+ T cells ( Fig 5C ) . As a result of this failed protection , all mice receiving IL-21R KO CD8+ T cells succumbed to infection while 100% of mice receiving WT CD8+ T cells with Th17 cells survived long-term ( Fig 5D ) . These studies clearly demonstrate that Th17-mediated protection in vivo operates through the secretion of IL-21 , and the resulting indirect T helper effects on CD8+ T cells . Given the improved protection afforded by Th17 over Th1 helper cells , we next explored the mechanisms responsible for the reason Th17-induced CD8+ T cells can control infection better than Th1-induced CD8+ T cells . Recovered spleen cells from RAG KO mice reconstituted with CD8+ T cells and Th1 or Th17 cells prior to T . cruzi challenge demonstrated that mice receiving Th17 cells had greater expansion of CD8+ T cells than mice receiving Th1 cells ( Fig 6A ) . In addition , CD8+ T cells primed by Th17 cells either in vivo ( Fig 6B ) or in vitro ( Fig 6C ) expressed higher levels of T-bet compared with Th1-primed cells , which has been associated with higher cytotoxic activity and IFN-γ production [30 , 31] . IFN-γ ELISPOT and ICS assays comparing spleen cells recovered from Th1- or Th17-transferred mice demonstrated that mice receiving Th17 cells also had >6-fold stronger total spleen cell responses to T . cruzi TS antigen , and >10 , 000-fold greater frequencies of antigen-specific CD8+ T cells than mice receiving Th1 cells ( Fig 6D and 6E ) . Overall , these results indicate that both quantitative and qualitative differences distinguish Th1- and Th17-induced CD8+ T cells . In order to test whether Th17 cells can have similar in vivo helper effects for CD8+ T cells in a physiological setting , where millions of CD4+ and CD8+ T cells of diverse specificities co-exist , we adoptively transferred immunocompetent WT BALB/c mice with TS-CD4-Tg Th1 or Th17 cells , then challenged these mice with highly virulent blood-form trypomastigotes . Total splenic cells and purified CD8+ T cells recovered from mice receiving Th17 cells contained 3- to 4-fold higher frequencies of antigen-specific total spleen cells ( Fig 7A ) and CD8+ T cells ( Fig 7B ) compared with mice given Th1 cells , consistent with our findings in RAG KO mice . These latter results indicate that superior Th17 helper effects were reproducible in a physiologic setting . In addition , WT BALB/c mice receiving adoptive transfer of either Th1 or Th17 cells developed higher titers of TS-specific antibodies than mice not receiving any cell transfer , indicating that , in addition to their robust helper effects on critically protective CD8+ T cells , Th17 cells also provide substantial help to T . cruzi-specific antibody-secreting B cells ( Fig 7C and 7D ) . Despite being highly protective , Th17 cells are not the predominant natural CD4+ T cell response to T . cruzi infection . However , Th17 cells could be targeted for induction by vaccination . In the past , we have successfully induced T . cruzi-specific Th1 cells in vivo by vaccinating wild-type mice with recombinant TS protein ( rTS ) and the TLR-9 agonist CpG as an adjuvant [19] . To induce Th17 cells , we vaccinated TS-CD4-Tg mice intramuscularly with rTS combined with the TLR-1/2 agonist and Th17-skewing adjuvant Pam3CSK4 instead [32] , and compared these to mice receiving rTS with CpG . T . cruzi-reactive Th1 responses could be detected by IFN-γ ELISPOT assay among splenic CD4+ T cells recovered from TS-CD4-Tg mice vaccinated with rTS and CpG , but not from mice vaccinated with rTS and Pam3CSK4 ( S8A Fig ) . On the contrary , the use of Pam3CSK4 as an adjuvant induced the generation of T . cruzi-reactive Th17 cells detectable by IL-17A ELISPOT , while the use of CpG as an adjuvant did not ( S8B Fig ) . These data indicate that endogenous , T . cruzi-specific Th17 cells can be induced in vivo by vaccination when the appropriate biasing adjuvants for promoting Th17 cell responses are used .
Recent studies demonstrating that IL-17 confers resistance against T . cruzi infection raised the possibility that Th17 cells could play a protective role , although Th1 cells are typically considered the most important CD4+ T cell subset for immunity against T . cruzi as well as other intracellular organisms . Th17 cells are known to be important for protection against a number of extracellular bacteria and fungi [13 , 33 , 34] , but are not believed to function significantly against intracellular pathogens . In addition to the protective roles of Th17 cells in the human immune system , Th17 cells are also known to drive pathology in many autoimmune and inflammatory diseases , including multiple sclerosis and rheumatoid arthritis [8] . Previous studies demonstrating a protective function for IL-17 utilized models of global IL-17 deficiency [5 , 6] , precluding any specific interpretation of the role of Th17 cells , as IL-17 is produced by multiple innate and adaptive cellular sources . Therefore , it remained unclear whether Th17 cells were protective or pathological during T . cruzi infection . Our lab recently generated a line of T cell receptor transgenic mice with CD4+ T cells specific for a single immunodominant epitope of the trans-sialidase virulence factor; immune responses directed against this immunodominant antigen can be highly protective . This critical reagent has greatly advanced our ability to fully explore the roles of various CD4+ T cell subsets in T . cruzi infection . We studied parasite-specific Th1 and Th17 cells in vivo in adoptive transfer experiments and in vitro using macrophage infection assays and CD8+ T cell activation assays . With these studies , we have created strong evidence that Th17 cells can orchestrate a powerful protective response to T . cruzi infection: 100% of RAG KO mice co-transferred with Th17 cells and CD8+ T cells survives a normally lethal T . cruzi challenge . Mechanistically , we discovered that through the secretion of IL-17A , Th17 cells trigger the function of NADPH oxidase to provide direct protection to infected cells . A previous report suggests that IL-17A may function to retain T . cruzi within the endosomes of host cells [35] . Our discovery that IL-17A acts through NADPH oxidase , involved in the phagocyte respiratory burst , indicates that this cytokine may direct synergistic mechanisms to promote immunity in infected macrophages . NADPH oxidase functions in the cell to produce superoxide and other ROS; these reactive products could protect against T . cruzi by directly killing intracellular pathogens or by functioning as signaling molecules to activate other protective pathways . Future investigations should focus on elucidating the detailed mechanisms of protection and the signaling events involved in IL-17 receptor triggering of NADPH oxidase activity . In addition , we demonstrated that the major mechanism of protection provided by Th17 cells in vivo is the help provided to critical CD8+ T cells , the major effector cells responsible for T . cruzi immunity , by the secretion of IL-21 . IL-21 is known to have several effects on CD8+ T cell function , including the enhancement and maintenance of antigen-specific proliferation and function , particularly during chronic infection [30 , 31 , 36 , 37] . Therefore , Th17-primed ( and thus IL-21-primed ) CD8+ T cells may offer several advantages over Th1-primed CD8+ T cells in the course of T . cruzi infection , making Th17 cells a highly relevant CD4+ T cell subset to target during the development of vaccines for Chagas disease , and perhaps for other pathogens . Additional experiments will investigate whether Th17 help for CD8+ T cell activation through IL-21 can generate more antigen-specific cells with more broadly reactive epitope specificity than Th1 priming . Interestingly , in our model , neutralization of IL-17 had no detrimental effects for mice receiving Th17 cell transfer , although previous studies found that IL-17-deficient animals were more susceptible to T . cruzi infection [5–6] . Our findings that IL-17 alone can protect infected macrophages in vitro are consistent with these previous reports that IL-17 has protective functions . However , in vivo , CD8+ T cells are the major effector cells protective against T . cruzi infection , and through the secretion of IL-21 , Th17 cells efficiently promote a robust CD8+ T cell response independent of IL-17 . In the context of this strong CD8+ T cell activation , the directly protective effects of IL-17 may be secondary or dispensable . In our adoptive transfer experiments , in vitro-differentiated Th17 cells representing an adaptive immune response are given to mice prior to T . cruzi infection , to provide a convenient model for studies of a vaccine-induced protective state . Importantly , we have demonstrated that Th17 cells can indeed be induced by vaccination in TS-CD4-Tg mice using Th17-skewing adjuvants . Taken altogether , these data advance our knowledge of the protective CD4+ T cell responses to T . cruzi infection , and illuminate the relevant T cell subsets and cytokine profiles to target in vaccination strategies . This novel understanding advances the goal of developing of a human Chagas vaccine , which could prevent 12 , 000 deaths per year and even more instances of disability . More broadly , these studies provide the first clear and strong evidence that Th17 cells can function significantly in intracellular as well as extracellular immunity , suggesting that Th17 cells may be an advantageous immune response to similar pathogens , and indicating the need for a revised Th1/Th2/Th17 framework . Given their newly identified role in intracellular immunity and their potent effects on CD8+ T cells , Th17 cells may prove to have a more diverse influence on the host immune response than previously thought . With increased understanding of their many roles and the mechanisms underlying their desired protective effects , Th17 cells could become an attractive target for host-directed therapies against various infectious pathogens .
All animal studies were conducted in accordance with the “Guide for the Care and Use of Laboratory Animals” manual published by the National Research Council and endorsed by the Assessment and Accreditation of Laboratory Animal Care ( AAALAC ) , with the approval of the Institutional Animal Care and Use Committee ( IACUC ) /Animal Care Committee ( ACC ) , under protocol #1106 , assigned by Saint Louis University’s IACUC . All mice were maintained in an AAALAC accredited facility at Saint Louis University . BALB/c mice ( NCI Charles River Laboratories , Frederick , MD ) , RAG KO mice ( The Jackson Laboratory , Bar Harbor , ME ) , and T . cruzi trans-sialidase amino acids 57–74 ( TSaa57-74 ) -specific TCR transgenic mice on the Thy1 . 1 BALB/c background ( TS-CD4-Tg , generated in the Hoft laboratory ) were used throughout this study . Tbx21-/- mice ( The Jackson Laboratory , Bar Harbor , ME ) and IL-21R KO BALB/c mice ( provided by Dr . Manfred Kopf , ETH Zurich ) were used to generate CD4+ and CD8+ T cells for adoptive transfer experiments . C57BL/6 wild type mice ( WT , NCI Charles River Laboratories , Frederick , MD ) and B6 . 129S6-Cybbtm1Din/J mice lacking the gp91phox catalytic subunit of NADPH oxidase ( gp91phox KO , The Jackson Laboratory , Bar Harbor , ME ) were used to generate bone marrow-derived macrophages . Mice were housed under specific pathogen free conditions . Sample sizes were based on previous experience , and sample size calculations were approved by the Saint Louis University Animal Care Committee and follow AAALAC guidelines and recommendations . All studies included age- and gender-matched groups . Although no formal blinding was done , all key experimental results were reproducible in multiple experiments and when conducted by different individuals in the laboratory . The Tulahuèn strain of T . cruzi was passaged through BALB/c mice and Dipetalogaster maximus insects . Blood form trypomastigotes ( BFT ) were collected from infected mice and used for systemic challenges in BALB/c mice . Culture-derived metacyclic tryptomastigotes ( CMT ) were generated as described [38] and used for in vitro studies and for the systemic challenges of RAG KO BALB/c mice . Systemic challenges were performed by subcutaneous ( s . c . ) injection of 5 , 000 BFTs or CMTs re-suspended in in 100 μl phosphate buffered saline ( PBS ) . In vitro infections of macrophages were performed by culturing cells with CMT at 5–10 multiplicities of infection ( MOI ) . We used a DNA vaccine encoding the consensus TS enzymatic domain as described previously [19] , originally provided by Dr . Maurício M . Rodrigues ( São Paulo , Brazil ) . BALB/c mice were immunized with 100 μg TS-DNA or control DNA intramuscularly ( i . m . ) . We also vaccinated mice intranasally ( i . n . ) with 50 μg recombinant trans-sialidase plus 10 μg CpG-containing oligodeoxynucleotides ( ODN ) 1826 ( Invivogen , San Diego , CA ) . For dendritic cell ( DC ) vaccines , splenic DC were isolated after intraperitoneal ( i . p . ) injection of BALB/c mice with 5x106 B16 cells expressing Flt-3 ligand as previously described [39] . DC were induced to mature in vivo with 2 μg lipopolysaccharide given intravenously ( i . v . ) and harvested from spleens 16 hours later using α-CD11c microbeads ( Miltenyi Biotec , Inc . , Auburn , CA ) . The purity and activation status of DC were determined by staining for CD11c , CD86 and MHC class II , and >90% pure CD11c+ DC were obtained . DC were pulsed with 100 μg/ml TS peptides p7 ( TSaa57-74 , aa sequence KVTERWHSFRLPALVNV ) and/or 100 μg/ml TSKd1 ( TSaa359-367 , aa sequence IYNVGQVSI ) for 90 minutes at 37°C , washed and then injected i . v . into BALB/c mice ( 1x106 cells/mouse ) . To induce Th1 and Th17 cells in vivo , TS-CD4-Tg mice were anesthetized with Ketamine/Xylazine , and 50 μg of rTS protein mixed with either 50 μg CpG or 50 μg Pam3CysSerLys4 ( Invivogen , San Diego , CA ) re-suspended in 100 μl volumes of PBS were injected into the left and right tibialis anterior muscles ( 50 μl per side ) . We screened various 18mer peptide sequences from the consensus catalytic region of TS using IFN-ɣ ELIPSOT assays and identified an immunodominant I-Ad-restricted CD4+ T cell epitope ( p7 , TSaa57-74 , aa sequence KVTERWHSFRLPALVNV , S1 Fig , S1 Table ) . We prepared T cell clones as described previously [2] . Briefly , we immunized mice with TS DNA 3 times , each 2 weeks apart . Two weeks after the third immunization , spleen cells were harvested and CD4+ T cells purified with α-CD4 microbeads ( Miltenyi Biotec , Auburn , CA ) . T cells were stimulated in vitro with recombinant TS protein or irradiated spleen cells pulsed with TSaa57-74 ( p7 ) in the presence of 10 U/ml IL-2 . After 2 cycles of antigen stimulation and IL-2 expansion , T cell cloning was performed by limiting dilution . We amplified clonal TCR-α and TCR-β chain cDNA from a TS p7-reactive T cell clone . We then subcloned these TCR-α ( Vα2 ) and TCR-β ( Vβ8 . 2 ) sequences into a CD2 vector ( VAhCD2 vector , kindly provided by Dr . Richard DiPaolo and Dr . D . Kioussis ) [40] . We digested and gel purified the TCR DNA fragments and injected them into the pronucleus of C57BL/6 x BALB/c fertilized eggs , with the help of Dr . Mike White at Washington University , Saint Louis , MO . We screened weanlings by PCR for the TCR chains Vα2 and Vβ8 . 2 . We backcrossed TCR Tg mice onto a congenic Thy1 . 1+ BALB/c background to generate TCR transgenic peripheral T cells capable of recognizing the I-Ad-restricted p7 epitope . Subsequent generations of TCR Tg mice were screened by flow cytometry for Vβ8 . 2 surface expression on peripheral T cells or MHC II:KVTERWHSFRLPALVNV tetramer staining ( NIH Tetramer Core Facility , Emory University , Atlanta , GA ) . Trans-sialidase-specific T cell lines were derived from naïve TS-CD4-Tg mice . Spleens from naïve Thy1 . 1+ transgenic mice were harvested , and CD4+ T cells were purified by positive selection using α-CD4 microbeads ( Miltenyi Biotec , Auburn , CA ) . The CD4+ T cells were stimulated with either α-CD3/α-CD28-coated plates or with irradiated CD4+ and CD8+ depleted Thy1 . 2+ BALB/c splenocytes pulsed with TSaa57-74 ( p7 ) at 2 . 5 μg/ml . Th1 biasing conditions included 10 ng/ml recombinant mouse IL-12 ( Genetics Institute , Cambridge , MA ) and 10 μg/ml anti-IL-4 neutralizing mAb 11B11 ( NCI Biological Resources Branch , Frederick , MD ) . Th17 biasing conditions included 1 ng/ml recombinant human TGF-β ( Biolegend , San Diego , CA ) , 50 ng/ml recombinant mouse IL-6 ( Invitrogen , Carlsbad , CA ) , 20 ng/ml recombinant mouse IL-23 ( R&D Systems , Minneapolis , MN ) , 10 μg/ml anti-IL-4 ( NCI Biological Resources Branch ) , and 10 μg/ml anti-IFN-γ neutralizing mAb XMG1 . 2 ( BD Pharmingen , San Diego , CA ) . Every 3 days , Th1 cell cultures received 10 U/ml IL-2 while Th17 cell cultures received 20 ng/ml IL-23 . Cells were stimulated again 1 week later in the presence of the same biasing cytokines and antibodies . IL-2 or IL-23 was again added 3 days after the second stimulation . Th1 and Th17 cells were used one week after the second stimulation . Polyclonal Th17 cells were generated similarly using wild-type CD4+ T cells; on days 0 and 7 , these cells were stimulated with immature DCs pulsed with the supernatants of repeatedly freeze-thawed T . cruzi trypomastigotes . IL-17+ cell purification was performed using Miltenyi IL-17 cell secretion purification kits according to manufacturer instructions . In vitro generated Th1 or Th17 cells were transferred i . v . into RAG KO mice , along with polyclonal CD8+ T cells purified from naïve BALB/c mice by positive selection using α-CD8 microbeads ( Miltenyi Biotec , Auburn , CA ) . RAG KO mice received 0 . 5–2 x 106 Th1 or Th17 cells , and 3–5 x 106 CD8 T cells , by tail vein injection and were challenged s . c . one day later with 5 , 000 CMT . Parasitemia was measured post-infection microscopically using 1 . 5 μl peripheral blood taken from the tip of the tail . Protection was also assessed by survival of infected mice . At 3 , 6 , 7 and 10 days post-infection , the spleens from representative mice were harvested for immune studies . CD8+ T cells were transferred along with TS-CD4-Tg Th17 cells into RAG KO mice that were infected s . c . the following day with 5 , 000 T . cruzi CMT . In order to neutralize IL-17A , mice were injected intraperitoneally ( i . p . ) with 100 μg of either anti-mouse IL-17A mAb ( clone 17F3 ) or mouse IgG1 isotype control ( clone MOPC-21 ) , both from Bio X Cell ( West Lebanon , NH ) . Mice were injected with antibodies every 48 hours , beginning 1 day before infection and continuing for 30 days post infection . A recombinant adenovirus expressing murine IL-17A was originally described by Schwarzenberger et al . [41] and generously provided by Dr . Shabaana Khader ( Washington University , Saint Louis , MO ) . A control recombinant adenovirus ( Ad5 ) expressing β-galactosidase ( β-gal ) was also used [42 , 43] . Viruses were propagated in HEK-293A cells at Saint Louis University using endotoxin-free conditions , purified by CsCl as previously described [44] , and stored at -80°C in 20mM Tris ( pH 8 ) with 4% sucrose . RAG KO mice were injected i . v . with 5x109 PFU of recombinant adenovirus encoding either mIL-17A ( IL-17A AdV ) or β-gal ( ctrl AdV ) one day prior to infection and 7 days post-infection . Mice also received CD8+ T cells i . v . and were challenged s . c . with 5 , 000 T . cruzi CMT . IL-17A was measured in the serum of infected mice by ELISA according to the manufacturer’s instructions ( BioLegend , San Diego , CA ) . Cell surface staining was performed according to standard procedures using antibodies against murine CD3 , CD8 , CD4 , CD19 , CD44 and Thy1 . 1 ( CD90 . 1 ) , all purchased from BD Pharmingen ( San Diego , CA ) . For intracellular staining , monoclonal antibodies against IFN-γ ( BD Pharmingen ) , IL-17A , IL-17F , MIP-1α , RORγt and T-bet ( eBioscience , San Diego , CA ) were used . For intracellular staining of in vitro cultures , cells were restimulated with phorbol-12-myristate 13-acetate ( PMA , 10 ng/ml ) and ionomycin ( 500 ng/ml ) for 3 hours . For intracellular staining of spleen cells post-infection , cells were restimulated ex vivo with A20-TS for 6 hours . All intracellular stained cells were cultured with monensin ( GolgiPlug , 1 μl/ml ) and brefeldin A ( GolgiStop , 0 . 67 μl/ml ) for the last 3 hours of stimulation at 37°C ( both from BD Pharmingen ) . After surface staining , cells were fixed and permeabilized with Foxp3/Transcription Factor Staining Buffer Set ( eBioscience ) before intracellular staining , according to manufacturer’s instructions . Cells were analyzed with a BD LSR II flow cytometer and FlowJo v7 software ( Tree Star , Inc . ) . To generate peritoneal exudate macrophages ( PEM ) , BALB/c mice were injected i . p . with 100 μg Concanavalin A ( Sigma Aldrich , Saint Louis , MO ) . Four days later , PEM were harvested from the peritoneal cavity of these mice and plated in 8-well tissue culture slide chambers ( Nunc LabTek from Thermo Scientific , Waltham , MA ) at 1 . 25x106 cells/well in DMEM with 10% FBS . After 2–5 hours , nonadherent cells were washed away and PEM were infected with T . cruzi CMT at an MOI of 5 ( 6 . 25x106 parasites/well ) for 3 hours . Extracellular parasites were then washed away and TS-CD4-Tg Th1 or Th17 cells added at a macrophage to T cell ratio of 25:1 ( 5x104 cells/well ) . Two days later , slide chambers were removed from slides , and slides were washed , dried , fixed and stained with a modified Wright Giemsa stain , Diff-Quik ( IMEB , Inc . , San Marcos , CA ) . The number of infected macrophages was determined microscopically . Variations of this assay included using neutralizing antibodies against IFN-γ ( clone XMG1 . 2 , BD Pharmingen ) or IL-17A ( clone TC11-18H10 , BD Pharmingen ) at 10 μg/ml , using the purified cytokines mIFN-γ ( 1000 U/ml , Genentech , San Francisco , CA ) or recombinant mouse IL-17A ( 100 ng/ml , R&D Systems ) instead of T cells , and inhibiting inducible nitric oxide synthase ( iNOS ) with N6- ( 1-Iminoethyl ) -L-lysine ( L-NIL , 1mM , Sigma Aldrich , Saint Louis , MO ) . Supernatants from the infected macrophage cultures were collected 24 and 48 hours post infection . Nitric oxide concentration was estimated by measuring nitrite concentration in supernatants with the Griess reagent system according to the manufacturer’s instructions ( Promega , Fitchburg , WI ) . For experiments with bone marrow-derived macrophages ( BMDM ) , bone marrow was harvested from the femurs and tibiae from C57BL/6 WT OR C57BL/6 gp91phox KO mice . Cells were cultured on 96-well and 6-well plates with 20 ng/ml recombinant mouse M-CSF ( PeproTech , Rocky Hill , NJ ) in DMEM with 10% FBS for 7 days . BMDM were then removed from culture plates with CellStripper non-enzymatic dissociation reagent ( Corning , Corning , NY ) , transferred to 8-well slide chambers at 2x105 cells/well , and infected with CMT at MOI of 10 . Three hours post infection , extracellular parasites were washed away and either IFN-γ or IL-17A was added . Two days later , slides were washed , dried , fixed and stained as described above . Polyclonal CD8+ T cells were purified from naïve BALB/c mice by positive selection with α-CD8 microbeads ( Miltenyi Biotec ) , labeled with the cell-staining dye CFSE ( Invitrogen ) and stimulated with a suboptimal dose of plate-bound α-CD3 antibody ( 1 μg/ml , clone 145-2C11 , BD Biosciences ) in 96-well plates with 2x105 cells/well . Further activation/co-stimulation was provided by DCs ( purified from spleens with α-CD11c microbeads , Miltenyi Biotec ) and/or TS-CD4-Tg Th1/Th17 cells ( 1x105 cells/well ) . Cellular responses were analyzed 5 days post-activation by re-stimulating with PMA and ionomycin and analyzing by flow cytometry to measure proliferation ( by CFSE dilution ) , CD44 surface expression , and production of IFN-γ and MIP-1α . Additional modifications of this assay included using supernatants from activated Th1 or Th17 cells with or without a neutralizing antibody directed against IL-17A at 10 μg/ml ( BD Pharmingen ) . CD8+ T cells were also activated with supernatants by purified recombinant mouse IL-17A , IL-17F , IL-21 , IL-22 or GM-CSF at 100 ng/ml ( all from R&D ) instead of by co-culture with CD4+ T cells . Results are presented as either frequency of CD8+ T cells positive for the indicated activation marker or as fold increase as compared to that with α-CD3 activation alone . IL-21 concentration in supernatants was measured by ELISA according to manufacturer’s instructions ( eBioscience ) . Millititer HA 96-well microtiter plates with nitrocellulose bases ( Millipore , Bedford , MA ) were coated with 10 μg/ml of either murine anti-IFN-γ mAb ( clone R46A2 , BD Pharmingen , San Diego , CA ) or anti-IL-17A mAb ( clone TC11-18H10 , BD Pharmingen ) overnight at 4°C . Plates were washed with PBS 4 times and blocked with RPMI + 10% FBS at room temperature ( RT ) for at least 2 hours . Total spleen cells from infected RAG KO mice , spleen cells from immunized BALB/c mice , CD4+ T cell clones or peripheral blood mononuclear cells ( PBMCs ) from TS-CD4-Tg mice ( 1-3x105 cells/well ) were stimulated with negative control A20 cells ( NC A20 ) , stably TS-transfected A20 cells ( TS A20 ) , A20 cells pulsed with TS p7 ( KVTERWHSFRLPALVNV at 2 . 5 μg/ml; A20 + p7 ) or A20 cells pulsed with other overlapping peptides ( at 2 μM ) . Some ELISPOT assays included stimulations of Tg cells with peritoneal exudate macrophages infected with T . cruzi . Additional stimulations were done with L-cells transfected with I-Ad or I-Ed molecules ( courtesy of Dr . Ron Germain , NIH ) . Numbers of IFN-γ- or IL-17A-producing cells were detected with biotinylated anti-IFN-γ Ab ( clone XMG , BD Pharmingen ) or anti-IL-17A ( clone TC11-8H4 , BD Biosciences ) , streptavidin conjugated to horseradish peroxidase ( HRP ) ( Jackson Immunoresearch Laboratories , West Grove , PA ) and 3-amino-9-ethylcarbazole substrate precipitation . Results are reported as number of spot-forming cells ( SFCs ) . Nunc Maxisorp flat-bottom 96-well plates ( eBioscience ) were coated with 5 μg/ml of recombinant trans-sialidase protein at 4°C overnight . Plates were washed with PBS-T and blocked with 10% FBS at room temperature , then mouse serum samples serially diluted in PBS were added to wells and incubated overnight . Plates were washed and incubated with an anti-mouse IgG antibody conjugated to horse radish peroxidase to detect TS-specific antibodies . Plates were developed with TMB substrate , and the reaction was stopped with H2SO4 . Plates were analyzed at 450 nm with a reference of 540 nm , and endpoint titers were determined as OD readings above the highest OD measured for uninfected controls . Statistical analyses were performed using Prism v4 software ( GraphPad Software , Inc . , La Jolla , CA ) . Mann-Whitney U tests or unpaired Student t tests were used to compare responses between groups . Log-rank ( Mantel-Cox ) and Gehan-Breslow-Wilcoxon tests were used to compare survival between groups . All data are presented with standard errors , ranges or scatter plots to allow easy assessment of the variation . | Chronic infection with the intracellular parasite Trypanosoma cruzi results in Chagas disease , an illness endemic in more than 20 countries that leads to life-threatening cardiac and gastrointestinal dysfunction . Although CD4+ Th1 cells are known to be protective against T . cruzi infection , less is known about the role of other CD4+ T cell subsets . We demonstrate that CD4+ Th17 cells are also highly protective against T . cruzi infection , even outperforming Th1 cells in protection from T . cruzi-related death , despite the standard conception that Th17 cells are only important in the defense against extracellular pathogens . The novel discovery that Th17 cells are significantly more protective than Th1 cells against T . cruzi infection has increased our understanding of the advantageous immune responses for this major human pathogen , and may guide future efforts toward vaccine development . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"blood",
"cells",
"t",
"helper",
"cells",
"medicine",
"and",
"health",
"sciences",
"immune",
"cells",
"enzyme-linked",
"immunoassays",
"respiratory",
"infections",
"immune",
"physiology",
"spleen",
"immunology",
"parasitic",
"diseases",
"parasitic",
"protozoans",
"pulmonology",
"protozoans",
"cytotoxic",
"t",
"cells",
"immunologic",
"techniques",
"research",
"and",
"analysis",
"methods",
"white",
"blood",
"cells",
"animal",
"cells",
"t",
"cells",
"immunoassays",
"trypanosoma",
"cruzi",
"trypanosoma",
"cell",
"biology",
"physiology",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"organisms"
] | 2016 | Th17 Cells Are More Protective Than Th1 Cells Against the Intracellular Parasite Trypanosoma cruzi |
Ebolaviruses cause hemorrhagic fever with up to 90% lethality and in fatal cases , are characterized by early suppression of the host innate immune system . One of the proteins likely responsible for this effect is VP24 . VP24 is known to antagonize interferon signaling by binding host karyopherin α proteins , thereby preventing them from transporting the tyrosine-phosphorylated transcription factor STAT1 to the nucleus . Here , we report that VP24 binds STAT1 directly , suggesting that VP24 can suppress at least two distinct branches of the interferon pathway . Here , we also report the first crystal structures of VP24 , derived from different species of ebolavirus that are pathogenic ( Sudan ) and nonpathogenic to humans ( Reston ) . These structures reveal that VP24 has a novel , pyramidal fold . A site on a particular face of the pyramid exhibits reduced solvent exchange when in complex with STAT1 . This site is above two highly conserved pockets in VP24 that contain key residues previously implicated in virulence . These crystal structures and accompanying biochemical analysis map differences between pathogenic and nonpathogenic viruses , offer templates for drug design , and provide the three-dimensional framework necessary for biological dissection of the many functions of VP24 in the virus life cycle .
The ebolaviruses and marburgviruses are enveloped , non-segmented , negative-strand RNA viruses that belong to the family Filoviridae . There are five antigenically distinct ebolaviruses that are ∼40% different in amino acid sequence , and are each named after the location of the outbreak during which they were first identified: Zaire ( now known simply as Ebola virus or EBOV ) , Sudan virus ( SUDV ) , Taï Forest virus ( TAFV ) , Reston virus ( RESTV ) and Bundibugyo virus ( BDBV ) . Marburgviruses and most ebolaviruses cause severe hemorrhagic fever in both humans and nonhuman primates , with fatality up to 90% . The exception is RESTV , which appears to be non-pathogenic in humans , although it remains pathogenic to non-human primates [1] , [2] . Reasons why RESTV has not caused disease in humans are unclear . However , microarray analyses have shown that RESTV has a reduced ability to suppress host immune responses [3] . For the pathogenic ebolaviruses , early suppression of host interferon ( IFN ) production and signaling plays a decisive factor in disease outcome [4] , [5] . Two proteins of the ebolaviruses are used in this strike . The protein VP35 blocks production of IFN-α/β [6] by binding dsRNA , a key hallmark of viral infection , and shielding it from recognition by host immune sensors such as RIG-I and MDA-5 [7] , [8] . By contrast , the protein VP24 inhibits signaling downstream of both IFN-α/β and IFN-γ by sequestering karyopherin α proteins ( α1 , α5 and α6 ) [9] . Binding to these proteins prevents them from shuttling otherwise activated , phosphorylated STAT1 to the nucleus [9]–[11] . STAT1 belongs to the STAT family of transcription factors , is a key mediator of the IFN response pathway [12]–[14] and plays an essential role in the immune response to viruses [15]–[17] . STAT1 predominately exists in an unphosphorylated form ( U-STAT1 ) . Numerous immune factors like type I and type II interferon [14] , [18] , [19] , interleukins like IL-6 and IL-10 [20]–[23] , growth factors [20] , [24]–[26] , angiotensin [27] , and TNFα [28] cause STAT1 to be phosphorylated ( P-STAT1 ) by the Janus family kinases ( JAKs ) . Upon phosphorylation , P-STAT1 either dimerizes or forms a complex with IFN α/β-stimulated gene factor 3 ( ISGF3 ) [18] , [29] , [30] , and is subsequently transported to the nucleus via karyopherin α proteins where it regulates genes involved in the immune response [31]–[33] . The importance of STAT1 to the antiviral response is underlined by the fact that viruses ( and other microbes ) have evolved proteins that inhibit every step of STAT1 activation [14] . As examples , the V proteins of Nipah and Hendra viruses and the P protein of rabies virus directly bind to P-STAT1 to sequester it in the cytoplasm [34]–[36] . By contrast , the P protein of measles virus and an unidentified protein of human metapneumovirus prevent phosphorylation of STAT1 [37] , [38] , while the VH1 protein of vaccinia virus and the NS5 protein of Japanese encephalitis virus actively dephosphorylate the P-STAT1 complex [39] , [40] , and the V protein of mumps causes ubiquitination and degradation of P-STAT1 [41] . The VP24 protein of ebolavirus , in yet another mechanism , prevents nuclear translocation of P-STAT1 by binding to the karyopherin α transporter proteins [9]–[11] . In addition to its role in interferon antagonism , ebolavirus VP24 has also been proposed to associate with membranes [42] , [43] , and is important for assembly and function of the viral ribonucleoprotein complex ( RNP ) [44]–[46] , where VP24 binds to the viral nucleoprotein NP [45] . VP24 has no sequence homology to any known protein and the molecular mechanisms by which VP24 suppresses immune signaling and contributes to RNP assembly are poorly understood . Here , we demonstrate that VP24 also binds directly to STAT1 itself , present the first X-ray crystal structures of VP24 from both Sudan and Reston viruses , and map a possible site of STAT1 interaction on the VP24 crystal structure by deuterium exchange mass spectrometry ( DXMS ) . The biochemical and structural analysis presented here identifies a new function by which VP24 may contribute to and/or prolong innate immunosuppression , and provides the necessary three-dimensional templates for understanding the multiple roles of VP24 in the ebolavirus life cycle and design of antiviral compounds against them .
Other negative-sense viruses encode proteins that suppress innate immune signaling by direct interaction with STAT1 . VP24 was previously known to indirectly affect STAT1 by binding karyopherins to prevent them from translocating phosphorylated STAT1 ( P-STAT1 ) . However , we wondered if VP24 could play a more direct role as well . To answer this question , we performed an ELISA to test binding of purified Ebola virus VP24 or Sudan virus VP24 to purified STAT11–683 ( truncated prior to its phosphorylation site at Tyr701 ) . BSA was used as a negative control . Indeed , VP24 is able directly associate with STAT11–683 ( Figure 1 ) . Although it remains to be determined if VP24 binds equally well to full-length P-STAT1 and U-STAT1 , the initial finding that VP24 binds STAT11–683 suggests an additional , unexplored way by which VP24 might contribute to innate immune suppression . In order to provide 3D templates for understanding VP24 and its many roles in immune evasion , replication and assembly , we crystallized VP24 from Sudan virus ( two versions crystallized: SUDV1–233 and SUDV11–233 ) and Reston virus ( one version crystallized: RESTV11–237 ) ( Figure S1a ) . We determined the structure of SUDV11–233 at 2 . 1 Å resolution by multiwavelength anomalous diffraction ( MAD ) using selenomethionine-incorporated protein expressed recombinantly in E . coli . We subsequently determined structures of SUDV1–233 and RESTV11–237 by molecular replacement , both at 2 . 0 Å resolution ( Table 1 ) . VP24 adopts a compact , single domain , α/β structure of novel fold ( DaliLite v . 3 [47] ) . The overall shape of VP24 resembles a triangular pyramid of dimensions 73 Å×30 Å×30 Å . The three faces of the pyramid are numbered 1 , 2 and 3 ( Figures 2 and S1b ) . A collection of α helices ( α1 and α5-10 ) and a small , three-stranded , antiparallel β sheet ( β1-3 ) form the top of the pyramid with the N-terminus at the apex . A five-stranded antiparallel β sheet ( β4-8 ) forms the center , while a second collection of α helices ( α2-4 ) forms the base . Portions of the C-terminal region resemble prior de novo predictions: as predicted , helices 5–8 are indeed observed , helix 8 is quite long , and a β sheet exists at the base of the structure . Differences between the prediction and the experimental structure are that a three-stranded sheet was predicted , but a five-stranded sheet is observed [48] and that an armadillo repeat-type domain structure was predicted , but no such domain is observed in VP24 . VP24 is 63% identical among ebolaviruses and ∼30% identical between ebola- and marburgviruses . Regions of high sequence conservation congregate on Faces 1 and 3 ( Figures 3 and S2a ) . The conserved center of Face 1 is formed by α5 , β3 , β5 and β8 . The conserved center of Face 3 is formed by α5 , α8 , β5 and β6 . The base of each of Faces 1 and 3 also contains a conserved cavity , and the two cavities are located adjacent to each other on the protein surface . The Face 1 cavity is hydrophobic and is 14×14×12 Å in size ( Figure 4a ) . The interior of the cavity is lined with five absolutely conserved leucine residues: L57 , L75 , L79 , L198 , and L221 . The entrance to the hydrophobic cavity ( 11 Å wide ) appears to be gated by two residues ( Y172 and M71 ) that point away from each other in RESTV VP24 but toward each other in SUDV VP24 , appearing to block the hydrophobic cavity ( Figure S3a–S3b ) . The Face 3 cavity is shallower than that of Face 1 ( 18×14×5 Å ) , and is hydrophilic rather than hydrophobic ( Figure 4b ) . Five residues that are conserved across all filoviruses ( S178 , E180 , I189 , T191 , and E200 ) populate the base of the cavity . Six conserved residues circle the rim ( P77 , T193 , K206 , and M209 are conserved across all ebolaviruses; H78 and N82 are conserved across all filoviruses ) . Also , three residues conserved across the filoviruses ( L75 , F76 , and L198 ) line the edge between the Face 1 and Face 3 cavities and are accessible from either side . Serial passage studies to confer lethality of EBOV to rodents resulted in five mutations in VP24 ( H186Y , T187I , M71I , L147P [49] , and T50I [50] ) , four of which lie in or near these cavities . H186 ( Y186 in SUDV ) , T187 ( A187 in SUDV ) and T50 ( N50 in SUDV and S50 in RESTV ) lie on the rim of the Face 3 cavity ( Figure 4c ) . M71 forms the “gate” to the Face 1 cavity . The fifth residue , L147 ( M147 in SUDV ) , is located toward the top of the pyramid in helix α8 , and is accessible from Face 3 . L147 is thought to be involved in karyopherin α binding ( Figure 4c ) . In infected cells , EBOV VP24 binds to karyopherin α1 , α5 , and α6 to prevent translocation of P-STAT1 into the nucleus [10] . Previous mutagenesis studies have shown VP24 residues W42 and 142–146 to be critical for karyopherin α1 binding [11] . W42 is buried in the interior of the single globular VP24 fold . Hence , mutagenesis of W42 most likely compromised the VP24 structure and affected karyopherin α1 affinity indirectly ( Figure 4c ) . By contrast , residues 142–146 are exposed to solvent and would be available to directly bind karyopherin α1 . As previously described , an L147P mutation ( α8 , adjacent to residues 142–146 ) in EBOV VP24 increases virulence in guinea pigs [49] . Unlike the ebolaviruses , the VP24 protein of Marburg virus does not block interaction of P-STAT1 with karyopherin α1 [51] . W42 is conserved between ebola- and marburgviruses , but residues 142–147 are not . Residues 142–147 are K ( E/D ) QLS ( L/M ) in the ebolaviruses but are GIYLTS in the marburgviruses ( Figure S2a ) . Deuterium exchange mass spectrometry ( DXMS ) is able to rapidly map footprints of protein-protein binding sites and offers a broader picture than analysis of point mutants alone [52]–[56] . In this method the ability of peptide amide hydrogens to freely and reversibly exchange with solvent deuterium is measured . Hydrogens for which mobility is restricted ( by conformational anchoring and/or ligand binding ) exchange more slowly . Hydrogens for which mobility is unrestricted ( conformational mobility ) exchange more rapidly . We performed comparative DXMS studies on VP24 alone and VP24 in complex with STAT11–683 . The resulting exchange maps identify some peptidic regions of VP24 that exchange with solvent less rapidly when in complex with STAT1 ( possible binding sites ) , and other regions of VP24 that exchange with solvent more rapidly when in complex with STAT1 , perhaps due to conformational change and increased mobility . In the presence of purified STAT11–683 , VP24 peptidic regions 96–98 and 106–121 demonstrate slower H/D exchange kinetics , suggesting a site of protein-protein interaction ( Figure 5 ) . By contrast , VP24 peptidic regions 71–79 and 181–198 demonstrated increased H/D exchange in the presence of STAT11–683 , suggesting possible conformational change with enhanced flexibility . The faster exchanging peptides 71–79 and 181–198 map to helix α3–4 and strands β5–7 , respectively . Both of these secondary structure elements exist in the polar cavity at the bottom of Face 3 and are highly conserved across the filoviruses . The slower-exchanging peptides 96–98 and 106–121 map to helices α5 and α6 , also on the conserved portion of Face 3 . Helices α5 and α6 are amphipathic in nature: the hydrophobic side of each coil points into the core of VP24 . Hydrophilic residues extend to the surface of Face 3 . The C-terminal region of α5 is negatively charged and the N-terminal portion of α6 is positively charged . Mapping of sequence differences between RESTV and the major pathogenic ebolaviruses ( SUDV and EBOV ) onto the RESTV11–237 structure indicates that RESTV VP24 differs in ∼30 sites ( Figure S4 ) . One of these sites , a cluster of residues L136 , R139 and S140 ( in RESTV ) , is next to the 142–146 loop , which is important for binding karyopherin α1 [11] ( Figure S4 and Figure S2a for sequence alignment ) . A second site is the cluster of residues L107 , H109 , T116 and G120 that exists within the 106–121 polypeptide that exhibits decreased H/D exchange in complex with STAT11–683 and may serve as a STAT1 binding site . A third site , the cluster of residues S184 , T185 , H186 , T187 and F197 , lies in the 181–198 polypeptide that undergoes enhanced H/D exchange in the presence of STAT11–683 . A fourth site , V201 , lies next to this region . A fifth site , residue S50 , was previously implicated in a serial passage study to confer lethality to mice [50] . Previous studies analyzing VP24 in the context of whole cell lysate found that the majority of VP24 was monomeric . A smaller portion appeared as a high molecular weight aggregate and a smaller oligomer , likely a tetramer [42] , [43] . We performed gel filtration analysis of purified , full-length SUDV , RESTV and EBOV VP24 ( whether produced in E . coli or 293T cells ) and find that purified VP24 from all three viruses is monomeric in solution ( Figure S5 ) . The significance of the multimerized portion observed in cell lysate is unclear . Perhaps a portion of VP24 homo-oligomerizes in cells , or perhaps factors present in whole-cell lysate are needed for VP24 to form oligomers . Crystal packing can sometimes illustrate biologically relevant assemblies , but no tetrameric or other higher oligomeric interactions are observed in crystals of SUDV or RESTV VP24 . One pairwise VP24-VP24 interaction is conserved in the crystal packing between the SUDV ( P3121 ) and RESTV VP24 ( P1211 ) structures , although it is currently unclear if it is biologically relevant ( Figure S3c ) . This interaction involves α1 , β1–3 and the N-terminal region of α8 , and buries residue L147 that was previously implicated for virulence [49] , although 142–146 remain solvent exposed . Another crystal lattice interaction observed in both RESTV and SUDV structures involves packing of the hydrophobic N-terminal regions of VP24 into the Face 3 pocket of a neighboring molecule ( Figure S3d–S3e ) .
The crystal structures presented here illustrate the novel , pyramidal fold of ebolavirus VP24 . In this work , we have also identified STAT1 as a new binding partner of VP24 and have used DXMS to suggest that residues 96–98 and 106–121 are contained in a putative binding site for STAT11–683 . Although VP24 differs by 37% in protein sequence among the ebolaviruses , there are large patches of complete conservation on Faces 1 and 3 including the two pockets in these faces at the base of the pyramid . Several residues , found in serial passage studies to increase virulence of Ebola virus , map to these sites , although the precise role of the conserved pockets remains unclear . Another residue identified in these studies maps to a site thought to be involved in binding host karyopherin α proteins . The putative STAT1-binding site identified by DXMS lies in the conserved region of Face 3 and is distinct from the site proposed to interact with karyopherin α1 . Crystal structures presented here include VP24 from an ebolavirus that is pathogenic to humans ( Sudan virus; SUDV ) and VP24 from an ebolavirus that thus far , appears nonpathogenic to humans ( Reston virus; RESTV ) , although it is lethal to nonhuman primates . The overall folds of SUDV and RESTV VP24 are similar , as expected ( r . m . s . d . of 0 . 81 Å; also see Figure S2b for structural alignment ) . Specific viral or host factors responsible for the differences in pathogenicity between these viruses have not yet been identified , but it has been proposed that RESTV has a diminished ability to suppress cellular IFN-α/β and IFN-γ responses [3] . Residues in VP24 that are unique to RESTV often colocalize with residues that appear to be important for karyopherin and STAT1 binding , or are important for virulence in rodents through an unknown mechanism . The location of these RESTV-specific amino acids invites speculation that RESTV VP24 and EBOV/SUDV VP24 could potentially bind immune factors like karyopherins and STAT1 with differing affinity . Here we have found that purified VP24 binds directly to purified STAT1 truncated prior to its phosphorylation site . In a healthy cell , STAT1 exists in an unphosphorylated form . During viral infection , production of interferons and cytokines leads to phosphorylation and homodimerization of STAT1 or heterodimerization of STAT1 with its β isoform . The resulting P-STAT1 dimer is then transported by karyopherin α proteins into the nucleus where it controls transcription-regulated genes . Interestingly , this P-STAT1 may have a different oligomeric structure than U-STAT1 [57] , [58] . U-STAT1 is not inactive , but rather , is also important in regulation of the immune response . Interestingly , U-STAT1 functions in different ways than its phosphorylated counterpart . U-STAT1 is transported into the nucleus [59] , [60] by direct involvement with nucleoporins [61] , and does not need transport by karyopherins . In the nucleus , U-STAT1 activates and prolongs the expression of a number of IFN-induced immune regulatory genes like IFI27 , IFI44 , OAS , and BST2 [62] . U-STAT1 functions independently of P-STAT1 and the set of genes on which it operates can be distinct from those of P-STAT1 [59] . U-STAT1 and P-STAT1 also differ temporally: the phosphorylation of STAT1 lasts for several hours , but the presence of U-STAT1 persists for several days [62] , [63] . In this way , U-STAT1 is likely to be able to prolong an antiviral state . Hence , both P-STAT1 and U-STAT1 play multiple roles in antiviral defense , and may play somewhat different roles in different cell types . By affecting both P-STAT1 ( by binding karyopherins and/or possibly by forming a karyopherin-STAT1-VP24 tertiary complex ) and U-STAT1 ( if it binds full-length U-STAT1 as well as unphosphorylated STAT11–683 ) , VP24 could prevent or dampen antiviral responses through multiple routes . The combination of both ebolavirus VP24 and ebolavirus VP35 ( which acts upon virally induced dsRNA ) in the infected cell offers greater coverage of the different pathways by which antiviral responses occur . Interestingly , plasmocytoid dendritic cells ( pDCs ) , which are major producers of type I interferon [64] , are insensitive to VP35 inhibition [65] . Perhaps VP35 and VP24 exert a synergistic effect , and/or VP24 functions in cells where VP35 does not . Although VP24 is key to the virulence of ebolaviruses , little is known about it due , in part , to the lack of any structural information on VP24 and the lack of any homology to other known proteins . We have shown that purified VP24 and purified STAT11–683 interact . The functional manifestation of this interaction remains to be determined . Does VP24 target STAT1 for degradation , sequester it in the cytoplasm or in high-molecular weight complexes , or prevent its phosphorylation ? Does VP24 also bind P-STAT , and does it exhibit a preference for one form over the other ? Does VP24 bind other STATs in addition to STAT1 ? Intriguingly , STAT3 shares about 72% sequence homology with STAT1 [66] , and operates in intestinal epithelia where it regulates mucosal wound healing [67] . Inactivation of STAT3 may contribute to colitis and clinical manifestations of Ebola virus infection like abdominal pain and bloody stools [68] , [69] . Another question is if any of the mapped differences between RESTV and EBOV/SUDV VP24 are linked to or are responsible for the differences in pathogenicity in humans . The structures presented here provide a framework for answering these and other questions about the multiple roles of VP24 in the viral lifecycle . These structures also provide the much-needed templates for design of antiviral drugs to inhibit key functions of VP24 in transcription , replication , and immunosuppression .
VP24 from Sudan virus ( SUDV1–233 and SUDV11–233 in pET46 Ek/LIC vector ) was expressed in E . coli Rosetta-gami 2 ( DE3 ) pLysS cells . Truncation of the C terminus permitted protein solubility in the absence of detergents . The N-terminal truncation used in the first SUDV constructs was based on general construct screening . Cultures were grown in LB medium supplemented with ampicillin ( 100 µg ml−1 ) , and expression was induced by the addition of 0 . 5 mM IPTG at 16°C . Harvested cells from overnight induction were resuspended in lysis buffer ( 50 mM NaH2PO4 , pH 8 . 0 , 0 . 3 M NaCl , 10 mM imidazole ) for lysis at 25 , 000 psi using a Microfluidizer processor . The lysed mixture was then centrifuged for 50 minutes at 16 , 000 r . p . m . in a JA-17 rotor ( Beckman Coulter ) . The supernatant was loaded on a HisTrap FF crude column ( GE Healthcare ) with a step gradient of 30 mM and 500 mM imidazole in lysis buffer . SUDV1–233 and SUDV11–233 VP24 were further purified by size exclusion on a HiLoad 16/60 Superdex 75 prep grade column ( Amersham Pharmacia ) in 10 mM Tris-HCl , pH 8 . 0 , 0 . 3 M NaCl . Full-length SUDV VP24 was expressed and purified essentially as above . Addition of 2 . 5 mM CHAPS throughout the purification enhanced solubility , and a Superdex 200 column was used for size exclusion . Selenomethionine-incorporated SUDV11–233 was expressed and purified as follows: 2 mL of an overnight culture in LB broth was transferred into 20 mL LB containing 0 . 4% glycerol and 100 µg ml−1 ampicillin . After a one-hour incubation , the cells were harvested by centrifugation at 3000 r . p . m . and resuspended in 20 mL M9 minimal media , then transferred into 1 L M9 media containing ampicillin . At OD600 0 . 4 , L-isoleucine , L-leucine , L-lysine , L-phenylalanine , L-threonine , and L-valine were added to final concentrations of 100 mg/L each , prior to addition of L-selenomethionine ( to 60 mg/L ) . The culture was induced with IPTG after 15 min . Cells were harvested after 4 hr and purified as described above . Full-length SUDV ( in pTriEx 5 vector ( Novagen ) ) was also transiently expressed in mammalian HEK293T cells in a five-layer CellStack ( corning ) . The cells were transfected at 60% confluency with 420 µg of DNA and 1 . 2 mg of PEI diluted in 42 ml of PBS . The PEI mixture was incubated at room temperature for 20 min . before adding to the cells . After 48 hours , cells were freeze-thawed three times and lysed in 10 mM Tris-HCl , pH 8 . 0 , 0 . 3 M NaCl . Protein was affinity purified with strep-tactin superflow beads ( Qiagen ) , then further purified by size exclusion on a Superdex 200 10/300 GL ( GE Healthcare ) . 10 µl of the peak fraction was run on SDS-PAGE , and probed by Western blot with an anti-strep antibody . All SUDV VP24 proteins contain a valine to alanine substitution at position 22 from the GenBank deposited sequence . Residue 22 is on helix α1 and is buried within the structure ( Figure S6 ) . The Sudan virus ( strain Boniface ) replicon was a gift of Dr . John M . Dye ( USAMRIID ) . Oligonucleotides were purchased from Valuegene Inc . Full-length Ebola virus ( Zaire , EBOV ) VP24 and both full-length and N- and C-terminally truncated Reston ( RESTV ) virus VP24 were expressed and purified in E . coli as previously described . Full-length EBOV VP24 was cloned into the pET46 Ek/LIC vector from cDNA that was a gift from Dr . Viktor Volchkov ( Claude Bernard Université de Lyon 1 ) . cDNA for RESTV VP24 was synthesized by GenScript ( Piscataway , NJ ) . Both full length RESTV and RESTV11–237 were subcloned into pET46 Ek/LIC for expression . Truncated , unphosphorylated STAT11–683 ( human ) in a pET46 Ek/LIC vector was expressed in E . coli Rosetta-gami 2 ( DE3 ) pLysS cells . Cultures were grown in LB medium supplemented with ampicillin ( 100 µg ml−1 ) , and expression was induced by the addition of 0 . 5 mM IPTG at 25°C overnight . Harvested cells were then resuspended in lysis buffer ( 50 mM NaH2PO4 , pH 8 . 0 , 0 . 3 M NaCl , 5 mM BME , 10 mM imidazole ) for lysis at 25 , 000 psi using a Microfluidizer processor . Next , the lysed mixture was centrifuged for 50 minutes at 16 , 000 r . p . m . in a JA-17 rotor ( Beckman Coulter ) . The supernatant was loaded on a HisTrap FF crude column ( GE Healthcare ) with a gradient of 10 mM to 500 mM imidazole in lysis buffer . STAT11–683 was further purified by size exclusion on a Superdex 200 10/300 GL ( GE Healthcare ) in 10 mM Tris-HCl , pH 8 . 0 , 0 . 3 M NaCl , 5 mM BME . STAT1 cDNA was a gift from Dr . Christopher Basler ( Mount Sinai School of Medicine ) . SUDV1–233 crystallized in 0 . 1 M HEPES , pH 7 . 5 , 0 . 1 M MgCl2 , and 8% ( v/v ) PEG 550 mme . SUDV11–233 crystallized in 0 . 1 M HEPES , pH 7 . 0 , 6% MPD , and 15% ( w/v ) D- ( + ) -sucrose . SeMet-derivatized SUDV11–233 crystallized in 0 . 1 M HEPES , pH 7 . 0 , 10% MPD , and 1 mM DTT . RESTV11–237 crystallized in 0 . 1 M Bis-Tris , pH 5 . 5 , 0 . 2 M NaCl , 8% ( w/v ) PEG 3350 , and 14% ( w/v ) D- ( + ) -sucrose . All crystals were grown by the hanging-drop vapor diffusion method at 22°C . Crystals were flash frozen in liquid nitrogen and cryoprotected with their reservoir solutions supplemented with 40% PEG 550 mme for SUDV1–233 , 40% sucrose for SUDV11–233 and RESTV11–237 , and 40% MPD for SeMet-derivatized SUDV11–233 . Both SUDV structures contain one molecule in the asymmetric unit while RESTV11–237 contains two molecules in the asymmetric unit . Diffraction data were collected at 100 K on SBC 19ID ( Advanced Photon Source , Argonne National Laboratory ) and BCSB 5 . 0 . 2 ( Advanced Light Source , Lawrence Berkeley National Laboratory ) , and were processed either with HKL2000 [70] or d*Trek [71] ( Table 1 ) . Experimental phases for SUDV11–233 were generated by MAD ( multiwavelength anomalous diffraction ) using Auto-Rickshaw [72] . Five of the seven internal SeMet residues were located and their locations were verified by hand through a difference Fourier anomalous electron-density map . Using the experimentally phased map , the orientations of two helices were determined in the initial partial model and the rest of the model was built using the MRSAD ( molecular replacement with single-wavelength anomalous diffraction [73] ) method in Auto-Rickshaw [72] . The structure of SUDV1–233 was determined by molecular replacement ( also Auto-Rickshaw [72] ) using SUDV11–233 as the search model . Refinement of both structures was performed with Phenix . refine [74] in PHENIX [75] and rebuilding was carried out in COOT [76] . Final rounds of refinement included TLS parameters [77] for SUDV1–233 and SUDV11–233 . The quality of the structures was validated with MolProbity [78] and Procheck [79] . 92 . 7% ( SUDV1–233 ) and 93 . 1% ( SUDV11–233 ) of residues are in the most favored region of Ramachandran plots , with no residues in the disallowed regions . The final model of SUDV1–233 contains residues 9–106 and 113–232 with residues 113 and 210–213 replaced with alanines . The final model of SUDV11–233 contains residues 13–61 , 71–107 , 114–209 , and 212–228 . Residue 209 was replaced with alanine . RESTV11–237 was determined by molecular replacement in Auto-Rickshaw [72] using SUDV11–233 as the initial search model . The structure was refined with Phenix . refine [74] in PHENIX [75] and rebuilt in COOT [76] . Separate NCS restraints and TLS parameters [77] over the two molecules of VP24 in the asymmetric unit were used during initial refinement . The quality of the structure was validated with MolProbity [78] and Procheck [79] . 92 . 3% of residues are in the most favored region of Ramachandran plots and no residues are in the disallowed regions . The final model contains residues 11–62 , 70–203 , and 217–231 in molecule A and residues 15–61 , 70–108 , 113–203 , and 216–231 in molecule B . Two residues ( 203 and 216 ) in molecule B were replaced with alanines . Figures were created using PyMol [80] ( Delano Scientific ) . Atomic coordinates and structure factors have been deposited in the Protein Data Bank under the accession codes 3VNE , 3VNF , and 4D9O for SUDV1–233 , SUDV11–233 , and RESTV11–237 , respectively . 50 µl of each VP24 ( SUDV1–233 and EBOV1–233 ) was bound to ELISA plates ( Corning Costar 3690 ) at 0 . 01 mg/ml in 10 mM Tris-HCl , pH 8 . 0 , 0 . 3 M NaCl , overnight at 4°C . Plates were then blocked for one hour at room temperature with 3% BSA . After washing with PBS containing 0 . 05% TWEEN 20 , 50 µl of STAT11–683 with a C-terminal HA-tag was added at 0 . 03 mg/ml and allowed to bind for two hours at room temperature . Plates were again washed , 50 µl of anti-HA ( Covance ) was added at 1 µg/ml and incubated at room temperature for one hour . After a third washing step , 50 µl of HRP-conjugated secondary antibody ( Thermo Scientific Pierce ) was added and allowed to incubate for one hour at room temperature . Plates were developed with TMB Substrate Kit ( Pierce ) and read at 450 nm . BSA was used as a negative control . Peptide amide hydrogens continuously and reversibly interchange with hydrogen present in water . In structured proteins , most amide hydrogens are infrequently exposed to water , and exchange only when dynamic fluctuations in the protein structure transiently reveal them to solvent . Changes in exchange rates after ligand binding or other protein perturbations allow detection and localization of protein binding surfaces and conformational changes . In DXMS , the aqueous phase is supplemented with deuterated water , so that each exchange event produces a 1 Dalton increase in the mass of exchanged peptide amide protons in the protein . An initial exchange-dependent labeling step is performed by adding deuterated water to a solution of the protein at physiologic pH , and ionic strength . As deuterium-labeling progresses , aliquots are exchange-“quenched” by shifting pH to 2 . 7 and cooling to 0°C or below , conditions that dramatically slow further exchange and loss of deuterium label from the protein even when the protein structure is subsequently disrupted . The site and amount of deuterium that exchanged onto the protein are quantified ( under continued quench conditions ) by rapid denaturation , optional disulfide-reduction and digestion by solid-phase pepsin into overlapping fragments of ∼3–15 amino acids . The perturbed masses of the resulting peptides , and therefore their deuterium content , are quantified by liquid chromatography- mass spectrometry . Prior to the deuteration studies , quench conditions that produced an optimal pepsin fragmentation pattern were established as previously described [52] , [53] , [55] , [56] . For SUDV VP24 ( 10 mg/ml stock solution ) and SUDV VP24-STAT1 ( 12 mg/ml stock solution ) , functional deuteration of proteins was performed by mixing 1 µl of stock solution with 1 µl of H2O buffer ( 8 . 3 mM Tris , 150 mM NaCl , in H2O , pH 7 . 2 ) and then diluted into 6 µl of D2O buffer ( 8 . 3 mM Tris , 150 mM NaCl , in D2O , pDREAD 7 . 2 ) at 0°C . At 10 s , 100 s and 1000 s , the deuterium exchange was quenched by adding 12 µl of optimized quench ( 1 . 6 M GuHCl , 0 . 8% formic acid , 16 . 6% glycerol ) and then samples were frozen at −80°C . In addition , nondeuterated samples ( incubated in H2O buffer mentioned above ) and equilibrium-deuterated samples ( incubated in D2O buffer containing 0 . 5% formic acid overnight at 25°C ) were prepared . The samples were later thawed at 5°C and passed over an AL-20-pepsin column ( 16 µl bed volume ( Sigma ) ) at a flow rate of 20 µl/min [81] . The resulting peptides were collected on a C18 trap ( Michrom MAGIC C18AQ 0 . 2×2 ) and separated using a C18 reversed phase column ( Michrom MAGIC C18AQ 0 . 2×50 ) running a linear gradient of 8–48% solvent B ( 80% acetonitrile and 0 . 01% TFA ) over 30 minutes with column effluent directed into an LCQ mass spectrometer ( Thermo-Finnigan LCQ Classic ) . Data were acquired in both data-dependent MS1:MS2 mode and MS1 profile mode . SEQUEST software ( Thermo Finnigan Inc . ) was used to identify the sequence of the peptide ions . The centroids of the isotopic envelopes of nondeuterated , functionally deuterated and equilibrium-deuterated peptides were measured using DXMS Explorer ( Sierra Analytics Inc . , Modesto , CA ) and then converted to corresponding deuteration levels [82] . | Ebolaviruses cause severe hemorrhagic fever that is exacerbated by immediate suppression of host immune function . VP24 , one of only eight proteins encoded by ebolaviruses , functions in virus replication and assembly , and is thought to contribute to immune suppression by binding to a certain class of molecules called karyopherins to prevent them from transporting a transcription factor termed STAT1 . Here we report that VP24 is also able to directly bind STAT1 by itself , and thereby likely contributes to immune suppression by an additional mechanism . Analysis of these multiple roles of VP24 and design of drugs against them have been hindered by the lack of structural information on VP24 and its lack of homology to any other known protein . Hence , here we also present X-ray structures of VP24 derived from two different ebolavirus species that are pathogenic and nonpathogenic to humans . These structures and accompanying deuterium exchange mass spectrometry identify the likely binding site of STAT1 onto VP24 , map sites that are conserved or differ between pathogenic and nonpathogenic species , and provide the critical 3D templates by which we may dissect and interpret the many roles that VP24 plays in the virus life cycle . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"biochemistry",
"immunology",
"biology",
"microbiology"
] | 2012 | The Ebola Virus Interferon Antagonist VP24 Directly Binds STAT1 and Has a Novel, Pyramidal Fold |
Bundled actin structures play a key role in maintaining cellular shape , in aiding force transmission to and from extracellular substrates , and in affecting cellular motility . Recent studies have also brought to light new details on stress generation , force transmission and contractility of actin bundles . In this work , we are primarily interested in the question of what determines the stability of actin bundles and what network geometries do unstable bundles eventually transition to . To address this problem , we used the MEDYAN mechano-chemical force field , modeling several micron-long actin bundles in 3D , while accounting for a comprehensive set of chemical , mechanical and transport processes . We developed a hierarchical clustering algorithm for classification of the different long time scale morphologies in our study . Our main finding is that initially unipolar bundles are significantly more stable compared with an apolar initial configuration . Filaments within the latter bundles slide easily with respect to each other due to myosin activity , producing a loose network that can be subsequently severely distorted . At high myosin concentrations , a morphological transition to aster-like geometries was observed . We also investigated how actin treadmilling rates influence bundle dynamics , and found that enhanced treadmilling leads to network fragmentation and disintegration , while this process is opposed by myosin and crosslinking activities . Interestingly , treadmilling bundles with an initial apolar geometry eventually evolve to a whole gamut of network morphologies based on relative positions of filament ends , such as sarcomere-like organization . We found that apolar bundles show a remarkable sensitivity to environmental conditions , which may be important in enabling rapid cytoskeletal structural reorganization and adaptation in response to intracellular and extracellular cues .
Understanding emergent behaviors in the actin cytoskeleton is important , since key biological functions such as cellular growth , division , and motility depend on cytoskeletal dynamics . Actin networks are transient and malleable within a single cell , constantly forming and remodeling various micro-architectures at different cellular locations . One salient class of such actin structures are actin bundles [1 , 2] , which appear both in the cell body as stress fibers [3–5] , or in specialized cellular processes such as filopodia [6–8] , stereocilia [9–11] and microvilli [11–13] . The formation and functionality of bundles is spatially and temporally regulated by various proteins that interact with actin , including nucleation factors [5 , 14] , crosslinkers [15 , 16] and molecular motors [17] . Depending on the cellular context and specific influence of actin regulatory proteins , different bundle structures emerge [18–20] , with distinct functional roles as elaborated next . First , owing to the polarity of individual actin filaments , an actin bundle may be unipolar , apolar , sarcomeric or of graded polarity ( S1 Fig in S1 Text ) [1] , being tethered at either one or both ends via a cross-linker or a molecular motor . Cellular structures of unipolar bundles are ubiquitous across living cells: they aid in cytoplasmic streaming in pollen tubes [21–23] as well as in fungal hypha formation [24] . Unipolar bundles are found in animal cells in filopodia [25] and proximal dorsal stress fibers [5 , 26] . These structures are usually crosslinked but do not contain myosin as active crosslinkers in vivo [27] . However , there is evidence for myosin minifilaments , which are polymers of 10–30 myosin units [28] , walking through tracks of unipolar bundles [29] under in vitro conditions . Non-muscle myosin II-A isoform minifilaments ( NMII-A ) [30] incorporate into dorsal stress fibers when these stress fibers are connected to transverse arcs [27] . Recently , NMI has also been implicated in filopodial force generation in neural growth cones [31] . Other actin bundle organizations , such as apolar bundles , underlie the portions of stress fibers near the cell center in fibroblasts [32] , in sections of contractile rings [33 , 34] and in sections of ventral and transverse stress fibers . Composite apolar bundles are mostly found as two unipolar precursor bundles interacting with one another due to filament sliding in response to NMII-A activity [29] . Finally , more finely organized polarity arrangements are found in sarcomeric ordering , which has great significance in generating contractility of stress fibers [1 , 2] . The large variety of actin fibers and their distinct functional roles have instigated a growing interest in how various factors , such as crosslinking activity and minifilament concentrations , affect the dynamics and stability of bundles having unipolar and apolar organization . Another salient property of many actin bundles namely , non-sarcomeric contractility [35–37] , has also been a focal topic of experimental and computational studies . An agent based model was employed by Zemel et al . to study one dimensional unipolar and apolar bundles of 10μm length under different concentrations of motors moving according to prescribed force-velocity relations [38] . They found that apolar bundles readily undergo sliding motions resulting in internal sorting . However , the lack of spatial details in their model did not allow predictions of the stability of bundle morphology under the influence of myosin . Dasanayake et al . [39] simulated actin bundles of 10μm length , modeled in 2D , considering explicit filament stretching and bending in addition to minifilament stretching forces in a confined volume . They found that apolar bundles exert higher wall stresses than parallel filaments . Nevertheless , it is still unclear how actin bundles behave at long time scales under a wide range of crosslinker and myosin concentrations , especially if they were to be modeled in three spatial dimensions . The latter point should be emphasized because studying actin bundles in 3D is crucial for reconciling with the phenomenology of bundles observed under in vivo cellular conditions . Recent studies in three dimensions by Kim and coworkers indicated that filament polarity plays an important role in bundle formation from disordered actin networks [40] and also in tension generation [40 , 41] . To shed further light on the mechanochemistry of actin bundles , in this work we set out to establish the fundamental principles that govern the stability of two fundamental bundle organizations , namely , purely unipolar and apolar bundles . Understanding conditions that either stabilize or destabilize various bundle geometries will shed light on the basic principles that govern cytoskeletal organization , bringing insights into complex in vivo processes , such as cytoplasmic disassembly [42] and actin network turnover [43–45] . For example , stability of untethered bundles is likely to depend on their internal polarity structure as well as crosslinker and myosin motor conditions . Furthermore , cells modulate actin filament turnover through a host of mechanisms such as filament severing [46–48] , branching [1 , 49] and capping [50 , 51] . As a result , in vivo turnover rates are orders of magnitude faster than those commonly observed in in vitro experiments [47 , 52–54] . In general , actin turnover plays a critical role in stress relaxation of entangled actin networks at longer timescales . For example , skeletal myosin is known to fluidize actin networks at very low mole ratios ( myosin head to total actin mole ratio , M:A 0 . 0039 ) [55] . On the other hand , myosin‘s catch bond behavior can lead to long residence times resulting in actin networks maintaining large internal stresses at high myosin concentrations . In addition , transient passive crosslinkers arrest network configurations and prevent relaxation , however , crosslinker unbinding events allow for slow reconfiguration dynamics [56 , 57] . The crosstalk between filament turnover and crosslinker mechanokinetics has not been comprehensively studied under a wide range of conditions . Hence , it is also necessary to explore stability of bundles under a broad set of treadmilling rates . To address the above outlined problems , we have used MEDYAN ( MEchanochemical Dynamics of Active Networks ) mechanochemical force field [58] to study the stability and dynamics of untethered actin bundles under a diverse set of polarity arrangements and levels of α-actinin crosslinker , myosin minifilament and treadmilling conditions . MEDYAN is further elaborated in the Methods section below . Our main finding is that unipolar bundles preserve bundle morphology at a wider range of crosslinker and , myosin concentrations than apolar bundles , because the latter experience a morphological instability due to being more susceptible to myosin induced intra-bundle filament sliding and shearing . We also show that three salient microarchitectures eventually emerge when simulating various untethered bundles under different conditions—bundles , asters and bundle-aster hybrids . Asters are filamentous networks exhibiting radial polarity sorting , with barbed ends clustered towards the aster center . We also investigated how the resulting phase diagrams depend on the speed of treadmilling . We found that network catastrophes , characterized by poorly crosslinked low density networks with obscure morphologies , occur when crosslinking cannot keep up with filament extension . Overall , our studies demonstrate that while a stand-alone apolar bundle is stable under a significantly narrower set of conditions compared to unipolar bundles , this geometric arrangement can serve as an important precursor to rich network remodeling phenomena such as global polarity sorting and sarcomeric organization .
MEDYAN is a mechanochemical force field for simulating active matter , including cytoskeletal networks . It deeply integrates chemical and transport dynamics with network mechanics , treating these phenomena on equal footing . MEDYAN has emerged from earlier efforts to model actin bundle growth in filopodia [59–63] , where both active and passive transport were shown to critically influence growth dynamics [60 , 62 , 63] . MEDYAN’s time evolution is based on alternating reaction-diffusion and mechanical equilibration steps , where the former events are stochastically generated according to the next reaction method [64] , while the conjugate gradient approach is used to achieve mechanical equilibration [58] . This propagation scheme takes advantage of the wide timescale separation between slow chemical processes and fast mechanical equilibration speeds within sub-micron length scale volumes containing a portion of an actin network [65] . Furthermore , MEDYAN can effectively model actin filament polymerization processes as well as explicit α-actinin ( crosslinker ) , myosin minifilament ( motor ) binding and unbinding events , in addition to myosin walking and mechanochemical feedbacks such as catch and slip bond behaviors . In contrast to MEDYAN , other cytoskeletal modeling strategies , such as Cytosim ( Cytoskeletal Simulation ) [66 , 67] , AFINES ( Active Filament Network Simulation ) [68] , and the model by Kim and coworkers [41] rely on the Langevin dynamics of cytoskeletal components , explicitly simulating thermal undulations at the expense of significant diminution of computational efficiency . Among these models , MEDYAN’s treatment of general reaction-diffusion processes is most comprehensive ( in particular , with regard to G-actin’s diffusion and reactions ) . In addition , MEDYAN’s mechanical potentials are the most elaborate , for example , having an analytical excluded-volume potential representing steric repulsions and also complex dihedral angle potentials at the dendritic actin network branch points . A detailed comparison of different cytoskeletal modeling approaches can be found in Popov et al . ( in particular , see Table S1 of [58] ) . In MEDYAN , the reaction volume is divided into voxels based on the Kuramoto length [58 , 69] . Diffusing molecules are assumed to be uniformly mixed within each voxel and can diffuse from one voxel to another . Actin filaments can polymerize and depolymerize from both plus and minus ends based on experimentally determined rate constants [70] . Filaments that polymerize towards the boundary experience a reduction in polymerization rate based on the Brownian ratchet model [71] . In MEDYAN , the growth propensity for an actin filamentous tip is based on the local , instantaneous concentration of diffusing G-actin in the tip’s neighborhood [58] , while , in comparison , actin filaments grow/shrink at a constant rate in Cytosim , [72] . Furthermore , MEDYAN can explicitly account for ATP , ADP . Pi , and ADP bound actin monomeric states , enabling more elaborate simulations of F-actin polymerization dynamics [73 , 74] . In summary , filament length fluctuations and filament treadmilling can be studied using MEDYAN at the resolution of a single actin monomer . As elaborated below , we found that these fluctuations play an important role in determining whether the bundle stays coherent or undergoes a morphological transformation . In MEDYAN , mechanical modeling of actin filaments is based on cylinder units with equilibrium spacing l0m≪lp ( l0m = 108 nm in this study , lp-persistence length ) connected with neighboring cylinders at flexible hinges . Persistence length of actin was reported from experiments as 17μm [75] . The MEDYAN force-field prevents filaments passing through each other via a novel cylinder-cylinder repulsion potential that is analytical , in contradistinction to the more widely employed technique of other comparable force fields , which relies on finding the closest distance between two cylinders that is used to compute their mutual repulsion [58] . Various MEDYAN mechanical potentials , such as intra-cylinder stretching and inter-cylinder bending are shown in Fig 1 . α-actinin and myosin molecules that are bound to actin filaments are modeled as springs connecting two actin filament sites within their respective binding distances ( α-actinin binding distance is in the range 30–40 nm , and minifilament binding distance is in the range 175–225 nm ) . Those reactions that are mechanosensitive , for example , unbinding kinetics of actin binding proteins such as crosslinkers and motors , are influenced by the local instantaneous stresses of the actin network , via corresponding modifications of the reaction rate constants . Motor/crosslinker binding and unbinding are modeled as a single step chemical reaction in MEDYAN , while in other force fields , for example , Cytosim , these processes occur via two elementary steps , comprising of separate reactions for each end of the motor or crosslinker [76] . In MEDYAN , crosslinker unbinding kinetics is modeled as a slip bond while myosin unbinding kinetics is modeled as a catch bond based on the parallel cluster model [77] . The motor walking rate is given by a linear force-velocity relationship for motor walking events . Further aspects of mechanical equilibration and mechanochemical feedback loops in MEDYAN as well as the implementation details of the chemical model are provided in Supporting Information ( Section 2 in S1 Text ) . Initial structures for all our simulations were based on 2 μm long actin bundles , comprising 30 actin filaments that correspond to in vivo stress fibers [32] , where the internal arrangements of filament polarities were in either unipolar or apolar geometries . Filaments were initially placed on a hexagonal lattice with a spacing of 35 nm as found in experiments [78 , 79] . Bundles were modeled with α-actinin and NMIIA minifilaments that can bind and unbind from actin filaments . Bundles were simulated at 7 different concentrations of α-actinin ( α-actinin to total actin mole ratio referred to henceforth as α:A 0 . 01 , 0 . 05 , 0 . 1 , 0 . 25 , 0 . 4 , 0 . 6 , 0 . 8 ) and 6 different concentrations of myosin ( myosin head to total actin mole ratio referred to henceforth as M:A 0 . 0225 , 0 . 045 , 0 . 09 , 0 . 18 , 0 . 225 , 0 . 675 ) . 8 trajectories , each 2000 seconds long , were generated for each of the 6x7 = 42 mole ratio pairs ( α:A , M:A ) studied . In addition , we also studied how the speed of filament treadmilling influences bundle stability . Myosin mole ratios of 0 . 0225 0 . 045 , 0 . 225 and 0 . 675 were considered at α:A 0 . 01 , 0 . 1 and 0 . 4 to investigate how all observed non-treadmilling morphologies behave under different treadmilling conditions . Treadmilling speed was varied by simultaneously altering polymerization and depolymerization rates at both filament ends by the same factor χ . As a reference , χ = 1 . 0 corresponds to the in-vitro treadmilling rate . We chose the following χ values , ( 0 . 1 , 0 . 3 , 0 . 6 , 1 . 0 , 3 . 0 , 6 . 0 , and 10 . 0 ) , hence mimicking treadmilling speeds that are both slower and faster than the in vitro rate . As a technical detail , in this work we have developed a flexible volume protocol that permits expanding and contracting the reaction volume along the X-axis , which allows avoiding artificial boundary effects on the bundle major axis in a computationally efficient manner . This technique is further elaborated in Supporting Information ( Section 2 . 5 in S1 Text ) . 7 trajectories were generated for each combination of the myosin mole ratio , α-actinin mole ratio and χ factor . In order to understand the underlying morphologies sampled , we devised a hierarchical clustering scheme . To carry out structural clustering of obtained bundle configurations , we first computed the distributions of plus end—plus end ( Dis++ ) , minus end—minus end ( Dis— ) , and plus end—minus end ( Dis+- ) Euclidean distances . Jensen Shannon divergences [80] between each of the 42 mole ratios taken pair-wise were used to construct initial -condition -specific dissimilarity matrices ( S12 Fig in S1 Text ) . The complete linkage method [81] results in a hierarchical cluster , which we visualized as dendrograms ( shown below ) . Supporting Methods Section 2 . 4 . 1 in S1 Text provides further details on the clustering algorithm .
A wide array of steady state network morphologies emerge when non-treadmilling bundles with unipolar initial organization ( non-treadmilling-BUInit ) and apolar initial organization ( non-treadmilling-BAInit ) were simulated under a broad set of α-actinin and myosin concentrations ( Figs 2 and 3 and S1 and S2 Movies ) . A trajectory is considered to have reached steady state when the network’s radius of gyration reaches a stationary value ( see S10 Fig in S1 Text ) . We found that for any combination of α-actinin and myosin ratios , marked differences are observed between steady state network morphologies of apolar and unipolar bundles . More specifically , antiparallel filaments show a strong tendency to mutually slide in response to myosin activity as a consequence of the latter being unidirectional walkers ( S2 Movie ) . To quantitatively characterize the resulting network morphologies , we applied a novel structure-based clustering analyses that we have developed in this work ( see the Methods section and the Supporting Information , Section 2 . 4 . 1 in S1 Text ) , which revealed dominant network morphologies preferred under various conditions , as elaborated below . The resulting dendrograms reveal three broad clusters closest to the root , colored in green , red and blue , pointing to three major network morphologies ( Figs 2–4 ) : bundle-like ( BL ) , aster-like ( AL ) and aster-bundle intermediate ( ABI ) states . We also considered an alternative morphology classification technique based on network radial distribution devised by Freedman et al . [82] ( see S3 Fig in S1 Text ) and a combination of nematic order and shape parameters ( see S2 Fig and Supplementary Methods 2 . 4 . 2 and 2 . 4 . 3 in S1 Text ) . Both order parameters delineated well the bundled morphologies from the aster-like morphologies . However , the hierarchical classification strategy introduced in this work was also able to identify the intermediate bundle-aster morphologies . Some reflection on the internal structure of these dendrograms shows that within each initial polarity arrangement , NMIIA concentration is the main driver of the resulting network morphology ( see Fig 4 ) . Specifically , the same highest level clusters are formed from configurations with similar M:A ratios , while finer-grained additional clustering is determined by other factors , such as the crosslinker ( α-actinin ) concentration . The effect of myosin is primary because at high motor concentrations inter-filament distances become significantly widened outside of the α-actinin binding compatibility zone of 30–40 nm . This , in turn , crucially depletes the network of crosslinker binding sites ( S4 Fig in S1 Text ) , hence , greatly diminishing parallel alignment among actin filaments and , subsequently , destabilizing the bundle phase . It is interesting to compare our finding of the depletion of crosslinkers with increasing myosin concentration with the recent simulations that studied sorting of two different crosslinkers along an actin bundle . Freedman et al . have established that a significant length difference between two actin binding proteins , when combined with a specific range of the filament bending moduli , can result in spatial sorting of crosslinkers along a bundle [83] . Here , we found that active myosin walking may significantly increase overall inter-filament separation , thereby reducing the number of sites available for crosslinker binding in the bundles studied here , which presumably should have important implications for the sorting phenomenon in the presence of molecular motors . When simulations were started with unipolar initial conditions , bundle-like states were stable when M:A ≤ 0 . 09 ( 21/42 cases = 50% of all M:A/α:A mole ratios studied ) . On the other hand , apolar initial arrangements result in stable bundle-like states only in ~20% ( 7/42 ) of the cases ( i . e . when M:A = 0 . 0225 ) . Thus , unipolar bundles are stable under a significantly wider range of conditions than their apolar counterparts . We tracked this difference primarily to myosin activity in apolar bundles giving rise to two thin polarity sorted sub-bundles that are together much longer than the initial bundle length and , furthermore , mutually interact via their barbed ends ( see Figs 3 and 4 ) . The resulting thinner bundles are more susceptible to bending deformations than the corresponding unipolar bundle . Consequently , at the time scale of about 30 minutes probed in this work , unipolar and apolar bundles arrive at different metastable morphologies despite being under the same crosslinker and myosin conditions . Presumably , if these systems were to be ergodic , then bundles are expected to eventually evolve to identical steady state configurations regardless of the initial polarity arrangement . However , as shown in our previous works [84 , 85] , cytoskeletal dynamics maybe sufficiently glassy such that only metastable states are reachable over laboratory timescale . Under very high myosin activity ( M:A = 0 . 675 ) , both unipolar and apolar bundles undergo a morphological collapse , preferring radially symmetric aster-like structures ( the last column in Figs 2 and 3 ) . We note in passing that in cells , asters are primarily found in microtubule networks as radially symmetric structures with filament plus ends spatially clustered together [86 , 87] . Actin networks are also expected to form radially polarity sorted asters [88] . Such structures have been found in in vitro treadmilling actin networks subject to skeletal muscle myosin ( M:A 0 . 1 ) , fascin and myosin [89] or just skeletal muscle myosin ( M:A 0 . 02 ) alone [90 , 91] , with the filament plus ends oriented towards the center of the aster . The difference in threshold myosin ratios between skeletal muscle myosin and NMIIA for the onset of aster-like structures is partly explained by the increased processivity of muscle myosin . Actin asters were observed in-vitro when cytoskeletal structures were destabilized using Cytochalasin D [42] . Recently , they were shown to be essential in fission yeast cells during fusion [92 , 93] . Fission cell actin asters are considered to be formed due to Fus1 nucleators and multimerization of Myo51 and Myo52 [93] . Overall , we found that experimentally observed salient network morphologies of treadmilling networks can also be sampled in our simulations under non-treadmilling conditions . Next , we investigated the combined effect due to network turnover and mechanokinetics ( crosslinker , and NMIIA ) by including actin filament polymerization and depolymerization processes in bundle simulations analogous to the systems discussed above . In the treadmilling study , steady state was defined when the average filament length fluctuations reach their stationary values ( see S11 Fig in S1 Text ) . To systematically modulate filament treadmilling , we varied polymerization and depolymerization rates at both barbed and pointed ends by a factor χ between 0 . 1 and 10 . 0 , where lower χ values cause slower treadmilling . We modeled a reaction volume having dimensions of 4 μm x 1 . 5 μm x 1 . 5 μm , with an initial total actin concentration of 5 μM , where the simulation box can expand and contract along the major axis based on the instantaneous bundle length ( more details are provided in the Supplementary Information , Section 2 . 5 ) . We investigated both unipolar and apolar bundles at M:A mole ratios of 0 . 0225 , 0 . 09 , 0 . 225 and 0 . 675 , whereas α:A was sampled at 0 . 01 , 0 . 1 and 0 . 4 . These values were carefully chosen to capture the different salient network morphologies manifested under non-treadmilling conditions ( non-treadmilling-BUInit and non-treadmilling-BAInit ) , namely , BL , AL and ABI states . 7 trajectories , each 2000 seconds long , were generated for each of the 84 triad conditions ( α:A , M:A , χ ) , starting from either unipolar ( χ-BUInit ) or apolar ( χ-BAInit ) initial conditions . Similar to non-treadmilling bundles , the above-mentioned classes of network morphologies were determined by clustering trajectories from 84 triads using the same clustering protocol ( S12 Fig in S1 text ) . We classify connected networks with morphologies that do not belong to BL , AL or ABI as either Type A catastrophes if the networks are fragmented into smaller clusters or Type B catastrophes , where filaments in the network are poorly connected ( S5–S9 Figs in S1 Text ) . We found that these catastrophes emerge from the interplay between inter-filament connectivity and treadmilling . Treadmilling is characterized by a net depolymerization at minus ends and filament growth at the plus ends . In particular , increasing χ leads to a faster rate of filament growth at plus ends ( and faster depolymerization at minus ends ) . In these systems , network stability is assured as long as the newly formed filament segments are effectively crosslinked . Under conditions where the latter does not take place , filament treadmilling dominates the system’s structural evolution , leading to poorly connected networks . When well-structured initial bundle configurations result in a highly fragmented network , we denote such transitions as catastrophes ( see S5 Fig in S1 Text ) . On the other hand , at higher mole ratios of α-actinin or myosin , the rate of inter filament connections is enhanced , which prevents network catastrophes . Having established how inter-filament connections influence network stability , we next investigated the effect of treadmilling speed . Our clustering analysis indicates that treadmilling networks starting from unipolar initial configurations attain BL and ABI morphologies ( i . e . non-aster , non-fragmented morphologies ) in 33/84 ( ~40% ) cases , compared to 19/84 ( ~21% ) cases for the apolar initial configurations . On the other hand , the apolar initial arrangements lead to aster-like structures in 39/84 ( 46% ) cases as opposed to 20/84 ( 24% ) cases for the unipolar cases . Taken together , these results demonstrate that treadmilling bundles that were evolved from unipolar initial configurations are less likely to undergo morphological collapse than those evolved from apolar initial configurations . Based on the network morphologies observed at the sampled triad combinations , we suggest the following two phase diagrams , shown in Fig 5 , indicating dominant network morphologies as functions of M:A and χ . To justify the choice of these two order parameters , we point out that α-actinin dynamics determine bundle stability in conjunction with myosin , however , affecting only very weakly the final network morphology of stable networks . Fig 5A suggests that treadmilling unipolar networks sample similar network morphologies to the non-treadmilling cases for χ ≤1 . 0 . At larger χ values , network treadmilling dominates as crosslinkers and minifilaments cannot effectively connect filament segments that are formed , leading to network catastrophes ( S6–S9 Figs in S1 Text , see snapshots from simulations with χ >1 . 0 ) On the other hand , treadmilling apolar bundles lead to aster-like states ( under non-catastrophic triads ) . The BL , ABI morphologies sampled by treadmilling apolar bundles have rich diversity due to the interplay between myosin activity and treadmilling as explained later . We finally discuss the weak influence of crosslinkers on the morphology of a treadmilling network . At α:A = 0 . 01 and under low myosin mole ratios ( 0 . 0225 , and 0 . 09 shown in Figures S6 and S7 ) , we found that unipolar bundles treadmilling at χ = 1 . 0 result in network catastrophes . However , significantly increasing crosslinker concentration ( e . g . between 10 and 40 folds ) can rescue such networks to prefer bundle-like/intermediate morphologies ( S3 and S4 Movies ) . These findings are in qualitative agreement with the observations by Bidone et al [40] that at high crosslinker concentrations bundles form robustly from networks obtained from a wide range of initial orientational biases ( at M:A = 0 . 08 ) . Under low myosin activity conditions , treadmilling bundles starting from an apolar arrangement generate a remarkably diverse set of final network morphologies , primarily based on how filament barbed ends are spatially localized ( see Fig 6 ) . These emergent network geometries arise from the tug-of-war between myosin activity and filament treadmilling . On the one hand , treadmilling apolar bundles are subject to the continuous process of plus end extension and minus end retraction . On the other hand , myosin activity drives mutual sliding of neighboring filaments . If the rate of plus end extension is slower than myosin sliding , myosin activity dominates , leading to two unipolar bundles connected at their plus ends ( Fig 6A ) or to networks with overlapping plus end segments ( Fig 6B ) . Conversely , under vigorous plus end extension conditions compared with myosin sliding , the network transitions to polarity sorted bundles interacting at their minus ends via myosin ( Fig 6C and S5 Movie ) . Modulation of χ also controls the overall distribution of myosin minifilaments in such bundles . The analysis of the computed spatial distributions of myosin minifilaments and α-actinin under low myosin concentrations ( Fig 6D ) , indicates that myosin spatially segregates close to pointed ends , characteristic of sarcomeric ordering [18 , 36 , 94] . The latter arises when myosin minifilaments interact with minus ends flanked on either side by plus ends .
Actin bundles are important for cellular stability , growth and mechanosensing . While prior experimental and modeling research has primarily focused on bundle formation processes [40 , 72 , 95 , 96] , in this work we have addressed the stability and temporal evolution of various bundle configurations . We used MEDYAN , a mechano-chemical forcefield based on molecular principles , to simulate bundle dynamics in 3D . The dimensionality of the model is crucial as filament deformations and mutual interactions are markedly dimension-dependent . In this study , we comprehensively analyzed how α-actinin and myosin influence the stability and morphological transformations of unipolar and apolar actin bundles . We discovered that at time scales of about 2000 seconds , non-treadmilling unipolar bundles are stable under a wider range of crosslinker and myosin mole ratios compared to apolar bundles . At high myosin mole ratios , we observed aster-like states characterized by interacting barbed ends grouped in the center of the cluster with radially emanating pointed ends . We also investigated how actin turnover affects bundle morphology fates , by developing and applying a simulation protocol that allows moving system boundaries . Our results indicate that treadmilling bundles , both unipolar and apolar , undergo network catastrophes when the network’s ability to form inter-filament connections is insufficient compared to the treadmilling speeds . In vivo cytoskeletal networks undergoing fast treadmilling may be able to avoid such undesired fragmentation using additional mechanisms such as filament capping and actin filament nucleators . Interestingly , at high myosin concentrations , even quick treadmilling does not rescue the network from transitioning to aster-like configurations . On the other hand , at low myosin activity , initially apolar bundles explore a diverse spectrum of network organizations , which we attributed to the tug-of-war between minifilament activity and filament treadmilling . Under certain conditions , interesting , biologically relevant architectures emerge , such as sarcomeric-like organization . We are not aware of prior models that resulted in the spontaneous assembly of sarcomeric arrangements without imposing spatial restrictions on crosslinkers . In particular , previous attempts to reproduce sarcomeric distribution of treadmilling apolar networks relied on various assumptions , such as preferential binding of passive crosslinkers near plus ends [18] or considered systems with both plus and minus end directed motors [97] . Finally , we reflect on the biological consequences that follow from this work . We found that treadmilling bundles with apolar initial configuration are poised to undergo a remarkable morphological response to the perturbations of the environment , such as alterations of myosin activity or treadmilling rates . On the one hand , this level of sensitivity to parameters might be potentially detrimental to the overall stability of the cellular actin network . On the other hand , if only a small fraction of the cytoskeleton is organized as apolar bundles , the latter can sensitively respond to various signaling cues that affect the local concentrations of actin binding proteins . Thus , we propose that apolar bundles might be crucial to cytoskeletal robustness and adaptation in scenarios that demand drastic structural reorganization . The ability to rapidly change network morphology might be important in certain cellular functions where force production or rapid cellular reorganization are necessary . Overall , the optimal choice of bundle architecture should be determined by the specific cellular processes that it affects: for example , in the case of cargo transport [98] or protrusive growth [99] , a structurally stable unipolar bundle would be the optimal choice , while contractile elements that require frequent reorganization would be more robust when apolar bundles are incorporated into their architectures [5 , 100] . In summary , we simulated actin bundles in 3D , while explicitly accounting for excluded volume interactions , diffusion of actin , crosslinker and NMIIA proteins , and numerous chemical and mechanical processes that enhance the model’s realism . We systematically studied the influences of initial bundle polarity , concentrations of myosin and α-actinin and the network turnover rate , finding a remarkably rich palette of bundle evolution trajectories , from stable bundle states to asters and sarcomeric organizations . In future works , additional effects may be considered , such as actin filaments transiently tethering to the substrate and nucleation of filaments via formins or Arp2/3 , which will bring us closer to achieving a more complete understanding of bundle dynamics under in vivo conditions . | Actin bundles are a salient feature of the cellular cytoskeleton offering structural integrity to the cell in addition to playing a key functional role in processes such as intracellular transport , mechanosensing and locomotion . Even though bundles have been studied extensively for years , a comprehensive picture of bundle stability is still lacking . In this study , we used MEDYAN , a publicly available software package for simulating active polymer networks at single-molecule resolution , to study how bundle’s initial geometry , molecular motors and crosslinkers determine the corresponding lifetime and eventual fate of the bundle . We found that unipolar bundles , where all filaments point in the same direction , are significantly more stable compared with mixed-polarity ( apolar ) bundles . We also show that while high myosin activity results in transition to aster-like morphology , increased treadmilling causes network collapse and fragmentation . Overall , bundle dynamics and morphological transformations are largely driven by the tug-of-war between filament treadmilling and myosin activity , leading in some cases to very interesting architectures , such as sarcomere-like organization . We propose that the ability of apolar bundles , in particular , to undergo remarkable structural transformations in response to perturbations of the environment may be crucial for cytosplasmic reorganization . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"cell",
"physiology",
"cell",
"motility",
"actin",
"filaments",
"cell",
"polarity",
"molecular",
"motors",
"actin",
"motors",
"cellular",
"structures",
"and",
"organelles",
"cytoskeleton",
"motor",
"proteins",
"dynamic",
"actin",
"filaments",
"contractile",
"proteins",
"polymer",
"chemistry",
"actins",
"proteins",
"chemistry",
"polymerization",
"biochemistry",
"cytoskeletal",
"proteins",
"cell",
"biology",
"myosins",
"biology",
"and",
"life",
"sciences",
"chemical",
"reactions",
"physical",
"sciences"
] | 2019 | Remarkable structural transformations of actin bundles are driven by their initial polarity, motor activity, crosslinking, and filament treadmilling |
Trafficking of lung dendritic cells ( DCs ) to the draining lymph node ( dLN ) is a crucial step for the initiation of T cell responses upon pathogen challenge . However , little is known about the factors that regulate lung DC migration to the dLN . In this study , using a model of influenza infection , we demonstrate that complement component C3 is critically required for efficient emigration of DCs from the lung to the dLN . C3 deficiency affect lung DC-mediated viral antigen transport to the dLN , resulting in severely compromised priming of virus-specific T cell responses . Consequently , C3-deficient mice lack effector T cell response in the lungs that affected viral clearance and survival . We further show that direct signaling by C3a and C5a through C3aR and C5aR respectively expressed on lung DCs is required for their efficient trafficking . However , among lung DCs , only CD103+ DCs make a significant contribution to lung C5a levels and exclusively produce high levels of C3 and C5 during influenza infection . Collectively , our findings show that complement has a profound impact on immune regulation by controlling tissue DC trafficking and highlights a potential utility for complement as an adjuvant in novel vaccine strategies .
Influenza is a global health problem and current vaccination strategies are still inadequate at providing protection against seasonal and epidemic outbreaks [1] . Vaccination strategies aiming to induce a protective CD8+ T cell response hold tremendous potential , since CD8+ T cells are able to recognize core epitopes conserved across a wide range of influenza strains [2] , [3] . Hence , there is a pressing need to improve our understanding on the mechanisms that contribute to the orchestration of CD8+ T cell responses during influenza infection . Influenza-specific T cell responses are initiated and maintained by lung dendritic cells ( DCs ) which are strategically localized within the respiratory tract to mediate this process effectively [4] , [5] . DCs comprise a heterogeneous population of antigen sensing and presenting cells that control the initiation of T cell responses thus bridging innate and adaptive immune responses [6] . Different subsets of DCs with unique homeostasis and immune functions had been described in both lymphoid and non-lymphoid tissues [7] . In this regard , lung resident DCs can be divided into CD103+CD11b− ( CD103+ DCs ) and CD103−CD11b+ ( CD11b+ DCs ) based on the expression of the integrins αEβ7 ( CD103 ) and ITGAM ( CD11b ) respectively [8] , [9] . During influenza infection , both types of lung resident DCs migrate to the dLN to prime T cells [10] henceforth referred to as migratory DC subsets ( mDCs ) . The complement system is an essential component of the innate immune network and has evolved as an important bridge between innate and adaptive immune systems , similarly to DCs , but at the molecular level [11] , [12] , [13] , [14] . In particular , complement component C3 has been shown to impact antiviral T cell immunity and allograft rejection in a mechanism independent of intrinsic C3 expression in T cells [15] , [16] , [17] , [18] . These observations suggested the involvement of an additional cell type expressing C3 that control T cell activation . Recent studies using bone marrow derived DCs have revealed that C3 is essential for DCs to efficiently stimulate T cells [19] , [20] , suggesting that DCs mediate these aforementioned complement effects [15] , [16] , [17] on T cells by acting both as complement producing and sensing cells . However , the in vivo functional relevance of C3 in peripheral tissue DCs , which play a central role in the induction of immunity by virtue of their location [21] , remains largely unexplored , necessitating further studies . Influenza infection is known to activate complement both locally in the lung and systemically [22] , [23] , but the biological significance of complement activation during influenza infection remains poorly understood . The functional importance of C3 in T cell immunity during influenza infection was initially demonstrated by Kopf et al . , who showed that C3−/− mice exhibited reduced T cell response to influenza infection and attributed this to a possible priming defect mediated by DCs [16] . Thus , we hypothesized that C3 may be critically involved in the control of the T cell priming function of lung mDCs during influenza infection . In this study , we used C3-deficient ( C3−/− ) mice to show that C3 was critical for survival during influenza infection , and C3 deficiency was associated with attenuated T cell priming in the dLN and reduced development of effector T cell responses in the lung as previously shown [16] . However , we found that this defective priming was not due to altered priming ability of C3-deficient DCs but rather due to a defect of mDC migration from the lung to the dLN in C3−/− mice . We further demonstrated that the direct interaction of complement activation products C3a and C5a with their receptors expressed on the surface of mDCs was critical for their migration . Finally , we identified CD103+ DCs as the sole mDC subset capable of secreting C3 and C5 and one of the major source of lung C5a during influenza infection . Altogether , our results establish a previously unidentified role for complement activation products C3a and C5a in mediating migratory function of mDCs and highlight the crucial role of CD103+ DC subset as an unique complement sensing and producing DC population controlling its own migration and the migration of CD11b+ DCs .
C3 has been shown to be important for eliciting T cell responses and viral clearance during influenza infection in mice [16] , although the impact of C3 deficiency on survival after influenza infection was not well reported . In order to evaluate this , we infected WT and C3−/− mice with a sub-lethal dose ( optimized in WT mice , 15 PFU for female and 25 PFU for male ) of influenza virus and monitored their weight loss and survival . WT mice showed ∼20% weight loss at the peak of infection and recovered with 20% mortality as previously reported [24] . In stark contrast , C3−/− mice showed greater weight loss on days 5 and 7 , and resulted in 100% mortality by day 12 ( Fig . 1A and B ) . Subsequent evaluation of effector T cell responses and viral load in the lungs recapitulated previous observations [16] , showing decreased effector T cell responses and viral clearance in the C3−/− mice ( Fig . S1 ) . Absence of C3 in T cells does not directly alter their function [16] , hence we reasoned that these defects in T cell responses were mediated by the lack of C3 in DCs , as suggested in previous reports [19] , [20] . In order to investigate whether priming of antigen-specific T cells in the dLN in response to influenza infection was altered in C3−/− mice , we adoptively transferred CFSE labeled OT-I CD8+ transgenic T cells into mice prior to infection with either wild type PR8 influenza virus or recombinant PR8 influenza virus containing the OVA epitope SIINFEKL ( PR8-OT-I ) . We then harvested lung dLNs at indicated time points post-infection to assess proliferation by CFSE dilution on SIINFEKL-Kb tetramer positive CD8+ T cells . As expected , proliferation was observed only when mice were inoculated with PR8-OT-I , and no proliferation was observed with PR8 influenza virus alone , reflecting the specificity of the response ( Fig . 1C ) . Strikingly , the level of proliferation was significantly reduced in the C3−/− mice in comparison with WT mice ( Fig . 1C ) . CD4+ T cell priming in the dLN was also evaluated using OT-II ( CD4 ) transgenic T cells and PR8 influenza virus containing the OVA epitope ISQAVHAAHAEINEAGR ( PR8-OT-II ) . Similarly , we found that the priming of CD4+ T cells was significantly reduced in the C3−/− mice when compared to WT controls ( Fig . 1D ) . Since C3 has been reported to influence the expression of co-stimulatory molecules on DCs [19] , we tested whether diminished levels of costimulation could explain the decreased T cell responses by measuring the expression of costimulatory molecules on mDCs in the dLN of both WT and C3−/− mice on day 2 post-infection . Resident and mDC subsets in the dLN were characterized as shown in Fig . S2 . We found comparable levels of CD86 and CD40 expression , while CD80 expression was even increased in the C3−/− mDCs ( Fig . 1E and F ) , excluding an intrinsic maturation defect in C3−/− DCs . Finally , to understand whether the T cell priming defect observed in vivo was due to a decreased priming ability by C3−/− mDCs ( Fig . 1C and D ) , we sorted CD103+ DCs and CD11b+ DCs on day 2 post infection with PR8-OT-I and evaluated their priming ability ex vivo . A strong proliferation of OT-I CD8+ T cells was observed only when CD8+ T cells were co-cultured with CD103+ DCs , but not with CD11b+ DCs , an observation supporting the predominant role of CD103+ DCs in early CD8+ T cell priming [10] . However , ex vivo priming ability was comparable between WT and C3-deficient CD103+ DCs ( Fig . 1G ) , demonstrating that there was no intrinsic priming defect in C3-deficient CD103+ DCs . Next , we investigated the composition of mDC subsets in the lung and lung dLN at steady state and during the course of influenza infection to test whether the priming defect could be explained by a defect of the mDC network in the lung and dLN of C3−/− mice . The gating strategy employed for characterizing lung mDCs is shown in Fig . S3 [25] , [26] . Under steady state conditions , the relative numbers of CD103+ DCs and CD11b+ DCs were comparable between C3−/− and WT mice , whereas enumeration of absolute numbers showed a slight decrease for CD103+ DCs only in the C3−/− mice ( Fig . 2A–C ) . Upon influenza infection , both the relative and absolute numbers of lung CD103+ DCs and CD11b+ DCs became comparable between C3−/− and WT mice during influenza infection ( Fig . 2A–C ) . These results show that the T cell priming deficit observed in C3−/− mice is not due to a defect in the differentiation of lung mDCs . Next , we assessed whether the migration ability of lung mDCs was compromised . Under steady state , lung mDCs constitutively migrate at a slow rate to the dLN , a process strongly increased under inflammatory conditions [8] . As expected , in WT mice during the course of infection , the absolute numbers of both CD103+ DCs and CD11b+ DCs increased in the dLN by day 1 post infection and further by day 2 post infection ( Fig . 2D–F ) . However , absolute numbers of both CD103+ DCs and CD11b+ DCs were significantly decreased in C3−/− mice , suggesting a decreased migration capacity of CD103+ DCs and CD11b+ DCs at both steady state and at all time points tested after influenza infection ( Fig . 2D–F ) . To confirm these data , the migration of mDCs from the lung to the dLN was tracked using CFSE delivered intranasally into the lungs at different time points after influenza infection [27] . As expected , the relative numbers of CFSE+ mDCs in the dLN increased after influenza infection , reaching a maximum by day 3 in the WT mice . However , CFSE+ mDC accumulation ( independent of subsets ) was significantly reduced in the C3−/− dLN ( Fig . 3A left panel and B ) . Upon enumeration of CD103+ DCs and CD11b+ DCs within the CFSE+ mDC population , we found that migration of both CD103+ DCs and CD11b+ DCs was severely compromised in the C3−/− mice ( Fig . 3A , C and D ) . In order to understand whether a strong inflammatory signal is able to overcome the defective signaling on mDCs due to the lack of C3 , we administered LPS intratracheally shortly after influenza infection and then evaluated mDC migration in C3−/− mice . Our observations indicated that LPS administration was not able to overcome the defective trafficking of both subset of lung mDCs in the C3−/− mice , although a mild effect was observed for CD11b+ DCs ( Fig . 3E ) . These observations indicate that the critical importance of complement mediated signaling in lung mDC trafficking is independent of inflammatory signals . Altogether , our observations demonstrate that migration of C3−/− mDCs to the dLN is significantly reduced compared to WT during inflammation . Complement component C3 mediated opsonization has been shown to affect viral uptake by DCs [28] . To test whether reduced viral uptake by C3−/− DCs could explain the defective mDCs migration in C3−/− mice , we examined their antigen uptake capacity using DiD labeled influenza virus as previously described [25] . We first infected mice with unlabelled PR8 influenza virus to establish an infection and its associated inflammation , and subsequently inoculated the same mice with DiD labeled PR8 influenza virus 16 hours before harvesting the lungs . DiD+ mDC subsets were comparable between C3−/− and WT mice suggesting that the lack of C3 did not affect antigen uptake ( Fig . 4A and B ) . Migratory DCs upon exposure to antigen under inflammatory conditions undergo maturation before migrating to the dLN and C3 is known to influence the expression of costimulatory molecules [29] . In order to understand whether the mDCs in the C3−/− mice produced sufficient inflammatory mediators comparable to the WT mice , we examined the expression levels of inflammatory cytokines in the mDCs after ex vivo and in vivo influenza infection . Expression levels of IL-1β and IL-6 were comparable in the mDCs under both ex vivo and in vivo conditions . However IL-12p40 expression were higher in the C3−/− mice when mDCs were infected under ex vivo conditions whereas comparable under in vivo conditions . ( Fig . 4C and D ) . Subsequently , we evaluated the maturation of mDC subsets by analyzing the expression of co-stimulatory molecules after influenza infection ( Fig . 4E ) . Expression levels of CD86 , CD80 and CD40 on mDCs were comparable between C3−/− and WT mice during the course of infection . CCR7 expression on mDCs has been shown to be important for their migration to dLN [29] , [30] , [31] , hence we evaluated the expression of CCR7+ at the surface of lung mDCs . Under steady state , the frequency of CCR7 expressing mDCs was lower in C3−/− mice ( Fig . 5A–C ) . After influenza infection , the relative numbers of CCR7 expressing mDCs increased in both groups to a similar extent . However due to the reduced initial frequency of CCR7+ mDCs in C3−/− mice at steady state , the proportion of CCR7+ mDCs upon flu infection remained lower than WT ( Fig . 5A–C ) . These observations suggest that lack of C3 did not affect the up- regulation of CCR7 expression on mDCs under inflammatory conditions . Altogether , these results show that C3-deficient mDCs are not defective in viral uptake and are fully competent in the expression of inflammatory mediators and co-stimulatory molecules for T-cell priming , but are less migratory . The absence of C3 causes a complete block in complement activation and hence C3−/− mice lack the ability to generate both C3a and C5a [22] . To ascertain whether these activation products are directly involved in mediating mDC migration , we first quantified the levels of C3a and C5a in the bronchioalveolar lavage fluid before and after influenza infection in WT mice by ELISA . At steady state , C3a levels were high , further increasing by approximately 4-fold on day 2 after infection and returning to background level by day 4 ( Fig . 6A ) . In the case of C5a , steady state levels were very low , increasing by approximately 7-fold and 17-fold on days 2 and 4 respectively post infection , before returning to background levels by day 7 post infection ( Fig . 6A ) . Next , we investigated the cellular source of C3a and C5a and assessed the ability of lung mDCs to synthesize and secrete C3 and C5 during infection . Lung DCs before and after influenza infection were sorted and C3 and C5 mRNA expression levels were determined by qRT-PCR . Both C3 and C5 mRNA levels were significantly upregulated after influenza infection in CD103+ DCs; C3 levels were upregulated by approximately 3-fold on day 1 post infection and 50-fold on day 3 after infection ( Fig . 6B ) , while C5 levels were upregulated 2-fold on day 1 post infection and 5-fold on day 3 ( Fig . 6C ) . No modulation of C3 and C5 mRNA expression was observed in CD11b+ DCs after influenza infection ( Fig . 6B and C ) . Since only CD103+ DCs showed increased C3 and C5 mRNA expression upon influenza infection , we evaluated the contribution of CD103+ DCs to the observed increase in C3a and C5a levels in the lung during influenza infection . For this purpose , we used langerin-DTR mice to specifically deplete CD103+ DCs [10] . Langerin-DTR mice specifically expressed DTR on CD103+ DCs in the lung and DT administration efficiently depleted lung CD103+ DCs ( Fig . S4 ) , and subsequent influenza infection in DT-treated langerin-DTR mice did not cause increased C3a and C5a levels in the lungs on days 2 and 4 when compared with CD103+ DCs sufficient mice ( Fig . 6D and E ) . Importantly , when depletion of lung CD103+ DCs in langerin-DTR mice was followed by infection with PR8-OT-I , priming of OT-I CD8+ T cells in the dLN was severely reduced when compared with PR8-OT-I infected WT ( Fig . 7A ) , similar to our observations in the C3−/− mice ( Fig . 1C ) . CD103+ DCs depleted langerin-DTR mice infected with influenza displayed greater weight loss ( Fig . 7B ) , higher mortality ( Fig . 7C ) , reduced lung effector T cell response ( Fig . 7D–G ) and increased viral load ( Fig . 7H ) as compared to WT , paralleling our observations in C3−/−mice ( Fig . 1A and B ) . Since DT administered langerin-DTR mice showed rapid mortality , we evaluated whether DT administration induced any toxicity during influenza infection in WT mice . Our observation did not indicate any toxicity for DT in WT mice during influenza infection ( Fig . S5 ) . Among the mDC subsets , only the CD103+ DCs were found to produce the complement components C3 and C5 upon influenza infection ( Fig . 6B and C ) and depletion of CD103+ DCs significantly reduced the availability of C3a and C5a in the lungs ( Fig . 6D and E ) . We hypothesized that CD11b+ DCs rely on complement produced by CD103+ DCs for their migration to the dLN . Therefore , we followed the migration of CD11b+ DCs in the langerin-DTR mice after depleting the CD103+ DCs . Our results suggested that depletion of CD103+ DCs in langerin-DTR mice significantly affected the migration of CD11b+ DCs , supporting our hypothesis ( Fig . 7I and J ) . Altogether , these observations suggest that C3a and C5a produced by CD103+ DCs are crucial for their migration to the dLN in order to initiate protective T cells responses and also control the migration of CD11b+ DCs . Our data suggest that sensing of complement activation products C3a and C5a by mDCs is crucial for their migration to the dLNs , which could be mediated by expression of the receptors C3aR and C5aR on mDCs during infection . Thus , we examined the expression of C3aR and C5aR mRNA in mDC subsets by qRT-PCR . In CD103+ DCs , there was a slight increase in the expression of C3aR mRNA on day1 post infection , and a 4-fold increase by day 2 post infection . C5aR also increased by 4-fold on day 1 and by 90-fold on day 2 post infection ( Fig . 8A and B ) . In CD11b+ DCs , the expression of both C3aR and C5aR was higher even under steady state as compared to CD103+ DCs and showed a modest increase on day 2 post infection ( Fig . 8A and B ) . To determine whether mDC migration to the dLNs was directly controlled through interaction of C3a and C5a with their receptors , we blocked C3aR and C5aR either alone or together in vivo using specific , high affinity competitive antagonists [32] , [33] . Compstatin was used as the control peptide since it is known to possess no complement inhibiting property in mice [33] . Treatment with antagonists started 2 days before infection and was continued daily . CFSE instillation was done 16 hours before sacrificing the mice . At day 2 post infection , the number of CD103+ DCs and CD11b+ DCs were quantified within the CFSE+ mDCs in the dLN . CD103+ DC numbers were marginally reduced when either C3aR or C5aR was blocked but were significantly reduced when both were blocked together ( Fig . 8C ) . A significant reduction in CD11b+ DC migration was observed when C3aR was blocked alone or together with C5aR ( Fig . 8D ) . These results indicate that direct signaling through both complement receptors C3aR and C5aR is critical for the migration of mDCs from the lung to the dLNs . To confirm that the C3aR and C5aR receptors on mDCs directly mediate their migration rather than by the receptors on other cells in the lungs , we purified total CD11c+ DCs from the lungs of WT and C3aR−/−C5aR−/− mice that had been flu infected 18 hours before , CFSE labeled and transferred them intratracheally to flu infected WT recipient mice . We then determined the number of donor CD103+ DCs and CD11b+ DCs in the dLN 18 hours after adoptive transfer . We observed a significantly reduced CD103+ DCs in the mice that received CD11c+ DCs from C3aR−/−C5aR−/− mice in comparison with mice that received CD11c+ DCs from WT mice . Although the CD11b+ DCs showed a similar trend , it is important to note that the CD11b+ DCs have delayed migration kinetics as compared to CD103+ DCs . These results suggest a direct role for these receptors on these mDCs in their migration to the dLN during influenza infection ( Fig . 8E–G ) .
Influenza infection is known to activate complement in the lung with an increase in complement and complement activation products , although the pathway through which it is activated remains unknown [22] , [23] . Our study has uncovered the role of lung resident DCs in this process and highlighted their importance as complement producing and sensing cells . Our observations showed that under steady state conditions , lung C3a levels are high and are likely contributed by cells other than CD103+ DCs . However , it is interesting to note that upon influenza infection , CD103+ DCs exclusively produced high levels of C3 and C5 , yielding activation fragments C3a and C5a . These anaphylotoxins , in turn , interacted with their receptors on both CD103+ DCs and CD11b+ DCs to promote DC migration . Furthermore , our studies using C3aRA and C5aRA demonstrate that C3a and C5a exhibit overlapping but not fully redundant functions because blockade of both their receptors has a significantly more profound effect on mDC migration than the blockade of either alone . Also adoptive transfer experiments using both receptor ( C3aR−/−C5aR−/− ) deficient lung DCs indicated that signaling mediated by both C3a and C5a on mDCs is required for their effective migration . Our results also highlighted the requirement of C3 in the maintenance of CD103+ DCs under steady state . Complement is known to regulate DC induced inhalation tolerance through C3's opsonisation property on innocuous antigen [34] . Hence in the absence of C3 , innocuous antigens are less efficiently taken up by CD103+ DCs , decreasing their ability to migrate ( our data shows decreased number mDC in the dLN under steadystate ) and thus proliferating less under steady state . However , the requirement of C3 in the maintenance of CD103+ DCs was limited to steady state , since under strong inflammatory conditions , deficiency of C3 seems not to affect CD103+ DCs number in the lung . The specific production of C3 and C5 by CD103+ DCs underlines their central role in transporting antigen from the lung to the dLN , allowing the priming of T-cells during the early phase of influenza infection [10] , [25] . It is also interesting to note that CD103+ DCs control the migration of CD11b+ DCs based on the fact that CD11b+ DCs do not produce C3and C5 upon influenza infection , and that their migratory function is facilitated through the production and generation of complement activation products from CD103+ DCs . Thus , to the best of our knowledge , our data demonstrates for the first time how one subset of DC is able to control the functioning of another in a complement dependent manner . Therefore , the defective priming observed in the C3−/− mice may be primarily due to a profound defect in DC-dependent transport of viral antigen to the dLN . In addition the frequency of resident DCs ( CD8α+ DCs ) in dLN were comparable between WT and C3−/− mice ( Fig . S2 ) , ruling out their contribution to this defect . This is further supported by our observations in CD103+ DC-depleted langerin-DTR mice which showed deleterious effects on survival and effector T cell responses similar to those observed in C3−/− mice upon influenza infection . Our results also indicated that the percentage of mDCs that express CCR7 was low under steady state in the C3−/− mice and remained low during influenza infection despite CCR7 upregulation . These results suggest that lack of complement affects CCR7 expression only under steady state , but that CCR7 upregulation during maturation was unaffected in the absence of complement mediated signaling ( Fig . 5A–C ) . Although CCR7 has been shown to be critical for mDC migration to the dLN , its upregulation alone was insufficient to compensate for normal mDC migration , suggesting that multiple signals control DC migration to the LNs [35] , [36] , [37] . Supporting this view , NLRP10 , a nucleotide-binding domain leucine-rich-repeat-containing receptors ( NLR ) has been recently implicated in DC migration , independently of CCR7 mediated signaling [38] . Of note , expression of factor-H and CD59a , genes that control complement activation was found to be altered in DCs from NLRP10−/− mice , suggesting a possible coordination between complement and NLRP10 in facilitating DC migration [38] . Previous studies have shown that DC-derived and -activated C3 and C5 could signal via C3aR and C5aR in an autocrine manner to promote T-cell activation during cognate interaction [19] . In addition , C5aR−/− DCs exhibited attenuated proinflammatory cytokine production , lower expression of MHC-II and costimulatory molecules in response to LPS challenge , as well as reduced capacity for allospecific T-cell stimulation [39] . Similarly , C3-deficient macrophages exhibited lower MHC-II expression and poor ability to expand alloreactive T-cells [40] . These studies , however , were limited by the fact that experiments were performed under in vitro conditions and were confined to bone marrow-derived DCs or macrophages . Furthermore , it is not clearly well established how much these in vitro derived cells are related to resident DCs within non-lymphoid organs . Tissue mDCs are derived from committed circulating DC precursors which migrate from the bone marrow to the periphery and differentiate into distinct DC subsets [7] , whereas bone-marrow DCs were proposed to be rather of monocytic origin [41] , [42] . Furthermore , monocyte-derived DCs or inflammatory DCs are absent at steady state and only appear during inflammatory conditions , which permits their differentiation from Ly6Chi blood monocytes [31] , [41] . Apart from their late appearance , these monocyte-derived DCs possess little ability to acquire viral particles , although their soluble protein uptake capacity is comparable with mDCs [43] . Furthermore , we observed that C3 does not play a costimulatory role on lung mDCs as the expression of maturation markers ( Fig . 1E , F , Fig . 4E and Fig . 5A–C ) and the priming ability of mDCs from the dLN were comparable between WT and C3−/− mice during influenza infection ( Fig . 1G ) . Supporting this similarly , the expression of costimulatory molecules on cDCs was not affected after infection with Listeria monocytogenes in C3−/− mice [44] , suggesting that complement mediated signaling is dispensable for DC maturation during infection . These results are in sharp contrast with previous observations using bone marrow-derived DCs , stimulated with soluble ovalbumin or LPS [19] , [20] . It is thus important to draw a distinction between these two DC subtypes as it is apparent that complement signaling may mediate different functions on these cells . We also observed that acute administration of LPS was unable to overcome the defective mDC migration in C3−/− mice suggesting that complement mediated signaling operate independently of the inflammatory signal in mediating lung mDC trafficking during influenza infection . CD8+ T cell responses are critical in the protection against influenza infection [45] , and it is noteworthy that C3 deficiency affects T cell immunity , viral clearance and survival [16] , [22] . Early lethality in CD103+ DC depleted langerin DTR mice paralleling C3−/− mice though surprising , highlights the critical importance of C3 and its contribution by CD103+ DCs in the control of early viral replication . C3−/− mice showed increased early viral replication kinetics as compared to WT mice ( data not shown ) suggesting a role for C3 in mediating innate antiviral immunity . Thus , our observations reiterate the critical importance of C3 during influenza infection , and we surmise increased viral load leading to pneumonia as a possible cause of mortality in the C3−/− mice . This is further strengthened by the report in the recent 2009 H1N1 pandemic patients wherein individuals who had severe disease had lower C3 levels in the serum , while the C3 concentrations were higher in moderately ill subjects [46] . These observations clearly suggest that complement has a key role in determining the outcome of influenza infection in both mice and humans . Because of its crucial role in protection against influenza , defects in many of the complement proteins , although rare , may be associated with increased susceptibility to influenza infection . Strikingly , deficiency or a defect in factor H and factor I are known to increase the susceptibility to bacterial infections , due to lack of C3 regulation [47] . Similarly , deficiency of mannose-binding lectin ( MBL ) , an activator of complement via the lectin pathway , is more prevalent as compared to other complement component deficiency and is known to be associated with increased susceptibility to upper respiratory tract infections in human and to influenza infection in mice [48] , [49] . Although complement deficiency is rare in humans , genetic polymorphisms in complement proteins such as factor I , factor B , C3 and factor H are known to affect their availability , activity , and susceptibility to chronic disease conditions [50] , [51] , [52] , [53] , [54] , [55] , [56] . Thus , it is possible that complement deficiency/polymorphisms which affect the level and functioning of C3 in an individual may be associated with higher susceptibility to influenza infection as a consequence of compromised DC migration , T-cell priming and decreased viral clearance resulting in severe disease outcome . By demonstrating a novel role for C3 in regulating tissue DC trafficking , our data may also provide a rationale for complement to be exploited as a target for controlling tissue immunity by modulating mDC emigration and as an adjuvant in novel vaccine strategies for enhancing mDC migration to the lymph node to initiate stronger T-cell responses .
Experiments were performed under the approval of the Institutional Animal Care and Use Committee in compliance with the Law and Guidelines for Animal Experiments of the Biological Resource Center ( BRC ) of Agency for Science , Technology and Research ( A*STAR ) , Singapore . These guidelines were established by the national advisory committee for laboratory animal research as per the Animals and Birds Act 2002 . C57BL/6 mice were obtained from the BRC , C3−/− C3aR−/− and C5aR−/− mice were purchased from The Jackson Laboratory . C3aR−/− C5aR−/− double receptor mice were generated by crossing C3aR−/− mice and C5aR−/− mice . Langerin DTR , OT-I Rag1−/− and OT-II Rag2−/− mice ( Taconic ) were obtained from Mutant Mouse Collection Core Service , Singapore Immunology Network ( SIgN ) , Singapore . Homozygous OT-II Rag2−/− mice were crossed once with homozygous CD45 . 1 mice ( Jax ) and the offspring from the F1 generation , referred to hereafter as CD45 . 1+OT-II , were used for adoptive transfer studies . Experiments were generally performed with sex matched mice at 6–10 week of age . Animals were bred under specific pathogen-free conditions at the BRC . Influenza virus strain A/PR/8/34 ( H1N1 ) was obtained from National Institute for Medical Research ( NIMR , London , UK ) . Recombinant influenza A/PR/8/34 strains containing the chicken OVA epitope SIINFEKL ( PR8 OT-I ) and chicken OVA peptide ISQAVHAAHAEINEAGR ( 323–339; PR8 OT-II ) were a gift of P . Thomas ( St . Jude Children's Research Hospital , Memphis , TN ) . Mice were infected intranasally with one of the influenza strains at 15 or 25 plaque forming units ( PFU ) in 25 µl for female or male mice respectively . The influenza dose was optimized to give ∼20% weight loss in WT mice during the peak of infection and recovery without incidence of mortality . In some experiments , diphtheria toxin was administered intraperitoneally ( at 7 ng/g of mouse weight ) before and/or during the course of influenza infection for depleting langerin-expressing CD103+ DCs . In some experiments LPS was administered intratracheally shortly followed by influenza infection . Anti-mouse Abs used for FACS analysis were CD3-APC from BD Pharmingen; CD11c- PerCPCy5 . 5 , MHCII-PB , CD103-PE , CD11b-PE-Cy7 , B220-APCCy7 , CD4-PE , CD86- FITC , CD80-FITC , CD40-APC and CD8-PE-Cy7 from eBioscience; Ly6G/Ly6C-APC , CCR7-APC and IFNγ-APC from Biolegend . For live cell gating , live/dead fixable dye ( Molecular Probes , Invitrogen ) was used . Lung DCs were identified as low auto fluorescent CD11chigh , MHC-II high , B220 negative and Ly6G negative cells . Detection was performed using secondary Ab , goat anti-rat-FITC or Chicken-anti-goat-Alexa Flour 647 from Jackson Immuno Research Laboratories , Molecular Probes and Invitrogen respectively . Intracellular staining of IFNγ in CD4+ T cells was performed by re-stimulating lung lymphocytes with 1 µM of Influenza virus nucleoprotein MHC-II restricted peptide ( 311–325 , QVYSLIRPNENPAHK ) overnight in the presence of brefeldin A ( Sigma ) . Cells were stained for surface markers and fixed with fixation/permeabilization buffer ( BD biosciences ) before staining for intracellular IFNγ . For enumerating virus specific CD8+ T cells , enriched lymphocytes from the lungs were stained with PE labelled H-2Db tetramer with the NP 366–374 epitope ASNENMETM ( Immudex ) for 15 min at room temperature followed by staining with the following antibodies: CD3e-APC and CD8 PE-Cy7 . For C3aR or C5aR staining , cells were incubated with primary Ab for 30 min , followed by secondary Ab for 30 min . All Ab staining was performed at 4°C . Fc blocking Ab ( Anti-Mouse CD16/CD32 ) was used during all FACS staining . Flow cytometric analysis was performed using BD FACS Canto or BD LSR II and analyzed using FlowJo software ( Tree Star , San Carlos , CA ) . Cell sorting of lung and dLN DCs was performed using MoFlo ( Beckman Coulter ) and BD FACS Aria . C3a and C5a levels in bronchioalveolar lavage fluid were quantified by a sandwich ELISA as described below . MaxiSorp immuno modules ( NUNC ) were coated with 100 µl rat anti-mouse C3a or C5a capture antibody ( BD pharmingen ) at 1 or 2 µg/ml in carbonate buffer pH 9 . 6 or phosphate buffer pH6 . 5 ( as recommended by the manufacturer ) respectively and incubated overnight at 4°C . The plates were then washed three times with wash buffer ( PBS with 0 . 05% v/v Tween 20 ) and blocked with 200 µl blocking buffer ( PBS with 10% FBS ) , following which the plates were washed 3 times before the addition of 100 µl standards or 1/50–250 diluted BAL fluid and incubated for 2 hours at room temperature ( RT ) . The plates were washed again , and then 100 µl of biotinylated rat-anti mouse C3a or C5a detecting Ab ( BD pharmingen ) was added at 2 µg/ml concentration to each well ( RT , 1 hour ) . After washing , 100 µl of horseradish peroxidase ( HRP ) conjugated streptavidin at 1 µg/ml ( Jackson Immuno research laboratories ) was added and incubated for 30 minutes at RT . Subsequently , the plates were washed and 100 µl TMB substrate solution ( Pierce Biotechnology Inc ) was added to each well , and incubated in the dark ( RT , 30 min ) . The reaction was stopped with the addition of 50 µl stop solution ( 2N H2SO4 ) to each well . The absorbance was read at 450 nm; reference wavelength 570 nm ( Tecan GENios/Magellan ) . C3a or C5a levels were expressed as ng/100 µg of BAL fluid protein . Total protein in the BAL fluid was quantified by Bradford method ( Bio-Rad ) following manufacturer's instructions . CD4+ T cells ( from OT-II mice ) and CD8+ T cells ( from OT-I mice ) were enriched using MACS beads ( Miltenyi Biotec ) from spleen and lymph node single cell suspension . Enriched cells were labelled with 5 µM of CFSE violet ( Molecular Probes , Invitrogen ) following manufacturer's instructions . Approximately 2×106 CD8+ T cells or 5×106 CD4+ T cells in 200 µl volume were injected retro-orbitally followed by intranasal flu infection . dLN were harvested on day 3 and T cell proliferation was determined by CFSE dilution . For ex vivo analysis mDC subsets were sorted from the dLN of OT-I PR8 infected mice and co-cultured in U bottomed plates at a 1∶10 ratio with CFSE labelled OT-I CD8+ T cells for 3 days . T cell divisions were measured by flow cytometry . The division index was calculated using Flow Jo software . Influenza infected mice were intranasally instilled with 25 µl 8 mM CFSE ( Molecular Probes , Invitrogen ) for labelling lung DC in vivo at indicated time points . Mice were sacrificed 16 hours later and the dLN were removed and analysed for DCs . In some experiments , mice were treated intraperitoneally with either C3aR antagonist - N2-[ ( 2 , 2-Diphenylethoxy ) acetyl]-L-arginine ( SB290157 , Calbiochem ) at 500 µg/mouse or C5aR antagonist- cyclic hexapeptide AcF[OPdChaWR] at 1 µg of peptide/g of mouse weight or both together . Compstatin at a concentration of 1 µg of peptide/g of mouse weight was used as the control . Treatment was started 2 days before influenza infection and continued daily until the mice were sacrificed . In some experiments CD11c+ DCs were purified from the pooled lungs of influenza infected mice , CFSE labeled and then administered into the lungs of influenza infected mice . The migration of recipient mDCs in the donor mice were analysed in the dLN 18 hours after adoptive transfer . Total RNA was obtained from lung tissue/sorted DC population using the RNeasy kit ( Qiagen ) , following which cDNA was then obtained using QuantiTect Reverse Transcription kit ( Qiagen ) . Both kits were used as per manufacturer's protocol . Real-time PCR was performed on an ABI7500 real time PCR system using SYBR Green ( Applied Biosystems ) Primers used for qRT-PCR are as follows: Influenza M-protein-forward: 5′-GGA CTG CAG CGT TAG ACG CTT-3′ and reverse: 5′-CAT CCT GTT GTA TAT GAG GCC CAT-3′; C3-forward: 5′-AAG CAT CAA CAC ACC CAA CA-3′ and reverse: 5′- CTT GAG CTC CAT TCG TGA CA-3′; C5-forward: 5′-GCA TTT CTG ACA CCA GGC TTC-3′ and reverse: 5′- AGC GCA CAG TCA GCT TCC A-3′; C3aR-forward: 5′-TGA AAG CAG GGA GTG TTG AG-3′ and reverse: 5′-TGC TCA CTT GCT CAC ATG AA-3′; C5aR-forward: 5′- CCA TGG ACG ACT CCT AAG GT-3′ and reverse: 5′-CTC CTC TAC ACC GCC TGA CT-3′ Data were analyzed using Prism GraphPad software . Statistical significance was determined by one-way ANOVA or the unpaired Student t test . | Influenza is a global health problem frequented by epidemics and pandemics . Current vaccines against influenza offer limited protection hence the need for reformulation and repeated vaccination . There is a pressing need to develop newer vaccines that are able to generate T cell response . In order to develop such vaccines , there is a need to understand how T cell responses are generated during influenza infection . Influenza specific T cell responses are generated by the dendritic cells ( DCs ) in the lung . Upon influenza infection , DCs in the lung carry viral peptides to the draining lymph node ( dLN ) to initiate an immune response . Thus , migration of DCs from the lung to the dLN is an important step in the initiation of influenza specific T cell response . We now show that activation products of the complement system interact with their receptors on the DCs , which signals for the DCs to migrate from the lung to the dLN . Thus , our results reveal a previously unknown function for complement in mediating lung DC migration during influenza infection and highlight its potential as an adjuvant in novel vaccine strategies . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"immunologic",
"subspecialties",
"immunology",
"biology",
"pulmonary",
"immunology"
] | 2013 | Complement Mediated Signaling on Pulmonary CD103+ Dendritic Cells Is Critical for Their Migratory Function in Response to Influenza Infection |
Although bipolar electrograms ( Bi-egms ) are commonly used for catheter mapping and ablation of cardiac arrhythmias , the accuracy and reproducibility of Bi-egms have not been evaluated . We aimed to clarify the influence of the catheter orientation ( CO ) , catheter contact angle ( CA ) , local conduction velocity ( CV ) , scar size , and catheter type on the Bi-egm morphology using an in silico 3-dimensional realistic model of atrial fibrillation . We constructed a 3-dimensional , realistic , in silico left atrial model with activation wave propagation including bipolar catheter models . Bi-egms were obtained by computing the extracellular potentials from the distal and proximal electrodes . The amplitude and width were measured on virtual Bi-egms obtained under different conditions created by changing the CO according to the wave direction , catheter-atrial wall CA , local CV , size of the non-conductive area , and catheter type . Bipolar voltages were also compared between virtual and clinically acquired Bi-egms . Bi-egm amplitudes were lower for a perpendicular than parallel CO relative to the wave direction ( p<0 . 001 ) , lower for a 90° than 0° CA ( p<0 . 001 ) , and lower for a CV of 0 . 13m/s than 0 . 48m/s ( p<0 . 001 ) . Larger sized non-conductive areas were associated with a decreased bipolar amplitude ( p<0 . 001 ) and increased bipolar width ( p<0 . 001 ) . Among three commercially available catheters ( Orion , Pentaray , and Thermocool ) , those with more narrowly spaced and smaller electrodes produced higher voltages on the virtual Bi-egms ( p<0 . 001 ) . Multiple factors including the CO , CA , CV , and catheter design significantly influence the Bi-egm morphology . Universal voltage cut-off values may not be appropriate for bipolar voltage-guided substrate mapping .
Bipolar catheters are commonly used for electrical mapping of cardiac substrates , with an application for the diagnosis and treatment of various cardiac arrhythmias . Low-voltage areas identified based on the bipolar electrograms ( Bi-egms ) typically correspond to scarred tissue and its border , and are frequently targeted for intervention [1–5] . Bipolar catheters are often preferred to unipolar ones because they provide electrophysiological information regarding the area covered by two metal electrodes , producing sharp electrograms . While bipolar catheters are commonly used in clinical settings , the Bi-egm morphology is known to depend on many factors including the catheter orientation relative to the direction of the activation wave [6–9] , catheter contact angle [7] , conduction velocity ( CV ) [8] , and catheter size and shape [8 , 10–12] . There have been attempts to understand the role of each such factor and develop a strategy to interpret the Bi-egms accordingly . It has been reported that the bipolar voltage increases as the catheter orientation relative to the activation wave direction changes from perpendicular to longitudinal , and decreases as the CV decreases [8] . Blauer et al . [7] used in silico modeling and simulation to examine the effects of the catheter orientation and contact angle on the accuracy of voltage mapping , and found that a higher contact angle with respect to the tissue may improve the voltage mapping accuracy . In the present study , we focused on the Bi-egm morphology clarifying the effects of each aforementioned factor using in silico modeling and simulation . In experimental or clinical studies , it is not feasible to reproducibly examine the effects of any single factor on the Bi-egm morphology while keeping other factors unchanged . On the other hand , in silico simulation enables a systematic quantification of the effects of each factor while controlling for all other factors . Importantly , we simulated three commercially available bipolar catheters and compared their Bi-egm amplitudes in various catheter orientations and contact angles . Our present findings confirmed the clinical observations regarding the effect of the various factors on the Bi-egm morphology and provided insight into the mechanism of the Bi-egm generation , which may be useful for a proper interpretation of the Bi-egm .
As the catheter orientation of the virtual Thermocool catheter changed from 0° to 90° , the bipolar amplitude decreased from 1 . 77±0 . 53 to 0 . 79±0 . 22 mV ( p<0 . 001; Fig 1A ) , while the peak-to-peak width decreased from 16±1 to 9±5 ms ( p<0 . 001 ) . Furthermore , as the catheter contact angle of the virtual Thermocool catheter changed from 0° to 90° , the amplitude decreased from 1 . 77±0 . 53 to 0 . 63±0 . 21 mV ( p<0 . 001; Fig 1A ) . These Bi-egms exhibited a single positive peak and single negative peak because the magnitude of the signal from the proximal electrode was relatively small due to the relatively long distance between the proximal electrode and atrial surface . As the CV decreased from 0 . 48 to 0 . 13 m/s , the bipolar amplitude decreased from 1 . 77±0 . 53 to 0 . 93±0 . 12 mV ( p<0 . 001; Fig 1A ) , while the peak-to-peak width increased from 16±1 to 60±9 ms ( p<0 . 001; Fig 1B ) . Representative Bi-egms for different catheter orientations , contact angles , and CVs are provided in Fig 1C . A representative Bi-egm strip obtained for the virtual Thermocool catheter in the 3D atrial model with induced AF is provided in Fig 2A . Compared to the sinus rhythm conditions , the AF conditions produced a lower bipolar amplitude ( 1 . 77±0 . 53 vs . 1 . 47±0 . 39 mV , p = 0 . 013; Fig 2B ) . That seemed to be due to the random direction of the activation wave with respect to the catheter orientation . On the other hand , the bipolar amplitude did not change significantly upon reducing the APD90 from 223 to 180 ms ( Fig 2B ) . Representative Bi-egms for the different APD90 and under AF conditions are provided in Fig 2C . Virtual scars were modeled as circular non-conductive areas with a diameter of 5 , 7 , or 9 mm on the anterior side of the left atrial model ( Fig 3A ) . The activation time map and positions of the virtual Thermocool catheter relative to the virtual scars are also shown in Fig 3A . Representative Bi-egms for each size of scar are provided in Fig 3B . The bipolar amplitude decreased ( p<0 . 001; Fig 3C ) while the peak-to-peak width increased ( p<0 . 001; Fig 3D ) as the scar diameter increased from 0 to 9mm . Three types of catheters corresponding to commercially available catheters ( Thermocool , Pentaray , and Orion ) were tested . The representative Bi-egms for each catheter type are provided in Fig 4A . The bipolar amplitude was highest for the Orion catheter , followed by the Pentary and Thermcool catheters ( p<0 . 001; Fig 4B ) . The contact with the atrial surface covered the entire electrode area of the Orion catheter because the electrodes of this catheter are flat; on the other hand , only portions of the electrodes of the Thermocool and Pentaray catheters contacted the atrial surface because the electrodes of those catheters are cylindrical . The shape and size of the electrode are the most likely factors explaining the high electrode signal provided by the Orion catheter compared to that provided by the other catheters . Upon examining the Bi-egms for the three catheter types at catheter orientations and contact angles of 0° , 30° , 60° , and 90° , we found that the bipolar amplitude decreased as the catheter orientation and/or contact angle increased ( Fig 4C ) . A clinically acquired voltage map is illustrated in Fig 5A . To facilitate a quantitative comparison between the clinically acquired and virtual maps , the left atrium was divided into 10 segments ( Fig 5B ) . The virtual voltage maps obtained for each type of bipolar catheter ( Fig 5C ) were compared against the clinically acquired map . The average voltage was higher for the virtual Orion and Pentaray catheters than the Thermocool catheter ( 5 . 71±0 . 92 and 3 . 15±0 . 36 mV , respectively , vs . 1 . 57±0 . 12 mV; p<0 . 001 for each comparison ) . The average voltage in each of the 10 segments of the left atrium ( Fig 5B ) showed a moderate correlation between the virtual and clinically acquired data ( r = 0 . 409 for Thermocool , r = 0 . 333 for Pentaray , and r = 0 . 348 for Orion; Fig 5D ) . Fig 6 shows the sensitivity analysis on the lower cut-off voltages to differentiate scar area for the three types of catheters ( Thermocool , Pentaray , and Orion ) . When catheter orientations were randomly selected , all the catheters in the scar area showed voltages less than 0 . 20 , 0 . 27 , and 0 . 63 mV for Thermocool , Pentaray , and Orion catheters , respectively . When catheter orientations were parallel to the wave direction , all the catheters in the scar area showed voltages less than 0 . 39 , 0 . 87 , and 2 . 14 mV for Thermocool , Pentaray , and Orion catheters , respectively . When catheter orientations were perpendicular to the wave direction , all the catheters in the scar area showed voltages less than 0 . 12 , 0 . 35 , and 0 . 87 mV for Thermocool , Pentaray , and Orion catheters , respectively ( Fig 6 ) .
In this study , we simulated the Bi-egm generation applying virtual bipolar catheters mimicking three commercially available catheters to examine the activation wave propagation in a realistic model of the left atrium . The factors associated with a reduced bipolar amplitude included an increased catheter orientation , increased contact angle , and decreased CV . Furthermore , the bipolar amplitude decreased under AF conditions . On the other hand , the change in the APD90 was not associated with a significant change in the bipolar amplitude . Increasing scar size was associated with a decreased bipolar amplitude and increased peak-to-peak width . Among the three types of commercially available bipolar catheters ( Thermocool , Pentaray , and Orion ) , the Orion catheter produced the greatest bipolar amplitude followed by the Pentaray and Thermocool catheters . There was a moderate correlation between the clinically acquired and virtual bipolar voltages in the left atrium . It has long been recognized that the Bi-egm morphology is affected by multiple factors including the catheter orientation , contact angle , and catheter geometry [6–12] . With the bipolar catheter , the two electrodes lie at a certain distance from each other , which inevitably causes the electrodes to sense the activation wave at different times depending on the catheter orientation [13] and on the inter-electrode spacing , with a greater spacing leading to a greater difference between the times at which the activation wave arrives at each electrode . Moreover , at contact angles greater than 0° , the signal intensity is stronger at the distal electrode than at the proximal electrode , which also contributes to the variation in the Bi-egm morphology . On the other hand , an increased CV is expected to decrease the difference between the times at which the wave arrives at the electrodes , regardless of the catheter orientation and contact angle . The catheter size and shape are also known to affect the Bi-egm morphology . Among the three commercially available catheter types tested in this study , the signal amplitude was smallest for the Thermocool catheter , possibly due to the specific size and shape of the electrodes used on that catheter . In the case of cylindrical electrodes , increasing the electrode size results in an increased absolute but not relative contact area with the atrial surface because the opposite side of the electrode does not contact the atrial surface and is moreover located farther from the atrial surface , which results in lower signal amplitude upon averaging over the entire electrode . The Bi-egm is commonly used for electrical mapping of the heart to identify scar areas that will be targeted for catheter ablation of arrhythmias . In this study , we confirmed that the signal amplitude from the mapping catheter is decreased in scar areas , and noted that the decrease is proportional to the scar size . Additionally , all geometrical factors affecting the Bi-egm morphology in the absence of a scar appear to have a similar effect in the presence of a scar . Thus , in clinical practice , low-voltage areas may indicate the presence of scar or may reflect an inappropriate orientation of the catheter . The findings of the in silico study by Blauer et al . [7] , which investigated the effect of the contact angle on the ability to identify scar areas via catheter mapping are informative in this context . As our present results confirmed , the catheter size is especially important for the ability to detect scar areas . Specifically , smaller electrodes tend to produce a higher voltage , whereas bigger catheters are more likely to cover the border of the scar and thus reflect attenuated values . Anter et al . [12] reported that the low-voltage area was larger and the mean voltage within the low-voltage area was lower when measured using a linear catheter with a 3 . 5-mm distal electrode than when measured using a 1-mm multielectrode-mapping catheter in patients with scar-related atrial arrhythmias . Taken together , the present findings and previous observations confirm the fact that different catheter types produce different voltage values , suggesting that it may be necessary to develop catheter-specific voltage criteria for a scar definition of electrical mapping . Traditionally , electrophysiology studies for arrhythmia mapping in the clinical setting have used activation , pace , and entrainment mapping methods [14–16] . However , if the activation sequence of the reentrant tachycardia is constantly changing or degenerating to chaotic fibrillation , it is difficult to map the driver of the arrhythmia using conventional approaches . In such cases , Bi-egm voltage-based substrate mapping , which can be acquired during sinus rhythm or with regular pacing , helps identify low-voltage areas corresponding to scar or its border zone and is useful for locating the slow conduction zone [4] . However , a substrate map based on the local bipolar voltage is influenced by multiple factors [6–12] , as was also shown in our present study . In the clinical electrophysiology laboratory , it is not feasible to generate homogeneous substrate maps with the same contact force , catheter orientation , and tissue conditions . The local CV should differ depending on the substrate remodeling and fibrosis . Moreover , under AF conditions , the catheter orientation relative to the wave direction cannot be controlled when conducting point-by-point mapping using single or multi-electrode catheters . Nevertheless , interventional electrophysiologists are using empirical voltage cut-off values to define low-voltage areas indicative of scar or scar border zones [4] , and it is clear that bipolar voltage-based substrate mapping provides very important information during complex arrhythmia mapping . However , 3D substrate mapping using the current technology employs spatiotemporal assumptions , and the operator should consider the various factors affecting the B-egm morphology . In particular , radiofrequency ablation guided by a 3D color substrate map alone may misidentify the target . Thus , simultaneous and careful monitoring of the electrograms provided by the ablation catheter tip are expected to facilitate a successful ablation . This study had some limitations . The 3D atrial model used for simulating the propagation of the activation wave was homogeneous in its thickness and did not consider transmural variations in the activation wave propagation . Another limitation was that this study did not test the mapping accuracy of the scar area for different types of catheters , which represented our focus for further research . Multiple factors including the catheter orientation , contact angle , CV , and catheter design have a significant influence on the Bi-egm morphology . It may not be appropriate to use a universal cut-off value for the bipolar voltage during voltage-guided substrate mapping .
Computed tomography images of the human left atrium were segmented to generate a 3-dimensional ( 3D ) model of the left atrium using the NavX system ( Abbott , Lake Bluff , IL , USA ) . A triangular mesh was generated on the atrial model for the calculation of the electrical potentials . The electrical wave propagation in the atrium was simulated by numerically solving the following reaction-diffusion equation [17]: ∂Vm∂t=1βCm{∇·D∇Vm−β ( Iion+Is ) } ( 1 ) where Vm is the membrane potential , β is the membrane surface-to-volume ratio , Cm is the membrane capacitance per unit area , D is the conductivity tensor , and Iion and Is are , respectively , the ionic and stimulation currents . A mathematical model of the human atrial Iion , as developed by Courtemanche et al . [18] , was adopted to determine the ionic currents at each computational node . For the Is , a current of -2900pA was applied for 1 . 5ms at each pacing time at a site corresponding to the location of Bachmann’s bundle . Models of three commercially available catheters were constructed by creating virtual objects with the corresponding shape and dimensions ( Fig 7A ) : ( i ) a virtual Thermocool catheter with cylindrical electrodes having a 3 . 5-mm tip and 2-mm inter-electrode spacing ( Thermocool; Biosense Webster , Diamond Bar , CA , USA ) ; ( ii ) a virtual Pentaray catheter with cylindrical electrodes having a 1-mm tip and 2-mm inter-electrode spacing ( Pentaray; Biosense Webster , Diamond Bar , CA , USA ) ; and ( iii ) a virtual Orion catheter with flat electrodes having a 0 . 4-mm2 electrode area and 2 . 5-mm inter-electrode spacing ( Intellamap Orion; Boston Scientific , Marlborough , MA , USA ) . The voltage measured by each of the two electrodes of the bipolar catheter was obtained by computing the extracellular potential for the respective electrode [19] . To examine the effects of the size and shape of the electrode on the Bi-egm morphology , the part of the model corresponding to each electrode was divided into many smaller regions ( ~0 . 07mm2 ) . The extracellular potentials were computed for each of these smaller areas and then averaged to obtain the voltage of the whole electrode . The Bi-egm voltage was obtained as the difference between the voltages noted for the proximal and distal electrodes , with a sampling frequency of 1kHz . To quantify the effects of the catheter orientation and contact angle on the Bi-egm morphology , the virtual catheter was placed on the surface of the 3D model of the atrium , and an electrical stimulus was applied at the location of Bachmann’s bundle using a pacing cycle length ( PCL ) of 600ms . The bipolar voltage and peak-to-peak width were measured as shown in Fig 7B for various catheter orientations and contact angles ( as defined in Fig 7C ) . For a catheter orientation and contact angle of 0° , CVs of 0 . 48 and 0 . 13m/s were tested . The effect of the action potential duration at 90% repolarization ( APD90 ) was also tested at a PCL of 600ms; two APD90 values were tested ( 223 ms and 180 ms ) . The APD90 was changed by adjusting the rapid and slow delayed rectifier K+ currents . To examine the Bi-egm morphology in atrial fibrillation ( AF ) , virtual AF was induced as described previously [20] . The bipolar voltage was measured in AF for 6 s . To examine the effect of the scar size on the Bi-egm morphology , a virtual scar was modeled as a circular non-conductive area , and three diameters were tested ( 5 , 7 , and 9mm ) . The virtual bipolar catheter was placed at the location of the virtual scar , and the Bi-egm morphology was examined at a PCL of 600ms . The location of fibrotic areas was determined based on a clinically acquired bipolar voltage map . First , the bipolar voltage data from the clinically acquired map were interpolated to the computational nodes on the 3D in silico atrial model . To determine the fibrosis status ( yes/no ) for each node , we used the following nonlinear relation between the bipolar voltage and probability of fibrosis . Y=−40 . 0X3+155X2−206X+99 . 8 ( 2 ) where Y is the probability that there is fibrosis at a given node , whereas X is the bipolar voltage at that node . Eq ( 2 ) was developed by comparing the predicted percentage of fibrosis across the 3D atrial model with the pre- and post-ablation fibrosis data reported in the literature [21] for patients with paroxysmal or persistent AF . For each node , the probability of fibrosis calculated based on clinically acquired bipolar voltage data ( Eq ( 2 ) ) was compared against a random number between 0 and 1 . If the random number was below the calculated probability of fibrosis , the node was considered to have a positive fibrosis status . The ionic currents and diffusion coefficients in the fibrosis area were adjusted as described elsewhere [22] . To generate virtual bipolar voltage maps , the activation wave propagation in the 3D atrial model was simulated at a PCL of 600ms . The virtual bipolar catheter was placed at 810 randomly selected nodes with a random orientation of the catheter . The virtual bipolar voltage was calculated at the 810 nodes , and those voltages were interpolated to all computational nodes to generate the virtual bipolar voltage map . Such a voltage map was obtained for the three types of virtual catheters ( Thermocool , Pentaray , and Orion ) . The 3D atrial model was divided into 10 sections , and the average voltage in each section was compared between the clinically acquired and virtual maps . A sensitivity analysis was performed on the lower cut-off voltages to differentiate scar area for the three types of catheters ( Thermocool , Pentaray , and Orion ) . A circular non-conductive zone ( Diameter = 18mm ) was created in the anterior region of a 3-dimensional model of the left atrium . Virtual catheter was placed at random locations in and around the non-conductive zone ( N = 303 ) . The numbers of locations at which both of the distal and proximal electrodes were fully included in the non-conductive zone were 34 , 58 , and 71 for Thermocool , Pentaray , and Orion catheters , respectively . An electrical stimulation was applied at a site corresponding to the location of Bachmann’s bundle with PCL of 600ms . Bipolar voltage was obtained at the eighth stimulation . Three catheter orientations with respect to the wave direction were tested: 0° , 90° , and random angles . Sensitivity analysis was performed with various threshold voltages until the voltages of all the locations in the scar area were less than the threshold voltage . | Cardiac arrhythmias are rhythm disorders of the heart leading to abnormal heart function . For the diagnosis and treatment of the arrhythmias , clinicians insert catheters into the heart and examine the electrical signal propagation in the heart . Among different type of catheters , bipolar catheters have two electrodes at the tip of the catheter with the signal being the difference between the two electrodes , which provides sharper signal than unipolar catheter . However , bipolar electrogram is dependent on many factors including catheter design and orientation , and consequently , knowledge of the determinants of the bipolar electrogram is needed for proper interpretation of the signal . In this study , we examined the effects of many factors on bipolar electrogram using computer simulation . Computer simulation is very useful in this type of study because , in clinical settings , it is not feasible to control each factor precisely . We quantitatively demonstrated the effects of catheter design and orientation , and cardiac wave propagation speed on bipolar electrogram . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"biotechnology",
"medicine",
"and",
"health",
"sciences",
"data",
"acquisition",
"engineering",
"and",
"technology",
"catheters",
"membrane",
"potential",
"fibrosis",
"electrophysiology",
"developmental",
"biology",
"waves",
"cardiology",
"bioengineering",
"computer",
"and",
"information",
"sciences",
"arrhythmia",
"atrial",
"fibrillation",
"chemistry",
"electrode",
"potentials",
"physics",
"wave",
"propagation",
"electrochemistry",
"physiology",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"medical",
"devices",
"and",
"equipment"
] | 2019 | Multiple factors influence the morphology of the bipolar electrogram: An in silico modeling study |
Mutations can originate from the chance misincorporation of nucleotides during DNA replication or from DNA lesions that arise between replication cycles and are not repaired correctly . We introduce a model that relates the source of mutations to their accumulation with cell divisions , providing a framework for understanding how mutation rates depend on sex , age , and cell division rate . We show that the accrual of mutations should track cell divisions not only when mutations are replicative in origin but also when they are non-replicative and repaired efficiently . One implication is that observations from diverse fields that to date have been interpreted as pointing to a replicative origin of most mutations could instead reflect the accumulation of mutations arising from endogenous reactions or exogenous mutagens . We further find that only mutations that arise from inefficiently repaired lesions will accrue according to absolute time; thus , unless life history traits co-vary , the phylogenetic “molecular clock” should not be expected to run steadily across species .
Because mutations are the ultimate source of all genetic variation , deleterious and advantageous , mutagenesis has been of central interest even before the discovery of DNA as the genetic material ( e . g . , [1] ) , and developing a model of mutational heterogeneity along the genome is a major focus of current disease mapping studies [2 , 3] . From many decades of research into mechanisms of DNA replication , damage , and repair , we know that mutations can arise from errors during replication , such as the incorporation of a non-complementary nucleotide opposite an intact template nucleotide during DNA synthesis [4] , or from DNA damage caused by exogenous mutagens or endogenous reactions at any time during normal growth of a cell ( Fig 1 ) . If uncorrected by the next round of DNA replication , these lesions will lead to arrested replication and cell death , or to mutations in the descendent cells ( either because of incorrect template information or due to lesion bypass by error-prone DNA polymerase ) [5] . While the fraction of mutations that is non-replicative in origin remains unknown , the common assumption is that mutations are predominantly replicative [6–9] . The basis for this assumption is a set of observations from disparate fields suggesting that , at least in mammals , mutations seem to track cell divisions . First , in phylogenetic studies , it has been observed repeatedly that species with longer generation times tend to have lower substitution rates , which under neutrality reflects lower mutation rates per unit time ( “the generation-time effect” ) ( e . g . , [7 , 10] ) . Second , based on comparisons of X , Y chromosomes and autosomes , it has been inferred that substantially more mutations arise in the male than in the female germline ( e . g . , [6 , 8 , 11] ) . In human genetics , pedigree resequencing studies have confirmed a male bias in mutation of approximately 3:1 at a paternal age of 30 , and revealed a linear increase in the number of mutations in the child with the father’s age ( e . g . , [12 , 13] ) . These observations are all qualitatively consistent with mutations arising from the process of copying DNA: all else being equal , organisms with shorter generation times should undergo more germ cell divisions per unit time; in mammals , oocytogenesis is completed by birth whereas spermatogenesis is ongoing since puberty throughout the male lifespan , resulting in more germ cell divisions in males than females ( Fig 2A ) [14 , 15] . An informative exception to the “generation time effect” seen in phylogenetic studies is transitions at CpG sites , which represent approximately a fifth of de novo germline mutations [12] , and show relatively constant substitution rates across species [16–18] . Their more “clock-like” behavior may reflect their distinct molecular origin [16] , as CpG transitions are believed to be due primarily to the spontaneous deamination of the 5-methylcytosine ( 5mC ) [19] . This case demonstrates the potential importance of non-replicative sources in germline mutations and raises the possibility that , despite the usual assumption ( e . g . , [20 , 21] ) , not all non-CpG mutations arise from mistakes in replication . A third argument for the preponderance of replication errors has been made recently in cancer genetics , on the basis of two observations: ( i ) that somatic mutations tend to accrue more rapidly in tissues with higher renewal rates [22] and ( ii ) that , across tissues , the lifetime risk of cancer is associated with the total number of stem cell divisions [9] . Together , these findings were interpreted as indicating that in humans , random errors that occur during DNA replication are the source of most somatic mutations , and hence the main determinant of the odds of developing driver mutations that lead to cancer [9] . However , sequencing of tumor samples also revealed characteristic mutation patterns ( “mutational signatures” ) that reflect known DNA damage processes by endogenous or exogenous sources [23] . Moreover , environmental mutagens are known to influence the incidence of a subset of cancers , implying a role of mutations of non-replicative origins ( e . g . , [24 , 25] ) . These apparently conflicting observations again point to the importance of understanding how mutations arise in somatic tissues as well as in the germline . Because , to date , arguments for the replicative origin of mutations have been qualitative and often based on implicit assumptions , we decided to model how the source of mutations relates to their rate of accumulation over cell divisions . For replication-driven mutations , we describe how mutations are expected to accumulate with age , and hence how the generation time relates to the yearly neutral mutation rate . This simple derivation allows us to show that , all else being equal , increases in the generation time will lead to decrease in the mutation rate only under very specific conditions on other parameters . For non-replicative mutations , we relate the mutation rate to rates of DNA damage , repair , and cell division . We show that only when the repair of DNA lesions is highly inefficient will mutations accrue according to absolute time . Otherwise , the accrual of mutations is expected to depend not only on absolute time but also on the rate of cell divisions—a feature previously thought to be specific to replication-driven mutations . By providing explicit expectations for how mutations should accumulate with sex , age , and cell division , these models provide a framework within which to interpret observations from evolutionary biology , human genetics , and cancer genetics .
The mutation rate per generation , i . e . , the total number of germline mutations between two consecutive generations , is the sum of mutations inherited from both parents , which arose in the lineages of germ cells that gave rise to the child . If mutations are introduced by replication errors , their accumulation will track rounds of DNA replication . In each developmental stage , the number of replication-driven mutations can then be expressed as the product of the number of cell divisions and the mutation rate per cell division . Although a constant mutation rate per cell division is often assumed , explicitly or implicitly [6 , 26] , this need not hold , especially when the cell lineage goes through different development stages , as do germ cells of multicellular organisms . Thus , we consider a more general case , allowing for variation in per cell division mutation rate ( e . g . , a higher mutation rate in early embryonic development ) [27] and describe the accumulation of replication-driven mutations as a piece-wise linear process ( following [18] ) . For simplicity , we divide germ cell development from fertilization to reproduction into four stages , separated by the settlement of primordial germ cells in the developing gonads ( which almost coincides with sexual differentiation ) , birth , and onset of puberty , respectively . Let dis and μis be the numbers of cell divisions and replication error rate in the ith stage ( i = 1 , 2 , 3 , 4 ) in sex s ( s ϵ{f , m} ) . Because there is no sex difference in the first stage , d1f = d1m and μ1f = μ1m , and we replace them by d1 and μ1 ( see Table 1 for a list of parameters involved in the model ) . Previous studies in Drosophila melanogaster suggest that the first division of a zygote has an extraordinarily high mutation rate [27 , 28] . Although the first division in Drosophila is quite distinct from that in mammals , it is possible that it would be more mutagenic in mammals as well , so we consider the first division separately as stage 0 , of which the mutation rate is μ0 for both sexes , and re-define stage 1 as from the second post-zygotic division to sex differentiation . The total number of replication-driven autosomal mutations from one parent to the offspring is then: MRs= ( μ0+μ1d1+μ2sd2s+μ3sd3s+μ4sd4s ) H , sϵ{f , m} where H is the total number of base pairs in a haploid set of autosomes . In mammals , all mitotic divisions of female germ cells are completed by birth of the future mother , so d3f = 0 and d4f = 0 , and the total number of replication-driven mutations inherited from mother is ( Fig 2B red line ) : MRf= ( μ0+μ1d1+μ2fd2f ) H . ( 1 ) In contrast , male germ cells undergo divisions in all stages outlined above; furthermore , the number of germ cell divisions after puberty ( d4m ) is not a fixed number , because after puberty , sperm are continuously produced through asymmetric division of spermatogonial stem cells , at a roughly constant rate . If we assume that males and females have the same ages of onset of puberty and reproduction ( denoted by P and G respectively ) , and that a spermatogonial stem cell undergoes cm divisions each year , the total number of paternal mutations is a function of reproductive age G ( Fig 2B blue line ) : MRm=[μ0+μ1d1+μ2md2m+μ3md3m+μ4m ( cm ( G−P−tsg ) +dsg ) ]H , ( 2 ) where tsg and dsg are the time ( in years ) and the number of cell divisions needed to complete spermatogenesis from spermatogonial stem cells . The two divisions in meiosis are counted as one here , because only one round of DNA replication takes place in meiosis . Summing Eqs 1 and 2 , the total number of autosomal replication-driven mutations inherited by a diploid offspring from both parents is ( Fig 2B purple line ) : MR=MRf+MRm=[2μ0+2μ1d1+μ2fd2f+μ2md2m+μ3md3m+μ4m ( cm ( G−P−tsg ) +dsg ) ]H . By dividing Eq 2 by Eq 1 , we obtain the ratio of male to female replication-driven mutations: αR=MmMf=μ0+μ1d1+μ2md2m+μ3md3m+μ4mdsgμ0+μ1d1+μ2fd2f+μ4mcmμ0+μ1d1+μ2fd2f⋅ ( G−P−tsg ) , which suggests that , keeping other parameters unchanged , increases in generation time G will lead to a stronger male bias in mutation , as expected intuitively ( Fig 2C ) . It follows that the average yearly mutation rate ( i . e . , the substitution rate if all mutations are neutral ) is a function of G: mR , y=mR , gG=2μ0+2μ1d1+μ2fd2f+μ2md2m+μ3md3m+μ4mdsg+μ4mcm ( G−P−tsg ) 2G⋅ ( 3 ) In order to explore the effect of generation time on the average yearly mutation rate , it is useful to reorganize Eq 3 as: mR , y=μ4mcm2+A*2G , ( 4 ) where A*=2μ0+2μ1d1+μ2fd2f+μ2md2m+μ3md3m−μ4m ( cmP+cmtsg−dsg ) , which is independent of G . Eq 4 suggests that if and only if A* = 0 will the yearly mutation rate be independent of G . Otherwise , mR , y will either increase or decrease monotonically with G , depending on the sign of A* . Changes in the timing of puberty ( P ) , in the number of cell divisions ( dis ) and in the replication error rate per cell division in each stage ( μis ) will also influence the dependence of mR , y on G . The relationship between mR , y and G can also be directly read off the curve in Fig 3 . The mutation rate per generation increases linearly with G after puberty , but this linear relationship does not apply to the period before puberty . If and only if the extended fitted line passes through the origin will the mutation rate per generation be exactly proportional to the generation time , and the average yearly mutation rate unaffected by G . If the intercept of the extrapolated line at age zero is positive , mR , y decreases with G , consistent with the observed “generation time effect” in primates . Conversely , if the intercept is negative , mR , y increases with G . In fact , the intercept obtained by extrapolation is exactly A* in Eq 3 , so interpretation from Fig 3 is equivalent to that suggested by Eq 4 . Although estimates of other parameters exist , little is known about the replication error rate per cell division in germ cells , so it is unclear whether A* is positive or negative . However , it seems highly coincidental that an expression that involves multiple variables would happen to equal zero . Therefore , we argue that there is almost certainly an effect of generation time on yearly mutation rate in humans , although the magnitude of the effect could be small . The magnitude of the paternal age effect in pedigree data suggests that there should be generation-time effect in humans ( see S1 Text ) . Our model further reveals that , all else being equal , a longer generation time can lead to either an increase or decrease in the average yearly rate at which replicative mutations accrue . Therefore , the general observation that substitution rate in mammals tends to decrease with increasing generation times [7 , 10 , 16] is not necessarily expected; in fact , its existence requires very specific conditions on ontogenesis to hold ( shown in Fig 3B ) . Moreover , given the current understanding of germ cell development in humans , the generation-time effect implies a higher mutation rate per cell division in early embryonic development than in spermatogenesis ( see S1 Text for a discussion of available data in humans and chimpanzees ) . Since mammalian species differ drastically in life history traits as well as development and renewal processes of germ cells [26 , 29] , Eq 4 implies that the yearly mutation rate likely varies among species ( even if per cell division mutation rates remain constant ) . As a result , unless life history traits co-vary in certain ways , we should not expect neutral substitution rates to be constant across mammalian species—or even along single evolutionary lineages . An important implication is that changes in life history among hominins [30] introduce uncertainty about dates in human evolution obtained under the assumption of a molecular clock [31] . DNA is subject to large numbers of damaging events every day as a result of normal cellular metabolism , and more DNA lesions may be generated by exogenous agents [32] . Typical DNA damage includes depurination and deamination due to DNA hydrolysis; alkylation and oxidation of bases induced by chemicals such as ethylmethane sulfonate or reactive oxygen species; pyrimidine dimers caused by ultraviolet radiation; and single- or double-stranded breaks produced by gamma and X-rays . Most single-stranded lesions cannot pair properly with any regular bases ( termed “noncoding lesions” ) and thus will block DNA replication if unrepaired ( Fig 1 ) . However , a few alterations to nucleotides can pair with bases different from the original Watson-Crick partners; such lesions ( termed “miscoding lesions” ) , if unrepaired before replication , will lead to irreversible replacement of a base pair after cell division ( Fig 1 ) [5] . To model the accrual of non-replicative mutations , we start by considering deamination of methylated CpG sites , which is the best understood example of miscoding lesions , and discuss more complex mutagenesis mechanisms in the S2 Text . This modification turns the methylated cytosine ( mC ) into a thymine ( T ) ; if uncorrected before DNA replication , an adenine instead of a guanine will be incorporated into the nascent strand , which results in a mutation in one of the two daughter cells . While DNA replication and cell division are obviously two distinct events , they are tightly coordinated such that DNA is replicated exactly once before each division ( other than in meiosis and under a few unusual conditions ) . In what follows , we therefore do not distinguish between the two events . We model the proportion of damaged base pairs at the time of cell division by considering the effects of both damage and repair ( Fig 4A ) . For simplicity , we assume that single-strand damage occurs at a constant instantaneous rate μ throughout cell cycle and that the repair machinery recognizes lesions at a constant rate r ( Fig 4A ) . Thus , the proportion of base pairs that carry a lesion at time t after the last cell division , p1 ( t ) , is described by a simple differential equation: dp1dt=μ ( 1−p1 ) −rp1 , with the initial condition p1 ( 0 ) = 0 . The solution to the differential equation is: p1 ( t ) =μμ+r ( 1−e− ( μ+r ) t ) . Because each unrepaired single-strand lesion leads to a base pair substitution in one of the two daughter cells , the average mutation rate in one cell division ( i . e . , the expected fraction of base pairs that differ between a daughter cell and its mother cell ) is: MNR ( T ) =12p1 ( T ) =μ2 ( μ+r ) ( 1−e− ( μ+r ) T ) , ( 5 ) where T is the time between two consecutive cell divisions ( Fig 4B ) . We assume that μ<<1/T for any biologically reasonable value of T , so even in the absence of DNA repair , the absolute mutation rate per base pair per cell division ( ≈½μT ) is very small . In addition , we focus on a single cell lineage and assume an infinite sites model , in which each genomic site can be mutated at most once . Thus , the total mutation rate over many cell divisions is simply the sum of the mutation rates for every division . A key feature of the result in Eq 5 is that the accumulation of mutations per cell division exhibits two different limiting behaviors , depending on the relative rates of cell division and repair . When the rate at which lesions are repaired is much slower than the rate of cell division ( rT<<1 ) , the number of mutations is approximately proportional to time between two rounds of DNA replication: MNR ( T ) =μT2 . ( 6 ) The intuition is that , for a cell under this condition , there is almost no time for the repair machinery to correct lesions , so almost all lesions result in mutations . Consequently , mutations accumulate at a constant rate regardless of the rates of cell division and repair ( Fig 4B , red box ) . In other words , non-replicative mutations that are inefficiently repaired will track absolute time . In contrast , in the other limit where the repair is highly efficient relative to the rate of cell division ( rT>>1 ) , the number of mutations approaches an equilibrium level by the time of cell division: MNR ( T ) =μ2 ( μ+r ) . ( 7 ) As a result , mutations accumulate at a rate that is roughly proportional to the number of cell divisions , regardless of absolute time ( Fig 4B , blue box ) . Here , the intuition is that when repair is highly efficient , the few lesions that have not been corrected tend to be those that arose right before the cell division , and therefore the time since the last division has little effect . Importantly , under this scenario , the accrual of mutations that arise from lesions mimics what would be expected from replication errors . We note that the existence of such an equilibrium comes from the assumption of no error in repair; however , even when errors in repair are taken into consideration , there exists a phase in which repair and damage roughly balance out , so the mutation rate is proportional to the cell division rate ( see S2 Text ) . To understand how the mutation rate of non-replicative mutations depends on absolute time and the rate of cell division in general , we derive the mutation rate per unit time as the product of mutation rate per cell division and the cell division rate ( c = 1/T>0 ) : m ( c ) =cMNR ( 1c ) =c2 ( 1+R ) ( 1−e− ( 1+R ) μc ) . ( 8 ) The mutation rate m ( c ) has two limiting behaviors when c approaches infinity and zero , respectively , which have the same intuitive explanations as Eqs 6 and 7 , respectively . Moreover , it can be shown that m ( c ) is a concave increasing function of c . In other words , in a given period of time , faster dividing cell lineages accumulate more non-replicative mutations than slowly dividing lineages , but the increase in the number of mutations is smaller than the increase in the cell division rate . Therefore , when repair is neither inefficient nor extremely efficient , and given fixed damage and repair rates , faster dividing lineages are expected to accumulate non-replicative mutations at a higher rate per year than more slowly dividing ones ( Fig 4C and see Table 2 for a list of parameters involved in the model ) . This model can be extended readily to incorporate more features , such as other types of non-replicative mutations as well as to understand phenomena such as the strand bias in mutations associated with transcription ( see S2 Text ) [33 , 34] . Although the quantitative results differ , the main conclusion holds: the accumulation of non-replicative mutations depends critically on the repair efficiency in relation to the cell division rate .
These results demonstrate the fundamental importance of repair efficiency in determining the dependence of mutation rates on age , sex , and cell division rate ( Fig 5 ) . When DNA repair is inefficient , we should expect a linear accumulation of damage-induced mutations , partially justifying the expectation that neutral substitution rates of non-replicative mutations should not depend on generation time or other life history traits , and hence may be constant across species . However , our model highlights additional conditions for this expectation to be met: in particular , it reveals that the clock-like behavior of CpG transitions in mammals not only requires a non-replicative origin but also implies both relatively low repair efficiency in germ cells and similar damage rates across mammalian species ( Fig 5A ) . A further implication is that the number of mutations of maternal origin should increase with the mother’s age for CpG transitions and other mutations that arise from inefficiently repaired lesions . In this regard , we speculate that the current lack of a detectable maternal age effect may be due to underpowered sample sizes ( notably because of the strong correlation between maternal and paternal ages ) . In any case , our model predicts that a maternal age effect should be detectable with sufficient data and reliable identification of parental origin of mutations ( e . g . , by sequencing of a third generation ) . Conversely , the detection of a maternal age effect on mutation rate would provide prima facie evidence for the existence of non-replicative mutations that are not efficiently repaired ( assuming no relationship between the age at which an oocyte is ovulated and the number of cell divisions experienced during oocytogenesis [35] ) . Also of note , lesions that have the same damage rate but are recognized by distinct repair mechanisms may differ not only in their absolute mutation rates but also in their time dependencies . Indeed , changes to the repair efficiency ( or to the division rate ) could alter the sex and time dependence of non-replicative mutations; for example , decreases in repair efficiency could lead mutations that previously tracked cell division rates to depend more on absolute time . Therefore , the phylogenetic molecular clock should not necessarily run at a steady rate even for mutations due to spontaneous DNA damage . Our modeling results also shed light on studies of somatic mutations . As an illustration , a recent single-cell sequencing study identified mutations in neurons from the cerebral cortex of three healthy individuals [36] . The numbers of mutations in each cell were similar regardless of the donor’s sex and age ( ranging from 15 to 42 years , Fig 5C ) [37] . The genome-wide distribution of the somatic mutations appeared to be associated with transcription , with most identified mutations being C to T transitions at methylated cytosines . These observations led the authors to conclude that the mutations that they observed were due to non-replicative damage that was poorly repaired [36] . However , if mutations are non-replicative in origin and not repaired , more DNA lesions should accrue in older individuals , even in post-mitotic cells . In light of our model , an explanation is that an equilibrium between DNA damage and repair was reached before adolescence , and thus that the number of mutations does not increase further with age ( Fig 5C ) . If this is the case , then there should be fewer somatic mutations in post-mitotic neurons from younger individuals , in which the equilibrium has not been reached . Similarly , the model helps to interpret patterns observed in tumor samples , in which the total number of somatic mutations increases with the age of patient at diagnosis and grows at higher rates in fast renewing tissues [22] . Deamination at CpG sites make substantial contribution to mutations in almost all cancer types and accumulate at constant yearly rates that appear to be positively correlated with the turnover rates of the corresponding normal tissues ( Fig 5B ) [23 , 38] . As we have shown , all else being equal , a positive correlation is expected even for mutations that arise from DNA damage , so long as lesions are not poorly repaired in all somatic tissues . Importantly , then , the recently reported correlation between number of stem cell divisions and lifetime risk of cancer across tissues is consistent with mutations of both replicative and non-replicative origins , and does not provide any evidence that most mutations are attributable to replication mistakes in stem cell divisions ( what the authors referred to as “bad luck” in [9] ) . Of course , tumorigenesis is a multistep process that depends not only on the accumulation of mutations but also on tissue architecture as well as the order and consequences of specific mutational events , and gaining insight into its causes will likely require consideration of all these facets . What our model makes apparent is that it will also be important to incorporate a realistic model for the source of mutations . Similar arguments apply to the male bias in mutation found by resequencing pedigrees and the generation time effect in phylogenetics: neither observation provides evidence for a replication-driven mutational process , as they could also reflect mutations arising from residual lesions left after efficient repair . Given these considerations , it becomes clear that , based on available data , we still do not know if a substantial proportion of human germline and somatic mutations—including those at non-CpG sites—are non-replicative in origin . In summary , we introduce a model that helps to interpret findings from studies of somatic mutations , human pedigrees , and phylogenies . Although very simple , its behavior appears to be robust . By making explicit the relationship between the genesis of mutations and their accumulation over ontogeny , the model reveals the critical importance of both the source of mutations and the repair efficiency of lesions . Because replicative mutations and non-replicative mutations can display similar properties when repair is efficient , none of the previous observations of correlations between mutation and cell division rates lends strong support to the commonly held belief that most mutations are replicative in origin . Further experimental work is therefore needed to distinguish between different sources of mutation . Notably , fitting models such as this one to growing data from diverse fields should provide a quantitative understanding of how DNA changes accumulate in somatic tissues during a lifetime and in the germline over evolutionary time scales . | We relate how mutations arise to how they accumulate in different sexes , with age and with cell division . This model provides a single framework within which to interpret emerging results from evolutionary biology , human genetics , and cancer genetics . We show that the accrual of mutations should track cell divisions not only when mutations originate during DNA replication but also when they arise through non-replicative mechanisms and are repaired efficiently . This realization means that previous observations of correlations between mutation and cell division rates actually provide little support to the commonly held belief that most germline and somatic mutations arise from replication errors . We further find that only mutations that arise from inefficiently repaired lesions will accrue according to absolute time; thus , without covariation in life history traits , the phylogenetic “molecular clock” should not be expected to run at constant rates across species . | [
"Abstract",
"Introduction",
"Results",
"Discussion"
] | [] | 2016 | Interpreting the Dependence of Mutation Rates on Age and Time |
Maternal homozygosity for three independent mutant hecate alleles results in embryos with reduced expression of dorsal organizer genes and defects in the formation of dorsoanterior structures . A positional cloning approach identified all hecate mutations as stop codons affecting the same gene , revealing that hecate encodes the Glutamate receptor interacting protein 2a ( Grip2a ) , a protein containing multiple PDZ domains known to interact with membrane-associated factors including components of the Wnt signaling pathway . We find that grip2a mRNA is localized to the vegetal pole of the oocyte and early embryo , and that during egg activation this mRNA shifts to an off-center vegetal position corresponding to the previously proposed teleost cortical rotation . hecate mutants show defects in the alignment and bundling of microtubules at the vegetal cortex , which result in defects in the asymmetric movement of wnt8a mRNA as well as anchoring of the kinesin-associated cargo adaptor Syntabulin . We also find that , although short-range shifts in vegetal signals are affected in hecate mutant embryos , these mutants exhibit normal long-range , animally directed translocation of cortically injected dorsal beads that occurs in lateral regions of the yolk cortex . Furthermore , we show that such animally-directed movement along the lateral cortex is not restricted to a single arc corresponding to the prospective dorsal region , but occur in multiple meridional arcs even in opposite regions of the embryo . Together , our results reveal a role for Grip2a function in the reorganization and bundling of microtubules at the vegetal cortex to mediate a symmetry-breaking short-range shift corresponding to the teleost cortical rotation . The slight asymmetry achieved by this directed process is subsequently amplified by a general cortical animally-directed transport mechanism that is neither dependent on hecate function nor restricted to the prospective dorsal axis .
Dorsoventral axis formation is a fundamental process in early vertebrate embryogenesis . In many vertebrates such as amphibians and teleosts , evidence indicates that maternally-supplied dorsal determinants trigger the formation of the future dorsal organizer . Embryological manipulations have indicated that the dorsal determinants are initially localized to the vegetal pole and subsequently translocate animally to the prospective dorsal side in a microtubule-dependent process in both amphibians ( reviewed in [1] ) and teleosts [2]–[4] . In amphibians , translocation of the signal from the vegetal pole to the dorsal side is initiated by cortical rotation , the microtubule-dependent movement of the egg cortex with respect to its core that is triggered by fertilization and is implemented by transport along microtubule tracks ( reviewed in [1] ) . Although not readily apparent in the zebrafish embryo , detailed analysis has indicated the existence of processes that share similarities with the amphibian cortical rotation . Early studies showed that fluorescent polystyrene beads injected at the vegetal pole were transported animally along microtubule-based cortical tracks in a microtubule dependent manner [2] , and that this movement had temporal dynamics and functional requirements similar to that of the movement of putative dorsal determinants as defined by embryological manipulations [3] , [4] . Subsequent work showed that maternal factors such as Syntabulin and Wnt8a , involved in axis induction , localized to the vegetal pole of the egg and upon fertilization and egg activation shifted animally towards the presumed prospective dorsal region of the embryo [5]–[7] . Further studies of microtubule rearrangements in live embryos confirmed that the tracks of bundled microtubules that form at the zebrafish vegetal pole upon egg activation become aligned in the direction of the future dorsal side of the embryo , and showed bulk cortical particle movement analogous to a cortical rotation [8] . The movement of the dorsal determinant results in the activation of the Wnt/βcatenin signaling pathway and the activation of β-catenin-dependent targets [1] . This well-known pathway is characterized by the activation of Frizzled receptors by the Wnt ligand , and an intracellular cascade involving the activation of Dishevelled and the downregulation of a β-catenin degradation complex that includes GSK3 , Axin and Adenomatous polyposis coli ( APC ) , leading to the accumulation of β-catenin in the nucleus [9] . Nuclear β-catenin in turn interacts with transcription factors of the Tcf family to activate transcription of target genes . Wnt/βcatenin pathway components and/or nuclear accumulation of β-catenin have been shown to be involved in embryonic axis determination in diverse deuterostomes such as fish , amphibians , mammals and amphioxus , as well as in lineages as basal as echinoderms , Cnidarians and planaria ( reviewed in [10] ) implying that the pathway was recruited for axis determination very early in animal evolution . Although the involvement of Wnt/βcatenin activation across species is well documented , the identity of the molecules that activate the pathway in the early embryo , often referred to as dorsal determinants , remains unknown in most cases . In Xenopus , wnt11 mRNA is first located at the vegetal pole and becomes enriched at the future dorsal side after fertilization , and depletion of wnt11 mRNA results in embryos defective in dorsal axis induction [11] . Thus , Wnt11 , together with ubiquitously present Wnt5 [12] , [13] has been proposed to be the dorsal determinant in this amphibian species . Studies in the zebrafish exclude a function for Wnt11 or Wnt 5 in axis induction but suggest a role for Wnt8a in this process [7] . Maternal zebrafish wnt8a mRNA is localized during oogenesis to the vegetal pole of the oocyte and , upon fertilization , wnt8a mRNA experiences a shift from its original location at the vegetal pole to an off-center region thought to correspond to the dorsal side [7] . These studies suggest that , while Wnt/β-catenin pathway activation may be highly conserved in axis induction across the animal kingdom , maternally-based mechanisms that lead to the activation of the pathway vary . Efforts from several laboratories have used forward genetics approaches to identify maternal factors essential for various aspects of early embryonic development in the zebrafish [5] , [14]–[18] . Several reports have documented maternal-effect mutations affecting zebrafish dorsal axis induction [5] , [15] , [17] , [18] . Mutations in three maternal-effect genes , ichabod , tokkaebi and hecate , cause specific ventralized phenotypes , consistent with a role for these genes in axis induction . Overexpression of Wnt signaling pathway components in ichabod mutant embryos indicate that this mutation acts within the Wnt/β-catenin signaling pathway at a level downstream of the β-catenin degradation complex [15] . These and other results show that ichabod corresponds to a β-catenin-2 gene expressed maternally and involved in axis induction [19] . Similar overexpression analysis has shown that tokkaebi and hecate , in contrast to ichabod , act upstream of the β-catenin-degradation machinery [5] , [20] . Specifically , overexpression of components that activate the pathway at multiple points can rescue the hecate and tokkaebi mutant phenotypes . Rescue by an exogenous Wnt ligand suggests that the Wnt/β-catenin pathway is intact in tokkaebi and hecate mutant embryos , and suggests that , rather than being required for an integral component of Wnt/β-catenin signaling , these genes are required to regulate an endogenous signal that activates this pathway . Consistent with such a proposed role , positional cloning reveals that tokkaebi corresponds to the syntabulin ( sybu ) gene , which codes for a linker of the kinesin I motor protein involved in cargo transport along microtubules [6] . Both sybu mRNA and protein are localized to the vegetal pole of the egg , and Sybu protein exhibits a slight off-center shift upon egg activation [6] . These data suggest a role for Sybu in the microtubule-dependent transport of vegetally localized dorsal determinants . Here , we present the molecular characterization of the zebrafish maternal-effect gene hecate ( hec ) . Maternal homozygosity for three independent mutant hec alleles results in embryos with reduced expression of dorsal organizer genes and defects in the formation of dorsoanterior structures ( [20]; this report ) . Positional cloning reveals that hec encodes the Glutamate receptor interacting protein 2a ( Grip2a ) . We find that grip2a mRNA , like wnt8a mRNA and Sybu protein , is localized to the vegetal pole of the oocyte and early embryo , where during egg activation it is shifted off-center corresponding to the previously proposed teleost cortical rotation [8] . The Drosophila Grip homologue has recently been shown to potentiate Wnt signaling at the neuromuscular junction by interacting with the Frizzled receptor on the cytoplasmic side of endocytosing membrane vesicles [21] , [22] , suggesting a potential mechanism of action for zebrafish Grip2a in axis induction at the level of Frizzled receptor regulation . Unexpectedly , however , we find that hec mutants show defects in the alignment and bundling of microtubules at the vegetal cortex , which result in corresponding defects in the asymmetric movement of wnt8a mRNA and are sufficient to explain the observed axis induction defects . The short-range shift in vegetally localized factors such as grip2a mRNA also led us to re-examine the functional significance of the previously observed animally-directed cortical transport on axis induction . We find that , although short-range shifts in vegetal signals are affected in hec mutant embryos , these mutants do not exhibit a defect in the long-range , animally directed translocation of cortically injected dorsal beads that occurs in lateral regions of the yolk cortex . Furthermore , we show that , contrary to our expectations , such movements are not restricted to a single arc corresponding to the prospective dorsal region , but occur in multiple meridional arcs even in opposite regions of the embryo . Together , our results propose a role for hec function in the reorganization and bundling of microtubules at the vegetal cortex to mediate a symmetry-breaking event likely corresponding to the teleost cortical rotation . This asymmetry is subsequently amplified by a cortical animally-directed transport mechanism that is neither dependent on hec function nor restricted to the prospective dorsal axis .
Embryos from mothers homozygous for the mutant hec allele , for simplicity referred to here as hec mutant embryos , display a range of ventralized phenotypes [17] , [20] , [23] ( Figure 1A–E; Table 1 ) . In the most severe cases , 24-hour mutant embryos exhibit a severe radial ventralization and lack all dorsoanterior-derived structures ( V4 class , according to criteria in [24]; Figure 1A ) . More moderate phenotypes can also be observed , such as embryos that lack all anterior structures as well as the notochord ( a dorsally-derived structure ) , and display an expansion of posterior somites ( V3 class; Figure 1B ) . More weakly ventralized embryos are also observed that lack the anterior-most head structures and the notochord and exhibit expanded posterior somites ( V2 class; Figure 1C ) , and embryos with reduced eyes and some notochord defects ( class V1; Figure 1D ) . In many clutches , a fraction of embryos from mutant mothers are indistinguishable from wild-type embryos ( Figure 1E ) . Such variability in phenotypes can be observed in maternal-effect mutants ( some examples can be found in [17] , [25] , [26] ) , in particular those exhibiting axis induction defects [5] , [15] , [27] , possibly by inherent variation in the maternal composition of individual eggs coupled to gene or pathway redundancy ( see also Discussion ) . In mutant clutches with a weak expression of the phenotype , a small and variable fraction of embryos exhibits axis duplication phenotypes ( Figure 1F ) instead of axis formation defects ( see Discussion ) . The hect2800 allele was originally isolated in an early-pressure-based screen for recessive maternal-effect mutations [14] , [17] and its effects described previously [20] , [23] . Two additional alleles , hecp08ajug and hecp06ucal were identified in another maternal-effect mutant screen , based on a four-generation scheme [16] , [18] . Using DNA markers linked to the hect2800 allele [20] , we determined that the hecp08ajug and hecp06ucal alleles were linked to the same SSLP markers , z59658 and z24511 on chromosome 8 . In addition , we carried out pair-wise crosses between individuals carrying the three mutant alleles to test for non-complementation . All crosses resulted in females that exhibited the hec mutant phenotype in their offspring in the expected proportions , i . e . Mendelian for F1 females ( approximately 50% in crosses between homozygous males and heterozygous females of all allelic combinations ) and maternal-dependent ( near 100% ) for F2 embryos ( Table S1 ) , indicating that all three mutations are part of the same complementation group . A comparison of the phenotypes for the three hec alleles suggests that they fall within an allelic series . Embryos from mutant females carrying each of the three alleles were classified and scored at 24 hours post-fertilization ( hpf; Table 1 ) . The hecp06ucal allele shows the strongest average phenotype among the three alleles , where most embryos ( 76 . 3% from 3-month females ) exhibit the strongest ( V4 ) phenotype , while hect2800 mutants exhibit intermediate phenotype ( 46 . 7% V4 class ) and hecp08ajug mutants show the weakest phenotype ( 10 . 9% V4 class ) . Double-axis embryos are observed only in offspring derived from females mutant for the two weaker alleles hecp08ajug and hect2800 , and only in clutches with weak penetrance and expressivity ( Table 1 and data not shown ) . In the case of ichabod and tokkaebi mutations [5] , [15] axis induction defects have been reported to vary with maternal age , with younger females exhibiting stronger phenotypes . We tested embryos from 3 month old and 12 month old mutant females and find a similar trend for the effects of the three hec alleles on axis induction ( Figure S1 ) . Previous studies have shown that the ventralized phenotype of hec mutant embryos is associated with a reduction in the embryonic shield and changes in patterns of gene expression in dorsal- and ventral-specific genes in the late blastula embryo . Expression of gene markers of the various germ layers , such as bmp2 and gata2 ( ectoderm ) , no tail ( mesendodermal precursors ) , and foxa2 ( endoderm ) was unaffected , other than regional differences due to predicted changes in dorsoventral specification [20] . As expected , embryos from females mutant for the newly isolated hec alleles , hecp06ucal and hecp08ajug , similar to the hect2800 allele [20] , exhibit a reduction in the shield region corresponding to the dorsal organizer at the incipient dorsal region ( Figure 1G–J ) , as well as a reduction in dorsal-specific gene expression ( goosecoid , Figure 1K–N; chordin , Figure 1O–R ) and a concomitant expansion of a ventrally-expressed gene ( eve 1 ) ( Figure 1S–V ) . Together , phenotypic , linkage and complementation analysis indicate that these three mutations are alleles of the same gene . To better understand the function of the hec gene , we determined its molecular identity using a positional cloning approach . We initially identified linkage of the hec locus between SSLP markers z59658 and z24511 on chromosome 8 through mapping of the hect2800 allele [20] . Homozygous mutant males were crossed to heterozygous females to generate large numbers of fish for fine mapping . Fine mapping analysis of 1762 meioses with newly identified RFLP markers further narrowed the critical region containing hec to a genomic region between gpd1a-1 and the zC150E8y RFLP in the Ensembl database , corresponding to an interval of 383 Kb ( Figure 2A ) . Five overlapping BAC clones were identified and aligned as a contig covering the whole critical region ( Figure 2B ) . Within this critical region , there are 11 predicted genes according to the Ensembl database of the zebrafish genome and the GENSCAN program . Sequencing of cDNA products from wild-type and mutant alleles revealed the presence of mutations in all three hec alleles in the gene glutamate receptor interacting protein 2a ( grip2a , NP_001116760 . 1 ) , one of two grip2 genes in the zebrafish genome [28] . The zebrafish grip2a gene has 16 exons , which produces a 3 , 124 bp transcript and a 736 amino acid protein ( Figure 2C ) . The three mutant alleles result in truncated forms of the Grip2a protein: hecp06ucal allele has a C-A transversion in codon 118 in exon 4 , hect2800 has a C-T transversion in codon 414 in exon 10 , and hecp08ajug has a C-T transversion in codon 499 in exon 12 , and these three mutations generate premature nonsense ( stop ) codons ( Figure 2D ) . A search in the Conserved Domain Database ( CDD ) in NCBI [29] indicates that Grip2a protein has four PDZ domains . The premature stop-codons for these different alleles delete the most C-terminal PDZ domain in the hect2800 and hecp08ajug alleles , and all four PDZ domains in the hecp06ucal allele ( Figure 2E ) . The retention of PDZ domains and size of the predicted truncated proteins roughly correlates with the observed phenotypic strength in the various mutant allele backgrounds ( Table 1 , Figure S1 ) , although we have not determined expression levels for the mutant proteins to confirm their relative activities . The identification of mutations in these three independently isolated alleles indicates that hec encodes Grip2a . This is further substantiated by the localization of grip2a mRNA in the region of the embryo affected by the hec mutation ( see below ) . Using BLAST searches on Ensembl and NCBI genome databases , homologous grip1 and grip2 genes were found for all vertebrate species , such as fish , amphibians , birds , and mammals . In invertebrate lineages , a distantly related Grip gene was identified only in Drosophila . Grip1 and Grip2 protein sequences among eight representative species were used to construct a phylogenetic tree using ClustalW ( Figure S2 ) . grip2 occurs as a single copy in amphibians , birds and mammals but is duplicated in the zebrafish and other fish species such as fugu and medaka , likely a consequence of an extra round of whole genome duplication in the ray-finned fish lineage [30] , [31] . Drosophila and all vertebrate Grip1 and Grip2 proteins contain PDZ domains but zebrafish Hec/Grip2a and Fugu Grip2b contain 4 predicted PDZ domains instead of the 7 PDZ domains predicted in other members of this family ( Figure S2 and not shown ) . Quantitative RT-PCR analysis of mRNA from wild-type embryos spanning early development indicates highest levels of grip2a mRNA in the 1-cell stage embryo , gradually declining to negligible levels at 50% epiboly ( 5 . 25 hpf ) and thereafter ( Figure S3 ) . In adults , expression can be detected in wild-type females and isolated ovaries , but not in males or female carcasses where the ovaries have been removed ( Figure S3 ) . Thus , at our level of analysis , hec/grip2a is specifically expressed in ovaries as a maternal-specific transcript , which is consistent with the strict maternal effect observed in females homozygous for the three hec mutant alleles . We examined the spatial expression pattern of hec/grip2a at various developmental stages during embryogenesis using whole mount in situ hybridization ( Figure 3 ) . grip2a mRNA is detected in the vegetal pole region of the yolk in early zygotes and cleavage-stage embryos ( Figure 3A–E ) . Similar to the case of wnt8a mRNA [7] and Sybu protein [6] , the grip2a mRNA localization domain is not precisely aligned with the vegetal pole in activated eggs or early embryos . Instead , grip2a mRNA is consistently located slightly off-center in the post-activation stages examined , from the early 1-cell stage embryo 10 minutes post-fertilization ( mpf ) until late cleavage stages ( Figure 3A–D ) . This off-center shift is not observed in manually extruded mature , inactive eggs , where the grip2a mRNA localization domain is instead located at the vegetal pole in a radially symmetric manner ( data not shown ) . In early embryos the extent of asymmetry of the grip2a mRNA localization domain , appears similar throughout the cleavage and blastula stages until mRNA levels become markedly reduced in the late blastula embryo ( sphere stage; 4 hpf; Figure 3E ) . Localized grip2a mRNA can no longer be detected starting at the onset of epiboly ( 30% epiboly; 4 . 66 hpf; not shown ) . In contrast to its Xenopus homologue [32] , zebrafish grip2a mRNA does not localize to the zebrafish germ plasm ( Figure 3C and data not shown ) , present at the furrows corresponding to the first and second blastomeric divisions [33]–[37] , nor does it become incorporated into the primordial germ cells ( Figure 3D , E and data not shown ) , which form four cell clusters during the late cleavage stages ( [33] , [35] , [38]; see below ) . To further confirm the slight , off-center shift in the grip2a mRNA localization domain , and to determine whether this shift , like that of Sybu protein and wnt8a mRNA , depends on an intact microtubule network , we tested the effect of early nocodazole treatment on the grip2a mRNA localization pattern in wild-type embryos . Embryos were treated at 5 mpf and fixed at 30 and 40 mpf for in situ hybridization . For both time points , control ( solvent-treated ) embryos show an off-center shift in the domain of grip2a mRNA localization so that it is located within an arc at 0–20° from the true vegetal pole of the embryo , while nocodazole-treated embryos do not exhibit a discernable shift in mRNA localization ( Figure 3H , 3I; Figure S4 ) . Thus , grip2a mRNA is located at the vegetal pole of the embryo in mature oocytes , but upon egg activation ( typically coupled to fertilization ) this mRNA exhibits a short-range , off-center translocation within the vegetal region of the embryo . Once in an off-center position grip2a mRNA remains static until its degradation at late cleavage stages . Whole mount in situ hybridization of embryos from mutant female mothers homozygous for each of the three hec alleles show significantly reduced levels of localized grip2a mRNA during the early cleavage stages , ranging from reduced to nearly undetectable levels ( Figure 3G and data not shown ) . In those embryos where grip2a mRNA localization can be discerned , the domain of localization appears centered at the vegetal pole , consistent with the absence of an off-center shift for vegetally localized products in hec mutants ( see below ) . Quantitation of total grip2a mRNA levels in mutant embryos at the 2-cell stage indicates that , for all alleles , grip2a mRNA abundance is drastically reduced to approximately 15–25% that in wild-type embryos ( Figure S3 ) . It is possible that hec/grip2a function is required for the localization of its own mRNA , which when not localized is unstable . Alternatively , the reduction in apparent grip2a mRNA localization in hec mutant embryos might be a consequence of a decrease in grip2a mRNA abundance in these embryos , possibly by non-sense mediated mRNA decay as has been proposed for other maternal transcripts [39] , [40] . To visualize how the pattern of grip2a mRNA localization is established during oogenesis , we carried out in situ hybridization analysis of wild-type oocytes at various stages using a grip2a antisense probe ( Figure 4 ) . In early stage I oocytes , grip2a transcripts appear sharply localized to a compact , spherical region at one region of the oocyte ( Figure 4A , 4D ) . By late stage I , grip2a mRNA acquires a more spread , subcortical localization pattern still centered in one region of the oocyte ( Figure 4A , 4E ) . This cortical localization is maintained until the end of oogenesis ( Figure 4A , 4F and data not shown ) . The localization pattern of grip2a mRNA in stage I oocytes is reminiscent of the zebrafish Balbiani body , a conserved aggregate of organelles present in animal oocytes shown to anchor subcellularly localized oocyte mRNAs [41]–[43] . We therefore tested whether grip2a mRNA localization is dependent on Balbiani body formation , using zebrafish mutants that affect this structure . Oocytes mutant for the gene bucky ball lack the Balbiani body [39] , [43] , and we found that bucky ball mutant oocytes lack grip2a mRNA subcellular localization during oogenesis ( Figure 4B , 4G , 4H ) . Moreover , oocytes mutant for the cytoskeletal linker protein magellan ( macf1 ) , which exhibit an enlarged Balbiani body with an abnormal location [40] , exhibit grip2a mRNA mislocalization ( Figure 4C , 4I , 4J ) similar to other Balbiani-localized transcripts . These results indicate that grip2a mRNA becomes localized to the vegetal cortex during oogenesis by a Balbiani body-dependent mechanism . The rescue of hec mutant embryos by overexpression of Wnt pathway components has suggested that hec likely activates signaling at an upstream step of the pathway [20] . Given that an early event in the pathway leading to Wnt signaling activation is the reorganization of microtubules at the vegetal pole required for the transport of local determinants , we tested whether this reorganization is affected in hec mutant embryos ( Fig . 5; Figure S5 ) . Consistent with previous studies [2] , [8] , [27] , we find that microtubules at the vegetal cortex in wild-type appear as parallel tracks of bundled microtubules at 20 mpf ( Figure 5A ) . In hec mutant embryos , such parallel tracks of bundled microtubules are not observed ( Figure 5B ) . In these mutants , unbundled microtubules typically appear to radially emanate from one or more aster-like structures at the vegetal pole region ( Figure 5C–F ) . Exposure to the microtubule-stabilizing drug taxol [44] during the first cell cycle ( 5 to 35 mpf ) does not influence the degree of residual axis induction in hec mutants ( Figure S6 ) , suggesting that the observed defects may not be simple consequence of altered microtubule dynamics . Labeling of the F-actin cortical network in the vegetal cortex region shows a similar appearance in wild-type and mutant embryos ( Figure S7 ) , including the presence of F-actin rich protrusions as previously described [45] . Since vegetal cortical microtubules have been proposed to mediate the off-center shift of factors involved in axis induction that are initially localized to the vegetal pole , such as grip2a mRNA ( this report ) , wnt8a mRNA ( [7]; Figure 6A , 6C ) and Sybu protein ( [6]; Figure 6E , 6G ) , we tested whether these shifts were affected in hec mutant embryos . As noted above the grip2a mRNA localization domain in hec mutants , when still detectable , fails to undergo an off-center shift ( Figure 3G ) . In situ hybridization analysis to detect wnt8a mRNA shows this mRNA also fails to undergo a noticeable off-center shift in one-cell ( 30 mpf ) hec mutant embryos ( Figure 6B ) . At the 4-cell stage ( 60 mpf ) , wnt8a mRNA localization at the vegetal pole is significantly reduced or undetectable ( Figure 6D ) , although this defect is associated with an overall decrease in the relative expression of wnt8a mRNA ( Figure S8 ) . In the case of Sybu protein , wild-type embryos show localization centered at the vegetal pole until 20 mpf ( Figure 6E ) and an off-center shift of protein localization by 30 mpf ( Figure 6G ) , as previously reported [6] . In hec mutants , the Sybu protein localization domain can be initially detected centered at the vegetal pole of hec mutants ( Figure 6F ) . However , Sybu protein is no longer detectable by 30 mpf ( Figure 6H ) , precluding testing an effect on Sybu protein off-center movement . These data indicate that hec/grip2a is essential for the short-range , symmetry-breaking transport of vegetally-localized factors , and are consistent with hec function being essential for microtubule reorganization in this region . The reduction in vegetally localized wnt8a mRNA and Sybu protein in hec mutants contrasts with the perduring vegetal localization of these factors in embryos with a perturbed microtubule network ( [6] , [7]; Figure 3I , 6I ) . The short-range off-center shift observed in the case of Sybu protein and grip2a mRNA , which occurs within the confines of the vegetal region , contrasts with the long-range transport thought to be involved in transporting a putative dorsal signal to blastomeres at the animal region [2] . wnt8a mRNA has been observed to reach the base of the blastomeres by the 16-cell stage ( 1 . 5 hpf; [7] ) , although in our experiments this RNA exhibits a relatively static off-center shift throughout the first 60 mpf ( Figure 6A , 6C ) , similar to the short-range movement of Sybu protein and grip2a mRNA . The animally-directed translocation of a putative dorsal signal is thought to be reflected in the microtubule-dependent , animally-oriented movement of small ( 0 . 2 µm ) polystyrene fluorescent beads during the first several cell cycles . When injected into the vegetal region , these beads reach the base of the blastomeres at the animal region by traveling through cortical paths [2] . Using the transport of microinjected fluorescent beads as an assay for this long-range transport mechanism , we tested whether long-range vegetal-to-animal movement along the cortex might be affected in hec mutant embryos . As previously reported [2] , in wild-type embryos bead movement from the vegetal region is observed along a meridional arc along the cortex reaching the base of the blastomeres at the animal pole ( 41% ( n = 87 ) ; Figure 7A , A′ ) . In hec mutant embryos , beads appear to be transported to a similar extent as in wild-type , also reaching the base of the blastodisc ( Figure 7B , B′ ) and at a similar observed frequency ( 39% ( n = 51 ) ) . Thus , hec function does not appear to be required for long-range animally-directed transport along the lateral cortex . The apparently normal movement of animally-directed beads in hec mutants appears to conflict with the role of this gene in early microtubule reorganization but is consistent with the observed presence of multiple microtubule populations [2] , [8]: aligned bundles of short microtubules at the vegetal region , which we find to be dependent on hec function , and a more dispersed and randomly oriented network in more medial regions , which appears unaffected in hec mutants ( Figure 5G , 5H and data not shown ) . The direction of aligned short microtubule bundles at the vegetal pole has been shown to correlate with the dorsal axis but only occurs in a limited area within the vegetal half of the early embryo [8] . Given that directionality of a long-range movement might depend on an earlier oriented symmetry-breaking event , we wondered whether the long-range animally directed movement was specific to the putative dorsal region , or whether the entirety of the cortex was competent to support such transport . To test whether long-range animally-directed transport was specific to the prospective dorsal region of the embryo or was a general property of the lateral cortex , we carried out two slightly off-center fluorescent bead injections on opposite sides of wild-type embryos . Such doubly-injected embryos show animally-directed bead transport along meridional arches in opposite regions of the embryo ( Figure 7C , C′ ) , an observation inconsistent with only the prospective dorsal region mediating long range vegetal-to-animal cortical transport . Instead , our data suggest that the entirety of the cortex can mediate long-range animally-directed movement . The apparently normal long-range transport of beads in hec mutant embryos , in the presumed absence of an early short-range symmetry-breaking process , may reflect asymmetries in the location of the beads during injection , followed by the action of this hec-independent long-range transport mechanism . Our data suggest that the transport of dorsal determinants to the prospective site of dorsal induction depends on two sequential processes: ( i ) an initial short-range transport dependent on hec-mediated formation of short aligned microtubule bundles that results in determinant asymmetry at the vegetal region of the embryo , and ( ii ) a subsequent long-range transport through lateral cortical regions , which is independent of hec function and not specific to the prospective dorsal region . In Xenopus , the homologous gene grip2 ( previously referred to as grip2 . 1 ) , like zebrafish hec/grip2a , is expressed maternally and its mRNA is localized to the vegetal pole of the egg and early embryo [32] . Following this vegetal localization , Xenopus grip2 mRNA becomes incorporated into primordial germ cells ( PGCs ) where it plays a role in their migration and survival [32] . In contrast , although localized to the vegetal pole , zebrafish grip2a mRNA does not localize to the zebrafish germ plasm or PGCs ( Figure 3 and data not shown ) . We used whole mount in situ hybridization to test whether germ plasm localization or PGC development may be affected in hec mutant embryos ( Figure 8 ) . The localization patterns of dazl mRNA , a germ plasm component initially localized to the vegetal pole of the egg ( [36] , [46] , [47]; Figure 8A ) , is not affected in these mutants ( Figure 8D ) . During the first two cell cycles , dazl mRNA localizes normally to the furrows in hec mutants ( Figure 8B , 8E ) , as does vasa mRNA ( Figure 8C , 8F ) , an animal germ plasm component already present in the animal cortical region during egg activation [33] , [36] , [38] , [48] . During embryonic development , although PGC migration is abnormal in strong hec mutants due to their radially symmetric , ventroposteriorized morphology , the average number of induced PGCs ( as determined by cells expressing vasa in the 10 . 5 hpf embryo [33] , [38] , [48] ) is similar to that in wild-type embryos ( Figure 8G , 8H; Figure S9 ) . Thus , as opposed to the case of Xenopus grip2 , our observations do not support a role for zebrafish hec/grip2a in PGC development . Two other zebrafish grip-related genes , grip1 and grip2b , exhibit maternal expression as assayed by RT-PCR of RNA from 1–4 cell embryos but do not show localization to the vegetal pole or germ plasm at these early stages , or in PGCs in the 24-hour embryo ( [28]; our own data ) . Together , these data indicate that , in spite of similar mRNA localization patterns at the vegetal pole of the egg , grip homologues in Xenopus and zebrafish have divergent roles in PGC development and axis induction , respectively .
In both teleosts and amphibians , dorsal determinants are initially localized at the vegetal pole and then translocate to the prospective dorsal side in a microtubule-dependent manner to initiate the dorsal cell fate program ( reviewed in [1] , [49] , [50] ) . Genetic and molecular searches have identified factors localized to the vegetal cortex with a role in axis induction in the zebrafish , such as the kinesin-1 linker Syntabulin [5] , [6] and the wnt8a mRNA [7] . We show that the mRNA product for zebrafish hec/grip2a is another zebrafish maternal factor involved in axis induction , whose product is also localized to the vegetal pole of the oocyte and early embryo . Zebrafish hec/grip2a is expressed solely during oogenesis as a maternal transcript , which is consistent with the identification of multiple maternal-effect mutant alleles of this gene , all of which lack associated zygotic defects . In particular , fish homozygous for the hecp06ucal allele , predicted to lack all four PDZ domains present in the wild-type protein and therefore likely a null , exhibit a highly penetrant maternal-effect phenotype yet are themselves viable . These observations indicate that hec/grip2a has a dedicated function in early axis determination . While we cannot rule out that hec/grip2a is expressed in older embryos or adults at levels below the sensitivity of our assays , other related genes such as grip1 or grip2b are expressed at later stages of development ( [28]; our own data ) when they may provide essential zygotic functions . Mechanisms inducing microtubule bundling and alignment , essential for the establishment of the primary body axis , remain incompletely understood . In Xenopus , cortical rotation and microtubule reorganization are dependent on both kinesin and dynein motor activity [51] , [52] and other factors such as Trim36 , a ubiquitin ligase whose mRNA is localized to the vegetal egg cortex [53] , Dead end , an RNA binding factor needed for trim36 mRNA vegetal cortex localization [54] and the lipid droplet component Perilipin 2 , whose mRNA is also localized to the Xenopus vegetal pole [55] , [56] . We find that zebrafish hec/grip2a function is required for the reorganization of vegetal cortex microtubules into bundles normally directed towards the prospective dorsal axis [2] , [8] . The lack of microtubule network alignment in hec mutants and associated defects in the transport of putative dorsal determinants such as wnt8a mRNA likely result in the axis induction defects observed in these mutants . Grip was originally identified as a factor interacting with AMPA-type glutamate receptors [57] and its multiple PDZ domains are thought to facilitate protein-protein interactions within large macromolecular complexes , including the surface presentation and trafficking of transmembrane proteins [58]–[60] . grip genes have been implicated in epithelial development in both mouse and zebrafish embryos ( [28] , [61] , reviewed in [62] ) . Other studies have implicated Drosophila Grip as a mediator of Wnt ligand activity in the postsynaptic terminal of the neuromuscular junction [21] , [22] . Our findings suggest parallels between subcellular transport at the vegetal pole of the zebrafish zygote and transport of neurotransmitter receptors in neurons . In the zebrafish zygote , transport of wnt8a mRNA depends on microtubules and occurs concomitantly with the movement of the kinesin adaptor Syntabulin [6] , [7] . Similarly , in dendrites Glutamate receptors associated with mammalian GRIP1 are driven by kinesin along microtubules [63] , and Syntabulin has been shown to be required for axonal transport [64]–[66] . In neurons , glutamate receptors and GRIP associate with membrane vesicles [67] , [68] . Although membrane vesicles have not been reported to be associated with dorsal determinants in zebrafish , studies in Xenopus implicate membrane vesicles in the transport of dorsal determinants [69]–[72] . Further studies will be required to determine mechanisms driving the reorganization of vegetal cortex microtubules , the precise role of Grip2a in this process and whether Grip factors have an analogous cytoskeletal restructuring function in other systems . Our previous studies have shown that manipulations to activate Wnt signaling , including the overexpression of wnt8 mRNA , can rescue the hec mutant phenotype , which suggests that hec acts in an upstream event required for Wnt signaling activation during axis induction [20] . Our identification of a role for Grip2a on cytoskeletal events needed for the relocation of dorsal determinants is consistent with such an upstream role . However , our studies do not rule out a more direct role for Grip2a as a regulator of Wnt pathway components . We note that the off-center shift of vegetally localized grip2a mRNA upon egg activation could provide an asymmetric source of Grip2a protein to influence Wnt signaling activity at the prospective dorsal region . Drosophila Grip is known to interact with the Wnt receptor Frizzled-2 , promoting the trafficking of a Frizzled-2 C-terminal fragment to the nucleus to activate target genes [21] , [22] , [73] , [74] , and it is possible that some of these interactions are conserved in the zebrafish embryo . It is also possible that Grip2a regulates non-canonical Wnt signaling , such as Wnt/calcium signaling , which in turn influences axis induction . hec mutants exhibit an increased frequency of intracellular calcium transients in blastula stage embryos ( 2 . 00–3 . 33 hpf; [20] ) , and the resulting intracellular calcium increase has been proposed to attenuate Wnt/β-catenin signaling pathway activity [75]–[79] . Further studies will be necessary to determine whether hec/grip2a , in addition to functioning in cytoskeletal organization in the early zebrafish embryo , has a direct role in the regulation of Wnt/β-catenin signaling and axis induction . Multiple studies in Xenopus have indicated the formation of long tracks of cortical microtubules associated with the cortical rotation [80]–[82] . While in this organism the cortical rotation involves the concerted movement of the cortex along a distance corresponding to a 30° arc , microtubule tracks and so-called fast transport of subcellular components , such as membrane organelles and specific factors , encompass a longer distance corresponding to a 60–90° arc ( reviewed in [1] ) . Thus , in Xenopus both the cortical rotation and fast transport may participate in the relocation of dorsal determinants . In zebrafish embryos , a cortical rotation-like process results in the displacement of granules along a 20° arc from the vegetal pole of the embryo [8] , with mediolateral regions of the cortex exhibiting a loose meshwork of microtubules independent of dorsoventral position [2] , [8] . The restriction of microtubule bundling and alignment to the vegetal region of the zebrafish embryo raises the question of how long-range transport of dorsal determinants to the animal pole is achieved . We found that injected beads reach the base of blastomeres in hec mutant embryos , which lack a cortical rotation-like movement , with a frequency similar to that observed in wild-type embryos . This suggests that transport of beads along the mediolateral cortex is independent of hec/grip2a function and aligned vegetal microtubules . Our finding that beads are able to move animally along opposite sides in multiply injected embryos , further suggests that the entirety of the mediolateral cortex , not just the prospective dorsal region , is competent for long-range vegetal-to-animal transport . Previous studies have identified animally-directed transport movement of cytoplasmic particles along cortical “meridional” streamers , hypothesized to mediate transport of vegetally-injected fluorescent beads [83] . This meridional transport along the mediolateral cortex may depend on various possible structures or processes , such as perpendicular bundles aligned along the animal-vegetal axis in deep regions of the cortex [8] , incipient yolk cytoplasmic layer ( YCL ) microtubules emanating from marginal blastomeres into the yolk [84] , an emerging property of the loose network of cortical microtubules found in this region of the embryo [2] , [8] , or other insofar unidentified cytoskeletal networks . Further analysis of the dynamic aspects of cytoskeletal networks in this region will be required to understand the mechanistic basis of this meridional transport system and its role in axis induction . In addition , our results do not exclude the possibility that diffusion of a translated protein such as Wnt8a may also contribute to long-range transport , as previously suggested [8] . Together , these data indicate that the transport of the putative dorsal determinant in the early zebrafish embryo involves at least two separate mechanisms: a short-range transport dependent on hec/grip2a function and aligned microtubule bundles , and a subsequent long range transport relying on the mediolateral cytoskeletal network ( Figure 9A ) . We hypothesize that the former generates an off-center , symmetry-breaking shift in initially symmetrically localized putative dorsal determinants , while the latter acts as a more general conduit that amplifies the early asymmetry . In hec mutants , the initial symmetry-breaking event is affected , so that even with a functional long-range animal transport mechanism , in most embryos an insufficient amount of dorsal determinants reaches the blastomeres at the animal pole ( Figure 9B ) . A dual mechanism of dorsal determinant transport may also explain how a fraction of embryos from females homozygous for the presumptive null allele , hecp06ucal , which lacks all 4 conserved PDZ domains , can develop a normal dorsal axis . In such embryos small fluctuations may occur in the position of the vegetally localized dorsal determinant , which could be amplified by the mediolateral transport system that is unaffected in these mutants . The presence of an embryo-wide pathway directing long-range transport towards the animal pole may also explain the appearance of double-axis embryos observed only in the weakest hec mutant clutches . In these cases , an aberrantly organized vegetal microtubule network may result in the off-center vegetal shift of dorsal determinants in more than one direction , leading to their animally-directed transport along multiple paths and resulting in supernumerary or expanded regions of axis induction . A wider distribution of dorsal determinants in weak hec mutants would result in their reduced concentration in animal regions and reduced Wnt pathway activation , consistent with the observed lack of anterior-most structures in the resulting double-axis embryos . This dual transport model suggests mechanisms by which small , directed changes , like the specific early short range symmetry-breaking event , can be amplified during early development by an embryo-wide mechanism to result in large differences in cell fate specification . Using BLAST searches on Ensembl and NCBI genome databases , homologous grip1 and grip2 genes were found for all vertebrate species . In Drosophila , there is an ancient single Grip gene . All other species containing recognizable grip genes are vertebrates . In amphibians , birds and mammals , there is one grip1 gene and one grip2 gene . On the other hand , zebrafish have one grip1 gene and two grip2 genes , likely due to an ancestral genome duplication in the teleost lineage [30] , [31] . Following published nomenclature [28] , we refer to the grip2 gene corresponding to hec , located in chromosome 8 , as grip2a , and the copy located in chromosome 22 , as grip2b . Phylogenetic analysis indicates that zebrafish grip2a is only present in fish species ( zebrafish , fugu and medaka ) . Drosophila Grip and all vertebrate Grip1 and Grip2 proteins ( including zebrafish Grip2b ) contain seven conserved PDZ domains , while zebrafish Grip2a contains only 4 PDZ domains . The amphibian grip2 homolog was identified as a novel vegetally-localized mRNA in Xenopus oocytes that is present in the mitochondrial cloud ( Balbiani body ) and subsequently in the germ plasm throughout oogenesis and early embryogenesis [32] , [85] , [86] . In Xenopus grip2 morphants , PGC numbers are significantly reduced , and PGCs are also present at ectopic locations along the anteroposterior axis in tailbud stage embryos [32] , [86] . However , unlike the case of Xenopus grip2 , zebrafish grip2a mRNA does not localize to the germ plasm or PGCs , and the number of PGCs is unaffected in strongly ventralized mutants ( i . e . radially symmetric ventralized embryos ) . These observations suggest that in zebrafish PGCs are determined independently of hec/grip2a function . Therefore , it appears that Xenopus grip2 and zebrafish grip2a mRNA , in spite of a similar localization at the vegetal pole of the oocyte , have distinct functions in early embryogenesis: Xenopus grip2 is involved in PGC development , while zebrafish grip2a appears to be devoted to dorsal axis formation . Despite this divergent function , the fact that zebrafish hec/grip2a and Xenopus grip2 are maternally expressed and localized to the vegetal pole during oogenesis , the site of localization of both germ plasm components and dorsal determinants in these two lineages , suggests that an ancestral grip gene may have functioned in both axis induction and PGC development . There is precedent for a relationship between these two processes , notably in Drosophila [87] but also in systems as basal as planaria [88] , [89] and annelids [90] . This relationship is also supported by the recent finding that the germ cell-specific factor Dead end is required for microtubule rearrangements and cortical rotation in Xenopus [54] . The presence of a single localization system involved in both the induction of the primary embryonic axis and the separation between germ cell and soma may constitute a simple mechanism for the species patterning and propagation . Studies of the role of grip genes in other organisms may shed light on the relationship between axis induction and PGC determination during evolution .
All animals were handled in strict accordance with good animal practice as defined by the relevant national and/or local animal welfare bodies , and all animal work was approved by the appropriate committee ( University of Wisconsin – Madison assurance number A3368-01 ) . Fish stocks were raised and maintained under standard conditions at 28 . 5° [91] . The hect2800 allele was originally isolated in an early-pressure-based screen for recessive maternal-effect mutations [14] , [17] , while the hecp06ucal and hecp08ajug alleles were found in a four-generation scheme based on natural crosses ( [16] , [18] , see also [92] ) . The hect2800 allele was induced in an AB/Tübingen hybrid background [14] , [17] , which was further hybridized with the WIK line during linkage mapping . The hecp08ajug and hecp06ucal alleles were induced in a Tübingen background , which was hybridized to an AB line in an F4 genetic screen coordinated with linkage mapping [16] , [18] . Homozygous mutant hec fish were identified by genotyping the flanking SSLP markers z59658 and z24511 , which are 1 . 2 cM apart on linkage group 8 and both of which were polymorphic for all three alleles . Mutant embryos were obtained by crossing homozygous hec females to AB males . Embryos from females homozygous mutant for the hecp06ucal allele were used unless otherwise specified . Wild-type control embryos were derived from either the AB line or heterozygous sibling females . Clutches were synchronized through 5-minute collections during natural spawning . Oocytes were collected from wild-type , bucky ballp106re [39] , [43] and magellanp6cv [40] mutants . Embryos were collected and developed in E3 embryonic medium [14] and were staged according to the age and morphological standards described in [93] . For complementation tests of hect2800 , hecp08ajug and hecp06ucal mutants , homozygous mutant males of one allele were crossed with heterozygous females of another allele to produce offspring , which were raised to adulthood . Female adult fish were crossed with wild-type males and phenotyped as wild-type or ventralized mutant by examining the resulting clutches at 24 hpf . Only those clutches producing more than 50 embryos were scored and non-complementation was indicated by the presence of ventralization phenotypes similar in expressivity and penetrance to those in clutches from mutant females from the original three mutant alleles on their own . Fish were anesthetized with MESAB ( 0 . 014% ) and the tail fin was clipped using a razor blade and placed into 100 ul DNA lysis buffer ( 10 mM Tris , pH 8 . 0; 10 mM EDTA , pH 8 . 0; 200 mM NaCl; 1% Triton X-100 ) containing 5 µg 10 mg/ml Proteinase K . Tissue lysates were incubated overnight at 55°C , and were incubated at 94°C for 10 minutes to inactivate Proteinase K . Lysates were diluted 1∶6 with water , and 2 . 5 µl of this genomic DNA diluted lysate was used per 10 µl PCR reaction . For a 10 µl PCR reaction , 2 µl of GoTaq green Buffer , 0 . 2 µl of dNTPs , 0 . 05 µl of GoTaq DNA polymerase ( Promega ) , 2 . 5 ul genomic DNA diluted lysate , and 1 µl each of 10 µM forward and reverse primers were used . For SSLP markers , PCR products were analyzed on a 2% high-resolution agarose gel right after the PCR reaction . For RFLP markers , PCR products were used for FastDigest Restriction Enzyme ( Fermentas ) digestion for 30 min , and then analyzed on a 1 . 5% regular agarose gel . Initial linkage was identified by bulk segregant analysis with SSLP markers . Once initial linkage of the mutation was obtained , genotypically identified homozygous mutant males were crossed to heterozygous females to generate large numbers of fish for fine mapping [92] . Chromosome walking was conducted by screening the CHORI-211 BAC library using marker z67047 and zC150E8z ( 0 recombination/1762 genomes ) . PCR-based screening of the primary pool and secondary pools identified 2 positive BAC clones . Individual BAC clones were ordered from BACPAC resources center ( http://bacpac . chori . org ) . To find the mutation in hec/grip2a , 5 fragments of hec/grip2a cDNA from mutant and wild-type 1-cell embryo cDNA were amplified by RT-PCR with 5 primer pairs , which cover the entire hec/grip2a coding region: 1 . 5′-ATGTCCTGCATCTTGCTTCCAGAG-3′ and 5′-CCTCAGTGGGAATCCCATTAATGG-3′ 2 . 5′-TGGAGTGTTACAAGTTGGCGACAG-3′ and 5′-TGAATGGCTTCGCTCAGAGGTTTG-3′ 3 . 5′-TTCATATCGGTGACCGAGTTTTGG-3′ and 5′-GACATTATTGTAGCCTCAAGCTCG-3′ 4 . 5′-GAGACCTGCGGTCAGTCAGAAATC-3′ and 5′-GTGCTCTGTGTTTCTCATTTGTGG-3′ 5 . 5′-AGGACACTTCCCAACAGTCTGCAC-3′ and 5′-ACCTGATCACTTCTAACCCAACAG-3′ All PCR products were cloned into pGEM-T easy vector and sequenced . For Phylogenetic analysis , homologous grip1 and grip2 genes were found using BLAST searches on Ensembl and NCBI genome databases . A phylogenetic tree was constructed and drawn using ClustalW in the MegAlign program from Lasergen . PDZ domains were identified using CD-search in the Conserved Domain Database ( CDD ) in NCBI [29] . Schematic diagram of the protein domain structures for each gene were drawn using DomainDraw [94] . Total RNA was isolated from whole embryos using TRIzol reagent ( Invitrogen ) . cDNA was synthesized using random primers ( Invitrogen ) and AMV Reverse Transcriptase ( Promega ) . RT-PCR reactions were performed with primer pairs derived from hec/grip2a and ef1α , using 30 cycles at an annealing temperature of 58°C ( in the semi-quantitative range ) . Absence of genomic contamination was verified by a negative control RT reaction without the Reverse Transcriptase . The following primers were used for the amplifications: hec/grip2a , 5′-GAGACCTGCGGTCAGTCAGAAATC-3′ and 5′-TATGAAGCTCTAGAGGCACTGACG-3′ , wnt8a , 5′-CGGAAAAATGGGTGGTCGTG-3′ and 5′-AGTCGACCAGCTTCGTTGTT-3′ , ef1α , 5′-ACCGGCCATCTGATCTACAA-3′ and 5′-CAATGGTGATACCACGCTCA-3′ . Quantitative ( q ) RT-PCR was performed on an iCycler machine ( Bio-Rad ) using iQ SYBR Green Supermix ( Bio-Rad ) . The thermal profile used for amplification is: 95°C for 3 min , 40 cycles of 94°C for 30 s , 58°C for 30 s and 72°C for 30 s . The relative mRNA level was quantified and normalized to ef1α . In situ hybridizations of embryos were carried out as described previously [95] . Probes for in situ included goosecoid [96] , chordin [97] , even skipped 1 [98] , vasa [33] , dazl [46] and wnt8a [7] , [99] . For the grip2a in situ probe , a fragment of grip2a cDNA was cloned into pGEM-T easy vector as described in the positional cloning section . Five different probes were tested against 5 different cDNA sequences , all of which showed the same expression pattern . Subsequently , all the expression data was acquired using one of the probes . Antisense digoxygenin probe was generated by linearizing and transcribing with SacII and SP6 RNA polymerase , while sense probe control was generated by linearization with SpeI and transcription with T7 RNA polymerase . Images were acquired with a Leica-FLIII microscope and a color camera ( Diagnostic Instruments Spot Insight ) . Ovaries were dissected from euthanized females and fixed overnight at 4°C in 4% paraformaldehyde . Fixed ovaries were then dehydrated in MeOH . Whole mount in situ hybridization of oocytes was performed as previously described [100] . Following staining , oocytes were embedded in JB-4 Plus Plastic resin and 7 micron sections were cut using a microtome . Stained sections were coated with Permount ( Fisher ) prior to addition of a coverslip . Antibody labeling of microtubules was as previously described [36] . Prior to the labeling procedure , embryos were dissected using fine dissecting forceps to generate two halves . To image cortical microtubules , bisections were carried out along an equatorial plane , the vegetal halves were labeled and mounted in 50% glycerol with DABCO reagent to prevent bleaching , with the vegetal cortex facing the coverslip . For imaging of mediolateral microtubules , bisections were carried out along a meridional plane and each mediolateral halves were labeled and mounted as above with the mediolateral cortex facing the coverslip . Images were acquired using an upright Zeiss LSM510 confocal microscope using an oil immersion 63× objective and collected as single 1 . 5 µm optical sections with a pinhole diameter of about 1 Airy unit , a low scan speed ( preset 6 ) and noise filtering through a 4-pass line mode average . The resulting images were analyzed with Fiji software . Antibody labeling of embryos to detect Sybu protein was carried in whole mount embryos using whole embryos as described previously [6] , and images were acquired using a Zeiss Axioplan2 fluorescent microscope and OpenLab software . Nocodazole treatment was carried out through exposure by 10 mpf of dechorionated embryos to a final concentration of 4 µg/ml nocodazole in E3 ( diluted from a 5 mg/ml solution in DMSO ) , followed by fixation at the indicated periods . Phalloidin labeling was carried out as in [36] , with the exception that dechorionated embryos were labeled whole and equatorially bisected prior to mounting with the vegetal pole facing the coverslip . Fluorescent beads injection experiments were carried as previously described [2] . A suspension of 0 . 2 mm fluorescent polystyrene beads ( 1 ul; Polysciences ) was diluted in 23 ml water and colored with trace amounts of phenol red ( 0 . 05% ) . The injection solution is microinjected into the embryos near the vegetal pole using a Phemtojet microinjector ( Eppendorf ) . Embryos had been injected at the 2-cell stage and imaged at 2 hours later . Embryos were mounted in methyl cellulose and imaged with an upright fluorescence microscope ( Zeiss , Axioplan II ) and a black and white digital camera ( Zeiss , Axiocam ) . | One of the earliest and most crucial events in animal development is the establishment of the embryonic dorsal axis . In amphibians and fish , this event depends on the transport of so-called “dorsal determinants” from one region of the egg , at the pole opposite from the site where the oocyte nucleus lies , towards the site of axis induction . There , the dorsal determinant activates the Wnt signaling pathway , which in turn triggers dorsal gene expression . Dorsal determinant transport is mediated by the reorganization of a cellular network composed of microtubules . We determine that hecate , a zebrafish gene active during egg formation that is essential for embryonic axis induction , is required for an early step in this microtubule reorganization . We find that hecate corresponds to glutamate receptor interacting protein 2a , which participates in other animal systems in Wnt-based pathways . We also show that the microtubule reorganization dependent on hecate results in a subtle symmetry-breaking event that subsequently becomes amplified by a more general transport process independent of hecate function . Our data reveal new links between glutamate receptor interacting protein 2a , Wnt signaling and axis induction , and highlights basic mechanisms by which small changes early in development translate into global changes in the embryo . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"genetics",
"biology",
"and",
"life",
"sciences",
"developmental",
"biology"
] | 2014 | Hecate/Grip2a Acts to Reorganize the Cytoskeleton in the Symmetry-Breaking Event of Embryonic Axis Induction |
Plants have evolved diverse cell types with distinct sizes , shapes , and functions . For example , most flowering plants contain specialized petal conical epidermal cells that are thought to attract pollinators and influence light capture and reflectance , but the molecular mechanisms controlling conical cell shaping remain unclear . Here , through a genetic screen in Arabidopsis thaliana , we demonstrated that loss-of-function mutations in ANGUSTIFOLIA ( AN ) , which encodes for a homolog of mammalian CtBP/BARs , displayed conical cells phenotype with wider tip angles , correlating with increased accumulation of reactive oxygen species ( ROS ) . We further showed that exogenously supplied ROS generated similar conical cell phenotypes as the an mutants . Moreover , reduced endogenous ROS levels resulted in deceased tip sharpening of conical cells . Furthermore , through enhancer screening , we demonstrated that mutations in katanin ( KTN1 ) enhanced conical cell phenotypes of the an-t1 mutants . Genetic analyses showed that AN acted in parallel with KTN1 to control conical cell shaping . Both increased or decreased ROS levels and mutations in AN suppressed microtubule organization into well-ordered circumferential arrays . We demonstrated that the AN-ROS pathway jointly functioned with KTN1 to modulate microtubule ordering , correlating with the tip sharpening of conical cells . Collectively , our findings revealed a mechanistic insight into ROS homeostasis regulation of microtubule organization and conical cell shaping .
Plants have evolved diverse epidermal cell types , that differ in size , morphology , and function , to adapt to life on land [1 , 2] . The Arabidopsis thaliana ( A . thaliana ) epidermis contain several well-known cell types , including interlocking jigsaw puzzle-shaped leaf pavement cells , tubular root hairs , and elongated hair-like trichomes . Using these three cell types as models to study cell shape , considerable progress has been made over the past two decades toward understanding how plant cells achieve their final morphologies [3 , 4] . Although similar molecular mechanisms controlling cell shape are shared between these cell types , certain mechanisms are used in a cell-type specific manner [2] . Despite this progress , the molecular mechanisms underlying plant cell shape formation remain to be further explored in specialized cell types . Conical-shaped cells are specialized epidermal cells decorated with radiate cuticular nanoridges and are usually found in the petal adaxial epidermis of most flowering plants [5–8] . Conical cells are thought to influence light capture and reflectance , temperature , and wettability , and are proposed to have a role in attracting bee pollinators by providing tactile cues [5 , 9–11] . However , the mechanisms underlying conical cell shaping have remained largely unknown . Most cell types in plants undergo anisotropic cell expansion that is driven by turgor pressure throughout the entire cell surface [12] , whereas the direction of expansion and the final cell shape largely depend on the patterning of cell wall architecture through the deposition and the orientation of cellulose microfibrils [13] . Cortical microtubules play an important role in plant cell shape determination by guiding cellulose microfibrils deposition in the cell wall [14–18] . It has been proposed that the co-alignment of circumferential microtubule arrays and cellulose microfibrils restrict cell expansion in girth and promote anisotropic cell expansion perpendicular to the orientation of cellulose microfibrils [14 , 19] . Genetic and live imaging studies have provided significant progress about the mechanisms underlying the organization of microtubule arrays [20] . We have recently established a confocal microscopy-based imaging approach to study the morphogenesis of conical cells and microtubule organization patterns in these cells [21] . Adaxial epidermal cells in the petal blades from flower development stage 8 are flat [21 , 22] , and undergo tip sharpening after stage 9 , which is correlated with the re-orientation of microtubules from disordered to well-ordered circumferential arrays at late developmental stages . Pharmacological evidence demonstrated that normal conical cell expansion requires an intact microtubule network [21] . Despite this progress , it remained elusive how cortical microtubule arrays are organized , and which upstream components are involved during conical cell development . The A . thaliana ANGUSTIFOLIA ( AN ) gene , encoding the plant homolog of the C-terminal binding protein/brefeldin A-ADP ribosylated substrate ( CtBP/BARS ) , has been shown to function in microtubule organization , vesicle budding from the Golgi , leaf pavement cell morphogenesis , and trichome branching [23–25] , but the underlying molecular mechanisms remain to be explored . In this study , through a reverse genetic screen , we showed that the AN knockout mutants an-t1 and an-t2 suppressed the tip sharpening of conical cells , correlating with the accumulation of significantly higher levels of reactive oxygen species ( ROS ) . ROS are known for their function as signaling molecules in plant organ growth and development and normal cellular processes [26–29] . Furthermore , mutations in KTN1 , encoding the p60 microtubule-severing protein [30–33] , caused enhanced conical cell phenotypes of an-t1 . AN acted in parallel with KTN1 to control conical cell tip sharpening . Together , our results suggest that the AN-ROS pathway jointly functions with KTN1 to modulate microtubule ordering and conical cell shaping .
To identify regulatory components involved in controlling conical cell expansion in the petal blade epidermis , we screened more than 500 mutant homozygous lines from the Arabidopsis Biological Resource Center ( ABRC ) for mutants with abnormal conical cell phenotypes [21] . From this screen , we identified one mutant line , SALK_026489C , previously named an-t1 [34] , with a T-DNA insertion causing a null allele of the gene AN ( AT1G01510 ) ( S1A and S1B Fig ) . The an-t1 mutant showed reduced tip sharpening of conical cells , exhibiting a phenotype of conical cells with wider tip angles ( Fig 1A–1E ) . As shown in Fig 1A , we quantified cone parameters ( cone angles and cone heights ) of conical cells from wild type with normal conical tips and mutant with swollen conical tips . Quantification data showed that conical cells of the an-t1 mutant had increased cone angles but no alternation in cone heights compared with those of the wild type ( Fig 1A–1E ) . To further confirm the role of AN in conical cell shaping , one additional T-DNA insertion null mutant of the AN gene , an-t2 ( WiscDslox329F05 ) ( S1A and S1B Fig ) , was obtained from the ABRC . The an-t2 mutant exhibited a conical cell phenotype similar to an-t1 ( Fig 1A–1E ) . Next , we performed the complementation experiment for an-t1 mutant line by introducing the coding sequence of the AN gene fused with the GFP tag under the control of AN’s promoter into an-t1 . We obtained more than 30 transgenic lines that could fully complement the previously reported an-t1 mutant phenotypes [23 , 24] , in terms of narrow cotyledons and leaves , less-lobed pavement cells , and reduced trichome branches . One transgenic line , an-t1 COM #1 , was selected for phenotypic analyses ( Figs 1A–1E and S1C–S1F ) , and was shown to express the AN gene fused with GFP ( S1G Fig ) . Expression of ANpro::GUS showed strong signals in petals throughout petal developmental stages ( S1H Fig ) . Furthermore , conical cells of the an-t1 mutant had similar geometric shape as the wild type at petal development stages 8–10 , but displayed increased tip angles at petal development stage 11 and beyond compared with the wild type ( S1I–S1K Fig ) . Taken together , these results demonstrated that AN plays a role in promoting the tip sharpening of conical cells at late developmental stages . Given that a previous report has shown that loss of AN function causes accumulation of high ROS levels in leaves by an unknown mechanism [34] , we investigated whether the wider-angled tips of conical cells of an mutants attributed to this abnormal ROS accumulation . Thus , we compared ROS levels between wild type and an mutants . ROS contain many molecules , with superoxide radical ( O2• – ) , the precursor for most other ROS , and hydrogen peroxide ( H2O2 ) , being the two major components [35] . We investigated O2• – and H2O2 distribution using nitroblue tetrazolium ( NBT ) [36] and 3 , 3’-diaminobenzidine ( DAB ) staining [37] , respectively . We found that both the an-t1 and an-t2 mutants accumulated higher levels of O2• – and H2O2 in their petals compared with those of the wild type ( S2A and S2B Fig ) . Furthermore , we examined the O2• – and H2O2 distribution at the cellular resolution . Using the fluorescent dye dihydroethidium ( DHE ) and 2' , 7'-dichlorofluorescein diacetate ( CM-H2DCFDA ) [38 , 39] to monitor O2• – and H2O2 production by laser scanning confocal microscopy , respectively , we found that both the an-t1 and an-t2 mutants’ mature conical cells had significantly higher levels of O2• – and H2O2 compared with the wild type ( Fig 1F–1I ) . Moreover , the an-t1 COM #1 complementation line could fully rescue the ROS levels of an-t1 ( Fig 1F–1I ) . Interestingly , qRT-PCR analysis revealed that the AN expression was down-regulated by H2O2 treatment ( S2C Fig ) . Consistently , western blot analysis demonstrated that AN protein was also down-regulated by H2O2 treatment ( S2D–S2F Fig ) , implying that H2O2 negatively regulates AN expression . Taken together , these results suggested that AN plays a role in conical cell shaping probably by negatively regulating ROS production . The above-mentioned results raised the possibility that ROS levels play crucial roles in normal conical cell expansion of petal adaxial epidermal cells over the course of development . Thus , we quantitatively examined the O2• – and H2O2 distribution in developing wild-type petals . Notably , we detected low levels of O2• – but not H2O2 in the petals at stage 8 ( S3 Fig ) , where petal adaxial epidermal cells display a flat shape . Remarkably , the levels of O2• – were slightly decreased in the petal adaxial epidermis at stages 9 and 10 ( S3 Fig ) , in which epidermal cells displayed roughly hemispherical morphologies , while H2O2 levels were almost not detected . At stage 10 and beyond , both the levels of O2• – and H2O2 became more and more apparent ( S3 Fig ) in the middle region of the petal blade’s adaxial epidermis , peaking at stage 14 , as the petal grew and conical cells became increasingly sharpened . This may indicate that O2• – plays crucial roles at both early and late stages during conical cell expansion and that H2O2 plays roles at later development stages . Notably , adaxial epidermal cells in top region of the petal claws displaying relatively flat but not conical shape , did not show increased ROS levels in the late phases of petal development ( S4 Fig ) . Therefore , these results demonstrated that ROS show a developmentally regulated accumulation pattern during conical cell expansion . We next compared ROS levels between wild type and an mutants at multiple developmental stages . Quantitative analysis showed that H2O2 levels in the an-t1 and an-t2 mutants were significantly higher at stage 10 and beyond compared with those of the wild type ( S5 Fig ) , this being consistent with the observation that AN affected conical cell development at late stages . Furthermore , the an-t1 COM #1 complementation line could fully rescue H2O2 levels ( S5 Fig ) . To investigate whether the AN-ROS pathway is a general mechanism important for cell shape in many different cell types , we next compared ROS levels in leaf pavement cells and trichomes between an mutants and the wild type . The an-t1 mutant accumulated higher levels of O2• – and H2O2 in their leaf pavement cells and trichomes compared with those of the wild type ( S6 Fig ) . These findings suggested that AN plays a general role in regulating ROS levels in diverse cell types . The observations that mature petals of an mutants displayed reduced tip sharpening of conical cells and increased ROS levels raise the hypothesis that ROS act downstream of AN to modulate conical cell expansion in petal adaxial epidermis . To test this hypothesis , we next sought to determine whether exogenously supplied ROS was sufficient to generate conical cells with wider tip angles similar to those of an mutants . We then used 50 mM and 100 mM H2O2 to treat flower buds of stage 7 from wild-type inflorescences , respectively . We repeated the H2O2 application four additional times , with 24 hours between each application , to generate long-term H2O2 effects . Indeed , analysis of conical cell phenotypes of mature petals after H2O2 treatment showed that , H2O2 application induced a wide-angled tip phenotype in conical cells in a H2O2 dose-dependent manner ( Fig 2A–2C ) . To further investigate the biological functions of ROS in conical cell shape control , we next reduced endogenous levels of O2• – and H2O2 , respectively . Wild-type flower buds of stage 7 were treated with 5 mM n-propyl gallate ( PG ) or 1 mM KI , respectively . To generate long-term drug effects , we repeated the treatments four additional times , with 24 hours between each application . The flower buds developed into stage 14 mature petals were used for phenotypic analyses of conical cell . We found that removal of either O2• – or H2O2 could result in conical cells with wider tip angles ( Figs 2D–2F and S7 ) . Surprisingly , removal of H2O2 by KI treatment caused accumulation of high levels of O2• – in mature petal epidermis ( S8 Fig ) . This result reflected that H2O2 negatively regulates O2• – biosynthesis during conical cell shaping through an unknown mechanism , consistent with a recent study showing that H2O2 negatively regulates O2• – levels in A . thaliana stem cells [39] . However , the conical cell phenotypes induced by eliminating endogenous ROS levels were less severe than those by exogenous H2O2 applications . Furthermore , consistently with the observation that H2O2 contents were hardly detected at early stages ( S3 Fig ) , treating wild-type flower buds with KI at stage 7 for one-time treatment led to no alternation of conical cell shape ( S7 Fig ) . In contrast , treating wild-type flower buds with KI at either stage 9 or stage 11 for one-time treatment resulted in wide-angled tips of conical cells compared with the mock treatment ( S7 Fig ) . Notably , treatment with PG at either stage 7 , stage 9 , or stage 11 for one-time treatment , respectively , could generate mature conical cells with wider tip angles ( S7 Fig ) . Taken together , these results suggested that endogenous ROS is required for normal conical cell expansion . To further confirm the role of ROS in modulating conical cell shape , we investigated whether endogenously increased ROS levels could also lead to the phenotype of wide-angled tips of conical cells . The plant Rho GTPase , which functions as a central molecular switch to control diverse cellular processes [40] , has been shown to positively regulate endogenous ROS levels through directly binding to NADPH oxidase [41] . Thus , we investigated conical cell phenotypes in transgenic lines overexpressing ROP2 . In agreement with the role of Rho GTPase in ROS production [41] , analyses of petals in transgenic Arabidopsis plants overexpressing ROP2 ( ROP2 OX ) or a constitutively active mutant ROP2 gene ( CA-ROP2 ) [42] under the control of the 35S promoter showed significantly increased ROS levels throughout the late phases of petal development ( S9 Fig ) . In agreement with the observation that wide-angled tips of conical cells are correlated with accumulation of high levels of ROS ( Fig 3A–3D ) , mature petals from both ROP2 OX and CA-ROP2 plants displayed swollen conical cells , with higher levels of both O2• – and H2O2 ( Fig 3E–3H ) . However , this phenotype was more severe than the an-t1 and an-t2 mutants . This may reflect that ROP GTPase signaling acts through multiple downstream components [40] . Given that both an mutants and CA-ROP2 plants exhibited a similar conical cell phenotype , correlating with accumulation of high ROS levels , we therefore explored the relationship between AN and ROP2 . Thus , we tested whether AN functions to affect ROS production by either directly or indirectly activating ROP2 . We then examined ROP2 activity in the inflorescences of wild type and an mutant lines . A comparison of the activation status of ROP2 in the wild type and the mutants an-t1 and an-t2 showed no significant differences ( S10 Fig ) , whereas the spk1-4 mutant with a mutation in SPIKE1 as a control had reduced ROP2 activity as described previously ( S10 Fig ) [43] . This result suggested that AN may not act upstream of ROP2 in regulating ROS production . To investigate the molecular mechanism by which AN regulates ROS levels , we identified potential interacting proteins of GFP-AN using a GFP-trap-immunoprecipitation based approach coupled with mass spectrometry ( MS ) -based protein identification . Proteins were isolated from whole-inflorescence tissues , excluding mature flowers and developing siliques , of a transgenic line expressing AN-GFP under the 35S promoter to pull-down GFP-AN protein complexes using GFP-binding agarose beads . Protein samples pulled-down from a transgenic line expressing GFP alone under the 35S promoter were used as a control . Co-immunoprecipitation ( Co-IP ) -MS analyses were conducted from three independent harvests of inflorescence tissues from both p35S:: GFP-AN line and p35S::GFP line to compose a biological triplicate . Gel analysis of the pull-down samples demonstrated a high specificity and efficiency for GFP-AN protein enrichment ( S11 Fig ) . Pulled-down protein samples were analyzed by MS , and stringent data analyses identified proteins that are exclusive in all three GFP-AN replicates but not present in any of the control pull-downs . To identify proteins present in GFP-AN complexes that can be relevant to a role in ROS production , we chose targets already annotated with gene ontology ( GO ) biological processes related to oxidation-reduction process and oxidoreductase . Notably , among the identified AN-interacting proteins ( S1 Table ) , we obtained 49 proteins directly responsible for ROS homeostasis ( S1 Table ) . These ROS-related proteins include CATALASE 2 ( CAT2 ) and CATALASE 3 ( CAT3 ) , which can catalyzes the breakdown of H2O2 [44] . In addition , several NAD ( P ) superfamily proteins involved in the oxidation-reduction process were identified as putative AN interactors ( S1 Table ) . To confirm the role of CAT2 in regulating conical cell shape , we identified one mutant line cat2-1 ( SALK_076998 ) from the ABRC stock center , with a T-DNA insertion causing a null allele of the gene CAT2 ( AT4G35090 ) ( S12A and S12B Fig ) . The cat2-1 mutant exhibited a phenotype of conical cells with wider tip angles compared with the wild type ( S12C–S12F Fig ) . We next sought to investigate whether AN and CAT2/CAT3 could physically interact . We used co-IP experiments to detect their interactions in vivo . We generated transgenic lines expressing CAT2-GFP and CAT3-GFP , respectively , and used GFP-Trap agarose beads to pull-down protein samples for the co-IP analyses . However , in our experimental conditions , the western-blotting results showed that AN may not physically interact with CAT2 and CAT3 in planta ( S13 Fig ) . Although the direct interactions between AN and ROS-related proteins need to be further determined , the list of potential AN interactors may suggest a regulation of ROS production by AN . To identify genetic components that function together with AN in controlling conical cell expansion , we mutagenized an-t1 seeds with ethyl methanesulfonate ( EMS ) and conducted a genetic screen for mutants with enhanced conical cell expansion defects of an-t1 . From this screen , we identified two enhancer mutants . Backcrosses to an-t1 and subsequent genetic analyses revealed that these two mutants were allelic and both harbored a recessive mutation . Sequencing of the KTN1 genomic DNA in these two mutants revealed a G-to-A mutation , resulting in amino acid alterations in the conserved ATPase domain of the KTN1 genes ( Fig 4A ) . Thus , these two mutants were named ktn1-7and ktn1-8 , respectively . Notably , these mutants displayed short roots , compact rosette leaves , and dwarf plants ( S14A–S14C Fig ) , reminiscent of the ktn1-4 mutant phenotypes [45] . The ktn1-8 allele had a weaker phenotype compared with those of the ktn1-4 allele ( S14A–S14C Fig ) , suggesting that it is a novel weak allele . A transgenic line expressing a KTN1pro::GUS construct showed strong GUS signals in petals throughout petal development stages ( Fig 4B ) , consistent with the role of KTN1 in petal conical cell shaping . Conical cells of the ktn1-7 and ktn1-8 single mutants had wider tip angles than the wild type , but displayed similar phenotypes to an-t1 ( Figs 4C–4F and S14D–S14G ) . Notably , both the an-t1 ktn1-7 and an-t1 ktn1-8 mutants had dramatically enhanced cell expansion defects , exhibiting extremely swollen petal adaxial epidermal cells , with a hemispherical shape instead of a conical shape , compared with the wild type and even the an-t1 mutant ( Figs 4C–4F and S14D–S14G ) . Taken together , these findings suggest that AN and KTN1 have a genetic interaction , and may act in parallel and functionally related processes during epidermal cell shaping . We next sought to investigate whether AN and KTN1 could physically interact . We used co-immunoprecipitation experiments to detect their interactions in vivo by transiently coexpressing 35S::GFP-AN with 35S::KTN1-Myc in Nicotiana benthamiana leaves . Transiently coexpressed 35S::GFP and 35S::KTN1-Myc were used as negative controls . In our experimental conditions , the result showed that KTN1-Myc was not detected in either the immunoprecipitated AN-GFP complex or the GFP complex ( S14H Fig ) , indicating that AN may not physically interact with KTN1 in planta . Consistently , we did not detect the KTN1 protein from our Co-IP MS analyses in the AN-GFP line . Recent reports have suggested that ROS influence cytoskeleton dynamics [46] , and that H2O2 directly activates the MAPK cascade to modulate the activities of MAP65 proteins , consequently affecting microtubule orientation [46–49] , although the underlying molecular mechanisms remain to be further explored . We next tested whether ROS are essential for cortical microtubule organization during conical cell shaping . Firstly , we investigated whether exogenously supplied H2O2 could lead to alterations in microtubule orientation using a microtubule reporter line expressing GFP-tagged α-tubulin 6 ( GFP-TUA6 ) [50] . We used 100 mM H2O2 to treat flower buds of stage 7 from inflorescences of the GFP-TUA6 reporter line , and the same treatment was repeated four additional times , with 24 hours between each application , to generate long-term H2O2 effects . After treatments , we visualized microtubule arrays from the top view of the non-folded mature petals in confocal Z sections , allowing for the top view of conical cells . We quantified the microtubule alignment with "OrientationJ" , a ImageJ plug-in , used for calculating the directional coherency coefficient of the fibers [51] . A coherency coefficient close to "1" represents a strongly coherent orientation of the microtubules . We found that adaxial epidermal cells without H2O2 treatments had well-ordered circumferential microtubule arrays aligned perpendicular to the axis of conical outgrowth ( Fig 5A and 5B ) , consistently with our previous report [21] . In contrast , adaxial epidermal cells of mature petals with H2O2 treatments displayed randomly oriented microtubules with reduced coherency ( Fig 5A and 5B ) . This result suggested that high levels of ROS accumulation inhibited microtubule ordering during conical cell shaping . Consistently with these findings , we observed randomly oriented microtubules in the mature conical cells of the CA-ROP2 line ( S15A and S15B Fig ) , which was shown to accumulate higher ROS levels . We next investigated the effects of eliminating endogenous ROS on microtubule organization . Notably , we found that eliminating either O2• – or H2O2 led to mature conical cells with disordered microtubule arrays with reduced coherency ( S15C and S15D Fig ) . Thus , these results suggested that ROS homeostasis mediated microtubule orientation into well-ordered circumferential arrays in petal conical cells , although the underlying mechanism remains to be further explored . To test whether ROS play a general role in regulating cell morphogenesis and microtubule organization in different cell types , we next investigated the effects on microtubule arrangements in both cotyledon pavement cells and leaf trichomes after exogenous H2O2 application . Notably , analysis of cotyledon pavement cell phenotypes after H2O2 treatment showed that , H2O2 application induced a phenotype of reduced lobe length in pavement cells , showing severe defects of interdigitated growth in a H2O2 dose-dependent manner ( S16A–S16D Fig ) . It has been reported that microtubule arrays play crucial roles in leaf pavement cell shaping [52 , 53] . Consistently , H2O2 application induced increased alignment of microtubules in pavement cells in a H2O2 dose-dependent manner ( S16E and S16F Fig ) . Exogenous H2O2 treatment leaded to generate smaller leaves with shorter trichomes but no alternation of trichome branching ( S17A–S17E Fig ) . Furthermore , H2O2 treatment resulted in increased alignment of microtubules in trichomes ( S17F and S17G Fig ) , which was similar to the result observed in leaf pavement cells . The observations that exogenous H2O2 treatment resulted in disordered microtubule arrays in petal conical cells and increased microtubule ordering in both pavement cells and trichomes , respectively , suggested that ROS may play diverse roles in modulating microtubule organizations in different cell types , although the underlying molecular mechanisms need to be further investigated . Based on the above-mentioned results showing that ROS played a role in mediating microtubule ordering , AN inhibited ROS production during conical cell development , and that AN and KTN1 acted in parallel pathways to modulate conical cell expansion , we hypothesized that AN acts through ROS to modulate microtubule orientation and that the AN-ROS and KTN1 pathways converge at a node to affect microtubule ordering during conical cell expansion . To test this hypothesis , we first compared microtubule organization patterns of the an-t1 mutant with the wild type . Microtubule arrays in wild-type cells became increasingly ordered over the course of conical cell development and displayed well-ordered circumferential arrays at stage 14 of petal development ( Fig 5C–5D ) , consistently with our previous report [21] . In contrast , an-t1 mutant conical cells that were shown to accumulate high ROS levels had randomly oriented microtubule arrays with reduced coherency throughout early and late developmental stages ( Fig 5C–5D ) , consistently with the observation that high levels of ROS accumulation caused wide-angled tips of conical cells with disordered microtubule arrays . Furthermore , in agreement with the previous report [21] , the ktn1-4 mutant conical cells had randomly oriented microtubule arrays , similar to those observed in the an-t1 conical cells ( Fig 5C–5D ) . Despite the conical cells of the an-t1 ktn1-4 double mutant displayed more severe defects than the single mutants , the an-t1 ktn1-4 double mutant conical cells displayed randomly oriented microtubule arrays , similar to those observed in the an-t1 or the ktn1-4 single mutant ( S18 Fig ) . Given that it is well established that KTN1 directly affects microtubule ordering through its severing activity at both microtubule nucleation and crossover sites [54 , 55] , KTN1 may function in microtubule orientation independently of ROS . As predicted , analysis of ROS accumulation in mature petals of the ktn1-4 mutant showed no significant differences as compared to the wild type ( S19 Fig ) . Taken together , these results suggest that the AN-ROS pathway and KTN1 acted in parallel to modulate microtubule organization and conical cell shaping .
ROS function as signaling molecules for organ growth and normal cellular processes such as cell growth and cell division and differentiation [26–29] . Our results provide definitive evidence for a role of ROS in modulating conical cell expansion of petal adaxial epidermal cells . Previous reports showed that AN promotes pavement cell interdigitation in leaves , correlating with negatively regulating cortical microtubule ordering [23 , 24] , although the detailed molecular mechanism is unclear . Our findings in this study showed that AN promotes tip sharpening of conical cells in petals , correlating with positively regulating microtubule ordering , and that AN negatively regulates ROS levels , which in turn affects microtubule organization . Notably , we demonstrated that AN also plays a negative role in regulation of ROS levels in leaf pavement cells and trichomes . Therefore , we proposed that the AN-ROS-microtubule pathway is a general mechanism important for cell shaping in many different cell types . In contrast to the role of AN in negatively regulating ROS production , ROP2 plays a role in positively regulating ROS production during conical cell shaping . AN and ROP2 may act antagonistically to regulate ROS homeostasis , although the molecular mechanisms by which AN suppresses and ROP2 promotes ROS production , respectively , remain to be further explored . Therefore , we propose that an endogenous balance between ROS accumulation and removal must be achieved and tightly regulated to generate microtubule reorientation and normal conical cell expansion of petal adaxial epidermal cells . Notably , loss of KTN1 function does not result in alternation in ROS levels in conical cells , suggesting that KTN1-mediated microtubule re-orientation may act in parallel with ROS signals during conical cell tip sharpening . Emerging evidence suggests that ROS and redox cues have effects on microtubule behaviors [46–49] . Consistently with these reports , we demonstrated that both H2O2 treatment and endogenously increased ROS levels induced by either loss of AN function or ROP2 overexpression resulted in reduced microtubule ordering of the conical cell . Furthermore , the effects of diverse ROS types on cell wall properties have been well studied [46] . ROS play critical roles in both cell-wall stiffening and loosening by promoting the formation of crosslinks between cell wall polysaccharides and glycoproteins , or by cleaving cell wall polysaccharides , respectively . Therefore , based on these findings , we cannot rule out the possibility that ROS also play a role in conical cell shaping by directly influencing cell wall properties . How ROS-mediated signaling regulates microtubule orientation remains unclear and will require extensive research in the future . Given that previous reports have shown that KTN1-mediated microtubule severing plays critical roles in promoting microtubule rearrangements in response to mechanical stress in both the A . thaliana shoot apical meristem and cotyledon pavement cells [56 , 57] , it is possible that mechanical forces could generate a signal to induce microtubule orientation in a KTN1-dependent manner during conical anisotropic expansion of petal adaxial epidermal cells . Interestingly , cells can respond to mechanical signals generated by cell geometry , providing a pervasive feedback on growth [58] . Future studies should aim to uncover the role of mechanical forces during conical cell shaping . On the basis of our findings , we proposed a model to explain the molecular mechanism underlying ROS-dependent microtubule orientation in the regulation of conical cell shaping ( Fig 6 ) . According to this model , the AN-ROS pathway cooperates with KTN1 to jointly reorient microtubules from random to well-ordered transverse arrays , which is critical for the tip sharpening of conical cells . We hypothesized that the re-orientation of microtubules from random to well-ordered arrays may orient the deposition of cellulose microfilaments and generate the cell wall reinforcement throughout conical cell development [12–19] , consequently maintaining conical cell expansion and forming the final characteristic cell shape . Also , we hypothesized that mechanical forces could generate a signal to induce microtubule orientation . Conical cells could respond to mechanical cues generated by cell geometry , providing a feedback loop to define the final cell shape .
All A . thaliana seeds used in this study were of the Columbia-0 ( Col-0 ) ecotype . The ktn1-4 ( SAIL_343_D12 ) , an-t1 ( SALK_026489C ) , and an-t2 ( CS851381 ) were obtained from the Arabidopsis Biological Resource Centre . The seeds were sown in petri dishes on Murashige and Skoog medium agar supplemented with 1% ( w/v ) sucrose . Plants were grown in a growth room at 22°C under 16-hr light/8-hr dark cycles . The sequences of primers used in this study are listed in S2 Table . The full length coding sequences ( CDS ) of the AN gene was amplified and cloned into PH35S-GFP-GW to construct p35S::GFP-AN . For Co-IP-LC-MS/MS analysis , the Ti plasmid expressing recombinant GFP-AN protein was introduced into rdr6-11 plants [59] , which suppresses gene silencing . For the complementation experiment , the AN promoter was amplified from wild-type genomic DNA , and was inserted at Hind III and Xba I sites of vector p35S::GFP-AN to replace the 35S promoter , generating pAN::GFP-AN . For GUS activity assays , the promoter of AN was amplified and cloned into pCAMBIA1301 . For co-IP assay , KTN1 CDS was cloned into pGWB517 to generate 35S::KTN1-Myc . CAT2 and CAT3 CDS was cloned into pGWB505 to generate 35S::CAT2-GFP and 35S::CAT3-GFP , respectively . The Ti plasmid expressing recombinant proteins were introduced into rdr6-11 to generate stable transgenic lines . Nitroblue tetrazolium ( NBT ) and dihydroethidium ( DHE ) were used for superoxide ( O2• – ) staining . 3 , 3’-diaminobenzidine ( DAB ) and CM-H2DCFDA were used for hydrogen peroxide ( H2O2 ) staining . For H2O2 treatments , the concentrations used were 50 mM or 100 mM H2O2 . For ROS scavenging treatments , O2• – was eliminated with 5 mM PG , and H2O2 eliminated with with 1 mM KI . For confocal imaging of conical cells from the side , petal blades were folded back , allowing for the side view of conical cells , and stained with a solution containing 10 μg/ml propidium iodide for more than 10 min . Petal samples were imaged with a Zeiss LSM 880 confocal laser scanning microscope ( excitation at 514 nm , emission 550-700nm ) . For live-confocal imaging of cortical microtubules , non-folded petals stably expressing GFP-TUA6 were imaged with a Zeiss LSM 880 confocal laser scanning microscope ( excitation at 488 , emission 500-570nm ) . Serial optical sections were taken at 0 . 5-μm increments with a 40 × water or 63 × oil lens , and then were projected on a plane ( i . e . maximum intensity ) using Zeiss LSM 880 software . For CM-H2DCFDA staining , petal samples were incubated in 50 mM phosphatic buffer solution ( PBS , pH 7 . 4 ) containing 10 μM CM-H2DCFDA ( Invitrogen , C6827 ) for 30 min , and then the samples were washed for three times with PBS , and observed with the Zeiss LSM 880 microscope ( excitation 488 nm , emission 500–570 nm ) or the Zeiss observer A1 inverted microscope . For Dihydroethidium ( DHE ) staining , petal samples were incubated into 50 mM PBS ( pH 7 . 4 ) buffer solution containing 40 μM DHE ( Sigma , D7008 ) for 30min , and then visualized with the Zeiss LSM 880 microscope ( excitation 514 , emission 520–600 nm ) or the Zeiss observer A1 inverted microscope . Petals from flower development stage 14 were dissected and directly observed with a TM-3030 table-top scanning electron microscope ( Hitachi ) . Approximately 10 , 000 seeds of an-t1 were mutagenized using ethyl methane sulfonate . M2 seeds were harvested from self-fertilized M1 plants individually , and M2 lines were screened for enhanced an-t1 petal conical cells phenotypes . Among 2 , 000 independent EMS-mutagenized an-t1 lines , two candidate enhancers were identified and described in this study . Candidate mutants were backcrossed to an-t1 three times before further phenotypic analyses . About 0 . 1 g of WT and mutant inflorescences were collected and frozen in liquid nitrogen , respectively . Total proteins were extracted using extraction buffer ( 25 mM HEPES , pH 7 . 4 , 10 mM MgCl2 , 10 mM KCl , 5 mM dithiothreitol , 5 mM Na3VO4 , 5 mM NaF , 1 mM phenylmethylsulfonyl fluoride , 1% Triton X-100 , and protease inhibitor cocktail ) . Twenty micrograms of MBP-RIC1-conjugated agarose beads were added to the protein extracts and incubated at 4°Cfor 2h on a rocker . The beads were washed four times in wash buffer ( 25 mM HEPES , pH 7 . 4 , 1 mM EDTA , 5 mM MgCl2 , 1 mM dithiothreitol , and 0 . 5% Triton X-100 ) at 4°C . GTP-bound ROP proteins associated with the MBP-RIC1 beads were boiled and used for analysis by western blot with a ROP2 antibody that was generated against the peptides QFFIDHPGAVPITTNQG ( Abicode , China ) . A . thaliana seedlings expressing GFP-AN ( in rdr6-11 background ) and , as control , GFP ( 35S promoter ) were grown under continuous light in MS medium . Two grams 5-day-old seedlings were collected and ground in liquid N2 , and total proteins were extracted with the buffer ( 25 mM Hepes-KOH , pH7 . 4 , 10 mM MgCl2 , 100 mM NaCl , 5mM NaF , 15% glycerol , 1% Triton X-100 , proteinase inhibitor cocktail ) . After centrifuging at 14 , 000g for 15 min at 4°C , the supernatant was mixed with GFP-Trap agarose beads ( gta-100 , Chromotek ) and rotated at 4°C for 4 hour . The immunoprecipitates were then separated in SDS-PAGE and digested with 0 . 025 mg/mL trypsin . The samples were put into a Thermo Scientific EASY trap column ( 100 μM × 2 cm , 5 μM , 100 Å , C18 ) for separation , and analyzed with Obitrap Fusion mass spectrometer ( Thermo Finnigan , San Jose , CA ) . Each sample was analyzed for 60 min with a resolution of 120 , 000 , the scanning range of 375-1500m/z , AGC target of 4e5 and injection time of 50 ms . Simultaneously , the MS2 scanning was performed with the following parameter: resolution ( 50 , 000 ) , activation type ( HCD ) , injection time ( 105ms ) , AGC target ( 1e5 ) . The raw data operated with Proteome Discoverer 2 . 1 ( Thermo Scientific ) were searched against with protein database ( TAIR10_pep_20101214 . fasta ) , and processed with FDR ( false discovery rate ) ≤0 . 01 at both the peptide and protein level . For the co-immunoprecipitation assay of AN and KTN1 , The Ti plasmids 35S::Myc-KTN1 and 35S::GFP-AN , or 35S::Myc-KTN1 and 35S:: GFP were transiently co-expressed in Nicotiana benthamiana leaves , respectively . Leaf tissue was ground in liquid N2 , and total proteins were extracted with co-IP buffer ( 25mM Hepes-KOH , pH7 . 4 , 10mM MgCl2 , 100mM NaCl , 5mM NaF , 15% glycerol , 1% Triton X-100 , proteinase inhibitor cocktail ) . After centrifuging at 14 , 000g for 10 min at 4°C , the supernatant was mixed with GFP-Trap agarose beads ( gta-100 , Chromotek ) and rotated at 4°C for 2 hour . Beads were then washed five times with ice-cold co-IP buffer ( without protease inhibitors ) . The bound proteins were eluted from the beads with 2×SDS-PAGE sample buffer by heating at 100°C for 5 min , and analyzed by immunoblot . The primary antibody used was: anti-Myc ( 1:3 , 000 , 05–724 , Millipore ) . For the co-immunoprecipitation assays of AN and CAT2/CAT3 , A . thaliana inflorescences expressing CAT2-GFP or CAT3-GFP ( in rdr6-11 background ) and , as control , GFP ( 35S promoter ) were used for protein extraction with the buffer ( 25 mM Hepes-KOH , pH7 . 4 , 10 mM MgCl2 , 100 mM NaCl , 5mM NaF , 15% glycerol , 1% Triton X-100 , proteinase inhibitor cocktail ) . After centrifuging at 14 , 000g for 15 min at 4°C , the supernatants were mixed with GFP-Trap agarose beads ( gta-100 , Chromotek ) and rotated at 4°C for 4 hour . The bound proteins were eluted from the beads with 2×SDS-PAGE sample buffer by heating at 100°C for 5 min , and analyzed by immunoblot . The primary antibody used was anti-AN ( 1:2 , 000 ) . For the generation of Anti-AN antibody , the amino acid sequence from 280 to 490 a . a . of AN was amplified and cloned into expression plasmid pGEX-4T ( GST tag ) , then transformed the vector into Escherichia coli for protein expression . The purified protein was injected into rabbits . Then , Antiserum was extracted and purified . Statistical analyses were performed using Mann–Whitney U test . Not significant difference P > 0 . 05; significant difference *P < 0 . 05 , **P < 0 . 01 , and ***P < 0 . 001 . Sequence data from this article can be found in the Arabidopsis Genome Initiative or GenBank/EMBL databases under the following accession numbers: AT1G80350 ( KTN1 ) , AT1G01510 ( AN ) , AT1G20090 ( ROP2 ) , AT4G35090 ( CAT2 ) , and AT1G20620 ( CAT3 ) . | Plants have diverse cell types with distinct sizes , shapes , and functions . For example , most flowering plants contain specialized conical petal epidermal cells that are thought to attract bee pollinators , but the molecular mechanisms controlling conical cell shaping remain unclear . Here , through a genetic screen in Arabidopsis thaliana , we demonstrated that loss-of-function mutations in ANGUSTIFOLIA ( AN ) , which encodes a homolog of mammalian CtBP/BARs , resulted in swollen conical cells , correlating with increased accumulation of reactive oxygen species ( ROS ) . Furthermore , through an enhancer screening , we demonstrated that mutations in katanin ( KTN1 ) enhanced conical cell phenotypes of the an-t1 mutants . Genetic analyses showed that AN acted in parallel with KTN1 to control conical cell shaping . We demonstrated that the AN-ROS pathway jointly functioned with KTN1 to modulate microtubule organization , correlating with the tip sharpening of conical cells . Collectively , our findings revealed a mechanistic insight into ROS-mediated regulation of conical cell shaping . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"plant",
"anatomy",
"trichomes",
"microtubules",
"pavement",
"cells",
"plant",
"cell",
"biology",
"brassica",
"plant",
"science",
"model",
"organisms",
"experimental",
"organism",
"systems",
"seedlings",
"plants",
"cellular",
"structures",
"and",
"organelles",
"flower",
"anatomy",
"cytoskeleton",
"research",
"and",
"analysis",
"methods",
"arabidopsis",
"thaliana",
"animal",
"studies",
"leaves",
"flowers",
"plant",
"cells",
"eukaryota",
"plant",
"and",
"algal",
"models",
"cell",
"biology",
"phenotypes",
"petals",
"genetics",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"organisms"
] | 2018 | Reactive oxygen species mediate conical cell shaping in Arabidopsis thaliana petals |
Candida albicans is a leading human fungal pathogen that causes life-threatening systemic infections . A key regulator of C . albicans stress response , drug resistance , morphogenesis , and virulence is the molecular chaperone Hsp90 . Targeting Hsp90 provides a powerful strategy to treat fungal infections , however , the therapeutic utility of current inhibitors is compromised by toxicity due to inhibition of host Hsp90 . To identify components of the Hsp90-dependent circuitry governing virulence and drug resistance that are sufficiently divergent for selective targeting in the pathogen , we pioneered chemical genomic profiling of the Hsp90 genetic network in C . albicans . Here , we screen mutant collections covering ~10% of the genome for hypersensitivity to Hsp90 inhibition in multiple environmental conditions . We identify 158 HSP90 chemical genetic interactors , most of which are important for growth only in specific environments . We discovered that the sterol C-22 desaturase gene ERG5 and the phosphatidylinositol-4-kinase ( PI4K ) gene STT4 are HSP90 genetic interactors under multiple conditions , suggesting a function upstream of Hsp90 . By systematic analysis of the ergosterol biosynthetic cascade , we demonstrate that defects in ergosterol biosynthesis induce cellular stress that overwhelms Hsp90’s functional capacity . By analysis of the phosphatidylinositol pathway , we demonstrate that there is a genetic interaction between the PI4K Stt4 and Hsp90 . We also establish that Stt4 is required for normal actin polarization through regulation of Wal1 , and suggest a model in which defects in actin remodeling induces stress that creates a cellular demand for Hsp90 that exceeds its functional capacity . Consistent with this model , actin inhibitors are synergistic with Hsp90 inhibitors . We highlight new connections between Hsp90 and virulence traits , demonstrating that Erg5 and Stt4 enable activation of macrophage pyroptosis . This work uncovers novel circuitry regulating Hsp90 functional capacity and new effectors governing drug resistance , morphogenesis and virulence , revealing new targets for antifungal drug development .
Hsp90 is an ATP-dependent molecular chaperone that stabilizes components of diverse signal transduction cascades , especially those involved in adaptation to stress [1] . Hsp90 function is modulated by co-chaperones , which are thought to mediate recognition of client proteins , many of which cycle dynamically through complexes with Hsp90 until their activation [2] . Hsp90 can be regulated in response to environmental stress at the transcriptional level , such as with the induction observed in response to high temperature via the transcription factor Hsf1 . Hsp90 can also be regulated by a dynamic code of post-translational modifications , including phosphorylation , acetylation , and nitrosylation [2] . Despite the complex cellular control of Hsp90 function , environmental stress can induce global protein misfolding that overwhelms the chaperone’s functional capacity [3] . As a consequence of Hsp90’s role as a central hub of protein homeostasis and regulatory circuitry , it provides an Achilles’ Heel that can be targeted to cripple diverse organisms , including fungal pathogens . Fungi are a leading cause of human mortality worldwide . These eukaryotic pathogens infect 1 . 7 billion people and kill more than 1 . 5 million people annually [4] . Candida albicans is a leading fungal pathogen of humans , accounting for 9–12% of all hospital-acquired bloodstream infections , with an attributable mortality rate of 38% , despite significant advances in diagnosis and the increased use of antifungal therapies [5] . In C . albicans , Hsp90 plays key roles in the pathogenesis-relevant phenotypes of morphogenesis , antifungal drug resistance , and response to host stresses [6–9] . Hsp90 inhibition abrogates resistance to the azoles and the echinocandins , which are clinically important antifungals that target the fungal cell membrane and the cell wall , respectively [8 , 10] . Genetic depletion or pharmacological inhibition of Hsp90 also induces a morphological transition from yeast to filamentous growth; the capacity to transition between these states is a key virulence trait [11 , 12] . Hsp90 regulates fungal drug resistance and morphogenesis by stabilizing key regulators of cellular signaling . Hypothesis-driven approaches have identified the protein phosphatase calcineurin and the terminal mitogen-activated protein kinase ( MAPK ) of the protein kinase C ( PKC ) cell wall integrity pathway , Mkc1 , as Hsp90 client proteins that are crucial for drug resistance [10 , 13] . Small scale screens covering ~3% of the genome have revealed that Hsp90 regulates morphogenesis through Ras1-PKA signaling [9] , and through a pathway that includes the cyclin Pcl1 , cyclin-dependent kinase Pho85 , and transcription factor Hms1 [14] . We also identified the cyclin-dependent kinase Cdc28 [15] , and three transcription factors , Cph2 , Hap5 , and Stp2 [9] , as important effectors through which Hsp90 controls morphogenesis . The pleiotropic effect of Hsp90 on cellular circuitry demands a systematic approach to explore Hsp90 genetic interaction networks to identify effectors through which it governs drug resistance and morphogenesis . We previously pioneered a chemical genomic approach to map the genetic interactors of Hsp90 in C . albicans , an organism for which classical genetic approaches are hampered by the lack of a complete sexual cycle [16] . Chemical inhibitors provide a powerful approach to detect genetic interactions when a mutant allele of one gene causes an unexpected phenotype in the presence of a chemical . Chemical genomics has revealed gene function and genetic networks in the model yeast Saccharomyces cerevisiae [17] , and provides a powerful approach to define HSP90 interactors [18 , 19] . Mapping HSP90 chemical genetic interactions is facilitated by the availability of potent and highly specific inhibitors of Hsp90 function , including the natural product geldanamycin , which binds in the unusual nucleotide binding pocket of Hsp90 , blocking ATPase activity and leading to the degradation of client proteins [20] . In our original study , we screened a C . albicans transposon mutant library covering ~10% of the genome for hypersensitivity to geldanamycin under standard growth conditions and five stress conditions , in order to identify environmentally contingent interactions [16] . We identified 226 HSP90 chemical genetic interactors , most of which were important for growth only under specific conditions . A small number of HSP90 interactors were identified in multiple stress conditions in addition to under standard conditions; these pleiotropic effects suggested that they function upstream of HSP90 . Consistent with this model , interactors identified in the majority of conditions included the transcription factor gene AHR1 , which regulates HSP90 expression , and the genes encoding regulatory subunits of protein kinase CK2 , which control Hsp90 phosphorylation and function [16] . This work illustrated the power of chemical genomics to map functional connections and genetic networks in C . albicans . Here , we build upon our previous HSP90 genetic interaction network by screening two additional mutant libraries covering 772 genes for hypersensitivity to geldanamycin , expanding our coverage to ~20% of the genome . From this , we identify 11 strong HSP90 chemical genetic interactors under basal conditions , and an additional 147 HSP90 chemical genetic interactors that are required for growth under stress . We focused on the sterol C-22 desaturase gene ERG5 and the phosphatidylinositol-4-kinase ( PI4K ) gene STT4 , and demonstrate that they regulate the functional capacity of Hsp90 , providing new avenues for antifungal drug development .
We extended our previous analysis of the HSP90 genetic interaction network in C . albicans , utilizing the same chemical genomic approach to screen two additional mutant libraries that cover 772 genes , including 566 ORFs distinct from the previous screen [11 , 21] . An advantage of these libraries over that used in our previous analysis is that these are composed of precise homozygous deletion mutants rather than transposon insertion mutants . Each strain was screened for growth under standard host-relevant conditions ( RPMI , 37°C ) in the presence and absence of a low concentration of the pharmacological Hsp90 inhibitor geldanamycin ( 3 μM ) that does not affect growth of the wild-type strain . This revealed 11 genes that are strong genetic interactors of Hsp90 ( Fig 1A ) , as the corresponding mutants had at least 50% reduction in growth in response to this low dose of Hsp90 inhibition . Of these , mutants lacking the sterol C-22 desaturase gene ERG5 and the phosphatidylinositol-4-kinase ( PI4K ) gene STT4 demonstrated the greatest sensitivity to Hsp90 inhibition . These genetic interactions were confirmed using Gene Replacement And Conditional Expression ( GRACE ) strains [22 , 23] , whose target gene expression is repressible using the tetracycline analog doxycycline ( S1 Fig ) . This confirms that the sensitivity of the stt4Δ/Δ and erg5Δ/Δ mutants to geldanamycin is due to the specific genetic perturbation and not to spurious mutations . Our previous analysis of Hsp90 genetic interactors suggested that many interactions are dependent upon additional stress to the cell , including ones that mimic stresses occurring in the human host . Therefore , we screened the mutant libraries [11 , 21] for hypersensitivity to geldanamycin under the general stresses of high temperature ( 41°C , human febrile temperatures ) , osmotic stress ( NaCl ) , and unfolded protein stress ( tunicamycin ) . We also used treatment with low concentrations of the commonly used antifungal drugs caspofungin , which targets the cell wall , and fluconazole , which targets ergosterol biosynthesis and causes cell membrane stress and toxic sterol accumulation . For these screens , we calculated whether there was a genetic interaction upon stress using the Product Multiplicative Model [24 , 25] . We used a conservative cut-off that defined a chemical genetic interaction if the observed fitness under the combined stress condition was less than half of the expected fitness determined from the singular treatment of stress or geldanamycin alone in at least two independent mutants for each gene Fitness ( stress+geldanamycin ) <Fitness ( stress ) *Fitness ( geldanamycin ) 2 . This was a more stringent cutoff than our previous screen , minimizing false positive interactions; however , it resulted in exclusion of some previously identified genetic interactors . For example , we observed that ORF19 . 1219 and ORF19 . 1496 are genetic interactors with HSP90 in response to caspofungin , similar to the previous screen , but they only had a 40% greater than expected reduction in fitness and were thus excluded from our more stringent set of genetic interactors . For the 11 mutants that were found to be hypersensitive to 3 μM geldanamycin alone , we calculated genetic interactions upon stress using a lower concentration of geldanamycin that had minimal impact on growth ( Fig 1A and 1B , thick outline ) . This allows us to determine whether these genetic interactions are also observed under other stress conditions , indicating that these genes have pleiotropic effects that could suggest a role upstream of Hsp90 . Together , these screens identified 158 chemical genetic interactors of Hsp90 ( S1 Table ) . The majority of the genetic interactions were only observed under a single growth condition , with caspofungin yielding the most genetic interactors . Six genes were HSP90 genetic interactors under three stress conditions , implicating them as high-connectivity interactors . Interestingly , the uncharacterized gene ORF19 . 194 , the septin gene CDC10 , and the transcription factor gene ZCF2 , which did not confer sensitivity to geldanamycin under standard conditions , were high-connectivity interactors under stress . Together , this highlights the environmental contingency of the Hsp90 chemical genetic interactions . We also examined the role of all of these genetic interactors in morphogenesis and drug resistance . As with Hsp90 , many of these genetic interactors were found to be important for filamentation or susceptibility to antifungal drugs ( S2 Fig ) , demonstrating that HSP90 genetic interactors regulate morphogenesis and cellular responses to antifungal drugs . For mutants that were hypersensitive to geldanamycin ( Fig 1A ) , we observed varying numbers of genetic interactions under the stress conditions ( Fig 1B ) . There are three models for why an Hsp90 genetic interactor under basal conditions would not also be a genetic interactor under stress conditions . First , false negatives can result if the mutant has a low fitness under the individual stress conditions , causing the fitness under the combined stresses to be below the limit of detection . Second , activation of specific stress responses may compensate for the loss of a gene . Third , distinct sets of Hsp90-dependent client proteins are crucial for cellular viability in different environmental conditions [19] . Screening for Hsp90 genetic interactions under multiple stress conditions therefore produces a comprehensive map of general and condition-specific Hsp90 genetic interactors . Based on our analysis , we identified STT4 , ERG5 , and the Set3C lysine deacetylase SNT1 as the strongest chemical genetic interactors of HSP90 under standard growth conditions ( Fig 1A ) that are also chemical genetic interactors under stress conditions , implicating them as high connectivity interactors in our network ( Fig 1B ) . To validate the genetic interactions between HSP90 and STT4 , ERG5 , and SNT1 , we created double mutant strains where the only allele of HSP90 was under the control of the tetO tetracycline-repressible promoter in the stt4Δ/Δ , erg5Δ/Δ , and snt1Δ/Δ mutant backgrounds . Construction of hsp90Δ/Δ deletion mutants is precluded by the fact that Hsp90 is essential in all eukaryotes tested [26] . We measured fitness of all of the strains in the presence of the tetracycline analog doxycycline to repress HSP90 expression . Using the Product Multiplicative Model of genetic interactions [24 , 25] , we determined that the fitness defects of the double mutant strains were more severe than the product of each single mutant fitness defect ( Fig 2 ) , confirming the genetic interactions . This genetic validation rules out the possibility that hypersensitivity of the stt4Δ/Δ , snt1Δ/Δ , and erg5Δ/Δ mutants to geldanamycin is due to alterations in geldanamycin accumulation in the cell or off-target effects of the drug , confirming that STT4 , ERG5 and SNT1 are bona fide genetic interactors of HSP90 . The genetic interaction network ( Fig 1 ) can be separated into two types of interactions—low connectivity , where the mutant is only hypersensitive to Hsp90 inhibition under one environmental condition , and high connectivity , where the mutant is sensitive to Hsp90 inhibition under multiple conditions . In our previous work , we demonstrated that high connectivity interactors are likely to function upstream of Hsp90 , and influence Hsp90 expression or function [16] . Based on this logic , we hypothesized that Stt4 , Erg5 , and Snt1 would function upstream of Hsp90 , due to their genetic interactions with Hsp90 in both basal and multiple stress conditions . To determine whether Stt4 , Erg5 , and Snt1 affect Hsp90 at the protein level , we measured Hsp90 protein levels in the mutant strains by Western blotting ( Fig 3A ) . Hsp90 levels were not decreased the stt4Δ/Δ , erg5Δ/Δ , and snt1Δ/Δ mutants , and were in fact slightly increased , suggesting that they must influence Hsp90 via another mechanism . To determine whether these genes affect Hsp90 function , we first monitored phosphorylation of the Hsp90 client protein Hog1 in the wild-type , stt4Δ/Δ , erg5Δ/Δ , and snt1Δ/Δ strains . Hog1 is a terminal mitogen activated protein kinase ( MAPK ) involved in adaptation to oxidative stress . Hog1 requires Hsp90 for stabilization and activation [16]; there is a decrease in the levels of total and phosphorylated Hog1 when Hsp90 function is compromised . We observed reduced levels of phosphorylated Hog1 in response to peroxide stress in the stt4Δ/Δ and erg5Δ/Δ mutant strains compared to the wild type ( Fig 3B ) , consistent with a decrease in Hsp90 chaperone function . However , Hog1 phosphorylation was not decreased in the snt1Δ/Δ strain . As a second reporter for Hsp90 function , we examined the capacity of Hsp90 to repress filamentation via the Ras signaling cascade [9] . Inhibition of Hsp90 using 10 μM geldanamycin is sufficient to relieve this repression and induce filamentous growth [9] . We reasoned that if Hsp90 function were compromised in the mutant strains , a lower concentration of Hsp90 inhibitor would be sufficient to induce filamentation in the mutants compared to the wild-type strain . Consistent with this hypothesis , we found that 2 . 5 μM geldanamycin induced robust filamentation in the erg5Δ/Δ and stt4Δ/Δ mutants but not in the wild-type strain ( Fig 3C ) . This demonstrates a divergence in the requirements for a phosphoinositide bis-phosphate ( PIP2 ) gradient during serum and Hsp90-compromise induced filaments , as C . albicans stt4Δ/Δ mutants are defective in filamentation in response to serum [27] . Additionally , the snt1Δ/Δ mutant behaved as the wild type , suggesting that Snt1 may either act in a parallel pathway to Hsp90 , or have specific functions in regulating Hsp90 that are not captured by our reporter assays . Together , these results are consistent with a model in which Erg5 and Stt4 regulate Hsp90 function or the cellular demand for Hsp90 . Consistent with our findings that ERG5 is an HSP90 genetic interactor , previous analyses of C . albicans strains resistant to the membrane disrupting antifungal drug amphotericin B demonstrated that deletion of the ergosterol biosynthesis genes ERG6 , ERG2 , or combined deletion of both ERG3 and ERG11 caused hypersensitivity to geldanamycin . Deletion of these ergosterol genes was found to increase basal stress levels , suggesting that the cellular reservoirs of Hsp90 were depleted [28] . We examined whether this was specific to strains that are resistant to amphotericin B , or whether perturbation of the entire ergosterol biosynthesis pathway would confer a similar hypersensitivity to Hsp90 inhibition . To do so , we tested 18 tetracycline-repressible conditional expression strains for ergosterol biosynthesis genes from the GRACE collection [22 , 23] for hypersensitivity to Hsp90 inhibition in the presence of doxycycline to repress target gene expression . We observed that the ERG24 , ERG2 , ERG3 , ERG5 and ERG6 conditional expression strains were all hypersensitive to geldanamycin ( Fig 4A ) . Interestingly , transcriptional repression of ERG10 , ERG20 , ERG7 , and ERG26 had little effect on geldanamycin sensitivity . However , many of the other ergosterol biosynthesis genes are essential , precluding their analysis for geldanamycin hypersensitivity [23] . Therefore , we used small molecule inhibitors of the ergosterol biosynthetic cascade to further explore the relationship between ergosterol biosynthesis and Hsp90 . As seen previously , synergy was observed between geldanamycin and the Erg11 inhibitor fluconazole ( FIC90 = 0 . 09 ) ( Fig 4B ) [7 , 8] . Synergy with geldanamycin was also observed between the allylamine terbinafine ( FIC90 = 0 . 25 ) , which targets Erg1 , and the morpholine fenpropimorph ( FIC90 = 0 . 016 ) , which targets Erg24 and Erg2 ( Fig 4B ) . In contrast , amphotericin B , which binds ergosterol and removes it from the membrane [29 , 30] , does not show synergy with geldanamycin ( FIC90 = 0 . 75 ) ( Fig 4B ) . Together , this suggests that it is not the lack of ergosterol per se that confers hypersensitivity to Hsp90 inhibition , but rather incorporation of specific sterol intermediates into the membrane . These altered sterols could cause altered membrane fluidity , which is important for temperature sensing , stress responses , and activation of the heat shock response [31] . To test this hypothesis , we supplemented the erg5Δ/Δ mutant strain with 50 μM ergosterol under conditions that allow for sterol uptake [32] . We also examined this relationship in the erg3Δ/Δ mutant strain , which does not produce the canonical toxic sterol intermediate , 14-α-methyl-3 , 6-diol [33] . We observed no rescue in geldanamycin hypersensitivity for either mutant strain upon addition of ergosterol ( Fig 4C ) , suggesting that incorporation of specific sterol intermediates into the cell membranes is sufficient to induce cellular stress and overwhelm the functional capacity of Hsp90 . C . albicans Stt4 is a type III phosphatidylinositol-4-kinase ( PI4K ) ; previous work in mammalian cells demonstrated that the type IIβ PI4K protein is stabilized by Hsp90 [34] . Therefore , we examined the stability of CaStt4 upon Hsp90 depletion . To do so , we engineered an N-terminally FLAG-tagged STT4 allele under the ACT1 promoter in the tetO-HSP90/hsp90Δ background , and monitored FLAG-Stt4 protein levels upon transcriptional repression of HSP90 with doxycycline ( Fig 5A ) . There was no reduction in FLAG-Stt4 levels upon Hsp90 depletion , indicating that that unlike in mammalian cells , Stt4 is not a client protein of Hsp90 in C . albicans . In S . cerevisiae , Stt4 function is controlled by two proteins , Ypp1 and Efr3 , which help localize Stt4 to the membrane [35] . To investigate the function of the entire PI4K complex in C . albicans , we examined the putative ORFs encoding Ypp1 ( ORF19 . 4163 , 26% protein identity with S . cerevisiae ) and Efr3 ( ORF19 . 4798 , 30% identity with S . cerevisiae ) . To determine if these proteins share a common function with Stt4 in C . albicans , we created the ypp1Δ/Δ and efr3Δ/Δ deletion strains . Similar to the stt4Δ/Δ deletion strain , the ypp1Δ/Δ and efr3Δ/Δ mutant strains were hypersensitive to geldanamycin ( Fig 5B ) , demonstrating that Ypp1 and Efr3 are also genetic interactors of Hsp90 . We then examined whether alterations in phosphatidylinositol signaling in general , or the production of phosphatidylinositol 4-phosphate ( PI4P ) specifically by Stt4 was important for geldanamycin sensitivity by testing four available phosphatidylinositol kinase and phosphatase mutants . However , the stt4Δ/Δ mutant was the only mutant with dramatic hypersensitivity to Hsp90 inhibition under basal conditions , suggesting that the genetic interaction between Stt4 and Hsp90 is specific ( S2 Table ) . We then took an unbiased approach to identify downstream effectors of Stt4 by examining C . albicans proteins with pleckstrin homology ( PH ) or pleckstrin homology-like domains ( S2 Table ) ; these domains bind phosphatidylinositols and help localize proteins to specific membranes . We also searched for proteins that are computationally predicted to bind to phosphatidylinositol-4-phosphate . We identified 42 genes encoding such proteins in the C . albicans genome , and examined 30 available mutants corresponding to these genes for geldanamycin sensitivity . Only the osh3Δ/Δ and tetO-WAL1/wal1Δ mutant strains were hypersensitive to Hsp90 inhibition , suggesting that modulation or perturbation of the localization of these proteins mediate the sensitivity of the stt4Δ/Δ mutant to geldanamycin ( Fig 5C and S2 Table ) . Wal1 is the homolog of the S . cerevisiae Las17 actin nucleation promotion factor , and is required for actin remodeling in C . albicans [36] . Although the PI4K complex is known to regulate actin polarization in S . cerevisiae [37] , ScLas17 does not contain a PH domain . We used rhodamine-phalloidin staining of F-actin in the wild type and the stt4Δ/Δ mutant strain , and observed a defect in polarized actin at the tip of elongating cells in the stt4Δ/Δ mutant ( Fig 5D ) . This was confirmed with the doxycycline-repressible STT4 mutant , demonstrating that the actin polarization defect can be attributed to loss of Stt4 ( S3 Fig ) . We observed a similar effect , though more drastic , upon doxycycline-mediated transcriptional repression of WAL1 in the tetO-WAL1/wal1Δ depletion strain ( Fig 5D ) . Together , this suggests that Stt4 is required for producing the PI4P used to localize Wal1 , which is a requirement for normal actin remodeling and stress tolerance . This led us to hypothesize that defects in actin organization cause increased cellular demand for Hsp90 , thus sensitizing the cell to Hsp90 inhibition . To test this , we examined the interaction between cytochalasin A , which prevents actin polymerization , and geldanamycin ( Fig 5E ) . We observed synergy ( FIC90 = 0 . 375 ) in the wild-type strain but not in the stt4Δ/Δ mutant ( FIC90 = 0 . 750 ) , suggesting that the primary cause of geldanamycin hypersensitivity in the stt4Δ/Δ mutant is due to perturbation of actin . Interestingly , many other genes that regulate cytoskeletal dynamics in C . albicans were not genetic interactors with Hsp90 ( S2 Table ) , suggesting that there is specificity in the interaction between Stt4 , Wal1 , and Hsp90 . Given that Hsp90 has a profound impact on virulence traits , such as response to stress and morphogenetic transitions , we next assessed the impact of the three Hsp90 genetic interactors , ERG5 , SNT1 , and STT4 , on virulence; our hypothesis was that defects in Hsp90 function would result in decreased virulence . To test this hypothesis , we utilized a macrophage model of virulence , where we assessed the capacity of the mutant strains to induce pyroptosis and kill macrophages [23] . We used bone-marrow-derived macrophages that have been stably transformed with an mCherry-labelled ASC protein . This protein oligomerizes upon NLRP3-dependent pyroptosis [38] , allowing us to precisely assess the ability of the C . albicans cells to induce macrophage pyroptosis . Notably , the stt4Δ/Δ mutant demonstrated slightly reduced filamentation within host cells ( Fig 6A ) , although the defect in filamentation in macrophages was not as severe as in serum [27] . Importantly , all three mutant strains were significantly attenuated for macrophage lysis compared with the wild-type strain ( p < 0 . 05 with Bonferroni corrections , Fig 6B ) , suggesting that these Hsp90 genetic interactors have a role in virulence . We validated our findings with the tetO-ERG5/erg5Δ and tetO-STT4/stt4Δ strains upon doxycycline treatment ( S4 Fig ) . This highlights the utility of Hsp90 genetic interactors as targets for antifungal drug development .
Our expanded map of the Hsp90 genetic interaction network in C . albicans reveals key perturbations in the genetic architecture of the cell that induce severe cellular stress , and further identifies novel effectors governing drug resistance , morphogenesis , and virulence . Our screen of 772 homozygous deletion mutants for hypersensitivity to geldanamycin delineated a network of 158 HSP90 chemical genetic interactors , most of which were specific for a single environmental condition ( Fig 1 ) . The low connectivity interactors are likely to enable growth in response to specific conditions , as with the PKC pathway MAP kinase MKC1 [13 , 39] or the Hog1 MAPKKK SSK2 , which are genetic interactors with NaCl treatment . The more pleiotropic high connectivity interactors that were identified in many conditions are likely to either operate upstream of HSP90 [16] , or to modulate the cellular demand for the chaperone machinery . Our cumulative screens to date now cover ~20% of the genome and have identified 352 distinct HSP90 genetic interactors . This work has established important principles through which network connectivity can reveal pathway order and functional relationships in the cell . Surprisingly , some of the HSP90 genetic interactors identified in basal conditions were classified as low connectivity interactors , as their genetic interaction was not maintained under stress conditions ( Fig 1 ) . This is likely due to condition-specific functions of Hsp90 or the gene of interest . For example , previous S . cerevisiae screens have found divergence in Hsp90 chemical genetic interactors under basal and stress conditions due to condition-specific functions of Hsp90 [19] . Under basal conditions , they found that Hsp90 function is important for the secretory pathway , protein transport , as well as forming and stabilizing oligomeric complexes; in stress conditions , Hsp90 is essential for control of cell cycle , meiosis , and cytokinesis [19] . Genetic interactions can also differ between environmental conditions based on condition-specific expression or activation of compensatory pathways . Our network provides a glimpse of the circuitry that modulates the functional capacity of Hsp90 . With STT4 , ERG5 , and SNT1 identified as HSP90 chemical genetic and bona fide genetic interactors under both basal and stress conditions ( Figs 1 and 2 ) , this suggests that they could operate upstream of HSP90 or that they modulate the cellular requirements for the chaperone machinery . The biochemical functions of phosphatidylinositol signaling and ergosterol biosynthesis are unlikely to directly impinge on regulation of Hsp90 function , but these are processes intimately coupled with modulating cellular integrity , signaling , and stress . Consistent with the model that perturbation of these pathways induces a state of cellular stress that overwhelms Hsp90’s functional capacity , we found that despite a modest increase in Hsp90 levels in the erg5Δ/Δ and stt4Δ/Δ mutants , Hsp90 function in stabilizing two distinct signaling cascades was compromised ( Fig 3 ) . In the snt1Δ/Δ mutant strain , however , we did not observe alterations in Hsp90 function , suggesting that Snt1 either acts in a parallel pathway with Hsp90 , or affects a subset of Hsp90 functions . Our work , and that of others has demonstrated that Hsp90 deacetylation is a key component of drug resistance [40 , 41] , suggesting that Snt1 , and the entire Set3C lysine deacetylase complex , may have specific roles in governing Hsp90 function in drug resistance . Overall , targeting core cellular homeostasis pathways provides a powerful strategy to induce cellular stress that can be exploited to cripple fungal pathogens . One core cellular process is the incorporation of ergosterol in the membrane . Impairing ergosterol biosynthesis can lead to accumulation of sterol intermediates in the membrane , which has profound impacts on membrane fluidity , membrane integrity and the functions of diverse signaling cascades [26 , 31 , 42] . We found that perturbation of ergosterol biosynthesis at multiple stages causes hypersensitivity to Hsp90 inhibition , consistent with previous findings that ERG1 , ERG2 , ERG5 , and ERG6 are HSP90 genetic interactors in S . cerevisiae [18 , 19] . This link between Hsp90 and ergosterol biosynthesis was further corroborated by synergy between an Hsp90 inhibitor and antifungals that target the ergosterol biosynthesis pathway ( Fig 4 ) . Indeed , mutations in the ergosterol biosynthesis pathway that confer resistance to amphotericin B create severe cellular stress that renders the fungus vulnerable to attack by host defenses [28] . By exploring the details of the genetic interactions between Hsp90 and the ergosterol biosynthetic cascade , we demonstrate that the incorporation of altered sterols into the membrane is sufficient to induce cellular stress , which overwhelms Hsp90’s functional capacity . Another core process regulating the cellular demand for Hsp90 and the tolerance of antifungal drugs is phosphatidylinositol-4-phosphate synthesis by Stt4 . We demonstrate that the phosphatidylinositol-4-kinase ( PI4K ) gene STT4 is a strong Hsp90 genetic interactor ( Fig 5 ) , as are two uncharacterized orthologs of additional components of the PI4K complex in S . cerevisiae , EFR3 and YPP1 [35] . In our previous study , we identified the VPS34 and VPS15 PI3 kinases as Hsp90 genetic interactors , but only under high temperature and high salt stress [16] . Based on analysis of downstream effectors of Stt4 , we propose a model in which PI4P created by Stt4 is required for localization of Wal1 and thus normal actin remodeling during growth . In the absence of Stt4 , actin remodeling is altered , resulting in stress that creates a cellular demand for Hsp90 that exceeds its functional capacity . One mechanism for this stress could be the altered regulation of glycolysis , leading to metabolic stress . In mammalian cells , PI3K signaling is required for actin remodeling , release of aldolase A , and enhanced glycolytic flux [43] . However , many of the other factors involved in this process , such as the GTPases that work with Rac1 , were not identified as genetic interactors with HSP90 in our study ( S2 Table ) . Targeting Hsp90 and key components of the chaperone genetic network provides a powerful strategy to treat life threatening infectious disease caused by diverse eukaryotic pathogens . Pharmacological inhibition of Hsp90 itself is facilitated by the phenomenal progress made in the development of potent compounds that selectively inhibit this chaperone , driven in large part by the desire to target Hsp90’s key role in stabilizing multiple oncogenic proteins and enabling malignant transformation [44 , 45] . These molecules can transform antifungals from ineffective to highly efficacious in mammalian models of fungal biofilm infections , where the infection and drug delivery are localized [46] . Beyond Hsp90 , other chaperone network components have emerged as promising targets for therapeutic intervention that are more divergent between pathogen and host , as is the case with Stt4 and Snt1 , which have only 45% and 30% sequence identity to their closest human ortholog . Recent work in antimalarials identified a Plasmodium type III PI4 kinase as a target of imidazopyrazines [47] , which could have broad relevance for fungal pathogens given the profound impact of perturbation of Stt4 function on cellular stress and virulence . Moreover , the alteration in Hsp90 functional capacity in the stt4Δ/Δ and erg5Δ/Δ mutants may minimize the potential for the evolution of antifungal drug resistance [8] . Fitness constraints associated with the evolution of resistance to Hsp90 inhibitors in combination with antifungals [48] , suggests promising strategies for the rational design of antifungal combinations that evade drug resistance .
The C . albicans homozygous deletion libraries were obtained from the Fungal Genetics Stock Center and maintained in cryo-culture at -80°C . Individual strains were inoculated into 96-well plates in 100 μL RPMI-1640 pH 7 ( 10 . 4 g/l RPMI-1640 , 3 . 5% MOPS , 2% glucose ) , supplemented with 5 mg/L histidine . The plates were sealed with Adhesive Plate Seals ( Thermo Scientific ) and incubated overnight at 37°C while shaking at 200 rpm . Cells were then diluted twice . First , using the VP408 96 Pin Multi-Blot Replicator ( VP Scientific ) , 0 . 5 μl of Candida culture was inoculated in 200 μl of 1X phosphate buffered saline ( PBS ) ( 1:400 dilution ) . This mixture was then diluted a further ten-fold into 200 μl RPMI-1640 , RPMI with 3 μM geldanamycin ( LC laboratories , G-4500 ) , RPMI with stressor ( S3 Table ) , or RPMI with both 3 μM geldanamycin and stressor in flat bottom 96-well plates . For the 11 mutants with strong hypersensitivity to Hsp90 inhibition , screening was repeated at the sub-inhibitory concentrations of 1 μM geldanamycin for all mutants except stt4Δ/Δ , which required 0 . 375 μM . Plates were incubated at the indicated conditions ( S3 Table ) and growth was measured by optical densities ( ODs ) at λ = 600 nm . All strains were maintained in cryo-culture at -80°C in 25% glycerol and passaged in YPD . All primer sequences are included in S4 Table . Individual strains are listed in S5 Table . Plasmids used for strain construction are included in S6 Table . Erg5 and Stt4 were tagged using a PCR-based strategy [49] . The HA C-terminal tag and selectable marker was amplified from pLC575 ( HIS3 ) using oLC4028+oLC4029 . The NAT-pACT1-Flag N-terminal tag was amplified from pLC620 using oLC3401+oLC3402 . The constructs were transformed into C . albicans strains using standard protocols . Proper integration was confirmed by PCR using primer pairs oLC1645+oLC2944 and oLC3464+oLC241 for Erg5-HA , and oLC3403+oLC275 and oLC3404+oLC274 for FLAG-Stt4 . Expression was confirmed by Western blots . The HSP90 double mutant strains were created by deleting one allele of HSP90 in the erg5Δ/Δ and stt4Δ/Δ strains by transforming in pLC62 digested with KpnI and SacII and confirming integration by PCR using primers oLC275+oLC276 and oLC274+oLC277 . The SAP2 promoter was then induced to drive expression of FLP recombinase to excise the NAT marker cassette . The strains were then transformed with the tetO-HSP90 promoter replacement construct amplified from pLC605 containing the NAT resistance marker using primers oLC3220 and oLC3390 . The NAT maker was then excised and lack of a wild-type HSP90 was confirmed by PCR using primers oLC294+oLC297 . The ypp1Δ/Δ , efr3Δ/Δ , and pikalphaΔ/Δ strains were created by sequential deletion and excision of the NAT resistance marker using standard PCR-based homologous recombination methods to generate precise gene deletions . The NAT cassette was amplified from pLC49 using primers oLC4020 and oLC4021 for Ypp1 , oLC4024 and oLC4025 for Efr3 , and oLC4629 and oLC4630 for Pikalpha . Proper integration was tested using oLC4022+oLC275 and oLC4023+oLC274 for Ypp1 , oLC274+oLC4026 and oLC275+oLC4027 for Efr3 , and oLC275+oLC4632 and oLC274+oLC4633 for Pikalpha . Absence of a wild-type allele was confirmed using oLC4022+oLC4023 for Ypp1 , oLC4052+oLC4053 for Efr3 , and oLC4685+oLC4632 for Pikalpha . The growth of each mutant in each of the stress conditions was compared to its growth under standard conditions to determine the relative fitness in each condition . The expected fitness of the stress+geldanamycin condition was calculated using the Product Multiplicative Model ( fitness in stress * fitness in geldanamycin ) . Chemical genetic interactions were defined if the observed fitness was less than half of the expected fitness in at least two independent mutants . For the mutants with strong hypersensitivity to geldanamycin , chemical genetic interactions were defined if the observed fitness was less than 90% of the expected fitness in at least two independent mutants . For Hsp90 levels and Flag-Stt4 levels , strains were inoculated overnight in YPD , subcultured to an OD600 = 0 . 1 in fresh YPD , incubated for 4 hours at 30°C with shaking , and then 1 mL of OD600 = 0 . 6 cells was collected . Pellets were washed once in 1X PBS before being resuspended in 50 μL of 2X laemmli sample buffer . The sample was then boiled for 5 min , spun at 13 , 000 rpm for 5 minutes , and the lysate was then loaded on a 6% SDS gel . The sample was transferred to a nitrocellulose membrane and blocked in 5% milk in PBS-T . Hsp90 levels were detected using a native antibody against C . albicans Hsp90 ( Gift from B . Larsen ) . Flag-Stt4 levels were detected using a mouse Monoclonal Anti-FLAG M2-Peroxidase ( HRP ) antibody ( Sigma-Aldrich ) . For pHog1 levels , strains were inoculated overnight in YPD , subcultured to an OD600 = 0 . 1 in RPMI , and incubated to mid log phase at 30°C with shaking . To induce pHog1 , cells were incubated with 10 mM hydrogen peroxide for 10 minutes . Proteins were collected as described above , and separated on a 12% SDS gel . The samples were then transferred to a nitrocellulose membrane and blocked in 5% BSA in TBS-T . pHog1 levels were detected using the anti-phospho-p38 MAPK T180/Y182 antibody ( Cell Signalling ) . Hog1 levels were detected using the Hog1 y-215 antibody ( SC-9079 Santa Cruz ) . Protein levels were normalized using an antibody against Tubulin ( AbD Serotec rat anti-tubulin MCA78G ) or Histone H3 ( Abcam 1791 ) . Protein level quantifications were performed using ImageJ , normalizing against the loading control , and relative levels were determined by comparison with the wild-type strain under the same conditions . Drug tolerance assays were performed in flat-bottom , 96-well microtiter plates ( Sarstedt ) using a modified broth microdilution protocol as previously described [10] . For target gene depletion in the tetO strains , cells were incubated overnight in 0 . 05μg/mL DOX before being assayed for drug sensitivity in the presence of 0 . 05μg/mL DOX . MIC tests were set up in a total volume of 0 . 2 ml/well with 2-fold dilutions of each drug in either RPMI or YPD , as indicated . Plates were incubated in the dark at either 30°C or 37°C , as indicated , before OD600 were determined using a spectrophotometer ( Molecular Devices ) . Each strain was tested in technical and biological replicates . MIC data were quantitatively displayed with color using the program Java TreeView 1 . 1 . 1 ( http://jtreeview . sourceforge . net ) . Error bars represent standard deviation of biological replicates . Dose response matrices were performed as previously described [50] . Fractional inhibitory concentrations were determined as previously described [51] , using the formula FIC = ( MICcombo / MICDrugA ) + ( MICcombo / MICDrugB ) . Rhodamine-phalloidin staining of actin was performed on fixed cells as previously described [15] . Briefly , cells were fixed in 4% formaldehyde after subculturing for 6 hours in RPMI at 37°C . For the GRACE strains , mutants were incubated in the presence or absence of 0 . 5 μg/mL DOX overnight before subculturing in the presence or absence of 0 . 5 μg/mL DOX . After fixation , cells were washed twice in PBS , incubated overnight in rhodamine-phalloidin at 4°C , washed twice in PBS , and imaged . Microscopic imaging of C . albicans was performed on a Zeiss Axio Imager . MI ( Carl Zeiss ) . Fluorescence microscopy of rhodamine-phalloidin was performed using an X-Cite series 120 light source with an ET HQ tetramethylrhodamine isothiocyanate ( TRITC ) /DsRED filter set from Chroma Technology ( Bellows Falls , VT ) . ASC-mCherry macrophages ( gift from Eicke Latz ) were infected at an MOI of 1:1 , as previously described [23] . Four hours post inoculation , infected macrophages were imaged using a Zeiss Axio Observer . Z1 at × 10 magnification . Pyroptosis events were determined by foci of red fluorescence , using ImageJ for quantification . Each experiment was performed in triplicate , with three biological replicates . Statistical significance ( P<0 . 05 ) was determined by unpaired t-tests . | Hsp90 is an essential and conserved molecular chaperone that is required for the folding and function of a wide range of client proteins , especially those involved in signaling and stress responses . In the human fungal pathogen Candida albicans , Hsp90 governs drug resistance , morphogenesis , and virulence . In our previous analysis , we developed a chemical genomic approach to map the HSP90 chemical genetic network in C . albicans , an organism for which classical genetic approaches are hampered by the lack of a complete sexual cycle . Here , we confirm the environmental contingency of the Hsp90 genetic network in C . albicans , identify novel genetic interactions , and demonstrate new circuitry that regulates Hsp90 functional capacity in the cell . In the context of treatment of infectious disease , the challenge in avoiding host toxicity and achieving specificity lies in either the development of pathogen-selective Hsp90 inhibitors or identifying pathogen-specific components of the circuitry through which Hsp90 governs virulence . This work provides insight into such circuitry in the leading fungal pathogen of humans . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods",
"and",
"Materials"
] | [
"medicine",
"and",
"health",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"pathogens",
"immunology",
"microbiology",
"genomic",
"library",
"screening",
"mutation",
"fungi",
"model",
"organisms",
"clinical",
"medicine",
"pharmacology",
"molecular",
"biology",
"techniques",
"genetic",
"interactions",
"hypersensitivity",
"fungal",
"pathogens",
"research",
"and",
"analysis",
"methods",
"contractile",
"proteins",
"mycology",
"antimicrobial",
"resistance",
"mutant",
"strains",
"actins",
"proteins",
"medical",
"microbiology",
"microbial",
"pathogens",
"molecular",
"biology",
"molecular",
"biology",
"assays",
"and",
"analysis",
"techniques",
"yeast",
"biochemistry",
"cytoskeletal",
"proteins",
"candida",
"clinical",
"immunology",
"library",
"screening",
"gene",
"identification",
"and",
"analysis",
"genetics",
"microbial",
"control",
"biology",
"and",
"life",
"sciences",
"yeast",
"and",
"fungal",
"models",
"drug",
"interactions",
"organisms",
"candida",
"albicans"
] | 2016 | Mapping the Hsp90 Genetic Network Reveals Ergosterol Biosynthesis and Phosphatidylinositol-4-Kinase Signaling as Core Circuitry Governing Cellular Stress |
Strains of simian immunodeficiency virus ( SIV ) that are limited to a single cycle of infection were evaluated for the ability to elicit protective immunity against wild-type SIVmac239 infection of rhesus macaques by two different vaccine regimens . Six animals were inoculated at 8-week intervals with 6 identical doses consisting of a mixture of three different envelope variants of single-cycle SIV ( scSIV ) . Six additional animals were primed with a mixture of cytoplasmic domain-truncated envelope variants of scSIV and boosted with two doses of vesicular stomatitis virus glycoprotein ( VSV G ) trans-complemented scSIV . While both regimens elicited detectable virus-specific T cell responses , SIV-specific T cell frequencies were more than 10-fold higher after boosting with VSV G trans-complemented scSIV ( VSV G scSIV ) . Broad T cell recognition of multiple viral antigens and Gag-specific CD4+ T cell responses were also observed after boosting with VSV G scSIV . With the exception of a single animal in the repeated immunization group , all of the animals became infected following an intravenous challenge with SIVmac239 . However , significantly lower viral loads and higher memory CD4+ T cell counts were observed in both immunized groups relative to an unvaccinated control group . Indeed , both scSIV immunization regimens resulted in containment of SIVmac239 replication after challenge that was as good as , if not better than , what has been achieved by other non-persisting vaccine vectors that have been evaluated in this challenge model . Nevertheless , the extent of protection afforded by scSIV was not as good as typically conferred by persistent infection with live , attenuated SIV . These observations have potentially important implications to the design of an effective AIDS vaccine , since they suggest that ongoing stimulation of virus-specific immune responses may be essential to achieving the degree of protection afforded by live , attenuated SIV .
The search for a safe and effective AIDS vaccine continues . While live , attenuated strains of SIV afford reliable long-term protection in animal models , at least against closely related challenge viruses , they have the potential to regain a pathogenic phenotype through the accumulation of compensatory genetic changes over prolonged periods of persistent replication in vivo [1]–[7] . Hence , there are legitimate safety concerns with the use of live , attenuated HIV-1 as a vaccine approach in people . Vaccine candidates based on recombinant DNA and/or viral vectors are safer and elicit potent cellular immune responses that effectively control virus replication after challenge with the simian-human immunodeficiency virus chimera SHIV89 . 6P [8]–[11] . However , these vaccines afford only modest protection against SIV challenge strains , such as SIVmac239 and SIVmac251 , that express neutralization-resistant , CCR5-tropic envelope glycoproteins typical of most primary HIV-1 field isolates [12]–[17] . The predictive validity of the more rigorous SIV challenge model as an indicator of vaccine efficacy in humans was recently supported by the failure of a replication-defective , recombinant adenovirus type 5 ( rAd5 ) vaccine candidate to protect against HIV-1 infection in a high profile clinical trial [18]–[23] . In a phase IIb proof-of-concept trial , nearly 3000 participants were immunized at 0 , 1 and 6 months with rAd5 vectors expressing HIV-1 clade B gag , pol and nef genes , or a placebo control [18] , [19] . The trial was halted after the data safety monitoring board , at its first interim analysis , determined that the vaccine not only failed to prevent infection , but failed to reduce viral loads in immunized individuals who later became infected [18] , [19] . These disappointing results have further diminished optimism that similar vaccine approaches might provide better protection in future trials [18] , [19] , [24] . Thus , there is an urgent need to continue to pursue innovative vaccine concepts that may afford more promising safety and efficacy profiles . It is presently unclear whether persistent , low-level virus replication , and associated stimulation of virus-specific immune responses , is a prerequisite for the robust protection afforded by infection of animals with live , attenuated strains of SIV . As an experimental AIDS vaccine approach designed to uncouple immune activation from ongoing virus replication and turnover of CD4+ lymphocytes , we and others have developed genetic systems for producing strains of SIV that are limited to a single cycle of infection [25]–[27] . Our approach is based on a two-plasmid system for producing Gag-Pol-complemented SIV with mutations in a cis-acting sequence required for ribosomal frameshifting between the gag and pol reading frames [25] , [26] . One plasmid carries a full-length proviral genome with mutations in the gag-pol-frameshift site to prevent Pol translation , and a second plasmid carries a Gag-Pol-expression construct that also has mutations in the frameshift site [25] , [26] . Co-transfection of both plasmids into the same cells results in the release of Gag-Pol-complemented virus particles that package a pol-deficient viral genome . Cells infected with this virus express all of the viral gene products except Pol and release immature virions that cannot complete subsequent rounds of infection [25] , [26] . Since neither construct retains a functional frameshift site , high-titer stocks of single-cycle SIV ( scSIV ) can be produced without generating replication-competent virus through recombination [25] , [26] . In preliminary studies , viral RNA loads reflecting the production of non-infectious progeny virus by scSIV-infected cells were measurable in plasma after inoculation of macaques with concentrated doses of scSIV [27]–[29] . Progressive reductions in these transient peaks of viremia were observed after three successive doses , and the rank order of peak viral loads after the last dose of scSIV was the same as the rank order of peak viral loads after challenge [28] . These observations suggested that protective immunity might improve with repeated immunization and that the ability to contain scSIV infection after repeated immunizations might ultimately be predictive of the ability to contain wild-type SIV infection after challenge; a hypothesis we have termed “immunization to extinction . ” Protective immunity may also be improved by maximizing the stimulation of virus-specific T cell responses . While a number of factors can influence the development of T cell memory , results from murine systems indicate that the size of the memory T cell population is ultimately determined by the number of activated T cells driven to proliferate during the process of clonal expansion [30] , [31] . All other factors being equal , the magnitude of antigen presentation largely determines the extent of T cell activation . Thus , we reasoned that approaches designed to maximize the infectivity of scSIV , and the frequency of scSIV-infected antigen presenting cells in immunized animals , should increase both the size and longevity of the memory T cell population . In the present study , we compared two different immunization regimens with scSIV for the ability to contain virus replication after an intravenous challenge with SIVmac239 . One group of animals was inoculated with six identical doses of the same cryopreserved stocks of scSIV to determine if repeated immunization would promote the maturation of virus-specific immune responses . A second group of animals was primed with scSIV strains expressing envelope glycoproteins with truncated cytoplasmic tails and boosted with VSV G trans-complemented scSIV to maximize infection , antigen presentation and the stimulation of virus-specific T cell responses . Despite differences in the magnitude of virus-specific T cell responses elicited , both immunization regimens resulted in statistically significant reductions in viral loads and better preservation of memory CD4+ T cell subsets after challenge compared to unimmunized control animals .
Rhesus macaques were immunized with single-cycle SIV by two different vaccine regimens to compare the effects of frequency of inoculation versus infectious dose of the inoculum on the development of protective immunity . One group of animals ( Group A ) was repeatedly immunized with the same dose of scSIV to determine if mimicking persistent infection through repeated antigenic stimulation would lead to the progressive maturation of virus-specific immune responses and incremental reductions in single-cycle viral loads predictive of the ability to contain wild-type SIV replication after challenge . A second group of animals ( Group B ) was immunized by a prime and boost regimen designed to maximize scSIV infection , antigen presentation and the stimulation of virus-specific T cell responses to determine if higher overall T cell responses might result in better control of SIV infection after challenge . Three different envelope variants of single-cycle SIV expressing full-length ( TMopen ) or truncated ( TMstop ) forms of the 239 , 316 and 155T3 envelope glycoproteins were used for immunization [32] . The 239 envelope uses CCR5 as a co-receptor for infection of predominantly memory CD4+ T cells [32] , [33] . The 316 envelope also uses CCR5 , but differs from the 239 envelope by 6 amino acids in gp120 which result in a 50- to 100-fold enhancement in infectivity for primary macrophages in culture [25] , [32] , [34] . The 155T3 envelope , which differs from the 239 envelope by 22 amino acids in gp120 , uses CXCR4 rather than CCR5 as a co-receptor for infection of both naive and memory CD4+ T lymphocytes [32] , [33] . These three envelopes were selected to ensure infection of a diverse population of antigen presenting cells by scSIV and to potentially broaden envelope-specific antibody responses . For each of these envelope variants , TMstop strains of scSIV were created by introducing a glutamic acid to stop-codon change at position 767 ( E767* ) in the gp41 cytoplasmic tail . The E767* mutation was present in the original isolate of SIVmac316 and may represent a naturally selected change to facilitate virus replication in macrophage [34] . Truncation of the gp41 cytoplasmic tail at this position was also shown to increase envelope glycoprotein incorporation into virions and to enhance virus infectivity [32] , [35] , [36] . To facilitate the stimulation of virus-specific CD8+ T cell responses , mutations in nef that eliminate residues required for MHC class I downregulation were also included in each strain [28] , [37] , [38] . In Group A , six animals were immunized intravenously at 8-week intervals with 6 identical doses of the same cryopreserved stocks of scSIV . Each dose contained a mixture of three scSIV strains expressing full-length envelope glycoproteins , scSIVmac239 TMopen , scSIVmac316 TMopen and scSIVmac155T3 TMopen ( Figure 1A ) [32] . In Group B , six additional animals were primed intravenously with a mixture of the envelope cytoplasmic tail-truncated strains , scSIVmac239 TMstop , scSIVmac316 TMstop and scSIVmac155T3 TMstop . The animals in Group B were then boosted intravenously with scSIV trans-complemented with the vesicular stomatitis virus glycoprotein ( VSV G ) on weeks 12 and 24 ( Figure 1B ) . Although it would have been possible to trans-complement any of the three envelope variants of scSIV with VSV G , only scSIVmac239 TMopen was used for these booster inoculations . The selection of scSIVmac239 TMopen was based on an effort to focus strain-specific antibody responses on the envelope glycoprotein of SIVmac239 . However , to prevent neutralization of the second boost by VSV G-specific antibodies , two different serotypes of VSV G were used for each inoculation [39] . The first boost was trans-complemented with the Indiana serotype of VSV G ( VSV GI scSIV ) and the second boost was trans-complemented with the New Jersey serotype of VSV G ( VSV GNJ scSIV ) . Plasma viral RNA loads were measured independently for each envelope variant of scSIV using a quantitative , multiplex , real-time RT-PCR assay specific for unique sequence tags cloned into the pol-locus of each strain [32] . Measurable levels of virion-associated viral RNA were transiently detectable in plasma after each dose ( Figure 2 ) . Since SIV particles are rapidly cleared from plasma with an estimated half-life of a few minutes [40]–[42] , these viral load measurements reflect the ongoing release of non-infectious particles from scSIV-infected cells . In Group A , viral loads peaked on day 4 after the first inoculation and declined below the limit of detection ( <30 RNA copy eq . /ml ) within 3 to 4 weeks ( Figure 2A ) . Geometric mean peak viral loads after the first dose were 1 . 34×104 , 1 . 49×103 and 1 . 48×104 RNA copy eq . /ml for scSIVmac239 , scSIVmac316 and scSIVmac155T3 respectively . After the second dose , peak viral loads occurred on day 2 and were 40-fold lower for scSIVmac239 ( P = 0 . 001 ) , 25-fold lower for scSIVmac316 ( P<0 . 001 ) and 50-fold lower for scSIVmac155T3 ( P<0 . 001 ) ( Figure 2A ) . Since the inocula were identical for each dose ( same volume of the same cryopreserved stocks ) , the lower peak and more rapid clearance of viremia after the second inoculation almost certainly reflects the elimination of scSIV-infected cells by virus-specific immune responses . Further reductions in peak viremia were also observed after the fifth and sixth inoculations for scSIVmac316 ( P = 0 . 008 and P = 0 . 02 ) , but not for scSIVmac239 or scSIVmac155T3 ( P>0 . 05 for all comparisons ) ( Figure 2A ) . Since SIVmac316 is considerably more sensitive to neutralizing antibodies than SIVmac239 or SIVmac155T3 , better containment of scSIVmac316 may reflect the affinity maturation of envelope-specific neutralizing antibodies [43] , [44] . However , this was not supported by the results of neutralization assays , since only one of the six animals in this group had detectable neutralizing antibody titers to SIVmac316 at the time of challenge . Therefore , other factors , perhaps related to the lower infectivity of scSIVmac316 , may have contributed to the better containment of viremia for this strain . In Group B , viral loads also peaked on day 4 after the first inoculation , and with the exception of scSIVmac155T3 TMstop , were resolved below the limit of detection within 3 to 6 weeks . Consistent with the infectivity enhancement associated with truncation of the gp41 cytoplasmic tail [35] , [36] , peak viral loads were 4 . 1- , 6 . 6- and 56-fold higher after the first dose for TMstop versus TMopen scSIVmac239 , scSIVmac155T3 and scSIVmac316 [32] . A delay in the clearance of viremia was also observed for scSIVmac155T3 TMstop , particularly in animals Mm 350-03 and Mm 465-02 ( Figure 2B ) . The more prolonged period of viremia for scSIVmac155T3 TMstop was significant based on area under the curve comparisons with scSIVmac239 TMstop and scSIVmac316 TMstop [32] , and may reflect preferential infection of a CD4+CXCR4+ target cell population that is less susceptible to the cytopathic effects of infection or more resistant to clearance by virus-specific immune responses . Trans-complementation of scSIV with VSV G resulted in a dramatic infectivity enhancement , presumably by enabling CD4− and chemokine receptor-independent entry of scSIV into cell types that are not normally permissive for SIV infection ( Figure 2B ) . Two days after boosting with VSV GI scSIV , peak viremia was 1 . 37×105 RNA copy eq . /ml . Relative to the first two doses of non-trans-complemented scSIVmac239 TMopen in Group A , this represents a 10-fold increase over the first peak ( P = 0 . 007 ) and a 415-fold increase over the second peak ( P<0 . 001 ) . The infectivity enhancement for VSV GNJ scSIV was less dramatic , but was still 4-fold higher than the second non-transcomplemented dose of scSIVmac239 TMopen in Group A ( P = 0 . 03 ) . SIV-specific CD8+ T cell responses to immunodominant epitopes in Gag , Tat and Nef were monitored in Mamu-A*01 and -A*02 positive animals by directly staining peripheral blood with MHC Class I tetramers . CD8+ T cell responses were detectable in Group A animals after the first two inoculations . However , subsequent rounds of inoculation failed to boost CD8+ T cell frequencies above the threshold of detection ( Figure 3A ) . Longitudinal analysis of virus-specific T cell responses by IFNγ ELISPOT assays revealed a similar pattern . Primary responses in Group A peaked within 2 to 3 weeks after the first inoculation ranging from 165 to 308 ( mean 261 ) spot-forming cells ( SFC ) per million PBMC to Gag and from 198 to 635 ( mean 327 ) SFC per million PBMC to Env ( Figure 4A and 4B ) . While recall responses of comparable magnitude were observed after the second inoculation , subsequent rounds of re-inoculation did not result in additional increases in these virus-specific T cell frequencies ( Figure 4A and 4B ) . Thus , despite detectable plasma viremia confirming the take of infection with each dose of scSIV , repeated inoculation did not drive additional expansion of virus-specific T cell responses . Primary SIV-specific CD8+ T cell responses in Group B were similar to primary responses in Group A . However , these responses increased dramatically after boosting with VSV G trans-complemented scSIV . One week after the first boost , the percentages of SIV-specific CD8+ T cells in peripheral blood increased 7- to 33-fold for Mamu-A*02 Nef159–167 ( 2 . 8% to 3 . 7% CD8+ T cells ) , 7- to 13-fold for Mamu-A*01 Gag181–189 ( 2 . 0% to 2 . 8% CD8+ T cells ) and 12- to 14-fold for Mamu-A*01-Tat28–35 ( 0 . 55% to 0 . 61% CD8+ T cells ) ( Figure 3B ) . Additional recall responses were observed after the second boost . However , with the exception of one animal that made a Mamu-A*01 Gag181–189-specific response that exceeded 13% of the circulating CD8+ T cell population ( Figure 3B ) , these responses were generally lower reflecting the lower take of infection for VSV GNJ scSIV than for VSV GI scSIV ( Figure 2B ) . Nevertheless , the majority of these CD8+ T cell responses remained above the threshold of detection until the time of challenge twelve weeks later , indicating the establishment of a memory CD8+ T cell population . Longitudinal analysis of IFNγ T cell responses in Group B also demonstrated significant expansion of virus-specific T cell responses after boosting with VSV G trans-complemented scSIV . Peak primary responses ranged from 122 to 506 ( mean 273 ) SFC per million PBMC to Gag and from 105 to 341 ( mean 230 ) SFC per million PBMC to Env ( Figure 4C and 4D ) . One week after the first boost , Gag-specific responses were 5 . 5-fold higher ( mean 1496 , range 350 to 2595 SFC per million PBMC , P<0 . 001 ) and Env-specific responses were 2 . 3-fold higher ( mean 518 , range 345 to 957 SFC per million PBMC , P = 0 . 007 ) than peak primary responses . Additional recall responses were observed after the second boost ( Figure 4C and 4D ) , and consistent with MHC class I tetramer staining , the animal with the highest IFNγ T cell response to Gag was the same animal with the unusually high frequency of Mamu-A*01 Gag181–189-specific CD8+ T cells at week 25 . Whole-proteome IFNγ ELISPOT assays also revealed broad T cell recognition of each of the viral gene products expressed by scSIV . The distribution of IFNγ T cell responses to each viral antigen for the animals in Group B is illustrated at week 13 ( Figure 5A ) and at week 25 ( Figure 5B ) . Although there was considerable animal-to-animal variation in the pattern of responses reflecting the MHC diversity of these outbred animals , the distribution of these responses was relatively stable after each boost . Dominant T cell responses were directed against Gag in three animals ( Mm 328-02 , Mm 350-03 and Mm 383-03 ) , Env in one animal ( Mm 295-00 ) and Nef in another ( Mm 465-02 ) . Only one animal ( Mm 512-02 ) exhibited a notable shift from a diverse , and relatively even distribution of responses to Gag , Nef , Vpr and Vpx after the first boost , to a predominantly Gag-specific response after the second boost ( Figure 5 ) . Hence , these results confirm the activation of broad T cell responses capable of targeting each of the 8 viral gene products expressed by scSIV . Virus-specific CD4+ T cell responses were also detectable after boosting with VSV G trans-complemented scSIV . PBMC were stimulated with a pool of overlapping Gag peptides in tubes coated with co-stimulatory antibodies to CD28 and CD49d , and CD4+ T cells expressing TNFα and CD69 were detected by intracellular cytokine staining according to methods described by Gauduin et al . [45] . Gag-specific CD4+ responses ranged from 0 . 10% to 0 . 27% ( mean 0 . 17% CD4+ T cells ) two weeks after the first boost ( Figure 6A ) . Increased Gag-specific CD4+ T cell frequencies were also observed in four of the six animals after the second boost ( Figure 6B ) . Virus-specific CD4+ T cell responses are often weak or undetectable in HIV-1 infected people and SIV-infected animals due to ongoing virus replication and turnover of CD4+ lymphocytes , [46] , [47] . Thus , the detection of CD4+ T cell responses following immunization with strains of SIV that are limited to a single round of infection may reflect the uncoupling of CD4+ T cell activation from ongoing infection and destruction of these cells . Antibody responses to SIV were monitored longitudinally by whole-virus ELISA . Plasma samples were tested at the time of each inoculation and four weeks later for antibodies capable of binding to plates coated with a lysate of purified virus particles . SIV-specific antibody responses were detectable in all of the animals by four weeks after the first inoculation ( Figure 7 ) . In Group A , binding antibody responses waxed and waned with each inoculation , but did not show an overall increase in titer ( Figure 7A ) . In Group B , significant increases in virus-specific antibody titers were observed after the first boost at week 12 ( P<0 . 001 ) followed by similar responses after the second boost at week 24 ( P = 0 . 04 ) ( Figure 7B ) . At the time of challenge , neutralizing antibody titers were measured against four different strains of SIV; a neutralization-sensitive , lab-adapted strain of SIVmac251 ( SIVmac251LA ) , and three primary isolates , SIVmac316 , SIVmac155T3 and SIVmac239 , matched in envelope with the scSIV strains included in the inoculum . Plasma samples from three of the animals in Group A and four of the animals in Group B neutralized SIVmac251LA ( <50% SEAP activity ) at titers >80 ( Figure 7C ) . Thus , both immunization regimens elicited antibodies capable of binding to the native , oligomeric conformation of envelope as it exists on virions . However , little or no neutralization was observed for the three primary isolates . While plasma samples from a couple of the animals in each group had measurable neutralizing activity against SIVmac316 and SIVmac155T3 at the lowest dilutions tested , none of the animals made detectable neutralizing antibody responses against SIVmac239 ( Figure 7C ) . The inability to detect neutralizing antibodies to SIVmac239 is not particularly surprising given the inherent resistance of the 239 envelope glycoprotein to antibody-mediated neutralization , even with plasma from animals persistently infected with this virus [44] , [48] . Plasma samples from Group B animals were also monitored for neutralizing antibody titers to VSV G . Ten-fold dilutions of plasma were tested for the ability to inhibit infection of CEM×174 cells by an env-deficient strain of SIV that was pseudotyped with either the Indiana or the New Jersey serotype of VSV G . While none of the animals had neutralizing antibody titers against either serotype at the time of the first boost ( week 12 ) , some non-specific inhibition of infectivity was observed at the lowest dilution of plasma tested ( Figure 8 ) . This effect was greater for virus pseudotyped with the New Jersey serotype than for virus pseudotyped with the Indiana serotype of VSV G , an observation that may account for the lower peak of viremia for VSV GNJ scSIV than for VSV GI scSIV ( Figure 2B ) . Nevertheless , four weeks after boosting with VSV GI scSIV ( week 16 ) , plasma samples from each of the six animals neutralized virus pseudotyped with the Indiana serotype of VSV G , but not the New Jersey serotype of VSV G ( Figure 8 ) . Conversely , four weeks after boosting with VSV GNJ scSIV ( week 28 ) , neutralizing antibody titers to the New Jersey serotype of VSV G ( 50% neutralization titer >500 ) were detectable in plasma from all six animals at a time when neutralizing antibody titers to the Indiana serotype were waning ( Figure 8 ) . These results are consistent with previous studies demonstrating that the Indiana and the New Jersey serotypes of VSV G do not elicit cross-reactive neutralizing antibodies in animals infected with VSV glycoprotein exchange vectors and validate our decision to change VSV G serotypes for each boost [10] , [39] . Twelve weeks after the last inoculation , each of the immunized animals in Groups A and B , and four unvaccinated control animals ( Group C ) , were challenged intravenously with 10 animal infectious doses of SIVmac239 . With the exception of a single animal in Group A ( Mm 416-02 ) , all of the animals became infected ( Figure 9A ) . This animal may in fact have been fully protected , since the same stock of SIVmac239 has been used extensively by our group , and by others , without a single failure to establish infection in an unvaccinated control animal by the intravenous route of challenge [12] , [28] , [49] . However , we could not identify any differences in the immune responses of Mm 416-02 that might account for its resistance to infection . Furthermore , single-cycle viral RNA load measurements after each dose of scSIV were similar to the other animals in Group A suggesting that there was no inherent genetic barrier to infection of this animal . Therefore , to avoid inappropriately biasing our interpretation of the outcome of challenge , this animal was not included in the statistical analysis of post-challenge viral loads and CD4+ T cell counts . Statistically significant reductions in viremia were observed in both immunized groups relative to the control group in both acute and chronic phases of infection . Area under the curve ( AUC ) analysis revealed significant containment of total viral loads for Group A ( P = 0 . 03 ) and for Group B ( P = 0 . 04 ) ( Figure 9 ) . Peak viral loads measured within the first 2 weeks post-challenge were 12-fold lower for Group A ( P = 0 . 005 ) and 16-fold lower for Group B ( P = 0 . 003 ) , and set-point viral loads measured weeks 12–16 post-challenge were 52-fold lower for Group A ( P = 0 . 009 ) and 33-fold lower for Group B ( P = 0 . 05 ) . In contrast to previous vaccine studies in which viral loads in immunized and control animals were indistinguishable by 40 weeks post-challenge [12] , [13] , these differences in chronic phase viral loads were stable for more than one year after infection ( Figure 9 ) . Indeed , comparisons of the geometric means of viral loads for each group over weeks 12–56 post-challenge using a linear mixed model analysis indicated that these chronic phase differences were statistically significant for both Group A ( P = 0 . 015 ) and B ( P = 0 . 014 ) . Thus , both immunization regimens afforded significant containment of virus replication after an intravenous challenge with SIVmac239 . Differences in the loss of memory CD4+ T cells were also observed for immunized versus control animals . There were no differences in the decline of total or naive CD4+ T cells ( P = 0 . 7 and P = 0 . 2 respectively ) ( Figure 10A–10D ) . However , AUC analysis indicated significantly better preservation of the memory CD4+ T cell subsets in each of the immunized groups ( Figure 10E–10J ) . The central , effector and CCR5+ memory CD4+ T cell populations were all significantly higher in Group A than in Group C ( P = 0 . 03 , P = 0 . 03 and P = 0 . 01 respectively ) . Likewise , the central and CCR5+ memory CD4+ T cell populations were significantly higher in Group B than in Group C ( P = 0 . 05 and P = 0 . 03 ) . Only the comparison of effector memory CD4+ T cell counts in Group B versus Group C fell short of statistical significance ( P = 0 . 06 ) . These results are consistent with the preferential infection and turnover of memory CD4+ T cells by SIVmac239 [33] , and provide additional evidence of partial protection by each immunization regimen . To identify potential correlates of protection , associations between the immune responses elicited by scSIV and post-challenge viral loads were also explored by linear regression analysis . IFNγ T cell responses to Gag and Env , at peak and at the time of challenge , were compared to peak and set-point viral loads after challenge . Neutralizing antibody titers to SIVmac251LA and binding antibody responses to whole-virus at the time of challenge were also compared to peak and set-point viral loads after challenge . Additional associations were tested between peak single-cycle viremia for each strain of scSIV and peak viral loads during the acute phase of infection . None of these comparisons revealed significant correlations between parameters measured during the immunization phase of the study and viral loads post-challenge . Therefore , similar to previous studies with live , attenuated SIV [50] , no correlations were observed between immune responses elicited by scSIV and the outcome of challenge .
More than 25 years after the emergence of the global HIV-1 pandemic , the development of a safe and effective AIDS vaccine remains an elusive scientific challenge . Although passive transfer studies have demonstrated that high concentrations of broadly neutralizing antibodies in plasma can provide sterilizing immunity to SHIV infection of macaques [51]–[54] , it is presently unclear how to elicit such antibody responses by vaccination . Thus , efforts of the vaccine community have focused on T cell-based vaccines designed to reduce viral loads in immunized individuals who later become infected with the goal of reducing the incidence of HIV-1 transmission to new hosts . This strategy is supported by epidemiological data showing that the risk of heterosexual HIV-1 transmission is directly related to viral loads of the donor , and that individuals with plasma viral loads less than 1500 RNA copies/ml rarely transmit their infections to their partners [55] . Several recombinant vaccine candidates have been developed based on this premise that are designed to stimulate potent cellular immune responses [8]–[11] , [16] , [17] . However , the extent of protection achieved by these vaccines has so far been disappointing [12]–[19] . Live , attenuated strains of SIV afford more reliable protection in animals , often achieving sterilizing immunity against closely related challenge viruses [1]–[3] . However , due to the potential to regain a pathogenic phenotype during persistent replication in vivo , there are justifiable safety concerns with the use of live , attenuated HIV-1 as a vaccine approach in humans [4]–[6] . To uncouple the activation of virus-specific immune responses from persistent virus replication , we devised a genetic system for producing strains of SIV that are limited to a single round of infection [25] , [28] . Single-cycle SIV retains many of the potentially advantageous properties of live , attenuated SIV , including the expression of 8 of the 9 viral antigens , the absence of any vector-derived gene products , and the expression of the native , oligomeric conformation of envelope on the surface of infected cells and virions . However , unlike attenuated vaccine strains , scSIV is not replication-competent and cannot revert to a pathogenic phenotype as a result of the accumulation of compensatory genetic changes during persistent replication in vivo . Contrary to our initial hypothesis of “immunization to extinction , ” repeated immunization with scSIV did not lead to the progressive maturation of virus-specific immune responses or successive decreases in single-cycle viremia . While SIV-specific T cell responses were detectable by MHC class I tetramer staining and IFNγ ELISPOT assays after the first and second doses , subsequent rounds of inoculation did not result in additional recall responses . These results are similar to previous observations [28] , but are nevertheless puzzling given that viral RNA load measurements in plasma indicate a consistent take of infection after each inoculation . One possibility is that the T cell responses elicited by the first two doses were sufficient to resolve later rounds of scSIV infection with faster kinetics , thereby curtailing additional T cell activation and proliferation . Alternatively , repeated immunization may have had a tolerizing effect , perhaps by eliciting regulatory CD4+ T cell responses or inducing a state of clonal anergy . While the immunological basis for this phenomenon is presently unclear , the mechanism ( s ) responsible for permitting a certain level of scSIV infection without a detectable expansion of virus-specific T cell responses may ultimately account , at least in part , for the inability of natural immunity to resolve wild-type HIV-1 or SIV infection . In contrast , boosting with VSV G trans-complemented scSIV dramatically increased virus-specific T cell frequencies . One week after the first boost , SIV-specific CD8+ T cell frequencies to immunodominant epitopes in Gag , Tat and Nef ranged from 0 . 55% to 3 . 7% of the CD8+ T cell population in peripheral blood . These responses were more than 10-fold higher than peak responses elicited after the initial priming dose , and are comparable to peak CD8+ T cell responses after boosting with recombinant poxviral or adenoviral vectors [8] , [11]–[13] . These responses are also similar to Gag181–189-specific CD8+ T cell frequencies following the acute phase of SIVmac239 infection , which typically range from 0 . 5% to 10% of circulating CD8+ T cells [56] . Since the particle doses of VSV G trans-complemented and non-trans-complemented scSIV were similar for each inoculation ( ranged from 10 to 15 µg p27 ) , the greater magnitude of T cell responses after boosting with VSV G scSIV presumably reflects increased virus infectivity , and hence greater numbers of scSIV-infected antigen presenting cells after immunization . However , we cannot exclude the possibility that VSV G may also have facilitated scSIV entry into dendritic cells , or other professional antigen presenting cells , that potently activate T cell responses . Gag-specific CD4+ T cell responses were also detectable by intracellular cytokine staining after boosting with VSV G trans-complemented scSIV . Although most HIV-1 infected patients and SIV infected animals make virus-specific CD4+ T cell responses , these responses are usually kept in check by ongoing virus infection [46] , [57]–[59] . Higher frequencies of virus-specific CD4+ T cells have been observed in certain individuals who are better able to control HIV-1 infection and in animals persistently infected with live , attenuated strains of SIV [57] , [59] . These observations suggest that virus-specific CD4+ T cells may be important for controlling HIV-1 and SIV infection , or alternatively , that better containment of virus replication may reduce CD4+ T cell turnover . By uncoupling CD4+ T cell activation from ongoing infection and destruction of these cells , scSIV may facilitate the development of virus-specific CD4+ T cell responses . Since CD4+ helper T cells play a central role in the maintenance of effective antibody and CTL responses to viral infections [60]–[65] , these responses are likely to be an important component of an effective AIDS vaccine . Both immunization regimens elicited similar neutralizing antibody titers to a lab-adapted strain of SIVmac251 , verifying the induction of antibodies capable of binding to the native , oligomeric conformation of envelope as it exists on virions . However , only a couple of the animals in each group made low-titer neutralizing antibody responses to SIVmac316 and SIVmac155T3 , and none of the animals had detectable neutralizing antibodies to SIVmac239 . These observations reflect the greater resistance of primary isolates to neutralizing antibodies , particularly for SIVmac239 [44] , and highlight the difficulty of eliciting even strain-specific neutralizing antibodies to viruses that express neutralization-resistant envelope glycoproteins typical of naturally transmitted HIV-1 field isolates . These results also suggest that envelope-specific antibodies were probably not a significant factor in protection against SIVmac239 at the time of challenge , although they do not preclude a role for anamnestic antibody responses in the control of virus replication after infection . Following an intravenous challenge with SIVmac239 , statistically significant reductions in viral loads and higher memory CD4+ T cell counts were observed for both groups of scSIV immunized animals . Relative to the unvaccinated control group , peak and set-point viral loads were 12- and 52-fold lower for the animals immunized by repeated inoculation ( P = 0 . 005 and P = 0 . 009 ) , and 16- and 33-fold lower for the animals immunized by the prime and boost regimen ( P = 0 . 003 and P = 0 . 05 ) . Furthermore , both groups of immunized animals maintained significantly better control of virus replication during the chronic phase of infection for more than one year after challenge . Based on prior associations of viral loads with HIV-1 transmission rates in humans , these reductions in set-point viral loads are in a range that might be expected to have a significant impact on heterosexual transmission [22] , [55] . While encouraging , the extent of protection achieved by immunization with scSIV in these studies was not as good as typically afforded by infection with live , attenuated SIV [1]–[3] . This difference may have important implications to our understanding of the mechanisms of protection by live , attenuated vaccine strains . In contrast to scSIV immunization , which activates virus-specific immune responses that appear to decline into memory after the clearance of productively infected cells , attenuated strains of SIV continuously stimulate antibody and T cell responses as a consequence of persistent virus replication . Persistent antigenic simulation may have significant qualitative and quantitative effects on virus-specific immune responses . Animals infected with SIVmac239Δnef develop unusually high virus-specific CD4+ T cell frequencies with predominantly effector memory phenotypes [59] , presumably reflecting ongoing antigenic stimulation . These responses , which can represent up to 4–10% of circulating CD4+ lymphocytes , are considerably higher than the CD4+ T cell responses elicited by scSIV immunization in this study , or indeed by wild-type SIV infection in previous comparisons [59] . By providing a constant source of infected antigen presenting cells , attenuated vaccine strains may also maintain CD8+ T cell responses in an activated state that allows for more rapid recognition and clearance of productively infected cells . In support of this , Rollman et al . observed that Gag-specific CD8+ T cells from macaques infected with an attenuated strain of SIV degranulated within 30 minutes after stimulation versus more than 3 hours for Gag-specific CD8+ T cells elicited by vaccination with recombinant poxviral vectors [66] . In the case of antibody responses , ongoing replication by live , attenuated viruses may drive the affinity maturation of envelope-specific neutralizing antibodies . Although passive transfer of serum from animals infected with attenuated viruses has not conferred protection [67] , the time dependence of protective immunity [3] , [68] , and associated changes in the avidity and conformational dependence of envelope-specific antibodies [69] , suggest that affinity maturation of neutralizing antibody responses may contribute to protective immunity . Thus , persistent virus replication and immune activation may be essential to achieve the degree of protection afforded by live , attenuated SIV . SIVmac239 is a notoriously difficult strain to protect against by vaccination , particularly by the intravenous route of challenge . Prime and boost vaccine regimens using recombinant DNA and either poxviral or adenoviral vectors that were able to effectively contain virus replication after challenge with SHIV89 . 6P afforded little or no protection against SIVmac239 [8] , [11]–[13] . In some cases , these vaccine approaches succeeded in reducing acute phase viral loads by approximately one log , but did not provide long-term control of SIVmac239 replication during the chronic phase of infection [12] , [13] . Similar reductions in acute phase viral loads without long-term control of virus replication have also been observed when protection was assessed by challenging with uncloned , pathogenic SIVmac251 [16] , [17] . In one study , better containment of SIVmac239 replication during the chronic phase of infection was achieved by vaccination of animals with a multivalent DNA-prime , rAd5-boost approach [14] . However , all of the animals in this study were Mamu-A*01 positive , and since Mamu-A*01 is associated with better control of SIV infection [70] , MHC class I immunogenetics may also have contributed to the better outcome of challenge in these animals [14] . Other vaccine studies that have reported more effective control of SIVmac239 replication have typically used rhesus macaques of Chinese or Burmese origin [71] , [72] . Since genetic evidence suggests that Chinese and Indian origin rhesus macaques represent distinct populations that separated approximately 162 , 000 years ago [73] , and Chinese origin rhesus macaques have significantly lower viral loads and slower courses of disease progression after SIVmac239 or SIVmac251 infection than Indian origin rhesus macaques [74] , [75] , genetic differences between these two populations may account for the better vaccine protection observed in Chinese origin rhesus macaques . Thus , with the notable exception of live , attenuated SIV , the control of SIVmac239 replication achieved in this study by immunization with scSIV was at least as good , if not better than , other vaccine approaches that have been evaluated in this challenge model . A better understanding of the differences in immune responses elicited by immunization of macaques with single-cycle SIV versus live , attenuated SIV may provide important insights into the design of more effective vaccines for protection against HIV-1 .
All of the animals used for these studies were Indian origin rhesus macaques ( Macaca mulatta ) . These animals were housed in a biosafety level 3 containment facility at the New England Primate Research Center ( NEPRC ) and were maintained in accordance with standards of the Association for Assessment and Accreditation of Laboratory Animal Care and the Harvard Medical School Animal Care and Use Committee . Animal experiments were approved by the Harvard Medical Area Standing Committee on Animals and conducted according to the principles described in the Guide for the Care and Use of Laboratory Animals [76] . All the animals selected for this study were negative for simian retrovirus type D , SIV , simian T lymphotrophic virus type 1 and simian herpesvirus B . All of the animals in this study were typed for the rhesus macaque MHC class I alleles Mamu-A*01 , -A*02 , -A*08 , -A*11 , -B*01 , -B*03 , -B*04 , -B*08 , -B*17 and -B*29 . The MHC class I alleles present in each animal are summarized in Table 1 . Since Mamu-A*01 is associated with better control of SIV infection [70] , the number of Mamu-A*01 positive animals assigned to each group was balanced to avoid a genetic bias in the outcome of challenge due to overrepresentation of this allele . Animals that were positive for both Mamu-A*01 and -B*17 , or Mamu-A*01 and -B*08 , and thus more likely to spontaneously control SIV infection independent of vaccination [70] , [77]–[79] , were excluded . MHC class I typing was performed by allele-specific PCR from genomic DNA using the primers and reaction conditions described by Kaizu et al . [80] . These assays were performed in Dr . David Watkins' laboratory at the Wisconsin National Primate Research Center ( WNPRC , Madison , WI ) . At the time that this manuscript was prepared , seven of the sixteen animals in this study had been euthanized with symptoms of AIDS . The week each animal died post-challenge , the experimental group the animal was assigned to , and the clinical conditions of the animal at the time of euthanasia are as follows . Mm 370-03 ( Group C ) was euthanized at week 27 post-challenge due to weight loss , persistent epistaxis and increased respiratory effort . Mm 465-02 ( Group B ) was euthanized at week 30 post-challenge with enlarged lymph nodes due to a progressive history of weight loss and diarrhea . Mm 350-03 ( Group B ) was euthanized at 49 weeks post-challenge due to periorbital edema , an enlarged spleen , an enlarged liver and agitated behavior . Mm 349-03 ( Group C ) was euthanized at 51 weeks post-challenge due to diarrhea and moderate weight loss with enlarged spleen and lymph nodes . Mm 377-03 ( Group A ) was euthanized at week 61 post-challenge with generalized lymphadenopathy and an enlarged spleen . Mm 295-00 ( Group B ) was euthanized at week 65 post-challenge with an enlarged spleen , lymphadenopathy and neurological symptoms . Mm 364-03 ( Group A ) was euthanized at week 68 post-challenge due to weight loss , diarrhea and lethargy . Six different envelope variants of scSIV were constructed as previously described [32] . These included strains expressing full-length ( TMopen ) and truncated ( TMstop ) forms of the 239 , 316 and 155T3 envelope glycoproteins . The TMstop strains contained a glutamic acid to stop-codon change at position 767 ( E767* ) that truncates the cytoplasmic domain of gp41 and results in increased envelope incorporation into virions and increased virus infectivity [35] . Unique 71–74 bp sequence tags selected from the Arabidopsis thaliana genome were introduced into the pol locus to allow independent quantification of viral RNA loads by real time RT-PCR analysis for each strain after mixed inoculation [32] . To facilitate the stimulation of virus-specific CTL responses , 26 residues from the C-terminus of the Nef protein that are required for MHC class I down regulation were eliminated by a glycine to stop-codon change at position 238 of Nef ( G238* ) followed by two single-nucleotide deletions ( Δ9791 and Δ9797 ) [28] , [37] . Single-cycle SIV was produced by co-transfection of 293T cells with the Gag-Pol expression construct pGPfusion and a full-length proviral DNA construct for each scSIV strain as previously described [25] , [28] , [32] . 293T cells were seeded at 5×106 cell per 100-mm dish in cell culture medium ( Dulbecco's modified Eagle's medium [DMEM] supplemented with 10% fetal bovine serum [FBS] , L-glutamine , penicillin and streptomycin ) and transfected the following day with 5 µg of each plasmid using TransFectin™ Lipid Reagents ( Bio-Rad , Hercules , CA ) . To produce VSV G trans-complemented scSIV , 5 µg of an expression construct for the Indiana or the New Jersey serotype of VSV G was included in the transfection mix . The cDNA clone for the New Jersey serotype of VSV G was kindly provided by Dr . John Rose ( Yale University School of Medicine , New Haven , CT ) . Twenty-four hours after transfection , the plates were rinsed twice with serum-free medium and the cell culture medium was replaced with DMEM supplemented with 10% rhesus serum ( Equitech-Bio , Kerrville , TX ) . Twenty-four hours later , the cell culture supernatant was collected and concentrated by repeated low speed centrifugation in YM-50 ultrafiltration units ( Millipore , Bedford , MA ) . One-milliliter aliquots of scSIV were cryopreserved at −80°C and the concentration of virus was determined by SIV p27 antigen capture ELISA ( Advanced BioScience Laboratories , Kensington , MD ) . One million CEM×174 cells were incubated with 100 ng p27 equivalents of scSIV in 100 µl volume for 2 hours at 37°C . Cultures were then expanded to a volume of 2 ml in R10 medium ( RPMI supplemented with 10% FBS , L-glutamine , penicillin and streptomycin ) and incubated in 24-well plates at 37°C for 4 days . Cells were treated with Caltag Fix and Perm reagents ( Caltag Laboratories , Burlingame , CA ) and stained with the FITC-conjugated SIV Gag-specific monoclonal antibody 2F12 ( provided by the DAIDS/NIAID Reagents Resource Support Program for AIDS Vaccine Development , under contract to Quality Biological , Inc . and Bio-Molecular Technology , Inc . ; Principal Investigator , Ronald Brown; Project Officer , Jon Warren ) . After staining , cells were fixed in 2% paraformaldehyde PBS and analyzed by flow cytometry to determine the frequency of SIV Gag-positive infected cells . Six macaques ( Group A ) were inoculated intravenously with 6 identical doses of scSIV at 8-week intervals . Each dose contained a mixture of scSIVmac239 TMopen , scSIVmac316 TMopen and scSIVmac155T3 TMopen ( 5 µg p27 eq . of each strain ) . Six additional animals ( Group B ) were inoculated intravenously with an initial priming dose that included scSIVmac239 TMstop , scSIVmac316 TMstop and scSIVmac155T3 TMstop ( 5 µg p27 eq . of each strain ) . These animals were then boosted on weeks 12 and 24 with VSV G trans-complemented scSIVmac239 TMopen . The first boost contained 10 µg p27 eq . of scSIV trans-complemented with the Indiana serotype of VSV G ( VSV GI ) and the second boost contained 13 µg p27 eq . of scSIV trans-complemented with New Jersey serotype of VSV G ( VSV GNJ ) . For each inoculation , 2–3 ml of concentrated scSIV was injected through a 22-gauge catheter placed aseptically in the saphenous vein of ketamine-HCI anesthetized animals ( 15 mg/kg intramuscularly ) . Virus specific CD8+ T cell responses were measured in the peripheral blood of Mamu-A*01 and -A*02 positive rhesus macaques . Whole blood ( 200 µl ) was incubated for 30 min at 37°C with one of the following APC-conjugated tetramers provided by Dr . David Watkins' laboratory at the Wisconsin National Primate Research Center ( University of Wisconsin , Madison , WI ) ; Mamu-A*01-Gag181–189 , Mamu-A*01-Tat28–35 and Mamu-A*02-Nef159–167 . The samples were then stained with anti-CD3-FITC ( clone SP34 , BD Pharmingen ) and anti-CD8-PerCP ( clone SK1 , BD Biosciences ) monoclonal antibodies for an additional 30 min at room temperature . After staining , the samples were treated with FACS Lysing solution ( BD Biosciences ) to eliminate red blood cells , washed and fixed in 2% paraformaldehyde PBS . Data was collected using a FACSCalibur flow cytometer ( BD Biosciences , San Jose , CA ) and the frequency of CD8+ T cells staining with each tetramer was determined by analysis using the FlowJo software package ( Tree Star , San Carlos , CA ) . Virus-specific T cell responses were measured using enzyme-linked immunospot ( ELISPOT ) assay . PBMCs were plated at 3×105 , 1×105 and 3×104 cells per well in multiscreen 96-well plates ( Millipore ) coated with an IFNγ capture antibody ( Mabtech , Mariemont , OH ) . PBMCs were stimulated in duplicate wells with peptide pools ( 15-mers overlapping by 11 amino acids , 2 . 5 µg/ml for each individual peptide ) representing the amino acid sequences of the SIV Gag , Env , Nef , Tat , Rev , Vpr , Vpx and Vif antigens . Plates were incubated overnight at 37°C and developed using an enzyme-linked , colorimetric assay for bound IFNγ ( Mabtech ) . Spots representing IFNγ-producing T cells were enumerated using an automated ELISPOT plate reader ( Zellnet Consulting , New York , NY ) . The frequency of spot-forming cells ( SFC ) per million PBMC was calculated by subtracting the number of background spots in medium control wells from the number of spots in peptide-stimulated wells and adjusting for the input cell number . SIV-specific CD4+ T helper cell responses were detected by intracellular cytokine staining ( ICS ) as previously described [45] . Polystyrene flow tubes ( 12×75 mm ) were coated overnight at a 5° angle at 4°C with 2 . 5 µg/ml goat anti-mouse IgG ( H+L ) ( KPL , Gaithersburg , MD ) . The next day , the tubes were incubated with 10 µg/ml mouse anti-human-CD28 ( clone CD28 . 2 , BD Pharmingen ) and mouse anti-human-CD49d ( clone 9F10 , BD Pharmingen ) antibodies at 37°C for 1 hour . Fresh PBMC ( 1 . 5–2×106 ) were stimulated at 37°C with the Gag peptide pool ( 2 µg/ml ) , SEB ( 100 ng/ml ) or complete R10 medium in the presence of cross-linked co-stimulatory CD28 and CD49d antibodies . Brefeldin A ( GolgiPlug , BD Pharmingen ) was added after one hour and the incubation was continued for another five hours . After 6 hours of antigen stimulation , PBMC were surface-stained with anti-CD3-FITC ( clone SP34 , BD Pharmingen ) and anti-CD4-PerCP monoclonal antibodies ( clone L200 , BD Pharmingen ) at 4°C for 30 min . The cells were then treated with Fix & Perm reagents ( Caltag Laboratories ) and stained with anti-CD69-PE ( clone FD50 , BD Biosciences ) and anti-TNFα-APC ( clone Mab11 , BD Pharmingen ) at room temperature for 50 min . Cells were then fixed in fresh 1% paraformaldehyde PBS . Data were collected using a FACSCalibur flow cytometer collecting >200 , 000 lymphocyte events per sample and analyzed using the FlowJo software package . SIV-specific binding antibodies were detected by whole-virus ELISA . Nunc-immunoplates ( Fisher , Pittsburgh , PA ) were coated overnight at room temperature with 0 . 1 µg p27 eq/ml whole-virus lysate prepared from aldrithiol-2 inactivated SIV CP-MAC ( AIDS Vaccine Program , NCI-Frederick , Frederick , MD ) [81] . Plates were blocked with a 1∶30 dilution of Kirkegaard and Perry BSA dilute/blocking solution concentrate ( KPL , Gaithersburg , MD ) and washed once with water . Duplicate 1∶20 dilutions of plasma were incubated in pre-treated wells for 1 hour . After 3 washes , 100 µl alkaline phosphatase conjugated Goat anti-human IgG ( Fc ) ( KPL ) at a dilution of 1∶100 was added to each well for 1 hour . The plates were then washed three times and 200 µl of phosphatase substrate solution ( KPL ) was added to each well . After 30 min , the reaction was stopped by the addition of 50 µl 3N sodium hydroxide and the absorbance was read at 405 nm . Neutralizing antibody titers were measured by the ability of plasma to block infection of target cells harboring a Tat-inducible secreted-alkaline phosphatase ( SEAP ) reporter gene [43] . Serial two-fold dilutions of plasma were incubated with lab-adapted SIVmac251 ( 0 . 25 ng p27 eq ) , SIVmac316 ( 5 . 0 ng p27 eq . ) , SIVmac155T3 ( 1 . 0 ng p27 eq . ) or SIVmac239 ( 1 . 0 ng p27 eq . ) in 96-well plates at 100 µl per well . After a one-hour incubaftion , 30 , 000 C8166 SIV-SEAP ( SIVmac251LA , SIVmac239 and SIVmac155T3 ) or CEM×174-SEAP ( SIVmac316 ) cells were added in an additional 100 µl R10 medium . SEAP activity was determined in culture supernatant collected on day 3 for SIVmac251LA , SIVmac239 and SIVmac155T3 and on day 5 for SIVmac316 using the Phospha-Light SEAP detection kit ( Applied Biosystems , Foster City , CA ) . Mock-infected cells and SIV-infected cells incubated in the absence of plasma were used to determine background and maximal SEAP production respectively . After subtracting the background activity , percent neutralization was calculated by dividing the mean SEAP counts for replicate wells at each plasma dilution by the maximal SEAP counts in the absence of plasma . VSV G-specific neutralizing antibody responses were measured by testing 10-fold dilutions of plasma for the ability to inhibit infection of CEM×174 cells by a VSV G-pseudotyped , env-deficient strain of SIV that expresses EGFP from the nef-locus ( SIVmac239ΔEnvEGFP ) [82] . The expression of the SIV envelope glycoprotein by SIVmac239ΔEnvEGFP was disrupted by introducing a combination of nucleotide substitutions that changed the second and third codons of env to stop-codons and introduced a single nucleotide frameshift deletion ( ATGGGATGTCTT -> ATGTGAT-AATT ) . VSV G-pseudotyped stocks of SIVmac239ΔEnvEGFP were produced by co-transfection of 293T cells with SIVmac239ΔEnvEGFP proviral DNA and expression constructs for either the Indiana or the New Jersey serotypes of VSV G . Ten-fold dilutions of plasma ( 1/20 , 1/200 and 1/2000 ) were incubated with 20 ng p27 eq . of VSV G-pseudotyped SIVmac239ΔEnvEGFP for one hour at 37°C in 100 µl R10 medium . CEM×174 cells ( 2 . 5×105 ) were added , and the incubation was continued for an additional 2 hours in a total volume of 200 µl R10 medium . The cultures were expanded to one ml and transferred to 48-well plates . Four days later , the cells were harvested , fixed in 2% paraformaldehyde PBS and the frequency of infected EGFP+ cells was determined by flow cytometry . Percent infectivity was calculated by dividing the mean percentage of EGFP+ cells at each dilution of plasma by the mean percentage of EGFP+ cells in the absence of plasma and multiplying by 100 . The immunized and control animals were challenged intravenously with SIVmac239 . A vial of SIVmac239 challenge stock prepared in activated rhesus macaque PBMC ( provided by Dr . Ronald Desrosiers , NEPRC ) was thawed and diluted to 10 animal infectious doses ( 1 . 5 pg p27 eq . ) per ml in serum-free RPMI 30 minutes prior to challenge . Under ketamine-HCl anesthesia ( 15 mg/kg , i . m . ) , each animal received one ml of the virus dilution through a 22 g catheter placed aseptically in the saphenous vein . Virus was recovered from 0 . 5 to 1 . 5 ml plasma collected in sodium citrate anticoagulant by centrifugation and viral RNA was extracted and reverse-transcribed into cDNA as previously described [83] . Single-cycle viral loads were measured using a quantitative , multiplex , real-time RT-PCR assay specific for unique sequence tags ( ggr , cao and gsa ) carried by each strain of scSIV . The primer/probe sets and reaction conditions for this multiplex assay are described in DeGottardi et al . [32] . Post-challenge viral RNA loads for SIVmac239 were measured using a standard quantitative , real-time , RT-PCR assay based on amplification of sequences in gag [83] . The nominal threshold of detection for this assay is 25 RNA copy eq . per ml and the interassay coefficient of variation is <25% . The loss of total , naive and memory CD4+ T cell subsets was monitored after challenge . Whole blood was stained with the monoclonal antibodies CD3-FITC ( clone SP34 , BD Pharmingen ) , CD4-PerCP ( clone L200 , BD Pharmingen ) , CD95-APC ( clone DX2 , BD Pharmingen ) , and CD28-PE ( clone CD28 . 2 , BD Pharmingen ) or CCR5-PE ( clone 3A9 , BD Pharmingen ) . Erythrocytes were eliminated by treatment with FACS Lysing solution ( BD Biosciences ) and the cells were fixed in 2% paraformaldehyde PBS solution . At each time point , the total number of lymphocytes was determined by complete blood count ( CBC ) analysis . Cell counts per µl of blood for each CD4+ T cell subset were calculated by multiplying the number of lymphocytes at each time point by the percentage of total , naive ( CD28+CD95− ) , central memory ( CD28+CD95+ ) , effector memory ( CD28−CD95+ ) and CCR5+ memory ( CCR5+CD95+ ) CD4+ T cells . Paired t-tests were used to assess differences in viral RNA load measurements and virus-specific T cell responses over time in the same animals . Independent sample t-tests were used to assess differences in these variables across groups at specific time points and to test differences between groups for area under the curve comparisons . Associations between immune responses elicited by scSIV and log-transformed , post-challenge viral RNA loads in plasma at peak and at set-point ( week 12 ) were examined by linear regression analysis . Linear mixed models were also applied to analyze differences between groups in mean viral RNA load measurements during the chronic phase of infection ( weeks 12–56 post-challenge ) [84] . This method makes efficient use of all data points available and accounts for correlations between repeated measurements on the same animals . In this analysis , each animal was assumed to be completely randomized to one of three experimental groups , and thus assumed to be independent from each other . Animal-specific intercepts and group-specific slopes over time were included in the models . Viral loads and immunological variables were transformed using 10-based logarithmic whenever appropriate . Analyses were carried out using SPSS 15 . 0 software ( SPSS Inc . Chicago , IL ) and Strata MP 10 . 0 ( Strata Corp . , College Station , TX ) . | AIDS vaccine candidates based on recombinant DNA and/or viral vectors stimulate potent cellular immune responses . However , the extent of protection achieved by these vaccines has so far been disappointing . While live , attenuated strains of SIV afford more reliable protection in animal models , there are justifiable safety concerns with the use of live , attenuated HIV-1 in humans . As an experimental vaccine approach designed to uncouple immune activation from ongoing virus replication , we developed a genetic system for producing strains of SIV that are limited to a single cycle of infection . We compared repeated versus prime-boost vaccine regimens with single-cycle SIV for the ability to elicit protective immunity in rhesus macaques against a strain of SIV that is notoriously difficult to control by vaccination . Both vaccine regimens afforded significant containment of virus replication after challenge . Nevertheless , the extent of protection achieved by immunization with single-cycle SIV was not as good as the protection typically provided by persistent infection of animals with live , attenuated SIV . These observations have important implications for the design of an effective AIDS vaccine , since they suggest that ongoing stimulation of virus-specific immune responses may ultimately be necessary for achieving the robust protection afforded by live , attenuated SIV . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"virology/vaccines",
"virology/immunodeficiency",
"viruses",
"immunology/immune",
"response",
"virology/animal",
"models",
"of",
"infection",
"immunology/immunity",
"to",
"infections"
] | 2009 | Immunization with Single-Cycle SIV Significantly Reduces Viral Loads After an Intravenous Challenge with SIVmac239 |
Schistosomiasis is one of the most important neglected tropical diseases and an effective control is unlikely in the absence of improved sanitation and vaccination . A new approach of oral vaccination with alginate coated chitosan nanoparticles appears interesting because their great stability and the ease of target accessibility , besides of chitosan and alginate immunostimulatory properties . Here we propose a candidate vaccine based on the combination of chitosan-based nanoparticles containing the antigen SmRho and coated with sodium alginate . Our results showed an efficient performance of protein loading of nanoparticles before and after coating with alginate . Characterization of the resulting nanoparticles reported a size around 430 nm and a negative zeta potential . In vitro release studies of protein showed great stability of coated nanoparticles in simulated gastric fluid ( SGF ) and simulated intestinal fluid ( SIF ) . Further in vivo studies was performed with different formulations of chitosan nanoparticles and it showed that oral immunization was not able to induce high levels of antibodies , otherwise intramuscular immunization induced high levels of both subtypes IgG1 and IgG2a SmRho specific antibodies . Mice immunized with nanoparticles associated to CpG showed significant modulation of granuloma reaction . Mice from all groups immunized orally with nanoparticles presented significant levels of protection against infection challenge with S . mansoni worms , suggesting an important role of chitosan in inducing a protective immune response . Finally , mice immunized with nanoparticles associated with the antigen SmRho plus CpG had 38% of the granuloma area reduced and also presented 48% of protection against of S . mansoni infection . Taken together , this results support this new strategy as an efficient delivery system and a potential vaccine against schistosomiasis .
Schistosomiasis remains one of the most prevalent diseases in the world and so a significant public health problem , especially in developing countries [1] . This parasitic disease affects more than 240 million people worldwide , with a further 700 million individuals living at risk of infection [2] and it causes up to 250000 deaths per year [3] . Current schistosomiasis control strategies are mainly based on chemotherapy but , despite decades of mass treatment , the number of infected people remains constant [4] . Extensive endemic areas and constant reinfection of individuals , together with poor sanitary conditions in developing countries , make drug treatment alone inefficient [5] . Many consider that the best long-term strategy to control schistosomiasis is through immunization combined with drug treatment [6] . An anti-schistosomiasis vaccine that induces even a partial reduction in worm burdens could considerably reduce pathology and limit parasite transmission [7] . The current schistosoma vaccine candidates prove not to be the most effective , so it is important to identify new antigens and to explore alternative vaccination strategies , including new adjuvants to improve vaccine efficacy [8] . In schistosomiasis , there is evidence indicating the involvement of low molecular weight proteins that bind GTP ( guanosine triphosphate ) during the process of maturation and deposition of eggs by the females of S . mansoni [9] . Over expression in female worms may be attributed to the involvement of Rho-GTPase in female reproduction processes , especially on vitelline cell maturation and/or egg laying . Immunolocalisation of S . mansoni Rho1 on the parenchymal cells surrounding the vitellaria adds support to this suggestion [10] . This brings an interest in understanding the role of this protein in immunological processes resulting from schistosomiasis and on the evaluation of its potential as a vaccine candidate . Considering that schistosome infection occurs predominantly in areas of rural poverty in sub-Saharan Africa , Southeast Asia and tropical regions of the Americas [11] a candidate vaccine that could be administered by oral route could offer an economical and effective solution to mass immunization . The main advantages presented by oral vaccine delivery are the target accessibility and enhanced patient compliance owing to the non-invasive delivery method . On the other hand , for effective oral immunization , antigens and plasmids must be protected from the acidic and proteolytic environment of the gastrointestinal tract and efficiently taken up by cells of the gut associated lymphoid tissue ( GALT ) . With this in mind , several studies have been done and showed that the association of antigens with nanoparticles increases the internalization by M cells and prevents the degradation in the gastrointestinal ( GI ) tract [12] . Another important aspect is that these carrier systems can act as immunostimulants or adjuvants , enhancing the immunogenicity of weak antigens [13] . Biodegradable and mucoadhesive polymeric delivery systems seem to be the most promising candidates for mucosal vaccines . Several polymers of synthetic and natural origin , such as poly ( lactic-co-glycolic acid ) ( PLGA ) , chitosan , alginate , gelatin , etc . , have been exploited for efficient release of mucosal vaccines and significant results have been already obtained [14] . Chitosan is the deacetylated form of chitin and has many properties suitable for vaccine delivery . It is a mucoadhesive polymer , biodegradable and biocompatible . In particular , its ability to stimulate cells from the immune system has been shown in several studies [15] , [16] , [17] , [18] . Nevetheless , the ability of chitosan in inducing a Th1 , Th2 or mixed responses is still controversial as also the type of immune response induced by different administration routes [19] , [20] . Additionally , chitosan is a cationic polymer , easily form complexes or nanoparticles in aqueous medium with the possibility to adsorb proteins , antigens and DNA [21] [22] that may protect them from degradation [23] . The oral administration of antigen adsorbed nanoparticles is demanding as processes like rapid antigen desorption from the particles or the attack of the antigens by enzymes or acidic substances from the GI fluids may occur . These obstacles may be overcome by coating those antigen loading particles with an acid resistant polymer , like sodium alginate [24] . Alginate coated chitosan nanoparticles was recently described [24] and it has the particular advantage of being constructed under very mild conditions ( aqueous medium and mild agitation ) , which is a great benefit for the encapsulation of proteins , peptides and antigens . Moreover , Borges and co-workers [25] have demonstrated that these coated nanoparticles were able to be taken up by rat Peyer's patches which is one of the essential features to internalize , deliver and target the intact antigen to specialized immune cells from the gut associated lymphoid tissue ( GALT ) [26] . Herein , we proposed to evaluate in vitro characteristics of chitosan nanoparticles associated to protein , for which the antigen Rho1-GTPase of S . mansoni was chosen , to be used as a candidate oral vaccine against schistosomiasis . Once in vitro characterization showed favorable data , its in vivo role was evaluated through mouse immunization . Added to that , chitosan was evaluated not only for its performance as a delivery system but also for its contribution due to its adjuvant properties . Additionally , since a mixed Th1/Th2 response seems to be optimal for a schistosomiasis vaccine , then bacterial CpG motifs ( which induce the production of IL-12 by DCs and macrophages that express the appropriate TLR95 ) can be used as adjuvants to boost immunity and also with the aim of inducing a TH1-like immune response that can prevent the normal Th1 to Th2 transition . With this in mind we investigated the co-administration of synthetic unmethylated oligodeoxynucleotides containing immunostimulatory CpG motifs ( CpG ODN ) , a TLR-9 ligand , with chitosan nanoparticles . With this in mind , on this work we will report a new strategy of vaccination against schistosomiasis based on chitosan nanoparticles associated to SmRho antigen plus the adjuvant CpG , and coated with sodium alginate .
Chitosan ( CH ) ( Chimarin DA 13% , apparent viscosity 8 mPa . s ) was supplied by Medicarb , Sweden . CH was purified by filtration of an acidic chitosan solution and subsequent alkali precipitation ( 1 M NaOH ) . The purified polymer was characterized by gel permeation chromatography ( GPC ) and Fourier Transform-Infrared Spectroscopy ( FT-IR ) . The average weight molecular weight of the material was found to be 1 . 2×105 ( GPC in 0 . 5 M CH3COOH - 0 . 2 M CH3COONa , 25°C ) . The degree of acetylation determined by FT-IR according to Brugnerotto et al . [27] was found to be 16% . Endotoxin levels of the purified chitosan extracts were assessed according to Nakagawa et al . [28] using the Limulus amebocyte lysate ( LAL ) QCL- 1000 assay ( Cambrex ) and found to be lower than 0 . 1 EU/mL , respecting the US Department of Health and Human Services guidelines [29] for implantable devices . Imidazole modified chitosan ( CHimi ) was prepared as previously described [30] and it was used to prepare DNA chitosan particles , with the aim of obtaining higher rates of transfection , as reported by Moreira and co-workers [31] and also by our group [31] . Low viscosity pharmaceutical grade sodium alginate was kindly donated by ISP Technologies Inc . , Surrey , UK ) . Class C , CpG ODN ( 2395 ) ( 5′-tcgtcgttttcggcgc:gcgccg-3′ ) was purchased from InvivoGen ( San Diego , CA , USA ) . All the others reagents used were of analytical grade . The SmRho cDNA sequence was amplified from an adult extract worm cDNA using specific oligonucleotides ( Table 1 ) designed and used in a PCR reaction to amplify the complete open reading frame of SmRho ( GenBank accession number AF140785 ) . PCR was performed using Platinum Pfx enzyme ( Invitrogen , Carlsbad , USA ) ; the reaction was initiated with one cycle of 2 min at 94°C , followed by 25 cycles of 15 s at 94°C , 30 s at 56°C , and 1 min at 68°C and finalized with a step of 68°C of 3 min . PCR products were cloned by a BP recombination reaction into pDONR 221 cloning vector ( Invitrogen , USA ) , according to manufacturer specifications . After producing the entry clone , a LR recombination reaction was performed with pDONR-rSmRho and pET-DEST42 to clone the full-length cDNA sequence of rSmRho into an expression vector ( Invitrogen , USA ) . The resulting clone was then sequenced to confirm its identity . To produce a recombinant ( r ) SmRho , the full-length DNA sequence was cloned into the expression vector pDEST42 ( to produce a protein that contains a C-terminal hexahistidine tag ) . The resulting plasmid was transformed into Escherichia coli BL21 pRARE for protein expression . The recombinant protein was then purified using HiTrap Chelating HP according to the manufacturer instructions ( Amersham Biosciences , Uppsala Sweden ) . The protein purity was assessed using sodium dodecyl sulfate polyacrylamide gel electrophoresis ( SDS-PAGE ) . After obtaining the purified protein SmRho , we wanted to confirm the ability of protein to be recognized by and to active the immune system after parasite infection . For this propose , sera of infected patient that had received drug treatment or not , compared to non-infected patient serum , were pooled and tested , by ELISA and Western Blotting . In this last assay , soluble extracts of schistosomula , eggs ( SEA – soluble egg antigens ) and adult worms ( SWAP – soluble worm antigen preparation ) were also evaluated , together with the protein SmRho , about their recognition against the sera tested . The preparation of this delivery system contains three main steps , the manufacturing of the chitosan particles , their recombinant protein loading by adsorption and finally the coating with sodium alginate . Nanoparticles were prepared by mixing , while vortexing , equal volumes of a 0 . 1% ( w/v ) of chitosan in 5 mM CH3COONa buffer , pH 5 . 5 and a solution containing 0 . 625% ( w/v ) of Na2SO4 , both previously heated at 55°C for 10 min . Nanoparticles were allowed to form overnight under stirring . In the following day , the suspension was centrifuged at 3200 g for 20 min at 18°C . The particles were resuspended in about 1/8 of the volume with ultrapure water ( Milli-Q , Millipore ) . The loading was done as described by Borges et al [24] , with some adaptations . Briefly , the solution of SmRho protein ( 1 µg/µL ) was incubated with chitosan particles under mild agitation at 18°C . The loading efficacy of the uncoated particles were calculated by an indirect way , quantifying the protein that remained in solution . After 1 h of incubation , an aliquot of the particle suspension was centrifuged at 15700 g for 30 min and the protein in supernatant was quantified by BCA-protein assay ( PIERCE , Rockford , USA ) using a microplate reader with a 590 nm filter ( Multiskan-EX 355 ) . The absorbance reading value was corrected subtracting the average absorbance reading obtained in the BCA-protein assay from that one of the supernatants of unloaded nanoparticles prepared exactly in the same conditions . The corrected OD value was then used to calculate the concentration using the standard curve prepared at same time from individual protein standards . The drug loading efficiency ( LE ) were calculated from the following equation: ( 1 ) The coating was performed as described by Borges and colleagues [24] . While vortexing , equal volumes of protein-loaded particle suspension and 1% sodium alginate solution . The suspension of the particles was maintained under agitation with a magnetic stirrer for 20 min at 18°C . The suspension was then centrifuged for 30 min at 370 g and the supernatant was discarded . To chemically cross-link the alginate at the particle surface , the particles were re-suspended in 0 . 524 mM CaCl2 solution and kept under agitation for another 10 min . The evaluation of the protein desorption during the coating procedure was performed during the incubation of the particles with sodium alginate . Aliquots of the particle suspension were collected , centrifuged at 15700 g for 30 min and the protein in the supernatant was assayed with a BCA protein assay as described previously . SmRho+CpG ODN loaded alginate coated chitosan NPs were prepared by adding the CpG and protein at same time during the loading process and before coating the nanoparticles . The loading of CpG with chitosan particles was assessed by electrophoresis in 1% ( w/v ) agarose ( Cambrex ) gel , with 0 . 05 µg/mL of ethidium bromide ( Q-BioGene ) and also by measuring the OD of the nanoparticle supernatants at 260 nm and calculate CpG by the difference . To eliminate background interference , the supernatant of unloaded particles were treated of the same way . The different formulations were administered orally with a gavage-feeding needle ( groups I–V ) and intramuscularly with an injection in the tibialis anterior muscle ( group VI ) . The primary immunization was followed by two immunizations with an interval of two weeks . Seven days after the last immunization mice were challenged and then , after 50 days mice were sacrificed . Blood samples were collected from tail veins from mice of each experimental group at two-week intervals and the sera were prepared by centrifugation and stored at −20°C until further analysis . Serum was collected from immunized and control mice to measure kinetics of SmRho specific antibodies . A measurement of specific anti-SmRho antibodies was performed using an indirect enzyme-linked immunoabsorbent assay ( ELISA ) as described elsewhere [32] . Briefly , maxisorp 96-well microtiter plates ( Nunc , Roskilde , Denmark ) were coated with 10 µg/mL of SmRho in carbonate-bicarbonate buffer , pH 9 . 6 , for 16 h at 4°C , followed by washes and then blocked for 1 h at 37°C with 200 µl/well of PBS-casein ( phosphate buffer saline , pH 7 . 2 with 1 . 6% of casein ) . Next , 100 µl of serum sample from individual mice diluted 1∶100 in PBS-casein was added to each well and was incubated for 1 h at 37°C . Plate-bound antibodies were detected by peroxidase-conjugated anti-mouse IgG ( SIGMA ) , IgG1 and IgG2a ( Southern Biotechnology Associates , Inc . , Birmingham , AL , USA ) diluted 1∶5000 in PBS with 0 . 25% casein . The plates were revealed by the addition of 100 µl of detection solution ( R&D systems , Minneapolis , USA ) containing tetramethylbenzidine ( Thermo Scientific Pierce ) and H2O2 in each well; after 20 min reactions were stopped with the addition of 50 µl of 5% ( v/v ) of sulfuric acid per well . The absorbance was read at 450 nm in an ELISA plate reader ( ELX 800 BIO-TEK Instruments Inc . ) . Seven days after the last immunization , mice were challenged through percutaneous exposure of abdominal skin to water containing 25 cercariae ( LE strain ) for one hour . 50 days after challenge , mice were sacrificed and adult worms were perfused from their portal veins [33] . Protection was calculated by comparing the number of worms recovered from each vaccinated group compared with control group using the following equation: ( 2 ) where C represents the worms recovered from the saline control group and I represents the worms recovered from the experimental group . Liver fragments from mice ( 5 mice per group ) of control and experimental groups immunized and infected , were collected 50 days post-infection in order to evaluate the effect of immunization on granuloma formation . Livers fragments were fixed in 10% paraformaldehyde . Fragments processed for paraffin embedding and histopathological sections were cut using a microtome at 5 µm . Sections were stained on a slide with hematoxilin-eosin ( HE ) . The areas of individual granulomas were obtained through the MacBiophotonics ImageJ software analyzer . Fifteen granulomas from each mouse with a single well-defined egg were randomly chosen using a microscope with the 4× objective lens; granulomas were then scanned using JVC TK-1270/RGB microcamera . Using a digital pad , the total area of each granuloma measured , and the results were expressed in square micrometers . Statistical analysis was performed using the ANOVA test using the Graph Pad Prism 5 software package . The Bonferroni test was used to compare subgroups with the level of significance set at p<0 . 05 . The studies regarding sera samples taken from patients were approved by Research Ethics Committee ( COEP ) of UFMG , the protocol number 523/07 , and a written informed consent was obtained from each patient before blood collection . Experiments with animals were performed in compliance with the guidelines of the Institutional Animal Care and Committee on Ethics of Animal Experimentation ( “Comitê de Ética em Experimentação Animal” – CETEA , national guidelines , Law number 11 . 794 , 8/10/2008 ) from Universidade Federal de Minas Gerais ( UFMG ) ; protocol number 204/2009 was approved on 24/03/2010 .
The full-length sequence of the S . mansoni cDNA encoding SmRho was obtained from an adult worm cDNA library using PCR with specific oligonucleotides . The resulting full-length cDNA displayed an ORF of 579 bp , encoding a protein of 193 amino acids with a predicted molecular mass of approximately 21 . 8 kDa and an isoelectric point of 5 . 70 . BlastP comparisons of the deduced protein sequence in GenBank exhibit a complete identity and similarity to S . mansoni Rho-GTPase protein ( EMBL-Bank CDS: AAD31508 . 1 ) ( data not shown ) . The nucleotide sequence of this protein was cloned into a pET-DEST42 expression vector and the protein was expressed in E . coli BL21 pRARE strain . Protein extract of transformed bacteria showed a band at ∼26 kDa when induced with IPTG , since the plasmid add a C-terminal hexahistidine tag in the protein expressed . The bacteria were then lysed and the lysate separated into soluble and insoluble fractions . The inclusion bodies were shown to contain the majority of the recombinant protein ( Figure 1A ) , which was mostly solubilized by extraction with 0 . 9% ( w/v ) N-Laurylsarcosine . The protein was bound to a nickel-charged column under denaturing conditions , and purified by affinity chromatography through an imidazole linear gradient . Eluted fractions containing rSmRho were pooled , and protein yield after purification was estimated to be around 3 mg/L ( Figure 1B ) . After buffer exchange the protein was used in further experiments . To confirm the immunogenicity of SmRho in S . mansoni infection , the recognition of the protein by sera of infected patients , who had received drug treatment or not , was evaluated . The results obtained in the ELISA test showed that SmRho was not recognized by sera of control patients , i . e . not infected with S . mansoni . On the contrary , sera of infected patients , in particular of drug treated ones , showed a specific reaction against SmRho ( Figure 2A ) . Additionally , the immunogenicity of SmRho was evaluated by Western blot assay corroborating the above-mentioned result ( Figure 2A ) . It was observed again the reactivity of the protein SmRho with sera of infected patients , drug treated and untreated , and no reaction in the control group . In the membrane incubated with sera of treated infected patient a protein band with a molecular weight close to SmRho band in the soluble extract of eggs ( SEA ) and adult worms ( SWAP ) was observed as well . This higher reactivity in treated group is probably due to a greater exposure of the protein after drug treatment which increases the recognition by immune system cells . Sera types incubated with each membrane are represented in Figure 2B . The preparation of the delivery system was performed as described by Borges and colleagues [24] and it contains three main steps , the manufacturing of the chitosan particles , their protein loading by adsorption and finally the coating with sodium alginate . The efficiency of this process was assessed by the quantification of non-bound protein that remained in the supernatant after the loading and coating steps . In Table 3 one can observe that the encapsulation efficiency of the protein in the chitosan nanoparticles before the coating with alginate was around 95% . After the coating around 76% of the protein remained bound to the particles . The association of CpG to alginate coated chitosan-SmRho nanoparticles was evaluated by two techniques using the supernatants of the particles . Both techniques indicated a CpGODN loading efficiency of about 100% since no DNA mobility in electrophoresis was verified as well as no absorbance at 260 nm was detected ( data not shown ) . The size of chitosan-SmRho particles , before and after coating , was determined by light scattering technique . One of the major currently described drawbacks of this methodology of particle preparation is the high polydispersity of the obtained nanoparticles [35] that results from particle aggregation during its formation . The average size presented by particles was approximately 750 nm scale before coating and after that a reduction of particles size was observed ( Table 4 ) . The unpredictable result of coated nanoparticles had a smaller size than those uncoated was most probably related with the aggregation phenomena of the particles . Therefore , average size values had the contribution of aggregates size , presented on particles suspension . These particles were also characterized in terms of zeta potential ( Table 4 ) . In the first step , zeta potential determinations have shown that an excess of polymer allowed the assembly of particles with a positive global net charge . During the coating procedure an inversion of the surface charge of the particles to negative values was observed due to the negative charge of sodium alginate . This zeta potential inversion is a strong indication of the presence of an alginate coating on the surface of the particles . Release studies were performed to evaluate the stability of the coated nanoparticles and the profiles of SmRho protein desorption from these nanoparticles in simulated gastric fluid ( SGF ) and simulated intestinal fluid ( SIF ) , at 37°C . It can be observed in Figure 3 that coated nanoparticles presented a great stability in SGF , with less than 40% of the protein released after 2 hours and it was even better in SIF , where less than 15% of the protein was released after 20 hours of assay . Further in vivo studies were performed to investigate the potential utility of rSmRho loaded chitosan-based nanoparticles in eliciting the production of antibodies or modulating the immune response following intramuscular or oral administration of the suspension of the particles . To evaluate the levels of SmRho-specific IgG antibodies serum samples from vaccinated animals from each group were tested by ELISA . The measures of IgG antibodies showed that nanoparticles were able to induce the production of specific SmRho antibodies mainly in the experimental group vaccinated with coated SmRho-chitosan nanoparticles by intramuscular route , which showed high levels of IgG that appeared at the day 45 ( 5 weeks after the first immunization ) and presented the highest level at the day 75 , compared to the control group ( Figure 4 ) . To determine the type of immune response induced after vaccination , the subclasses IgG1 and IgG2a were also analyzed . For this same experimental group , the levels of specific anti-SmRho IgG1 antibodies were increased since day 30 until day 90 , when the animals were sacrificed , and a peak of IgG1 antibodies was observed at the day 60 , after performed the three rounds of immunization . The levels of IgG2a were also increased after the three rounds of immunization , with a peak at the day 60 , which remained elevated until the day 90 . The high levels of IgG1 and IgG2a showed a mixed Th1/Th2 profile . In relation to the remaining groups , there was no significant production of specific SmRho antibodies during the period evaluated . To investigate the protective activity induced by vaccination with different formulations of chitosan nanoparticles in murine model of S . mansoni infection , immunized mice were challenged with 25 cercariae . The difference in the number of adult worms recovered in the experimental groups compared to control group was calculated 50 days post-challenge ( Table 5 ) . Groups immunized with coated SmRho nanoparticles , associated or not to CpG also presented the highest level of protection protection of 48% and 55% , respectively . Interesting , the group of animals immunized with CH nanoparticles without SmRho protein and challenged with cercariae showed a significant reduction of 47% in adult worm burden . This result suggests that chitosan has an important role in inducing nonspecific immunity against S . mansoni infection , and that the adjuvant CpG did not have a considerable contribution in reducing worm burden in this experiment . To evaluate the effect of the proposed vaccine on reducing granuloma reactions , histological analysis was performed by digital morphometry . Seven days after the third immunization , mice were challenged with 25 cercariae . After 50 days of challenge infection , mice were sacrificed and liver samples were taken for histological analysis . Hematoxilin and eosin stained liver sections were then used to measure the size of individual granulomas . Vaccination with coated SmRho-CpG-chitosan nanoparticles by oral route reduced liver granuloma area by 38 . 4% ( Table 5 ) , compared with mice that were immunized with PBS . The remaining groups also presented a considerably granuloma area reduction , however , these granulomas were not as small as those observed in groups above mentioned . These findings suggest that the antigen SmRho associated with CpG in nanoparticles was important to induce this anti-pathological effect . Cytokine profile evaluation was performed using splenocyte cultures from individual mice immunized with chitosan-based nanoparticles . The production of IFN-γ , IL-10 and TGF-β was measured in the supernatants of spleen cells cultured only with complete RPMI medium or in the presence of SmRho , SEA ( soluble egg antigens ) or SWAP ( soluble worm antigen preparation ) . The highest levels of the immunomodulatory cytokine IL-10 were produced by SEA-stimulated splenocytes from mice immunized with coated SmRho-CpG-chitosan nanoparticles , compared with the control-stimulated splenocytes . In the others experimental groups significant levels of IL-10 were also observed , although their levels were not so high compared with those above described ( Figure 5A ) . Significant levels of IFN-γ , a cytokine typical of Th1-type immune response , were produced by SmRho-stimulated splenocytes from groups immunized with coated SmRho-chitosan and coated SmRho-CpG-chitosan ( Figure 5B ) . Taken together , these results show that groups which presented the highest levels of the immunomodulatory cytokine IL-10 , also were able to achieve a significant protective response and a reduced liver pathology , which is probably related to the prevention of an excessive Th1 and/or Th2 response by IL-10 .
Schistosomiasis is one of the most important neglected tropical diseases and an effective control is unlikely in the absence of improved sanitation and vaccine . The antigens tested , so far as vaccine candidates , prove not to be so effective as desirable , consequently , it is important to continue identifying new target antigens [36] . The selection of a suitable delivery system and an adjuvant to aid in the stimulation of the appropriate immune response is a critical step in the path to the development and employment of successful anti-schistosome vaccines , and a number of approaches are being tested , with some success [8] . Here we proposed a candidate vaccine formulation based on SmRho antigen loaded chitosan nanoparticles , coated with alginate , as an alternative strategy to induce protection against S . mansoni infection . This vaccination strategy offers many technical advantages , including the possibility of administration by oral route , which makes the vaccine safer than injectable vaccines and facilitates its use mainly in underdeveloped areas [37] . The recombinant expression of SmRho was optimized using an E . coli pRARE lineage , which co-expresses rare tRNAs required for the synthesis of some eukaryotic proteins . The expression and purification of the protein of interest was obtained with high yield and showed a protein with approximately 26 kDa of molecular mass . The immunogenicity of SmRho protein was confirmed through the reaction with sera of patient infected with S . mansoni . Besides that , the presence of SmRho in soluble egg and adult worms extracts was verified , which is supported by Vermeire and co-workers reports [10] . The methodology for the preparation of coated chitosan nanoparticles was successfully adapted from Borges and co-workers [24] as demonstrated by particle characterization results . The protein loading of the nanoparticles was done by adsorption process based on electrostatic interaction [38] and this process was favored considering that SmRho has an isoeletric point of 6 . 5 . Consequently , in a physiological solution that has a pH of 7 . 4 , SmRho is negatively charged and can easily interact with the positively charged chitosan nanoparticles . In the present work , SmRho loading efficiency of uncoated particles was 95% , which are better than those results found in literature [39] [24] . After coating , the loading efficiency showed a significant decrease to 76% . Nevertheless , the result obtained here was still better than those found by Borges and co-workers [24] and Li and co-workers [39] , which was 60% and 66% , respectively , for ovalbumin . A disturb of the protein adsorption equilibrium can occur and a new equilibrium can be established having alginate as a direct competitor for positive charges on chitosan surface which explain our results . In vitro release studies were performed in SGF and SIF medium at 37°C in order to evaluate the release profiles of protein SmRho and also to assess the stability of the designed delivery system when submitted to medium with different pH , ionic strength at physiological temperature . In SGF , there was a gradual release of the protein , and after 2 h , about 30% of the protein had been released . After this period , it is believed that the particles have passed through the stomach therefore , it can be deducted that nanoparticles are quite resistant to the influence of acid environment . Despite using a similar system , Borges et co-workers ( 2006 ) [25] obtained a protein release rate of over than 90% in SGF medium after the same period of the release test . The protein release profile in SIF had an even better result , since after 20 h of assay , only about 15% of the protein had been released . This result was also better than that obtained by Borges and colleagues [25] and Li and colleagues [39] . Again , the system under study remained stable in SIF medium conditions , what is desirable for an efficient antigen delivery . After characterization and based on the results which showed that the nanoparticles have suitable features to be delivery orally , the immunization was realized to investigate the effect of coated chitosan nanoparticles loaded with antigen on mice immune system and their potential to prevent infection of S . mansoni . Our results demonstrated that an anti-pathological or a protective response induced against infection with cercarie are not necessary correlated with high levels of specific antibodies . This can be presumed because only the group intramuscularly immunized with coated chitosan nanoparticles presented high levels of SmRho specific IgG1 and IgG2a antibodies , and it did not show any significant reduction in granuloma area or in worm burden . On the other hand , SmRho-CpG-chitosan nanoparticles administered by oral route reduced liver granuloma area by 38 . 4% and were not able to induce systemic specific antibodies . This result confirm our previous work , recently published [31] where the immunization with chitosan-DNA nanoparticles did not induce antibodies , however was able to reduced liver pathology . Nevertheless , this group ( CH-Rho-Alg i . m . ) showed an important role of alginate coated chitosan nanoparticles as adjuvant , considering that any other adjuvant or immunopotentiator was used and high levels of antibodies were produced . It is a consensus that the development of new and safe adjuvants is necessary not only for parenteral vaccination but also for the more challenging mucosal routes of administration in order to maximize the effectiveness of new antigens as well as those already available [40] . Within this perspective , and due to their unique and interesting properties recently reviewed in several scientific journals [41] , [42] , [43] , chitosan-based nanoparticles have been used associated with various antigens as carrier system for vaccination and administered by different mucosal routes [12] , [44] , [45] . However , its potential as an adjuvant to parenteral vaccination has been less studied . In a previous study [16] a solution of chitosan was explored as an adjuvant for mice subcutaneous immunization with a model antigen . It was shown that chitosan was able to increase more than five times antibody titers and the proliferation of specific CD4+T cells more than six times . With respect to sodium alginate , some studies show that alginate microparticles are internalized by M cells of the mucosa [46] and were able to transport antigens to mucosa associated lymphoid tissues , inducing systemic and mucosal immune response in a variety of animal species following oral administration [47] . Moreover , alginate microparticles can be used not only as a carrier system , but also as an adjuvant , as it was shown to induce the production of cytokines such as TNF-α , IL-6 and IL-1 [48] and it increases the antibodies production similarly to other adjuvants , such as incomplete Freund's adjuvant and aluminum hydroxide [47] . Mata and colleagues also showed that immunization studies in Balb/c mice by intradermal route using alginate as adjuvant elicited higher humoral and cellular immune responses leading to more balanced Th1/Th2 profile [49] . Pathology resulting from granuloma formation around the eggs in murine schistosomiasis is characterized by Th2-type of immune response and the granuloma size can be reduced by neutralization of IL-4 [50] . Thus , morbidity and mortality in murine schistosomiasis were hypothesized to be developed as a direct consequence of the egg-induced Th2 type of immune response . Herein , we suggest that the addition of CpG to nanoparticles formulation was important to induce , even in a small proportion , a shift from Th2 to Th1 immune response and also due a regulatory role for IL-10 on CpG–ODN-induced Th1 immune response , as recently reported by Jarnicki et al [51] . They found that TLR ligands , including CpG–ODNs promoted IL-12 and IL-10 production from dendritic cells . This resulted , with both components , Th1 immune response and induction of the IL-10-secreting T regulatory cells ( Treg ) could help to prevent an exacerbated granuomatous reaction . The role of CpG as an immune modulator was also described by Slütter and Jiskoot [52] and the mechanism by which CpG induces a reduced granuloma formation is probably the same as observed when CpG is used in therapy for asthma/allergy as has been widely reported [53] . In the present study the group immunized with formulations CH-Rho-CpG-Alg , which showed a reduced granuloma area of 38% , also showed higher levels of the immunomodulatory cytokine IL-10 when splenocytes were stimulated with SEA . This probably contributed to reduce inflammation and liver pathology observed in this group . It has been reported that IL-10 plays a key regulatory role in preventing the development of severe pathology due to excessive Th1 and/or Th2 responses [54] . Additionally to these balance of cytokines produced in groups immunized with coated chitosan nanoparticles containing CpG , the presence of the antigen SmRho is also supposed to have an important role in the modulation of immunopathological responses of S . mansoni infection . This is likely due to the induction of INF- γ production that can prevent the normal Th1 to Th2 transition that occurs in infected hosts after the onset of egg production by the parasites , and therefore acts preventing the development of severe chronic morbidity [55] . These observations revealed an anti-pathological role of SmRho but more studies are required to obtain a better understanding of the involvement of this protein in the process of infection of S . mansoni . To determine whether coated chitosan nanoparticles conferred protection against S . mansoni infection , immunized mice were challenged with cercariae and worm burdens were assessed . All groups immunized with coated chitosan nanoparticles by oral route showed a significant reduction in worm burden and even the group immunized with chitosan nanoparticles without protein showed a 47% of protection . It suggests that chitosan has an important role in inducing a protective immune response against schistosomiasis that is likely due to its immunostimmulatory properties . Nevertheless , the mechanism behind the ability of chitosan in inducing protection is still unknown . Its role is more likely related to trigger an innate immune response , since chitin , which is a progenitor to chitosan , has been demonstrated to function as a PAMP and it is linked with the accumulation of innate cells including alternatively activated macrophages , eosinophils and basophils [56] [57] . At sites of infection with chitin-containing agents anti-infectious immune responses and local chitinases are believed to induce chitin fragmentation . The ability of chitin to induce an acute inflammatory response is already described , but it seems to act by different pathways depending of the size of the fragment and the time point of assessment . Da Silva and co-workers demonstrated that chitin fragments induced a macrophage- and neutrophil-rich inflammatory response with only a modest degree of eosinophil infiltration , while Reege et al . showed mainly eosinophil-based response [17] , [58] . When viewed in combination , these studies suggest that chitin induces an inflammatory response that is initially neutrophilic and becomes eosinophilic over time . In the early times , macrophages were shown to be stimulated by chitin particles in vitro and in vivo , by a TLR-2 dependent mechanism that utilizes a MyD88-dependent pathway to induce IL-17 elaboration and enhance the expression of the IL-17AR . These studies also demonstrated that this novel innate immune pathway plays an essential role in the regulation of macrophage cytokine production and the induction of acute inflammation [58] . Additionally , it is well described that IL-4/IL-13-activated alternative macrophages are essential for surviving acute schistosomiasis [59] . These facts cooperate with the convincing evidence that immune elimination of challenge parasites occurs in the lungs and macrophages is expected to mediate the protective response [60] . At later times Yasuda et al . , suggested that the effects of IL-33 for the expansion and the activation of eosinophils might aid to expel infected worms from the lungs [18] . These reports support an important theory by which chitosan induce a protective immune response against S . mansoni infection , however it needs to be further investigated . Mice immunized with coated chitosan nanoparticles associated or not with CpG also showed high percentage of protection 48% and 55% , respectively . The antigen and CpG seems not to have a high contribution in conferring protection against worm infection , nevertheless , they demonstrated important roles in granuloma down-modulation , as discussed before . As a final conclusion of this work , we believe that the combination of chitosan nanoparticles associated to the antigen SmRho plus CpG is an efficient vaccine formulation candidate against schistosomiasis in light of the data obtained from murine studies , which was able to modulate the granuloma area , that represents the major pathological response in schistosomiasis and also to induce protection against infection of S . mansoni . It is important to highlight that these results were obtained with oral administration of the formulation and can be compared to those obtained from conventional routes of administration . Comparing with our previous work , in which DNA-chitosan nanoparticles were explored , this system achieved better results to be used as a vaccine because induced both protection as well as reduced the granulomatous reaction , while the first presented just an anti-pathological effect with granuloma modulation . Chitosan-based nanoparticles were also found to play an important role as adjuvant and this characteristic should be more explored with other antigens . Furthermore , the role of chitosan in inducing a protective immune response against schistosomiasis deserves special attention and requires more studies to confirm and understand this feature . Taken together , these results support this new strategy to find a safe and efficacious vaccine against schistosomiasis . | Schistosomiasis is one of the most important neglected tropical diseases and an effective control is unlikely in the absence of improved sanitation and vaccine . The selection of a suitable delivery system and an adjuvant to aid in the stimulation of the appropriate immune response is a critical step in the path to the development and employment of successful anti-schistosome vaccines . Here we propose a candidate vaccine based on chitosan nanoparticles associated with the antigen SmRho and coated with alginate , as an alternative strategy to induce protection against S . mansoni infection . This vaccination strategy offers many technical advantages , including the possibility of administration by oral route , which makes the vaccine safer than injectable vaccines and facilitates its use mainly in underdeveloped areas . Chitosan nanoparticles were prepared and characterized; the results showed that the formulation has features suitable to be delivery orally . Immunization studies suggest that the combination of chitosan nanoparticles associated to the antigen SmRho and CpG is an efficient vaccine candidate against schistosomiasis , which was able to modulate the granuloma area , that represents the major pathological response in schistosomiasis and also to induce protection against infection of S . mansoni . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"biotechnology",
"materials",
"science",
"immunology",
"biology",
"biomaterials"
] | 2012 | A New Strategy Based on Smrho Protein Loaded Chitosan Nanoparticles as a Candidate Oral Vaccine against Schistosomiasis |
Tainan experienced the most severe dengue epidemic in Taiwan in 2015 . This study investigates the association between the signs and symptoms at the time of reporting with the adverse dengue prognoses . A descriptive study was conducted using secondary data from the Dengue Disease Reporting System in Tainan , Taiwan , between January 1 and December 31 , 2015 . A multivariate stepwise logistic regression was used to identify the risk factors for the adverse prognoses: ICU admissions and mortality . There were 22 , 777 laboratory-confirmed reported cases ( mean age 45 . 6 ± 21 . 2 years ) , of which 3 . 7% were admitted to intensive care units ( ICU ) , and 0 . 8% were fatal . The most common symptoms were fever ( 92 . 8% ) , myalgia ( 26 . 6% ) , and headache ( 22 . 4% ) . The prevalence of respiratory distress , altered consciousness , shock , bleeding , and thrombocytopenia increased with age . The multivariate analysis indicated that being in 65–89 years old age group [Adjusted Odds Ratio ( aOR ) :4 . 95] , or the 90 years old and above age group ( aOR: 9 . 06 ) , and presenting with shock ( aOR: 8 . 90 ) and respiratory distress ( aOR: 5 . 31 ) were significantly associated with the risk of ICU admission . While old age ( aOR: 1 . 11 ) , respiratory distress ( aOR: 9 . 66 ) , altered consciousness ( aOR: 7 . 06 ) , and thrombocytopenia ( aOR: 2 . 55 ) were significantly associated with the risk of mortality . Dengue patients older than 65 and those with severe and non-specific signs and symptoms at the time of reporting were at a higher risk of ICU admission and mortality . First-line healthcare providers need to be aware of the varied presentations between the different age groups to allow early diagnosis and in-time management , which would prevent ICU admissions and fatalities in dengue patients .
Dengue is a mosquito-borne viral disease that has become a major public health problem owing to its wide geographical extension , high incidence , and disease severity [1 , 2] . Southeast Asia and the Western Pacific are most seriously affected by dengue , with 75% of the current globally reported outbreaks [1] . Taiwan has experienced three dengue outbreaks in recent years: the first in Penghu County in 2011 ( prevalence rate 101 per 100 000 population ) , the second in Kaohsiung City in 2014 ( prevalence rate 540 per 100 000 population ) , and the most recent outbreak in Tainan in 2015 ( prevalence rate 1 , 208 per 100 000 population ) [3] . The 2015 Tainan dengue outbreak , caused by dengue serotype 2 , was the most severe epidemic in Tainan’s history [4] , with 2 . 99–14 . 9% of dengue patients being admitted to the intensive care unit ( ICU ) [5 , 6] and 0 . 52% dying [7] . Patients aged over 70 years had the highest prevalence rate [3] , which was in contrast to the predominantly younger dengue patients in other Southeast Asian countries [8–14] . Dengue presentations are diverse and non-specific and often have unpredictable clinical progression and outcomes [1] . While most patients recover from DF , approximately 0 . 3–14 . 9% develop severe manifestations that result in ICU admission [5 , 6 , 13 , 15] , and 1–5% die without early recognition and proper treatment [1] . Timely access to proper treatment for dengue patients by primary healthcare professionals not only reduces the number of unnecessary hospital admissions but also lowers fatality rates below 1% [1] . The sensitivity of the World Health Organization ( WHO ) 2009 classification systems was 52% in differentiating patients who required ICU admissions at first presentation [13] . However , according to the WHO 2009 classification systems , as sensitivity decreases with age , it is difficult to differentiate dengue from other clinically febrile diseases in older patients [16] . Therefore , a better understanding of the signs and symptoms at the time of reporting associated with poor prognoses may assist first-line healthcare providers to focus on patients who at higher risk and enable timely treatment , especially for the aging dengue population . Only one case-control study in Singapore has reported clinical factors associated with ICU admission in dengue patients; 50–59 year age group , diabetes , the WHO 2009 classification of dengue severity , hematocrit change greater or equal to 20% concurrent with platelets less than 50 , 000/μl , hypoproteinemia , hypotension , and severe organ involvement [13 , 17] . However , there have been several studies that have identified the risk factors associated with dengue mortality: old age [18–21] , being female [18] , and presenting with symptoms such as nausea and vomiting [18] , bleeding [18 , 22] , gastrointestinal bleeding [20] , hematuria [20] , thrombocytopenia [20] , leukocytosis [23] , altered mental status [18 , 22] , plasma leakage [18 , 21] , cavity effusions [19] , tachycardia [24] , and shock [18 , 22] . Most of these studies were limited , however , because of small sample sizes [13 , 23] and single hospital studies [22 , 24] , or were focused primarily on severe dengue patients [19–21] . A study in Malaysia examined the national registry data of 43 , 347 dengue patients in 2013; however , the dengue diagnosis was verified using WHO 1997 criteria [18] , which had a dengue severity sensitivity and specificity lower than the WHO 2009 [25 , 26] , and only 30 . 2% of patients with DF were serologically confirmed [18] . In addition , the mean age of patients with DF in Malaysia was 30 [18] , which was much younger than the majority of the dengue population in Taiwan [3] . There is a lack of information about the signs and symptoms at the time of reporting associated with ICU admissions and mortality in DF patients in aging societies such as Taiwan . This study seeks to describe the signs and symptoms at the time of reporting in dengue patients across different age groups and identify the signs and symptoms associated with ICU admission and mortality in 2015 dengue patients in Tainan , Taiwan .
Using registry data from the Dengue Disease Reporting System from January 1 and December 31 , 2015 , the retrospective cohort study included all 2015 dengue patients in Tainan , Taiwan . In Taiwan , according to the Law on the Control of Communicable Diseases , all suspected dengue cases must be reported to the Health Department at the Tainan and Taiwan Centers of Disease Control ( Taiwan CDC ) within 24 hours [4] . The dengue-infected cases were diagnosed from laboratory results based on the following criteria; a reverse-transcription polymerase chain reaction ( RT-PCR ) , an evaluation of the anti-dengue virus IgM and IgG , and dengue viral isolation from serum or tissue [27] . The specimens were confirmed by the approved laboratories at Taiwan CDC after which final confirmations were performed [28] . The decision to admit a DF patient to ICU or not was judged by the treating physician and was not documented by the reporting system . However , Taiwan national guidelines on dengue fever management are available and clinicians could treat the critical dengue patients accordingly [28] . This study was exempt from a full review by the Institutional Review Board of National Cheng Kung University Hospital ( no . A-ER-104-386 ) as the database consisted of de-identified data and a confidentiality agreement with Tainan City Government was signed by all researchers using the dataset . The data included age , gender , the status of the patient at the time of reporting , the region of patients who lived when being reported , the levels of hospitals/clinics reporting dengue cases , dates for the dengue onset , reporting dates to Taiwan CDC , dates of the dengue confirmations by the Taiwan CDC , dates of the mortalities , the signs and symptoms at the time of reporting , the presence or absence of severe symptoms , information about ICU admission , and the mortality reported by the physicians . The reporting data were updated and confirmed by the medical officers at Taiwan CDC . Patients without complete data were excluded from analysis . The primary outcomes included ICU admission and mortality . ICU admission was defined as the ICU admission details in dengue-infected people on the latest version before the data analysis , mortality was defined as the death registration of the confirmed dengue patients , and accuracy was verified by cross matching the discharge data from the three major hospitals in Tainan , which between them had seen 9 , 816 dengue patients . The independent variable was the signs and symptoms in dengue-infected people at the time of reporting . The signs and symptoms reported by physicians were verified by the Tainan Health Department and were classified into thirty-seven categories based on the WHO ( 2009 ) classification system [1 , 4] . To avoid misclassification , the classification system was further validated by a programmer , an epidemiologist , and an infectious disease clinician ( S1 Table ) . The other variables included age: <15 years , 15–39 , 40–64 , 65–89 , and over 90 years , gender , the status of patients at the time of reporting , the region of patients who lived when being reported , the levels of hospitals/clinics reporting dengue cases , dates for dengue illness onset , reporting dates to Taiwan CDC , dates of dengue confirmation by Taiwan CDC , dates of ICU admission and death , the presence or absence of severe symptoms . The differences between the groups ( different age group , ICU and non-ICU , survivors and non-survivors ) were examined using t-tests or median test for the continuous variables , and an χ2 test for the categorical variables . Patients who were admitted to ICU or died during reporting were excluded from multivariate analysis , and then the univariate logistic regression was performed using the demographic variables ( age and gender ) , the levels of hospitals/clinics reporting dengue cases , and the signs and symptoms as the independent variables and ICU admission and mortality as the dependent variables . A p value less than 0 . 05 was considered potentially significant and was further analyzed with multivariate stepwise logistic regression using the Allen-Cady modified backward selection method to identify the significant demographic variables and the signs and symptoms at the time of reporting that led to ICU admission and mortality [29] . An odds ratio ( OR ) and 95% confidence interval ( CI ) were considered significant at a p value ≤ 0 . 05 . IBM SPSS Statistics 19 software was used for the analyses in this study .
A total of 22 , 777 laboratory-confirmed dengue patients in Tainan were included in this study . Of these , 22 , 737 ( 99 . 8% ) patients reported the presence of at least one sign and symptom at the time of reporting to Taiwan CDC . Of the confirmed dengue patients , 3 . 3% ( 396/11 , 922 ) had severe symptoms , 3 . 7% ( 337/9197 ) were admitted to the ICU . Among them , 1 . 3% ( 131/9816 ) died at the three major hospitals in Tainan , and 0 . 8% ( 189/22777 ) died during the 2015 dengue outbreak in Tainan . Table 1 describes the characteristics of the confirmed cases in the 2015 dengue outbreak in Tainan . The mean age was 45 . 6 years ( standard deviations [SD] = 21 . 2 ) , 50 . 4% were female , 44 . 7% were reported at the local clinics , and 30 . 8% were reported at the emergency departments of either regional hospitals or medical centers . The majority of the reported cases ( 74 . 1% ) lived in the urban region of Tainan , and 39 . 5% were reported by the regional hospitals . The median days between illness onset and reporting to Taiwan CDC was 1 day ( interquartile range [IQR] , 1–3 ) , and 2 days between illness onset and dengue confirmation by Taiwan CDC ( IQR , 1–4 ) . The median days between dengue confirmation and ICU admission was 0 days ( IQR , -1–2 ) , and 3 days ( IQR , 1–13 ) between dengue confirmation and death . Patients admitted to the ICU admission were predominantly male , older , and more likely to be the residents in the rural region and being reported by medical centers . Of note , they were apt to have severe symptoms , and had longer days between illness confirmation and death . Patients who died were older , more likely to be reported at emergency departments and medical centers . Similar to patients admitted to ICU , they were more likely to have severe symptoms , but had shorter days between illness onset and dengue reporting or confirmation ( Table 1 ) . Table 2 shows a summary of the signs and symptoms at the time of reporting . Fever ( 92 . 8% ) , myalgia ( 26 . 6% ) , and headache ( 22 . 4% ) were the most common symptoms . Fig 1 shows the distribution of the signs and symptoms at the time of reporting in dengue patients in the different age groups . In patients aged older than 65 , the proportion with fever , myalgia , headaches , and skin rashes was significantly lower with an increase in age ( p = 0 . 0001 ) . The following signs and symptoms were more prevalent with an increase in age: nausea and vomiting , poor appetite , fatigue , thrombocytopenia , bleeding , respiratory distress , altered consciousness , shock , gastrointestinal symptoms , and chest tightness/pain . Patients who presented with fever , headache , or myalgia at the time of reporting had shorter days between illness onset and reporting than those without these symptoms ( fever: 2 . 1 ± 2 . 1 vs . 2 . 7 ± 2 . 9 , p = 0 . 0001; headache: 1 . 9 ± 1 . 8 vs . 2 . 2 ± 2 . 2 , p = 0 . 0001; myalgia: 2 . 0 ± 1 . 8 vs . 2 . 2 ± 2 . 2 , p = 0 . 0001 ) . Patients who died and patients admitted to the ICU were significantly less likely to have fever , myalgia , headaches , bone and joint pain , skin rashes , and retro-orbital pain at the time of reporting compared to the survivors and non-ICU admitted patients . Patients who were admitted to the ICU had higher proportions of the following signs and symptoms at the time of reporting: gastrointestinal symptoms , fatigue , dizziness , thrombocytopenia , bleeding , gastrointestinal bleeding , respiratory distress , shock , abnormal liver function , altered consciousness , and pneumonia on X-rays . The patients who died had higher proportions of following signs and symptoms at the time of reporting: gastrointestinal symptoms , fatigue , thrombocytopenia , bleeding , respiratory distress , chest tightness/pain , shock , altered consciousness , abnormal heart rhythm , severe bleeding , and hepatosplenomegaly ( Table 2 ) . Table 3 shows the multivariable analysis of the factors associated with ICU admission in the 2015 dengue outbreak in Tainan ( n = 9 , 087 ) . Of dengue-infected people , 10 patients died and 110 patients admitted to ICU during reporting were exclude from analysis . The multivariate analyses showed that an age equal to or greater than 65 , and having shock and respiratory distress at the time of reporting were more likely to be admitted to the ICU . Patients who were not reported at medical centers , with bone and joint pain , and skin rash were negatively associated with ICU admission . Multivariate analysis showed that increasing age , respiratory distress , altered consciousness , and thrombocytopenia at the time of reporting were independent factors associated with mortality ( Table 4 ) . In contrast , myalgia , as the typical DF symptom was negatively associated with mortality . Patients who were reported at regional hospital or local clinics were negatively associated with mortality .
In the 2015 dengue outbreak in Tainan , patients older than 65 and those with severe and non-specific signs and symptoms at the time of reporting were at higher risk of ICU admission and mortality . Patients with adverse prognoses were of older age , had critical presentation on diagnosis , and had a rapid disease progress . First-line healthcare providers need to identify patients who are potential ICU admissions or have the possibility of dying as early as possible and be aware of atypical dengue presentations in the elderly . Moreover , preventive strategies as well as treatments specific to dengue infection in elderly people needs further study . | Clinical presentations of dengue fever ( DF ) are diverse and non-specific and often have unpredictable progression and outcomes . The patients in the 2015 dengue epidemics in Taiwan were predominantly much older than in other countries in Southeast Asia . However , limited data are available in Taiwan on dengue patients with adverse prognoses who were admitted to ICU or on those with fatal complications . All suspected dengue cases were reported to the Dengue Disease Reporting System in Taiwan . We examined patients who had laboratory-confirmed dengue in Tainan during the 2015 dengue outbreak and analyzed secondary data from the reporting system . Of these patients , 3 . 7% were admitted to ICU , and 0 . 8% died . Patients who received adverse prognoses were generally of older age , had a critical presentation at the time of reporting , and experienced a rapid progress of the disease , highlighting that people with severe and non-specific signs and symptoms at the time of reporting were at a higher risk of ICU admission and mortality . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"medicine",
"and",
"health",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"elderly",
"tropical",
"diseases",
"geographical",
"locations",
"health",
"care",
"age",
"groups",
"signs",
"and",
"symptoms",
"neglected",
"tropical",
"diseases",
"infectious",
"diseases",
"taiwan",
"intensive",
"care",
"units",
"dengue",
"fever",
"pain",
"hospitals",
"people",
"and",
"places",
"diagnostic",
"medicine",
"asia",
"health",
"care",
"facilities",
"hemorrhage",
"population",
"groupings",
"myalgia",
"viral",
"diseases",
"geriatrics",
"vascular",
"medicine"
] | 2017 | Symptoms associated with adverse dengue fever prognoses at the time of reporting in the 2015 dengue outbreak in Taiwan |
Neuronal differentiation is exquisitely controlled both spatially and temporally during nervous system development . Defects in the spatiotemporal control of neurogenesis cause incorrect formation of neural networks and lead to neurological disorders such as epilepsy and autism . The mTOR kinase integrates signals from mitogens , nutrients and energy levels to regulate growth , autophagy and metabolism . We previously identified the insulin receptor ( InR ) /mTOR pathway as a critical regulator of the timing of neuronal differentiation in the Drosophila melanogaster eye . Subsequently , this pathway has been shown to play a conserved role in regulating neurogenesis in vertebrates . However , the factors that mediate the neurogenic role of this pathway are completely unknown . To identify downstream effectors of the InR/mTOR pathway we screened transcriptional targets of mTOR for neuronal differentiation phenotypes in photoreceptor neurons . We identified the conserved gene unkempt ( unk ) , which encodes a zinc finger/RING domain containing protein , as a negative regulator of the timing of photoreceptor differentiation . Loss of unk phenocopies InR/mTOR pathway activation and unk acts downstream of this pathway to regulate neurogenesis . In contrast to InR/mTOR signalling , unk does not regulate growth . unk therefore uncouples the role of the InR/mTOR pathway in neurogenesis from its role in growth control . We also identified the gene headcase ( hdc ) as a second downstream regulator of the InR/mTOR pathway controlling the timing of neurogenesis . Unk forms a complex with Hdc , and Hdc expression is regulated by unk and InR/mTOR signalling . Co-overexpression of unk and hdc completely suppresses the precocious neuronal differentiation phenotype caused by loss of Tsc1 . Thus , Unk and Hdc are the first neurogenic components of the InR/mTOR pathway to be identified . Finally , we show that Unkempt-like is expressed in the developing mouse retina and in neural stem/progenitor cells , suggesting that the role of Unk in neurogenesis may be conserved in mammals .
Neural progenitors in the developing human brain generate up to 250 , 000 neurons per minute . After differentiating from these neural progenitors , neurons migrate and are then integrated into neural circuits . Temporal control of neurogenesis is therefore critical to produce a complete and fully functional nervous system . Loss of the precise temporal control of neuronal cell fate can lead to defects in cognitive development and to neurodevelopmental disorders such as epilepsy and autism . Mechanistic target of rapamycin ( mTOR ) signalling has recently emerged as a key regulator of neurogenesis [1] . mTOR is a large serine/threonine kinase that forms two complexes , known as mTORC1 and mTORC2 [2] . mTORC1 is rapamycin sensitive and is regulated upstream by mitogen signalling , such as the insulin receptor ( InR ) /insulin like growth factor ( IGF ) pathway , amino acids , hypoxia , cellular stress and energy levels [3] . mTORC1 positively regulates a large number of cellular processes including growth , autophagy , mitochondrial biogenesis and lipid biosynthesis and activation of mTOR has been linked to cancer . Hyperactivation of mTOR signalling in neurological disease is best understood in the dominant genetic disorder tuberous sclerosis complex ( TSC ) , which causes epilepsy and autism [4] . mTOR signalling has also been shown to be activated in animal models of epilepsy and in human cortical dysplasia [5]–[7] . The control of neurogenesis by the InR/mTOR pathway was first discovered in the developing Drosophila melanogaster retina , where activation of the pathway caused precocious differentiation of photoreceptor neurons and inhibition caused delayed differentiation [8]–[10] . Subsequent in vitro studies demonstrated that insulin induces neurogenesis of neonatal telencephalonic neural precursor cells in an mTOR dependent manner and that Pten negatively regulates neuronal differentiation of embryonic olfactory bulb precursor cells [11] , [12] . More recently , in vivo studies have shown that inhibition of mTOR suppresses neuronal differentiation in the developing neural tube [13] . Furthermore , knock-down of the mTOR pathway negative regulator RTP801/REDD1 causes precocious differentiation of neural progenitors in the mouse embryonic subventricular zone ( SVZ ) , while overexpression of RTP801/REDD1 delays neuronal differentiation [14] . Loss of Pten , Tsc1 , or overexpression of an activated form of Rheb , also cause premature differentiation of neurons in the SVZ [15]–[17] . These studies have demonstrated that InR/mTOR signalling plays a conserved role in regulating neurogenesis in several different neural tissues . However , the downstream effectors of InR/mTOR signalling in neurogenesis are completely unknown . To identify neurogenic downstream regulators of InR/mTOR signalling we screened genes that were previously shown to be transcriptionally regulated by mTOR in tissue culture cells [18] , for in vivo neurogenic phenotypes in the developing Drosophila retina . From this screen we identified the zinc finger/RING domain protein Unkempt ( Unk ) as a negative regulator of photoreceptor differentiation . Loss of unk phenocopies the differentiation phenotype of InR/mTOR pathway activation and Unk expression is negatively regulated by InR/mTOR signalling . Importantly , unk does not regulate cell proliferation or cell size and so uncouples the function of InR/mTOR signalling in growth from its role in neurogenesis . We also identified the evolutionarily conserved basic protein Headcase ( Hdc ) [19] , as a physical interactor of Unk and show that loss of hdc causes precocious differentiation of photoreceptors . Hdc expression is regulated by the InR/mTOR pathway and by unk , demonstrating that Hdc and Unk work together downstream of InR/mTOR signalling in neurogenesis . Unk also regulates the expression of and interacts with D-Pax2 , suggesting a model for the regulation of neurogenesis by the InR/mTOR pathway . We also show that one of the mammalian homologs of Unk , Unkempt-like , is expressed in the developing mouse retina and in the early postnatal brain . We have thus identified the Unk/Hdc complex as the first component of the InR/mTOR pathway that regulates the timing of neuronal differentiation .
The eight photoreceptors ( R1-R8 ) that constitute each ommatidium ( facet ) of the Drosophila compound eye ( Figure S1A ) differentiate in a stereotypical sequence that is initiated by R8 in response to signalling events around the morphogenetic furrow ( MF ) . The MF is a physical indentation that traverses the eye imaginal disc from posterior to anterior during the final 48 hours of larval development and the early pupal stage ( Figure 1A ) [20] . Photoreceptors differentiate posterior to the MF in rows that are aligned along a differentiation front ( Figure 1A , dotted line ) , with each new row forming every two hours . Adult ommatidia are organised in rows forming a mirror image about the equator ( Figure S1A , dotted line ) . We have previously shown that the timing of differentiation of R1/6/7 and cone cells , but not R3/4 and R2/5 , is regulated by InR/mTOR signalling [8] , [10] . In clones that are mutant for Tsc1 , in which mTOR signalling is activated , R1/6 differentiate two to three rows ahead of the differentiation front ( Figure 1B , arrows ) , whereas differentiation is severely delayed in clones that are mutant for Rheb ( Figure 1J ) , in which the pathway is inhibited . To identify novel factors that regulate photoreceptor differentiation downstream of mTOR , we screened genes that are transcriptionally regulated by mTOR in Drosophila S2 cells [18] for differentiation phenotypes in vivo by RNAi . Flp-out clones expressing dsRNAs against 28 mTOR regulated genes ( Table S1 ) , were generated to test for photoreceptor differentiation phenotypes . Using this approach we identified unkempt ( unk ) as a potential negative regulator and three molecular chaperone encoding genes ( Hsc70Cb , Hsp60 and Hsp83 ) as potential positive regulators of R1/6 differentiation ( Table S1 and Figure S1B ) . Hsp83 mutant clones are cell lethal , suggesting that the delayed differentiation phenotype caused by RNAi of Hsp83 observed in the screen , and potentially also Hsc70Cb and Hsp60 , may be due to cell death rather than a genuine delay in differentiation . These chaperones were not studied further and we focused on unk as a potential negative regulator of photoreceptor differentiation . The Unk protein contains an N-terminal zinc finger domain and C-terminal RING domain ( Figure 1C ) . Unk physically interacts with mTOR , Raptor and 4E-BP in Drosophila Kc167 cells and the strength of these interactions is regulated by insulin [21] . Moreover , phosphopeptides corresponding to one of the two mammalian homologs of Unk were identified in mouse and human cells in both of the recent mTOR phosphoproteome studies [22] , [23] . Null alleles for unk are lethal at around mid-pupal development , while hypomorphic alleles develop to become adults with an ‘unkempt’ phenotype ( small rough eyes , held out wings and crossed scutellar bristles ) [24] . However , nothing else is known about the function of unk . The previously isolated unk mutants are no longer extant . Therefore , we generated novel mutations in unk ( Figure 1C , see Materials and Methods ) . Animals homozygous for unkex13 , unkex24 , unkDf , unke01984 or heteroallelic combinations of these alleles are pupal lethal and mutant clones showed a complete absence of Unk protein expression ( Figure S1C ) , suggesting that they are null alleles . unk mutant clones cause precocious differentiation of R1/6/7 and cone cells ( Figure 1D , E , F , arrows ) , very similar to the phenotype seen in Tsc1 mutant clones ( Figure 1B ) . Moreover , unk mutant clones have no effect on the differentiation of R3/4 ( Figure 1G ) , or R2/5 ( Figure 1H ) . Co-staining for the pan-neuronal marker Elav , or markers of R3/4 and R8 , with either Bar or Prospero also demonstrated that the precocious differentiation in unk mutant clones is not due to ectopic expression of Bar or Prospero ( Figure S2 A–C ) . Thus , loss of unk phenocopies the precocious photoreceptor differentiation phenotype caused by activation of InR/mTOR signalling . This suggests that unk activity is normally repressed by mTOR during photoreceptor differentiation . The R1/6 precocious differentiation phenotype in unk mutant clones was rescued by overexpression of unk cDNA ( Figure S1D ) , demonstrating that the precocious differentiation phenotype is specifically due to loss of unk . To test whether increased expression of unk is sufficient to delay photoreceptor differentiation , MARCM ( mosaic analysis with a repressible cell marker ) clones were generated that overexpressed unk . No change in the timing of differentiation of R1/6 was seen in these clones ( Figure S1E ) . Therefore , unk is necessary but not sufficient to regulate the timing of differentiation of R1/6/7 and cone cells . In the adult eye unk mutant clones had a striking phenotype . Mutant ommatidia in the anterior half of the eye had a normal structure but had rotation defects ( Figure 1I , black arrows ) , similar to Tsc1 and Pten mutant clones [8] . Ommatidia in the posterior half of the eye were missing photoreceptors ( red arrow ) and contained photoreceptors with elliptical ( black arrowheads ) and split ( red arrowheads ) rhabdomeres ( Figure 1I ) . This phenotype is typical of perturbation of the F-actin cytoskeleton and Pten and Tsc1 mutant clones cause similar defects in photoreceptor apical membrane morphogenesis ( [25] , [26] and Figure S1F ) . In summary these data demonstrate that unk is necessary for the normal timing of differentiation and morphogenesis of photoreceptors . To test whether unk acts genetically downstream of InR/mTOR signalling , we generated double mutant clones that lacked both unk and Rheb . Compared to Rheb clones , which cause a strong delay in the differentiation of R1/6 ( Figure 1J ) , differentiation in unk , Rheb clones appeared normal ( Figure 1K ) . Thus , the strong delay caused by the loss of Rheb was suppressed , but photoreceptors in these double mutant clones did not differentiate precociously as in unk mutant clones ( Figure 1D ) . In the adult eye the elliptical and split rhabdomere phenotype seen in unk mutant clones was suppressed in unk , Rheb mutant clones ( Figure S1H ) . However , both Rheb and unk , Rheb mutant clones contained mis-rotated ommatidia and missing photoreceptors ( Figure S1G , H ) . Therefore , although unk suppresses the delay in photoreceptor differentiation caused by inhibition of InR/mTOR signalling , there may be an additional factor ( s ) that regulates R1/6/7 and cone cell fate and acts in parallel with unk ( see Discussion ) . Unk is a ubiquitously expressed cytoplasmic protein ( Figure 2A , B ) . In the eye disc Unk is expressed in undifferentiated cells anterior to the MF , in photoreceptor precursors posterior to the MF , photoreceptors and cone cells ( Figure 2A , B and 3B , C ) . Moreover , Unk is expressed more strongly posterior to the MF in the apical plane of the disc containing differentiated photoreceptors and cone cells ( Figure 2A and S3B ) . Although localised to the cytoplasm , Unk has a partially punctate distribution ( Figure 2B ) . In accordance with the negative regulation of unk expression by mTOR in S2 cells ( [18] and Table S1 ) , Unk expression is reduced in Tsc1 clones posterior to the MF both in differentiated photoreceptors and undifferentiated photoreceptor precursors ( Figure 2C , D , arrows ) . However , Unk expression is unchanged in Tsc1 clones anterior to the MF ( Figure 2D , arrowheads ) . Thus , Unk expression is negatively regulated by the InR/mTOR pathway in differentiating photoreceptor neurons and photoreceptor precursors . Inhibition of InR/mTOR signalling , using Dp110 , or Rheb mutant clones , did not cause an increase in Unk expression ( Figure S4 ) and so Unk expression is not positively regulated by inhibition of mTOR signalling . Several lines of evidence demonstrate that unk does not regulate growth . First , eye discs from unkex24 homozygous larvae , or an unk transheterozygous mutant , had similar levels of phospho-histone H3 expression posterior to the MF ( arrowheads ) as wild-type discs ( Figure 3A–C ) . Second , unlike Tsc1 mutant clones ( Figure 3F , G ) , unk mutant clones were a similar size to the control ( Figure 3D , E , G ) . Third , unlike cells mutant for Tsc1 ( Figure 3K , H ) , cells mutant for unk in the eye disc were a similar size to controls ( Figure 3I , J , H ) . Phospho-S6K or phospho 4E-BP antibodies did not work as a direct readout of mTOR activity by immunofluorescence . We therefore used phospho-AKT ( P-AKT ) expression as an indirect readout of mTOR pathway activity through negative feedback regulation via S6K [27] . As expected , expression of P-AKT was decreased in Tsc1 mutant clones ( Figure 3N ) . However , no change in P-AKT expression was observed in unk mutant clones ( Figure 3M ) , suggesting that unk is not required for the canonical regulation of S6K by mTOR . Together these data show that unk plays no role in the regulation of growth . Therefore , unk is the first gene to uncouple the function of InR/mTOR signalling in growth control from its role in regulating the transition of a post-mitotic precursor cell to a neuronal fate . To provide insight into how Unk regulates neurogenesis we interrogated protein interaction maps to identify proteins that might interact with Unk . Headcase ( Hdc ) was identified as interacting with Unk in two independent protein interaction maps and in a study that characterised the Drosophila InR/mTOR pathway interaction proteome [21] , [28] , [29] . Hdc is an evolutionarily conserved basic protein with no conserved motifs [19] . hdc mRNA generates two overlapping proteins , a short form ( HdcS ) and a full-length form ( HdcFL ) , as a result of a novel translational readthrough mechanism [30] . To provide direct evidence for the physical interaction between Unk and Hdc , we expressed epitope-tagged forms of these proteins in S2 cells . Immunoprecipitation assays in both directions showed that Venus-Unk and FLAG-HdcS co-immunoprecipitate ( Figure 4A ) . These assays confirm that Unk and Hdc physically interact in S2 cells . R1/6/7 and cone cells differentiated precociously in clones that were mutant for a null allele of hdc ( Figure 4B , C , D , arrows ) , but the differentiation of R3/4 and R2/5 was not affected ( Figure 4E , F ) . Also there was no ectopic expression of markers for R3/4 or R8 in hdc mutant clones ( Figure S2D–F ) . In the adult hdc mutant ommatidia in the anterior half of the eye had a normal structure but had mis-rotation defects ( Figure 4G , black arrows ) , while ommatidia in the posterior half of the eye had missing photoreceptors ( red arrows ) , elliptical ( black arrowheads ) and split rhabdomeres ( red arrowhead ) ( Figure 4G ) . These phenotypes are similar to but weaker than those observed in unk mutant ommatidia ( Figure 1I ) . Loss of hdc also had no effect on photoreceptor proliferation or cell size ( Figure 3G , H ) . Together these data demonstrate that hdc is necessary to regulate the timing of photoreceptor differentiation and photoreceptor morphogenesis . Hdc is a ubiquitously expressed cytoplasmic protein and in the eye disc , similar to Unk , is expressed more strongly posterior to the MF in the apical plane of the disc containing differentiated photoreceptors and cone cells ( Figure 5A and S3B ) . Hdc is evenly distributed throughout the cytoplasm and does not have any punctate localisation ( Figure 5B ) . In unk mutant clones Hdc expression was increased in a specific spatiotemporal pattern; Hdc expression began to increase two to three rows posterior to the differentiation front for R1/6 ( Figure 5C , D , H ) . This dynamic increase in Hdc expression was also observed in Tsc1 mutant clones ( Figure 5E , F , I ) . Thus , Hdc expression is positively regulated by mTOR signalling and negatively regulated by unk , but only after the R1/6 differentiation front . Hdc expression was also slightly decreased in Rheb mutant clones , in which InR/mTOR signalling is inhibited ( Figure 5G , J ) . We also tested whether hdc regulates Unk expression . Unk expression was decreased in hdc mutant clones both anterior and posterior to the MF ( Figure 4C′′′ and 5K ) . Therefore , Unk and Hdc are mutually dependent on each other to maintain their expression levels and the precocious differentiation phenotype caused by loss of hdc may be due to its requirement to maintain the expression of Unk . To test whether unk and hdc act together double mutant clones were generated . unk , hdc double mutant clones caused precocious differentiation of R1/6 two to three rows ahead of the differentiation front ( Figure 6A , arrow ) , similar to the phenotype of either single mutant . unk , hdc double mutant clones in the adult eye also caused a similar phenotype to unk and hdc mutant clones ( Figure 6B ) . Therefore , unk and hdc act in the same pathway to regulate the timing of photoreceptor differentiation and photoreceptor morphogenesis . Overexpression of cDNAs for either unk or hdcFL ( that expresses both the short and full length forms of hdc ) alone in clones did not affect photoreceptor differentiation ( Figure S1E and 6C ) . However , combined overexpression of unk and hdcFL delayed the differentiation of R1/6 ( Figure 6D , arrow ) , but did not delay the differentiation of R3/4 ( Figure S5F ) , or affect cell growth ( Figure 3H ) . Moreover , overexpression of either gene individually in all cells posterior to the MF using GMR-Gal4 had no effect in the adult eye ( Figure 6F , G ) , but combined overexpression of unk and hdcFL caused the eye to have a glassy appearance , indicating a strong defect in eye development ( Figure 6H ) . Neither the delay in differentiation , nor the rough eye phenotype are due to increased apoptosis ( Figure S5A−E ) . Importantly , while overexpression of unk in Tsc1 mutant clones did not affect the precocious differentiation phenotype ( Figure 6I ) , co-overexpression of unk and hdcFL completely suppressed the precocious differentiation of R7 and cone cells caused by loss of Tsc1 ( Figure 6J ) . These data strongly suggest that unk and hdc act together to negatively regulate photoreceptor differentiation downstream of mTOR . D-Pax2 is a paired domain transcription factor and is the main regulator of cone cell fate in the developing eye [31] . The timing of cone cell differentiation is regulated by InR/mTOR signalling and cone cells precociously differentiate in unk and hdc mutant clones ( Figure 1E , F and 4C , D ) . We noticed that D-Pax2 expression increased in unk , hdc and Tsc1 mutant clones ( Figure 1F , 4D and S6A ) , demonstrating that the Unk/Hdc complex and mTOR signalling regulate D-Pax2 protein level in developing cone cells . Pax8 , part of the Pax2/Pax5/Pax8 subgroup of paired domain transcription factors that is homologous to D-Pax2 , physically interacts with the human Unk homolog [32] . We therefore tested whether Unk and D-Pax2 physically interact by expressing epitope-tagged forms of these proteins in S2 cells . Immunoprecipitation assays showed that Venus-Unk and FLAG-D-Pax2 co-immunoprecipitate ( Figure S6B ) . Therefore , the regulation of D-Pax2 protein levels by mTOR signalling and Unk/Hdc may contribute to the rate of cone cell differentiation though a direct physical interaction with D-Pax2 . There are two homologs of Unk in vertebrates , known as Unkempt ( Unk ) and Unkempt-like ( Unkl ) . Mouse Unk and Unkl both have a similar degree of identity to Drosophila Unk ( 45% , Figure S7 and S8 ) . By staining cells overexpressing an HA-tagged version of mouse Unkl we found that an antibody generated against human Unkl also recognises mouse Unkl ( Figure 7A ) . Staining of mouse embryonic day 14 . 5 ( E14 . 5 ) tissue with this antibody showed that Unkl is strongly expressed in the developing retina ( Figure 7B ) , suggesting that Unk may have a conserved role in eye development in Drosophila and mammals . mTOR signalling has recently been shown to be active in the early postnatal mouse subventricular zone ( SVZ ) , where it regulates neural stem cell ( NSC ) self renewal and differentiation [33] . Staining of the early postnatal brain showed that Unkl is expressed throughout the brain ( Figure S9A ) . Unkl is expressed throughout the SVZ , but is most strongly expressed in the cells close to the ventricle , similar to phosphorylated 4E-BP ( P-4E-BP ) , a marker of mTOR pathway activity ( Figure 7C , D ) . Further analyses showed that glial fibrillary acidic protein ( GFAP ) positive NSCs , Mash1 positive transit amplifying progenitors ( TAPs ) and neuroblasts ( identified by the expression of doublecortin ( Dcx ) ) all express Unkl ( Figure 7E–G and S9B–D ) . These data suggest that Unkl may play a role in mTOR-dependent neural stem/progenitor cell differentiation in the mammalian CNS . Thus , Unk may act downstream of InR/mTOR signalling to regulate neuronal differentiation in both Drosophila and mammals .
Several lines of evidence together demonstrate that unk and hdc act downstream of InR/mTOR signalling to negatively regulate the timing of photoreceptor cell fate . First , loss of either unk or hdc causes precocious differentiation of the same cells and to the same degree as activation of InR/mTOR signalling . Second , the expression of both Unk and Hdc are regulated by InR/mTOR signalling . Third , loss of unk suppresses the strong delay in photoreceptor differentiation caused by inhibition of the InR/mTOR pathway and combined overexpression of unk and hdc suppresses the precocious photoreceptor differentiation caused by loss of Tsc1 . Fourth , although Unk has been shown to physically interact with mTOR [21] , neither unk nor hdc regulate cell or tissue growth . Taken together these data show that unk and hdc are novel downstream components of the InR/mTOR pathway that regulate the timing of neuronal differentiation ( Figure 8 ) . InR/mTOR signalling is a major regulator of cell growth . In Drosophila activation of InR/mTOR signalling by loss of either Tsc1 , Tsc2 , Pten , or overexpression of Rheb causes increased cell size and proliferation [34]–[37] . In the genetic disease TSC , which is caused by mutations in Tsc1 or Tsc2 , patients develop benign tumours in multiple organs including the brain [4] . The previously identified components of the InR/mTOR pathway regulate both growth and neurogenesis in Drosophila and vertebrate model systems [8] , [10] , [13] , [14] , [16] , [17] , [38] . unk and hdc therefore represent a branchpoint in the pathway where its function in neurogenesis bifurcates from that in growth control ( Figure 8 ) . Moreover , our analysis of unk and hdc demonstrates that regulation of cell growth can be uncoupled from and is not required for the function of InR/mTOR signalling in the temporal control of neuronal differentiation . At the protein level we show that Unk and Hdc physically interact in S2 cells . Although this interaction remains to be demonstrated in vivo , the additional observations that they both regulate each other's expression and act synergistically in vivo strongly support the model that they physically interact ( Figure 8 ) . Moreover , Unk and Hdc have also previously been shown to physically interact by yeast-2-hybrid and co-immunoprecipitation [21] , [28] , [29] . Unk and Hdc are both expressed in all developing photoreceptors and so we hypothesise that they control the timing of differentiation through the regulation of neurogenic factors whose expression is restricted to R1/6/7 and cone cells ( Figure 8 ) . Loss of unk causes increased expression of D-Pax2 , the main regulator of cone cell differentiation . hdc and Tsc1 mutant clones also cause a similar increase in D-Pax2 expression . Overexpression of D-Pax2 alone is insufficient to induce cone cell differentiation , which requires overexpression of both D-Pax2 and Tramtrack88 ( TTK88 ) [39] . Thus , regulation of D-Pax2 expression by mTOR signalling may contribute to the rate of cone cell differentiation , while overall control would require the regulation of additional factors such as TTK88 ( Figure 8 ) . Pax8 , part of the Pax2/Pax5/Pax8 paired domain transcription factor subgroup that is homologous to D-Pax2 , has been shown to physically interact with one of the two human homologs of Unkempt [32] . We find that Drosophila Unk physically interacts with D-Pax2 , demonstrating that the physical interaction between Unk and this group of transcription factors is conserved . We suggest that D-Pax2 may be one of several neurogenic factors regulated by InR/mTOR signalling , through a physical interaction with the Unk/Hdc complex , to control the timing of R1/6/7 and cone cell fate ( Figure 8 ) . Unk has been shown to physically interact with mTOR and the strength of this interaction is regulated by insulin [21] . This suggests the intriguing possibility that the inhibition of Unk activity by InR/mTOR signalling is dependent on the strength of the physical interaction between Unk and the mTORC1 complex . Unk was also identified as part of the mTOR-regulated phosphoproteome in both human and murine cells [22] , [23] . Thus , Unk may potentially be regulated by mTOR through phosphorylation . Future studies will fully characterise the mechanism by which mTORC1 regulates Unk activity . Our study represents the first demonstration of a role for unk in specific developmental processes . By contrast , hdc has previously been shown to regulate dendritic pruning during metamorphosis and to act as a branching inhibitor during tracheal development [30] , [40] . A screen for genes affecting tracheal tube morphogenesis and branching recently identified Tsc1 [41] , suggesting that InR/mTOR also regulates tracheal development . Thus , hdc and unk may act repeatedly as downstream effectors of the InR/mTOR pathway during Drosophila development . The one previous study of either of the mammalian Unk homologs showed that Unkl binds specifically to an activated form of the Rac1 GTPase [42] . If this function is conserved in Drosophila then the defects in photoreceptor apical membrane morphogenesis caused by activation of mTOR signalling or loss of unk/hdc may be mediated through Rac1 . The function of the two unk homologs , unk and unkl , in mammalian development is not known , but unk has been shown to be expressed in the mouse early postnatal mouse retina [43] . We find that Unkl is also expressed in the developing mouse retina , suggesting that Unk may play a conserved role in eye development in both flies and mammals . InR/mTOR signalling acts as a pro-survival pathway preventing retinal degeneration [44] , but its role in mammalian eye development has not been characterised . By contrast InR/mTOR signalling has a well characterised role in NSC self-renewal and differentiation in the mouse SVZ . Loss of Tsc1 or expression of a constitutively active form of Rheb in neural progenitor cells in the postnatal mouse SVZ causes the formation of heterotopias , ectopic neurons and olfactory micronodules [15] , [16] . Furthermore , individuals with TSC , which results in activated mTOR signalling , have aberrant cortical neurogenesis and develop benign cortical tumours during foetal development and throughout childhood [4] , [45] . mTOR signalling has been shown to be active in proliferative NSCs and TAPs in the neonatal SVZ and inhibition of mTOR signalling prevents NSC differentiation [33] . We find that Unkl is expressed in both NSCs and TAPs in the early postnatal SVZ . Thus , Unkl may regulate NSC differentiation downstream of mTOR signalling in the mammalian brain . Unkempt may therefore play a conserved role in regulating the timing of neural cell fate downstream of mTOR signalling in both flies and mammals .
Flies were maintained on standard yeast , glucose , cornmeal , agar food at 25°C unless stated otherwise . Fly stocks were FRT82B , Dp110A [46] , FRT82B , Tsc1Q600X [34] , FRT82B , Tsc1Q87X [37] , FRT82B , Rheb2D1 [36] , FRT82B , hdc43 [19] and UAS-hdcFL [30] and UAS-unk ( this study ) . For clonal experiments the stocks used were y , w , hs-flp; FRT82B , ubi-GFP , y , w , hs-flp; FRT82B , M[95A]Rps63 , ubi-GFP and y , w , hs-flp;tub-Gal4 , UAS-mCD8GFP;FRT82B , tub-Gal80 ( MARCM stock ) . RNAi lines were obtained from the Vienna Drosophila Resource Centre . All other stocks were obtained from the Bloomington Stock Centre . For mosaic analysis mutant clones were generated by Flp/FRT mediated recombination using heat-shock-flp or by MARCM [47] . For clonal analysis in eye discs larvae were heat-shocked for 1-1 . 5 hours at 37°C 24 hours after egg laying ( AEL ) . For adult clones larvae were heat-shocked for 30 minutes at 37°C 24 hours and again at 48 hours AEL . Adult eye sections were prepared as described previously [48] . The line P[EPgy2]unkEY03956 was used to generate mutants unkex13 and unkex24 through imprecise P-element excision [49] . The deleted region in both unkex13 and unkex24 begins 557 bp into the second intron and for unkex13 ends 219 bp into the third exon and for unkex24 ends 150 bp into the fourth intron . The PiggyBac lines , PBac[RB]unke01984 and PBac[WH]f03929 were used to generate unkDf using a Flp/FRT –based precise excision strategy [50] . Mutation break points and deleted regions were confirmed or determined by PCR and sequencing . The Unk antibody was generated using a GST-fusion protein as described in Text S1 . The antibodies used were: rat anti-Unk3 ( 1∶500 , this study ) , mouse anti-Hdc ( a gift from Robert White , 1∶5 ) [19] , rat anti-Bar ( 1∶500 ) [10] , rat anti-Elav ( DSHB , 1∶100 ) , mouse anti-Prospero ( DSHB , 1∶100 ) , mouse anti-Rough ( DSHB , 1∶10 ) , rabbit anti-Spalt ( a gift from R . Barrio , 1∶500 ) , guinea pig anti-senseless ( a gift from Hugo Bellen , 1∶500 ) , rabbit anti-cleaved caspase 3 ( Cell Signalling , 1∶100 ) , rabbit anti-D-Pax2 ( a gift from M . Noll , 1∶20 ) [31] , rabbit anti-P-AKT ( Cell Signalling , 4045S , 1∶200 ) , rabbit or mouse anti-GFP ( Life Technologies , 1∶1000 ) , rabbit anti-PH3 ( Upstate , 1∶50 ) , rabbit anti RING finger protein unkempt like ( Abcam ab155197 , 1∶300 ) , rat anti HA ( Roche ) , mouse anti-Mash1 ( a gift from Francois Guillemot , 1∶20 ) , mouse anti GFAP ( Sigma , G3893 , 1∶1000 ) and goat anti Dcx ( Santa Cruz Sc8066 , 1∶200 ) . For immunofluorescence , dissected third instar larvae were fixed for 30 minutes in 4% paraformaldehyde , then washed five times for 10 minutes in PBS/0 . 1% triton ( PBST ) , before blocking in PBST/1% normal goat serum ( PBST-NGS ) for an additional hour . Eye discs were incubated with primary antibody overnight in PBST-NGS at 4°C . After five to six washes of 10 minutes in PBST discs were incubated for two hours with secondary antibody at room temperature , washed five to six times in PBST and then mounted in Vectashield ( Vectalabs ) . Phospho-AKT staining was performed exactly as described [51] . P0 mouse CNS tissue was fixed overnight in 4% paraformaldehyde then embedded in paraffin and 6 µm sections were cut . To generate primary neural progenitor monolayers , subventricular zone fragments from P1 mice were triturated with 2 ml of HBSS containing 0 . 25% trypsin ( Gibco ) and 40 µl of DNAse l ( 1 mg/ml , Worthington ) and incubated at 37°C for 2 minutes . The trypsin was inactivated with 5 ml of DMEM ( Gibco ) containing 10% FCS and the solution was centrifuged at 1 , 500 rpm for 5 minutes . After another two washes with DMEM/10% FCS to remove any traces of trypsin , the pellet was re-suspended in pre-equilibrated ( at 37°C/5% CO2 ) Neurobasal complete medium ( Gibco ) containing B27 supplement , 2 mM L-glutamine ( Invitrogen ) and 0 . 6% glucose ( Sigma ) . Cells were plated onto 24 well plates ( 2 . 5×106 cells/well ) on glass coated with polyornithine ( 0 . 5 mg/ml , Sigma ) . Cells were maintained in Neurobasal complete medium at 37°C/5% CO2 for 24 hours then fixed and stained as for COS-7 cells ( below ) . COS-7 cells were fixed for 30 minutes in 4% paraformaldehyde , washed several times in PBS , blocked for one hour in PBS/1% bovine serum albumen/0 . 2% triton/0 . 02% sodium azide , then incubated overnight at 4°C in the primary antibody diluted in blocking buffer . Secondary antibodies were FITC donkey anti-mouse , Cy3 donkey anti-rat , Cy5 donkey anti-rat and Cy5 donkey anti-mouse ( Jackson Immunolabs ) ; Alexa488 anti-rabbit , Alexa594 anti-rabbit and Alexa594 anti-mouse ( Life Technologies ) . Images were acquired on a Zeiss LSM710 confocal microscope and processed in Photoshop CS4 ( Adobe ) . The differentiation front was marked with a dotted line positioned just ahead of the most anterior row of photoreceptors . Details of the cloning of unk , hdc and D-Pax2 are described in Text S1 . UAS-unk transgenic lines were generated by germline transformation ( BestGene Inc . ) . Overexpression of Unk using these lines was confirmed by immunostaining . For immunoprecipitation ( IP ) experiments Venus-unk , FLAG-hdcS and FLAG-D-Pax2 were expressed in S2 cells ( DGRC ) cultured in SF9-S2 medium ( PAA laboratories ) . For IP experiments , cells were seeded in six well plates at a density of around 1 . 5 million cells per well . Cells were then transfected with 4 . 5 µg of Venus-unk , with or without 0 . 5 µg FLAG-hdcS or 0 . 5 µg FLAG-D-Pax2 using transfectin ( Biorad ) according to the manufacturer's instructions . The IP protocol was adapted from [52] . For each condition after 48 hours cells from two wells were manually detached , washed in cold PBS and lysed for 1 . 5 hours in lysis buffer ( 25 mM Tris pH 8 , 150 mM NaCl , 5% glycerol , 1% triton , 1 mM PMSF , 1× protease inhibitor cocktail ( Roche ) ) at 4°C . 800 µg of protein was then used for IPs . After clearing , lysates were incubated overnight with the antibody at 4°C . Venus-Unk was immunoprecipitated with 4 µg of rabbit anti-GFP antibody ( Life Technologies ) . FLAG-HdcS and FLAG-D-Pax2 were immunoprecipitated with 3 µg of rabbit anti-FLAG ( Fisher ) . The complexes were then immunoprecipitated for 3 hours using Protein G agarose beads ( Thermo Scientific Pierce ) . After washes in 25 mM Tris pH 8 , 150 mM NaCl , immunoprecipitated proteins were recovered by boiling the beads in 2× SDS-PAGE loading buffer and then subjected to SDS-PAGE followed by immunoblot with rabbit anti-GFP ( 1∶1000 ) and mouse monoclonal anti-FLAG M2 ( 1∶500 , Agilent ) . For overexpression of HA-Unkl COS-7 cells were grown in Dulbecco's modified Eagle's medium ( Life Technologies ) supplemented with 10% fetal bovine serum ( Life Technologies ) , in a humidified atmosphere of 5% CO2 at 37°C . For transfection COS-7 cells were seeded at a density of 2 . 5×105 cells per well in Opti-MEM Reduced Serum Medium ( Life Technologies ) and transfected with 5 µg of pcDNA3-HA-unkl [42] using Lipofectamine 2000 ( Life Technologies ) according to the manufacturer's instructions . For quantification of photoreceptor cell areas confocal images of MARCM clones stained with Bar were used . CD8-GFP expression was used to identify individual cell membranes , which were manually outlined at the level of R1/6 nuclei ( identified by Bar staining ) and cell areas were calculated using ImageJ . Clone and twin spot areas were manually outlined and the areas calculated using ImageJ . The numbers of active caspase 3 positive cells in eye discs were quantified using the quantification tool in Volocity ( Perkin Elmer ) from three dimensional confocal images of the whole disc with scans every 1 µm . Expression levels were determined in ImageJ using the Measure tool and four mutant clones/four adjacent areas of wild-type tissue were quantified for each genotype . Statistical analysis was performed in Graphpad Prism 5 . Statistical significance was determined using an unpaired Student's t test for pairwise comparisons , or one way analysis of variance ( ANOVA ) with Dunnett's multiple comparison post hoc test for multiple comparisons to the control . | The development of a functional nervous system requires that nerve cells are generated at exactly the right time and place to be correctly integrated . Defects in the timing at which nerve cells are generated , or ‘differentiate’ , lead to neurological disorders such as epilepsy and autism . However , very little is known about the identity of the genes that control the timing of nerve cell differentiation . Using developing photoreceptor nerves in the eye of the fruit fly , Drosophila , as a model , we showed previously that a molecular pathway known as ‘mTOR signalling’ is a key regulator of the timing of differentiation . In this study we have identified two new genes , unkempt and headcase , which control the timing of photoreceptor differentiation in Drosophila . The activity of unkempt and headcase is controlled by mTOR signalling and it is through these genes that mTOR is able to control nerve cell differentiation . The proteins encoded by unkempt and headcase form a complex and act synergistically to control the development of Drosophila photoreceptors . mTOR signalling controls a number of important cellular processes , but unkempt and headcase are the first components of this pathway to be identified that control nerve cell differentiation . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"developmental",
"biology",
"developmental",
"neuroscience",
"cellular",
"neuroscience",
"neurogenesis",
"molecular",
"neuroscience",
"biology",
"and",
"life",
"sciences",
"neuronal",
"differentiation",
"neuroscience",
"cell",
"differentiation"
] | 2014 | Unkempt Is Negatively Regulated by mTOR and Uncouples Neuronal Differentiation from Growth Control |
The entanglement of lignin polymers with cellulose and hemicellulose in plant cell walls is a major biological barrier to the economically viable production of biofuels from woody biomass . Recent efforts of reducing this recalcitrance with transgenic techniques have been showing promise for ameliorating or even obviating the need for costly pretreatments that are otherwise required to remove lignin from cellulose and hemicelluloses . At the same time , genetic manipulations of lignin biosynthetic enzymes have sometimes yielded unforeseen consequences on lignin composition , thus raising the question of whether the current understanding of the pathway is indeed correct . To address this question systemically , we developed and applied a novel modeling approach that , instead of analyzing the pathway within a single target context , permits a comprehensive , simultaneous investigation of different datasets in wild type and transgenic plants . Specifically , the proposed approach combines static flux-based analysis with a Monte Carlo simulation in which very many randomly chosen sets of parameter values are evaluated against kinetic models of lignin biosynthesis in different stem internodes of wild type and lignin-modified alfalfa plants . In addition to four new postulates that address the reversibility of some key reactions , the modeling effort led to two novel postulates regarding the control of the lignin biosynthetic pathway . The first posits functionally independent pathways toward the synthesis of different lignin monomers , while the second postulate proposes a novel feedforward regulatory mechanism . Subsequent laboratory experiments have identified the signaling molecule salicylic acid as a potential mediator of the postulated control mechanism . Overall , the results demonstrate that mathematical modeling can be a valuable complement to conventional transgenic approaches and that it can provide biological insights that are otherwise difficult to obtain .
The complex , interwoven structure of lignin , cellulose , and hemicellulose polymers in plant cell walls is the main cause for the recalcitrance of lignocellulosic feedstocks to microbial and enzymatic deconstruction towards fermentable sugars . This recalcitrance , in turn , accounts for the high cost of biofuel production from renewable sources [1] . In current technologies , the release of polysaccharides from the entanglement with lignin demands a thermo-chemical pretreatment that is expensive and has undesirable side effects during the later fermentation steps . Recent efforts aimed at decreasing the lignin content with transgenic techniques suggest that it might be feasible to reduce or even obviate the need for pretreatment [2] , which would permit the inclusion of polymer separation in downstream biomass processing technologies [3] and thereby make the cost of biofuel production competitive with that of fossil fuels . Reflecting the substantial impact of lignin on forage digestibility [4] , pulping efficiency [5] and sugar release from biomass [2] , considerable effort has been devoted towards a better understanding of monolignol biosynthesis in situ . Most pertinent genes have been identified in model species with complete sequence information , and knowledge from these genomes is currently being used for homology searches in species where sequencing efforts are ongoing . Such homology investigations are often effective , but caution is necessary , because multiple genes with similar sequences and annotations pointing to the same enzyme may possess distinct expression patterns and substrate preferences [5] . As a consequence , monolignol biosynthesis in vivo might be strikingly different among species and depend not only on gene sequences , but also on the tissue or even cell type of interest . Thus , before genetic modification strategies that had proven effective in some species are implemented in another species , it is prudent to consider and account for contextual differences . As a case in point , alfalfa ( Medicago sativa L . ) , the organism used for our analysis , exhibits substantial differences in cell wall composition among the different internodes of young plants . In most woody plants , the biochemical pathway of monolignol biosynthesis leads to three building blocks of lignin , which are known as p-hydroxyphenyl ( H ) , guaiacyl ( G ) and syringyl ( S ) monolignols ( Figure 1 ) . In potential bioenergy crops like poplar and switchgrass , lignin consists principally of G and S units , while H units are present in low to negligible quantities . In other plants , including some alfalfa transgenics , H can be present in significant amounts . Although the generic sequences of metabolic reactions within the monolignol pathway have been identified , it is becoming increasingly clear that critical details of the pathway structure and its regulation are not entirely understood . As a case in point , Chen et al . [6] recently introduced systematic , transgenic alterations in alfalfa ( Medicago sativa L . ) plants by independently modifying the activities of seven key enzymes of monolignol biosynthesis . While many of the results were easily explained , down-regulation of caffeoyl coenzyme A 3-O-methyltransferase ( CCoAOMT ) had little effect on S lignin , an observation that is conceptually inconsistent with the commonly accepted pathway structure ( Figure 1; black colored arrows ) . A recent study identified two isoforms of cinnamoyl CoA reductase ( CCR ) , MtCCR1 and MtCCR2 , in Medicago truncatula [7] . Furthermore , an earlier finding had suggested that caffeyl aldehyde is one of the preferred substrates for caffeic acid 3-O-methyltransferase ( COMT ) in alfalfa [8] . Taken together , these findings could imply an alternative route for S lignin synthesis ( Figure 1; red colored arrows ) upon CCoAOMT down-regulation [8] , [9] . However , they cannot explain why only G lignin is decreased because feruloyl-CoA is a common precursor of both G and S lignin . In dicotyledonous plants like alfalfa , the stem consists of many segments , called internodes . During maturation , all internodes grow asynchronously and thus independently represent different developmental stages . This phenomenon suggests a customized modeling approach: Instead of studying the pathway within a single developmental context , it seems advantageous to launch a systematic investigation that simultaneously encompasses dozens of internodes from seven wild-type or transgenic plants . This comprehensive approach supports the fact that lignin biosynthesis is tightly coordinated by a hierarchy of transcription factors during secondary wall thickening [10] . It also circumvents the potential problem that regulatory mechanisms might escape discovery during an analysis based on singular phenotypic datasets , such as lignin content and monomer composition , if only one internode or one transgenic line is studied at a time . This potential failure to detect regulatory signals is exacerbated in the lignin system by the fact that several enzymes in the pathway catalyze multiple steps , which makes intuitive analyses difficult . With a comprehensive analysis of several datasets as the target , we propose here a novel modeling approach that integrates the data in a semi-dynamic fashion . First , flux balance analysis ( FBA ) [11] is applied independently in each individual internode of the wild-type plant . In contrast to microbial systems , where maximization of the growth rate is usually assumed to be the species' overall objective , we use the monolignol production as the objective function for FBA . Second , for every internode of a lignin-modified line , we use the method of minimization of metabolic adjustment ( MOMA ) [12] to characterize the altered flux distribution in relation to the corresponding FBA solution for the same wild-type internode . Specifically , the relative proportions of the fluxes leading to three lignin monomers are constrained at experimentally-observed values to improve the prediction . Finally , we perform a Monte Carlo-like simulation of randomly parameterized kinetic models in cases where the results arising from the static models may have alternative , kinetics-based explanations . This combined modeling approach represents , to the best of our knowledge , the first computational study of lignin biosynthesis in angiosperm stem tissues and , more generally , of secondary plant metabolism in angiosperms . As we will discuss later , the model analysis resulted in six postulates concerning the metabolic control of monolignol biosynthesis that had not been considered at all or at least not in detail . These postulates address the reversibility of some enzymatic reactions , shed light on the hypothesis of independent pathways for the synthesis of G and S monolignols , and suggest a novel feedforward regulatory mechanism exerted by a cinnamic acid-derived compound . Of note is the fact that evidence in support of this last postulate has subsequently been obtained in laboratory experiments . By critically evaluating the transgenic data against a revised pathway structure in alfalfa , we hope these postulates will not only serve as guidelines for directing future experiments , but also provide mechanistic insights that will aid the design of combined genetic modification strategies toward the generation of bioenergy crops with reduced recalcitrance .
Accounting for recent experimental observations , we adopted a revised pathway structure of monolignol biosynthesis in alfalfa stems that includes the CCR2-catalyzed reduction of caffeoyl-CoA to caffeyl aldehyde and the subsequent synthesis of coniferyl aldehyde by COMT ( Figure 1: black and red colored reactions ) , as explained earlier . The pathway of monolignol biosynthesis contains a fairly small number of branch points , and it is known that flux partitioning at these branch points determines the ultimate transport fluxes v6 , v15 and v19 and thus the relative amounts of lignin monomers ( cf . [13] ) . The FBA-derived steady-state flux analysis for wild-type plants supports this argument . It suggests that variation in lignin composition from young to mature internodes is accomplished by modulating the flux partitioning at three principal branch points: p-coumaroyl-CoA , coniferyl aldehyde , and coniferyl alcohol . As a paradigm illustration , the proportion of H lignin declines from 7% of the total monomer yields in the first two internodes to 1% in the eighth internode . This decline is singularly achieved through a monotonic decrease in v4 ( Figure 2A ) . A parallel increase in the ratio of S to G lignin—commonly termed the S/G ratio—from 0 . 09 in the first two internodes to 0 . 64 in the eighth internode requires a combined effort of flux adjustments at coniferyl aldehyde and coniferyl alcohol ( Figure 2B ) . Since F5H controls the first committed steps ( i . e . , v16 and v20 ) towards the synthesis of S lignin , one would expect to see its expression being up-regulated in mature versus young internodes , which has recently been validated by microarray analysis ( Table 4 of [14] ) . For a systemic analysis of the pathway we used the results of a gene modification study in alfalfa where genes encoding for PAL , C4H , HCT , C3H , CCoAOMT , F5H , and COMT were independently down-regulated . With the exception of F5H-modified lines , which did not permit measurements of the targeted enzyme activity , we applied MOMA to each strain and each internode and predicted the new steady-state flux distribution ( see Materials and Methods ) . A very interesting result is the fact that no feasible solution exists for four of the six transgenic plants , if the revised metabolic map is correct ( Figure 1; black and red colored arrows ) . For example , if C4H activity is down-regulated to 45% of its wild-type level , it is analytically impossible to derive a set of fluxes that satisfies the mass balance at cinnamic acid as well as the observed lignin composition , if the supply of phenylalanine is constant . To remedy this situation , it seems to be necessary to add to the pathway structure three “overflow” fluxes counteracting the potential accumulation of the intermediate metabolites cinnamic acid , p-coumaryl aldehyde , and 5-hydroxyconiferyl alcohol ( blue arrows v22 , v23 , v24 in Figure 1 ) . This proposed amendment is at least partially supported by observations . First , salicylic acid ( SA ) , an essential signaling molecule for systemic acquired resistance against pathogen attack , can be formed from cinnamic acid [15] , [16] , [17] , although it may also originate from the shikimate pathway via isochorismate [18] . Second , the biosynthesis of all flavonoids begins with the condensation of p-coumaroyl-CoA and three molecules of malonyl-CoA by the enzyme chalcone synthase [19] . And third , incorporation of 5-hydroxyconiferyl alcohol into lignin polymer is found in the COMT-deficient alfalfa [20] . Thus , we included these additional effluxes , and the expanded system ( Figure 1; v1 to v24 ) permitted feasible solutions in all cases tested . In wild-type plants , the FBA-derived steady-state values of the three added fluxes are minimized to prevent lignin precursors from being channeled into peripheral pathways producing SA or flavonoids . In the transgenic plants , these auxiliary fluxes are no longer restricted to small values and thus can be raised to substantial levels to facilitate the re-distribution of fluxes . However , the assumption that the peripheral fluxes are minimized in wild-type plants must be handled with caution: although the phenylpropanoid pathway in cells undergoing secondary wall thickening may evolve towards maximizing the synthesis of lignin precursors , this is apparently not the case when biosynthesis of flavonoid-derived products , which may function as floral pigments or as anti-microbial agents , becomes the plant's top priority . The MOMA analysis revealed flux distributions for all transgenic lines and their individual internodes . Figure 3 shows the developmental evolution of flux patterns in CCoAOMT-deficient plants; similar plots for other transgenic plants are given in Figures S1 , S2 , S3 , S4 , and S5 . Of note is that all computed fluxes exhibit strong and essentially monotonic trends: for each transgenic line , the flux partitioning at important branch points follows clear trends throughout the internodes rather than jumping in value from one internode to the next . This result is surprising and encouraging , because MOMA simply assumes that the fluxes undergo a minimal re-distribution when the pathway system is perturbed . Because these perturbations occur independently for each internode , there is no mathematical guarantee that individual fluxes would follow any smooth trend from internode to internode . In other words , the collective results , while fitting into the context of a gradual change in lignification pattern during stem development , are by no means “automatic , ” because no external constraints or conditions were imposed or enforced on the transition from one internode to the next . The computed trends are summarized in Table 1 . The following paragraphs are structured as follows . First , we re-evaluate the gene knock-down data in a systematic way across different stages of growth and formulate four postulates that actually do not require a full model analysis , but emerge from the “logic” of the pathway . Second , we discuss two postulates regarding novel mechanisms of metabolic regulation that result from our comprehensive model analysis . Third , we present new experimental results that directly support one of the model-based postulates . The total lignin production is driven by the availability of phenylalanine rather than by enzymatic limitations . This conclusion results from the observation that the down-regulation of PAL has much less effect on total lignin content and/or lignin composition in young internodes with small amounts of lignin than in mature internodes with high lignin production ( Table S3 in Text S1; [6] ) . Expressed differently , PAL is not acting at capacity when the demand for lignin is relatively low , as is the case in young internodes . This conclusion is also supported by the observation that lignin production is not enhanced proportionately when PAL enzyme is over-expressed in transgenic plants [21] . In transgenic plants where C3H is down-regulated , the proportion of H lignin among total monomer yields is significantly increased over control plants , especially in mature internodes ( Figure 4A ) . This finding is at first puzzling , because it is unlikely that the cell can detect changes in C3H activity and adapt accordingly by exerting appropriate flux control at an earlier branch point ( i . e . , p-coumaroyl-CoA ) within the network . Arguably the simplest explanation is that HCT ( possibly along with other plant acyltransferases ) is reversible [22] . If so , the following scenario is possible: as p-coumaroyl-shikimate accumulates due to a reduced C3H activity , HCT converts it back to p-coumaroyl-CoA in the presence of free CoA , thereby allowing the cell to escalate the production of H lignin beyond the wild-type level . The catalytic efficiency of HCT acting on p-coumaroyl-shikimate as substrate remains to be experimentally determined , along with the possible competition for CoA between two shikimate esters ( i . e . , p-coumaroyl-shikimate and caffeoyl-shikimate ) . The hypothesis of HCT being reversible prompts us to investigate whether C3H , which controls the material flow between two HCT-catalyzed steps , also permits catalysis in both directions . A slightly increased proportion of H lignin in CCoAOMT-deficient plants ( Figure 4A ) seems to suggest that C3H is mildly reversible and that part of the accumulated caffeoyl-CoA is therefore converted back to p-coumaroyl-CoA and subsequently channeled towards H lignin , a scenario which seems unlikely based on the known catalysis by cytochrome P450 enzymes . However , the amounts of H lignin determined by thioacidolysis appear to be unaffected by the low CCoAOMT activity despite a noticeable decrease in total lignin content ( Table S3 in Text S1; [6] ) . One plausible explanation is that thioacidolysis yields are highly correlated with the in vivo abundance of S lignin [5] , which might suggest that plants may in effect produce more H lignin than was measured against the down-regulation of CCoAOMT . If both HCT and C3H are reversible , the two CCR-catalyzed reactions—v10 and v13—can be regarded as the “committed” steps ( i . e . , they are essentially irreversible ) , because manipulation of any downstream enzyme , such as COMT and F5H , has no substantial effect on H lignin ( Figure 4B ) . Interestingly , the postulate seems to echo the conclusion from a previous enzyme assay [23]: CCR purified from poplar stems was able to catalyze the conversion of coniferaldehyde into feruloyl-CoA in the presence of other co-factors but preferentially reduced CoA-esters , as judged by the calculated equilibrium constants . In addition to a modest increase in H lignin , down-regulation of CCoAOMT leads to a noticeable increase in the S/G ratio of all internodes except for internodes 1 and 2 ( Figure 5A ) . This finding is puzzling because coniferyl aldehyde is a common precursor to both S and G lignin and one would therefore expect a similar effect on both . The analogous situation arises in COMT-deficient plants , where the S/G ratio is reduced ( Figure 5A ) . This case , however , is not quite as clear-cut because COMT also shows activities towards downstream intermediates like 5-hydroxyconiferyl aldehyde and 5-hydroxyconiferyl alcohol . Thus , in this case of COMT deficiency , the S/G ratio might not be a good indicator of the flux partitioning at coniferyl aldehyde towards G and S lignin . As an explanation for the altered S/G ratios in cases of CCoAOMT or COMT down-regulation , we postulate that the enzymes controlling v12 and v16 ( and maybe even v10 and v17 ) are organized into a functional complex through which the intermediates are channeled without much leakage . Similarly , we postulate that v13 and v14 form a corresponding complex without much leakage . This dual postulate for crossing channels is supported indirectly by literature information and by findings from our flux analysis , as outlined below . First , an analysis of mature stems ( internodes 6–9 ) collected from CCoAOMT down-regulated transgenic lines indicated that the levels of G lignin were greatly reduced , whereas those of S lignin were nearly unaffected ( cf . CCOMT antisense line ACC305 in Table 1 of [24] ) . Similarly , down-regulation of CCR1 , which actively catalyzes the subsequent reduction of feruloyl-CoA to coniferyl aldehyde , also resulted in an increased S/G ratio in mature internodes of alfalfa stems [25] , again with G lignin being more strongly reduced than S lignin . Although the existence of the CCR2-COMT pathway helps sustain the lignin content in either CCoAOMT or CCR1 down-regulated lines , the findings do not explain why S lignin is synthesized at the expense of G lignin upon genetic modifications of the CCoAOMT-CCR1 pathway . Nevertheless , the findings are entirely consistent with the postulate of crossing channels . Second , one of the constituent enzymes , F5H , is localized to the external surface of the endoplasmic reticulum [26] , so that the proposed channel may exist in the form of an enzyme complex anchored in the endomembrane . Indeed , a labeling experiment in microsomes extracted from lignifying alfalfa stems suggested such a co-localization of COMT and F5H [27] . It showed that caffeyl aldehyde , when incubated with [methyl-14C]-labeled S-adenosyl L-methionine ( a co-substrate necessary for COMT-mediated O-methylation ) and NADPH ( the reducing agent for F5H ) , is converted to coniferyl aldehyde , 5-hydroxyconiferyl aldehyde , and a small amount of sinapyl aldehyde . Finally , our flux distribution analysis reveals a strong correlation between the computed flux values of v13 and v14 for all but the CCoAOMT-deficient plants ( Pearson correlation coefficient ρ = 0 . 9952; p-value<0 . 001 ) ( Figure 6 ) . This correlation suggests that there is normally almost no exchange of products between v12 and v13 , and that most of the coniferyl aldehydes produced through the CCR2-COMT shunt are directly utilized by F5H without having the opportunity of diversion into G lignin biosynthesis . A notable exception seems to be the situation where CCoAOMT is significantly down-regulated . In this case , caffeoyl-CoA tends to accumulate at least in the short term , thus providing the CCR2-COMT pathway and the associated metabolic channel with an abundance of substrate . The predicted flux distribution ( Figure 3 ) and the observed lignin composition ( Table S3 in Text S1 ) indicate that CCoAOMT-deficient plants produce a considerable amount of G lignin , although the levels of S lignin are comparable to those in the controls , which implies that only some of the extra caffeoyl-CoA can be converted efficiently into S lignin through the proposed channel . Overall , the proposed functional channels seem to be consistent with results of the flux analysis as well as with earlier discussions in the literature [8] , [9] . The correlation between v12 and v16 is less pronounced , which is presumably due to the fact that F5H and COMT catalyze parallel pathways , with the latter ( v20 and v21 ) buffering changes in earlier precursors . An alternative explanation for an increased S/G ratio upon modifications of the CCoAOMT-CCR1 pathway could be that the kinetic features of the enzymes that catalyze coniferyl aldehyde and coniferyl alcohol are fine-tuned such that they could permit the adjustment of fluxes leading to G and S lignin and thus change the S/G ratio . For instance , given that down-regulation of CCoAOMT or CCR1 may alter the intracellular level of coniferyl aldehyde , the relative values of v14 and v16 at steady state could depend on whether the respective enzyme works within the linear or saturation region of its kinetic profile . To investigate this alternate hypothesis , we designed and analyzed a kinetic Michaelis-Menten model that contains the two alternative pathways from caffeoyl-CoA to coniferyl aldehyde as well as the two principal branch points where the fluxes leading to G and S lignin diverge ( see Text S1 ) . The model was simulated 10 , 000 times with randomly sampled kinetic parameter values , as described in Materials and Methods and Text S1 , and we recorded the percentage of admissible parameter sets that yielded a significantly increased S/G ratio in response to a 80% reduced CCoAOMT or CCR1 activity . We first examined the case where CCoAOMT is down-regulated . Only ∼5% of all admissible systems ( see Text S1 for definition ) yielded a significantly increased S/G ratio , whereas nearly half of all systems resulted in an S/G ratio that differed by less than 5% . The few cases of significant increases in the S/G ratio did not reveal particular patterns , which may not be too surprising because the system involves 16 kinetic parameters that affect each other in a nonlinear fashion . Intriguingly , for the scenario of CCR1 down-regulation , none of the admissible systems showed a significant increase in S/G ratio; in fact , all changes in S/G ratios were less than 0 . 5% . Replacing the Michaelis-Menten kinetics with cooperative Hill kinetics allowed more flexibility . Still , only ∼3% of all admissible systems exhibited an increase in S/G ratio upon CCR1 down-regulation . Taken together , it seems that , theoretically , some precisely tuned sets of kinetic parameters could lead to the observed effects on the S/G ratio . However , these sets are rare and do not seem to be robust enough to render the kinetics-based hypothesis viable . One of the most paradoxical findings among the collective results from the transgenic plants is the opposite effect on lignin composition ( and specifically the S/G ratio ) when either PAL or C4H is down-regulated . It seems that these alterations should not differentially affect monolignol biosynthesis , because both occur before the first branch point , but they do . Closer inspection of the data from different internodes reveals that the S/G ratio is consistently increased in PAL-deficient plants but decreased in C4H-deficient plants ( Figure 5B ) . While experiments with tobacco have suggested that the differential co-localization of PAL isoforms and C4H might be the underlying cause of such observations [28] , there is as yet no direct evidence for this intracellular association in alfalfa or other related legume species . In accordance with the proposition of separate metabolic channels for G and S lignin , we postulate that the different effects of PAL or C4H down-regulation on the S/G ratio are due to feedforward regulation . Specifically , we suggest that this regulation is mediated by a downstream product of the cinnamic acid degradation pathway , which is represented collectively as v22 in Figure 1 . Notice that this feedforward regulation had not been recognized by the scientific community and was postulated by the model analysis purely with computational means . Consistent with the observation of all transgenic experiments , an appropriate control strategy by this unknown compound X is summarized in Figure 7 and discussed below . In the case of PAL-deficiency , where the biosynthesis of cinnamic acid from phenylalanine declines , a diminished pool of X could directly or indirectly reduce the expression of CCoAOMT/CCR1/CAD and/or activate the expression of CCR2/COMT/F5H , thereby altering the channeling towards G and S lignin and increasing the S/G ratio . Intriguingly , this proposed inhibition of CCoAOMT expression following PAL down-regulation is supported by a strong correlation of the proportion of G and S lignin in total monomer yields in internodes 4–8 of the PAL- and CCoAOMT-deficient plants ( Figure 8 ) . In the case of C4H deficiency , however , the production of X through v22 is likely to increase because the consumption of cinnamic acid through a competing branch v2 is not as effective as in wild-type plants . Thus , an accumulation of X could in turn activate the expression of CCoAOMT/CCR1/CAD and/or reduce the expression of CCR2/COMT/F5H , leading to a smaller S/G ratio . Salicylic acid ( SA ) is a notable endogenous signaling molecule that is known to be derived from cinnamic acid [29] . Down-regulation of one pathway enzyme other than C4H ( e . g . HCT [30] ) had recently been shown to lead to elevated levels of SA . To investigate whether SA is the postulated signaling compound X , we measured its intracellular levels in many independent transgenic alfalfa lines in which different monolignol biosynthesis genes had been down-regulated . Indeed , the results show that the intracellular levels of SA are highly proportional to the extent of lignin reduction ( Figure 9 ) . Based on our postulated feedforward regulation , this effect can be explained through the participation of SA in the inhibition of the metabolic channel committed to S lignin biosynthesis , thus reducing the total lignin content .
Functional genomics is a premier tool for identifying metabolic pathways in sequenced model species and for pinpointing genes involved in them [31] . However , it is known that many enzymes coexist in multiple isoforms with unique expression patterns and substrate specificities . A pertinent example seems to be the recent discovery of two CCR isoforms with distinct catalytic properties towards major CoA-esters in Medicago [7] . Steady-state flux analysis of an extended pathway system that accounts for the isoforms reveals that the alternative path is dispensable in wild-type plants , but that it may rise to significant levels in specific transgenic lines . Indeed , CCoAOMT-deficient plants support a much higher lignin production than lines where HCT or C3H is down-regulated ( Table S3 in Text S1; [6] ) . The intricate differences in pathway operation among otherwise very similar transgenic lines point to the need of investigating flux patterns not only in different plants , but also in different strains , lines and even different internodes and tissues . The results shown here furthermore demonstrate that subtle variances among tissues and lines are difficult to discern with intuition alone , but that computational analyses can serve as objective and rigorous tools for explaining such differences . Specifically , the new integrative modeling approach proposed here combines static flux-based models and a Monte Carlo simulation of randomly parameterized kinetic models . This approach has the advantage that it allows the collective analysis of many experimental results and sheds light on pathway features that are particularly important for functionality under normal and altered conditions . The analysis here revealed a quantitative trend of flux patterns during development , which in turn allowed the identification of principal branch-point metabolites at which internode-specific flux partitioning patterns control the observed mode of lignification . While it is relatively easy to single out principal metabolites in linear or slightly branched pathways , the system studied here is confounded by the plant's employment of the same enzymes , such as CCR and CAD , in different key positions . Due to this multiple use , manipulating the flux partitioning pattern towards a desired mode of lignification may incur undesired “side effects . ” The computational analysis indicates that a single flux analysis just for wild-type plants is insufficient for understanding because even a seemingly simple pathway like monolignol biosynthesis requires relatively minor , yet important , extensions to account for the overflow of some intermediate metabolites that only occurs in transgenic plants . At the same time , the analysis also demonstrates that the simultaneous analysis of several independent datasets , in this case transgenic lines and sequential internodes , can lead to insights that otherwise would have been difficult to obtain . Here , it led to several postulates that are specific enough for experimental validation or refutation . Some model-free postulates refer to the need for reversibility or committedness of key reactions , which might not be too surprising . Two further postulates are more intriguing . They refer to the functional channeling within the pathway and its mechanistic control . Based on the observation of an increased S/G ratio in CCoAOMT or CCR1 down-regulated lines , the computational results suggest an S lignin-specific channel capable of converting caffeyl aldehyde directly into 5-hydroxyconiferyl aldehyde or sinapyl aldehyde . Different experiments in the literature suggested the co-localization of COMT and F5H in lignifying alfalfa stems [27] and the localization of F5H to the external surface of the endoplasmic reticulum [26] . These and our findings would imply the likely location for a functional S-channel complex to be associated with the endomembrane . While the proposed membrane-bound channel for synthesizing S lignin could constitute an important control mechanism , it may only have comparatively limited capacity because even in CCoAOMT down-regulated lines G lignin is generated in a higher proportion of total monomer yields than S lignin ( Table S3 in Text S1; [6] ) . One likely cause is that different O-methyltransferases ( OMTs ) are involved in converting caffeyl aldehyde to coniferyl aldehyde . These OMTs may have distinct sub-cellular localization ( to cytoplasm or endomembrane ) and therefore a different affinity to F5H . Thus , it could be that the cytosolic OMT in the transgenic lines with reduced CCoAOMT expression is up-regulated and helps consume extra caffeyl aldehyde outside the proposed channel . A corresponding labeling experiment in alfalfa [27] confirmed that only a small proportion of total cellular COMT activity against caffeyl aldehyde is associated with the microsomal membrane , and that adding excess recombinant COMT has little effect on the metabolism of caffeyl aldehyde by microsomes . To examine whether the observed increase in the S/G ratio upon modifications of the CCoAOMT-CCR1 pathway could be explained alternatively by a kinetically-controlled mechanism , we generated 10 , 000 ODE models for a reduced pathway system ( Text S1 ) and simulated both down-regulation schemes . Among all sampled parameter sets , only a minute percentage of systems had the ability to increase their S/G ratio significantly in either case . Although the results neither reject the possibility of a kinetically-controlled S/G ratio nor directly corroborate our channeling postulate , they do suggest that purely kinetic control might be unlikely , because it would require rather precise implementations of specific parameters in different tissues , which seems to compromise the robustness of the system . As shown in a structural study of alfalfa COMT [32] , mutations of some key residues lining the active site result in significantly different substrate binding and/or turnover rate . Moreover , it is likely that the kinetic properties of other enzymes may also exhibit a similar , if not more severe , susceptibility to genetic perturbations ( e . g . , [33] , [34] ) . Since the variation in the S/G ratio is typically small ( s . d . ≈0 . 03 in two control lines; [6] ) , the proposed functional channeling mechanisms seem to offer a more robust option to help maintain a physiologically proper S/G ratio . The observed decrease in the S/G ratio of COMT down-regulated lines alone is not sufficient to prove the existence of a G lignin-specific channel , because a reduced COMT activity affects all fluxes that are specific for the synthesis of S lignin , thus leading to a smaller S/G ratio . Nevertheless , the strong correlation between v13 and v14 that emerged from our computations for most transgenic experiments lends further credence to such an inference . This correlation not only supports the operation of a G lignin-specific channel , but also hints at the possibility of CCR1 and CAD ( and maybe CCoAOMT ) being complexed or co-localized on internal membranes . One option for testing this postulate would be to down-regulate CCR2 and record if the strain exhibits a greater decrease in S lignin than in G lignin , giving a smaller S/G ratio . Surprisingly , knocking out CCR2 in M . truncatula , a species closely related to alfalfa , leads to an increased S/G ratio , whereas M . truncatula CCR1 knock-out mutants show a reduction in the S/G ratio [7] . However , in spite of their close taxonomic relatedness , the operation and control of monolignol biosynthesis might be quite different in tetraploid alfalfa ( M . sativa L . ) and diploid M . truncatula . For instance , the S/G ratio in wild-type alfalfa stems ( 0 . 62; internodes 1–8 ) is approximately twice as large as that in wild-type M . truncatula stems ( 0 . 29; internodes 1–7 ) . Consequently , further experimental work is required to validate or reject the postulate that a G lignin-specific channel is operational in alfalfa . If the postulates of specific channels towards the synthesis of G and S lignin are valid , one may further surmise that the opposite effects of PAL or C4H down-regulation on lignin composition are the results of differential gene or enzyme expression , which could be mediated by a cinnamic acid derivative . However , the model could not identify this molecule , leading us to call it Compound X . Supporting this hypothesis , the transgenic experiments used here have shown that down-regulation of CCoAOMT , which we postulate to be involved in the G lignin-specific channel , yields similar proportions of G and S lignin among total monomers as does the down-regulation of PAL , which is postulated to inhibit and/or activate the functioning of the G lignin- and S lignin-specific channels , respectively ( Figure 8 ) . Salicylic acid ( SA ) , a phenolic phytohormone derived from phenylalanine , was proposed as a potential candidate for this unknown Compound X . Intriguingly , post-hoc experiments showed that the intracellular levels of SA are indeed highly proportional to the extent of lignin reduction in transgenics where different pathway genes are down-regulated ( Figure 9 ) . This result fits directly into the context of our feedforward control postulate . At the same time , it makes us wonder why putting a block on monolignol biosynthesis could affect the homeostasis of SA , especially if the blockage is located away from the pathway entrance . Based on previous findings that SA can be derived both from cinnamic acid and from isochorismate via the shikimate pathway [29] , and that HCT uses shikimate as a preferred cofactor ( Figure 10 ) , we propose the following scenario: when the flux going through the pathway is decreased due to some genetic manipulation , fewer shikimates will be trapped in those shikimate esters ( p-coumaroyl-shikimate and caffeoyl-shikimate ) and thus become available to make SA . In other words , the shikimate recycling facilitated by HCT enables the shikimate pool to work as a sensor of the flux into lignin . Future in-depth studies , whether they are experimental or computational , are required to justify this hypothesis . It is noteworthy , however , that the reason why plants shuttle monolignol pathway intermediates between Coenyme A and shikimate esters has yet to be explained . In conclusion , our analysis shows that a combined modeling effort can uniquely and effectively complement experimental studies of the type used here . In contrast to analyzing one dataset at a time , it allowed us to integrate all results from a comprehensive experimental investigation of various transgenic lines and internodes . This integration , in turn , revealed dynamic , developmental patterns and their dependence on key enzymes . Together , the analyses uncovered elusive control of monolignol biosynthesis and led to testable hypotheses regarding various pathway aspects that should be clarified before one attempts to generate and optimize viable , productive “designer” crops with minimal recalcitrance .
In a previous study [6] , lignin content and composition were analyzed in transgenic alfalfa plants in which seven enzymes were independently down-regulated ( cf . Figure 1 ) . These enzymes were: L-phenylalanine ammonia-lyase ( PAL ) , cinnamate 4-hydroxylase ( C4H ) , hydroxycinnamoyl CoA:quinate/shikimate hydroxycinnamoyl transferase ( HCT ) , coumarate 3-hydroxylase ( C3H ) , caffeoyl coenzyme A 3-O-methyltransferase ( CCoAOMT ) , ferulate 5-hydroxylase ( F5H ) , and caffeic acid 3-O-methyltransferase ( COMT ) . Each transgenic plant was cultivated to early flowering stage , and the mature stem consisting of eight internodes was harvested and divided into individual segments; all internodes were numbered according to their maturity , with internodes 1–2 representing the pooling of the two uppermost stem segments . The lignin content and monomer composition for each internode were determined for each transgenic line via established protocols [6]; the results are summarized in Table S3 in Text S1 . The activities of all targeted enzymes were also measured and summarized elsewhere , with the exception of F5H , which showed no activity towards any documented substrates when assayed in crude alfalfa extracts in vitro ( Table 2c of [6] ) . Thus , the F5H-deficient line is excluded from the following analysis . | Cellulose-based biofuels presently offer the most environmentally attractive and technologically promising alternative to fossil fuels . To be viable , biofuels must be derived from non-food crops , such as grasses , wood , bark , and plant residues . Techniques for releasing the energy stored in these renewable materials must first untangle a very recalcitrant scaffold of interlinking molecules inside the plant cell walls , which is very costly . Much of the recalcitrance is due to the natural polymer lignin , which hardens the cell walls and is composed of three different building blocks , called monolignols . Modern transgenic techniques have yielded plant lines whose cell walls are easier to break down , but some of these modified plants have exhibited unexplained and undesired features . Here , we present new computational methods for analyzing monolignol biosynthesis in unprecedented detail . The analysis simultaneously accounts for lignin biosynthesis in various transgenic lines and different developmental stages and yields six novel , testable postulates regarding the metabolic control of the pathway . The results suggest new , targeted experiments towards a better understanding of monolignol biosynthesis and issues of recalcitrance reduction . More generally , the results highlight the genuine benefits of using computational methods as companions and complements to experimental studies . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"systems",
"biology",
"computer",
"science",
"plant",
"science",
"computer",
"modeling",
"theoretical",
"biology",
"plant",
"biology",
"biology",
"genetics",
"and",
"genomics"
] | 2011 | Integrative Analysis of Transgenic Alfalfa (Medicago sativa L.) Suggests New Metabolic Control Mechanisms for Monolignol Biosynthesis |
HRMs ( hypoxia-responsive miRNAs ) are a specific group of microRNAs that are regulated by hypoxia . Recent studies revealed that several HRMs including let-7 family miRNAs were highly induced in response to HIF ( hypoxia-inducible factor ) stabilization in hypoxia , and they potently participated in angiogenesis by targeting AGO1 ( argonaute 1 ) and upregulating VEGF ( vascular endothelial growth factor ) . Here we constructed a novel computational model of microRNA control of HIF-VEGF pathway in endothelial cells to quantitatively investigate the role of HRMs in modulating the cellular adaptation to hypoxia . The model parameters were optimized and the simulations based on these parameters were validated against several published in vitro experimental data . To advance the mechanistic understanding of oxygen sensing in hypoxia , we demonstrated that the rate of HIF-1α nuclear import substantially influences its stabilization and the formation of HIF-1 transcription factor complex . We described the biological feedback loops involving let-7 and AGO1 in which the impact of external perturbations were minimized; as a pair of master regulators when low oxygen tension was sensed , they coordinated the critical process of VEGF desuppression in a controlled manner . Prompted by the model-motivated discoveries , we proposed and assessed novel pathway-specific therapeutics that modulate angiogenesis by adjusting VEGF synthesis in tumor and ischemic cardiovascular disease . Through simulations that capture the complex interactions between miRNAs and miRNA-processing molecules , this model explores an innovative perspective about the distinctive yet integrated roles of different miRNAs in angiogenesis , and it will help future research to elucidate the dysregulated miRNA profiles found in cancer and various cardiovascular diseases .
When cells are exposed to low oxygen tension , cellular adaptation occurs by transcriptionally activating a variety of genes that participate in pathways involving angiogenesis , metabolism and proliferation/survival [1] . The oxygen-sensitive transcription factor HIF-1 ( hypoxia-inducible factor 1 ) , with over 1000 putative targets in human , is the master mediator of this response [2 , 3] . HIF-1 is a heterodimer of HIF-1β subunit , which is constitutively expressed regardless of O2 availability , and HIF-1α subunit , whose expression is highly dependent on O2 levels . In normoxia , HIF-1α protein levels are undetectable [4]; they are rapidly hydroxylated by PHDs ( prolyl hydroxylases ) and FIH-1 ( factor inhibiting HIF-1 , abbreviated throughout this paper as FIH ) , followed by polyubiquitination by pVHL ( von Hippel-Lindau ubiquitin E3 ligase ) complex that marks HIF-1α for proteasomal degradation [1 , 5] . In hypoxic conditions , HIF-1α protein is stabilized and it translocates from the cytoplasm into the nucleus , where it binds to HIF-1β to form the heterodimer HIF-1 complex . The dimer complex associates with HREs ( hypoxia-responsive elements ) located in the promoters of target genes and tethers transcriptional coactivators , such as CBP ( CREB-binding protein ) and p300 , to activate gene expression . Many of the genes targeted by HIF-1 encode proangiogenic factors including VEGF-A ( vascular endothelial growth factor A , abbreviated throughout this paper as VEGF ) and EPO ( erythropoietin ) [1 , 6 , 7] . MicroRNAs ( miRs ) are endogenous , small , non-coding RNA molecules ( ~22 nt ) that mediate gene expression at the post-transcriptional level . RNA polymerases II and III participate in the transcription of microRNA genes to produce miR primary transcripts ( pri-miRs ) that are usually several hundred nucleotides in length and contain conserved stem loops [8–10] . These pri-miRs are processed by the RNase III enzyme Drosha into stem-loop intermediates ( ~60–70 nt ) which are termed precursor miRs ( pre-miRs ) , and pre-miRs are actively transported out of the nucleus via a nucleocytoplasmic shuttler Exportin-5 ( XPO-5 ) assisted by the GTP binding nuclear protein Ran [11 , 12] . The pre-miRs in the cytoplasm are cleaved by another RNase III enzyme Dicer to become miR duplexes which are then incorporated into the miR-induced silencing complex ( miRISC ) [11] . Within the miRISC , proteins of the argonaute family ( AGO ) are essential for miR function in human as they facilitate the activation of miRISC by catalyzing the dissociation of miR guide strand ( mature miR ) from the passenger strand ( cleaved later ) ; only AGO1 and AGO2 , among the eight AGO proteins in human , can mediate such strand dissociation during miR maturation [13 , 14] . AGO1 is identified to be associated with miR-mediated translational repression; however , only AGO2-containing RISC is capable of catalyzing the cleavage of target mRNAs [15 , 16] . Although previous research has extensively investigated the role of AGO in coordinating miRISC activities , very limited knowledge exists about the expression of AGO in response to cellular stress and its physiological importance in the remodeling of vasculature . Many miRs have been confirmed to have links with the pathophysiology of various cardiovascular diseases . Since endothelial cells ( ECs ) control the formation of new blood vessels ( angiogenesis ) which is critical for vascular homeostasis , dysfunction of ECs in response to adverse hemodynamic alterations and pathological stimuli , such as inflammation or chronic hypoxia , would lead to inadequate or anomalous angiogenesis that predisposes to the development of many vascular diseases including PAD ( peripheral arterial disease ) and CAD ( coronary artery disease ) [17] . MicroRNA let-7 ( lethal-7 ) family is among the most promising miR candidates as novel regulators of angiogenesis considering its high expression in ECs and it directly targets several angiogenesis-related factors such as TSP-1 ( thrombospondin 1 ) , TIMP-1 ( tissue inhibitor of metalloproteinases 1 ) and TGFBR1 ( transforming growth factor beta receptor 1 ) [18–20] . A recent work by Chen et al . revealed that the HIF-1-let-7-AGO1-VEGF signaling pathway is essential in the control of EC angiogenesis in hypoxia [21] . Members of the let-7 miR family are identified as HRMs ( hypoxia-responsive microRNAs ) whose levels are robustly upregulated by HIF-1 transcription factor in hypoxia . Mature let-7 targets the mRNA of AGO1 and reduces the level of miRISC formed by AGO1 and other miRs that target VEGF , therefore freeing VEGF from translational repression to promote angiogenesis ( Fig 1 ) . Validated by in vitro and in vivo experiments , these findings supported the argument of an important angiogenic axis connecting HIF , miRs and AGO1 in ECs that may potentially serve as a valuable target for pro- and anti-angiogenic therapies [21] . Though many of the molecular components that are involved in the miR control of the HIF-VEGF pathway in ECs have been characterized , the detailed dynamics of how they mechanistically interact with each other within the signaling network are barely understood . In this sense , a computational model constructed from the perspective of systems biology would provide dynamic understanding and mechanistic insights of the complex cellular response to hypoxia , as the model relies on basic biophysical principles and biochemical reactions to describe relevant molecular interactions within a cell [22] . However , mathematical models of miRs are very limited in literature; the available models , such as the model of miR-193a in ovarian cancer and the model of miR control circuits in epithelial-mesenchymal transition , focused on predicting connections between certain expression patterns of miR-related molecules and disease-related physiological phenotypes [23 , 24] . On the other hand , Kim et al . integrated the miR-451-mTOR signaling pathway into a multiscale hybrid model that described the complex processes of glioma cell proliferation and migration in great detail [25] . Most of these recent models have not considered time-course experimental data available from related studies in their validations and predictions , which may undermine the predictive power of computational models since any important details hidden in the dynamical responses would be easily overlooked . In this study , we have developed a mechanistic model describing the miR regulation of the HIF-VEGF signaling pathway that , for the first time at the molecular level , unveils the critical role of miR in the complex process of hypoxia-driven angiogenesis . The model incorporates biophysical details of miR biogenesis and considers cellular compartmentalization that were absent in previous miR pathway models . We have employed the model to study how different gene overexpression/silencing strategies in ECs would affect the overall cellular adaptation to hypoxia , quantitatively and qualitatively , by analyzing the dynamics of several signature proteins . Assisted by the model , we have identified various characteristics in cellular oxygen sensing mechanisms and in miR regulatory network that controls the canonical HIF-VEGF pathway . The model predicts that HIF-1α stabilization obeys a hypothetical switch-like mode and it is negatively regulated by an mRNA destabilizer in hypoxia [26] . To address a major focus of the study , we show that let-7 and AGO1 are the initiators and coordinators of VEGF release , whereas they negatively exert feedback control on each other and are capable of minimizing the impact of possible outside perturbations; we also illustrate the role of miR-15a as a final effector molecule which is under control of AGO1 , since abundance of miR-15a directly determines how much VEGF mRNA is available for translation . From these observations , we propose a potential mechanism that may contribute to impaired angiogenesis during recovery in patients with peripheral arterial disease . Another key focus of the study is the extension of our model analysis into potential clinical settings , where we evaluate different pathway-based therapeutic strategies designed to differentially regulate angiogenesis in highly hypoxic conditions , which are commonly observed in tumor and ischemic tissues . Together , these findings reveal an integrated image of multiple miRs , each with different targets , that work cohesively with miR-processing proteins ( e . g . Dicer , AGO1 ) to counteract adverse physiological stresses by promoting VEGF synthesis and angiogenesis . This study should stimulate future research to investigate , both experimentally and computationally , the mechanistic signaling networks that contribute to the dysregulated miR expressions in cancer and in ischemic vascular disease .
The model we constructed , as shown in Fig 2 , describes the regulation and coordination by miRs in in vitro hypoxia-induced HIF stabilization and VEGF synthesis in ECs . The model consists of a cytoplasmic and a nuclear compartment , and it is functionally divided into four modules: oxygen sensing/HIF stabilization , HIF dependent gene transcription , miR-15a targeting of VEGF , and let-7 biogenesis/targeting . The oxygen sensing module is a selected integration of two established HIF models: Qutub and Popel’s work which considers the participation of iron and 2-oxoglutarate ( 2-OG ) during HIF stabilization , and the model by Nguyen et al . that includes FIH-mediated events in HIF hydroxylation [27 , 28] . Although the ascorbate binding suggested by Qutub and Popel as well as the process of nuclear HIF stabilization are not included in order to maintain a moderate complexity , our carefully integrated model is able to capture the essential oxygen-sensing behaviors of ECs that are needed to address our research focus . For the same reason we do not include the effects of reactive oxygen species and succinate on HIF-1α considered in subsequent papers of Qutub and Popel [29 , 30] . The influence of HIF/PHD feedback is not considered since the model assumes that PHD2 concentration is in excess [31] . Although this feedback mechanism is absent , the model with the current parameter set and reactions is able to capture the core dynamics of the distinct HIF-1α behaviors in normoxia and in hypoxia; according to model simulations presented in this work , the high initial concentrations of PHD2 contributes to the early suppression of HIF-1α after its rapid induction in hypoxia , which agrees with an assumption made by Bruning et al . about the temporal role of HIF/PHD feedback loop [32] . To describe the negative feedback control of HIF in hypoxia , however , we included the mechanism of HIF-1α mRNA destabilization by TTP ( tristetraprolin ) identified by Chamboredon et al . and assumed a HIF-dependent TTP production , since hypoxic exposure is experimentally shown to induce TTP [33 , 34] . Similarly , VEGF protein synthesis is also downregulated by TTP accumulated in hypoxia [35] . The module describing HIF activation of its targets , including the genes of VEGF and let-7 , details the process of stabilized HIF-1α being transported into the nucleus , dimerizing with HIF-1β subunit and promoting the transcription of these genes containing HREs [21 , 36] . We assumed that the step of HIF-1 complex binding with the coactivators CBP/p300 was included in the process of HIF-1α/HIF-1β dimerization . Interestingly , HIF-2α ( hypoxia-inducible factor 2 alpha ) is also shown by Chen et al . to transcriptionally induce the same group of HRMs upon its induction in hypoxia , and it is known that HIF-1α and HIF-2α share not only highly similar protein structures but also various common target genes including VEGF [21] . In skeletal muscle and especially ECs , HIF-2α signaling seems to be rather ancillary to the predominant regulatory potential of HIF-1α that primarily modulates the cellular angiogenic and migratory activities [37 , 38] . Therefore , for the scope of this study we decided not to distinguish between these two molecules in the model but to represent them both in terms of HIF-1α , which is more prevalently expressed across different cell types than HIF-2α [39] . The model currently considers let-7 and miR-15a as key miR regulators of the hypoxia-driven VEGF desuppression process . The production of miRs in the model followed a well-established miR biogenesis pathway that undergoes transcription , nuclear-cytoplasmic transport , endonucleolytic processing and miRISC loading [11] . The model combines Drosha processing and XPO-5 transport into a one-step reaction , and miRISC formation along with miR duplex dissociation is simplified as one reversible association process between AGO1 protein and miR . The complex formed by AGO1 protein and let-7 can travel back into the nucleus and promote the processing of pri-let-7 which constitutes a positive auto-regulatory loop [40] . Let-7 represses the translation of two confirmed targets , AGO1 and Dicer , and this silencing negatively feeds back to the maturation and stabilization of let-7 [21 , 41] . The mRNAs of AGO1 and Dicer are processed by the let-7 miRISC and directed to cytoplasmic domains called p-bodies [42 , 43] . Since p-bodies are found to be involved in general mRNA turnover , we assumed that once mRNAs entered the p-bodies , they would be stored , inaccessible to translation with a significantly slower degradation rate compared to that of cytoplasmic mRNAs , while a very small fraction of them could still exit p-bodies and re-enter the translational machinery [44] . Since the let-7/AGO1 axis would influence the expression of a group of miRs leading to altered dynamics of many target genes , we selected VEGF , due to its crucial importance in angiogenesis and extensive literature data support , as an epitome gene to demonstrate mechanistically the details of how this cascade controls specific gene expression during a pro-angiogenic response . For the current purposes of the model , miR-15a is selected to represent a group of VEGF-targeting-miRs as miR-15a has been experimentally validated to directly repress VEGF synthesis and markedly affect angiogenesis in ECs [45] . In addition , hypoxia is shown to weaken the association of AGO1 with many VEGF-targeting miRs including miR-15a and cause significant downregulation of these miRs; these evidence further links the dynamics of miR-15a to the coordination by the let-7/AGO1 axis [21 , 46] . VEGF mRNAs targeted by miR-15a also undergo a series of steps including p-body storage similar to the mechanism of let-7-mediated mRNA silencing . Details including the mathematical formulation of the biochemical reactions in the model and parameter optimization are discussed in the Methods section . The simulations of the model were compared with the data from independent experiments performed by different research groups . The first form of validation focused on the oxygen sensing module and compared the model’s prediction of HIF-1α accumulation in hypoxia with the quantified Western blot data in ECs ( Fig 3A ) [21] . Experimental data suggest that HIF-1α accumulation in vitro is most significant at O2 levels between 0 . 5–6% , and this is reflected in the model by setting the initial O2 concentration to 19 . 9 μM which corresponds to 2% ambient oxygen [27 , 47] . The level of HIF-1α predicted by the model was the sum of both the free form and bounded proteins . HIF-1α concentration at time zero was taken as a reference measure which represents the normoxic ( 21% O2 ) steady state level . In agreement with the experimental result , the simulation showed a quick induction of HIF-1α during the first few hours followed by a gradual decrease to the steady state level . Similarly , Western blot data of HIF-1α in different cell types from other research groups also indicate that , in hypoxia , HIF-1α protein is induced rapidly while its expression peaks and gradually decreases after a few hours , suggesting that the cascade of TTP-mediated HIF-1α mRNA destabilization , as one of the major mechanisms which downregulate HIF-1α via negative feedback , should be incorporated as a fundamental part into the model [33 , 48 , 49] . In addition , the predicted time course abundance of AGO1 protein was compared with the experimental quantification using Western blot in ECs in hypoxia ( Fig 3B ) [21] . The predicted AGO1 level was also a summation computed in a similar way of how HIF-1α was defined above . To obtain the relative expression over time , the absolute level of AGO1 was normalized with respect to its initial concentration ( steady state level in normoxia ) , which was obtained by simulating the model at an O2 concentration 209 μM ( 21% O2 ) for a long enough time span [27] . After an initial delay during which let-7 was accumulated in hypoxia and let-7 miRISC were formed , AGO1 level started to decrease because of a rapid decline in the amount of AGO1 mRNAs that were available for translation ( S1 Fig ) . Chen et al . also demonstrated that this response was not specific to ECs: cells from different organs/tissues including liver , kidney and muscle all displayed significant AGO1 downregulation in response to hypoxia [21] . The VEGF protein production curve predicted by the model was compared with experimental data from two different groups . Although endothelial cells , compared to other cell types , may not be the biggest contributor of VEGF secretion in response to low oxygen tension , adequate autocrine VEGF signaling was proven to be critical in the maintenance of vascular homeostasis [52] . Zhou et al . measured the VEGF protein expression at different time points in whole-cell extract of SHEP cells ( a human neuroblastoma cell line ) that were cultured in hypoxic conditions ( 1% O2 ) [50] . A two-fold increase in VEGF level was predicted after a simulated 8-hour exposure to 1% O2 tension ( Fig 3C ) . Liu et al . analyzed the impact of CoCl2-induced hypoxia on the expression of VEGF proteins using Western blot in HepG2 cells ( a human hepatocellular carcinoma cell line ) [51] . CoCl2 ( cobalt chloride ) is one of the hypoxia-mimetic agents; it stabilizes HIF-1α in normoxia by directly inhibiting the process of PHD/FIH-mediated HIF-1α hydroxylation ( see reactions in Fig 2 ) [53 , 54] . The model assumes an initial CoCl2 concentration of 200 μM in the simulation and the relative expression was normalized to VEGF level at time zero ( Fig 3D ) . Since the current model reactions and parameters are established to describe signaling events specifically in ECs , the fit between model predictions and experimental results in Fig 3C and 3D does not imply that ECs , HepG2 and SHEP cells have the same dynamics of intracellular signaling and VEGF production . Likely , different parameter sets would be needed in order for the model to make more accurate predictions of VEGF synthesis in other cell types ( e . g . tumor , muscle , stromal cells ) as they are the more significant sources of VEGF secretion compared to ECs [55] . Previous studies have quantified the induction of VEGF in stromal and tumor cells and found a 2 to 6 fold increase of VEGF in stromal cells and a 3 fold increase in a breast cancer cell line after 24 hours of hypoxia treatment; our EC-based model predicts a 3 . 5 fold increase in the intracellular VEGF level after 24 hours of simulation at 2% O2 ( S2C Fig ) [56 , 57] . In this sense , the model described in this work is able to predict VEGF dynamics that are comparable , both quantitatively and qualitatively , to biological VEGF data in response to hypoxia in different types of cells , which allows for further extension of the HIF-let-7-AGO1-VEGF framework in other cell models . The step of oxygen sensing determines how much HIF-1α will be stabilized and then dimerize with HIF-1β to form active transcription factors at different O2 levels . Initially , HIF-1α expression is low in normoxia and transcriptional activities of let-7 , VEGF and TTP are insignificant . As oxygen availability decreases , hydroxylation of HIF-1α by PHDs and FIH is also reduced , allowing more HIF-1α to escape from VHL-mediated degradation and enter the nucleus [1] . Fig 4A shows the overall oxygen dependent response of HIF-1α . For small enough oxygen levels 2% and 1% in Fig 4A , accumulated HIF-1 reaches a maximum ( overshoot ) at around 10 hours and then slowly declines to a steady level . Since the model assumes that initial PHD2 concentration is in excess in order to capture switch-like responses in HIF-1α hydroxylation ( Fig 4B ) , changes in PHD2 dynamics when hypoxia is sensed should take place very quickly at a time point much earlier than the overshoot [26 , 27] . The same reason justifies that FIH does not cause the overshoot , so we hypothesize that it is TTP which creates the initial overshoot , since hypoxia promotes TTP synthesis which destabilizes the mRNAs of HIF-1α and downregulates its translation [34] . Results in Fig 4C show that in silico knockdown of TTP mRNA effectively prolongs the initial overshoot in HIF-1α stabilization curves . The time course profiles of HIF-1α also strongly depend on the rate of cytoplasmic-nuclear trafficking ( Fig 4D ) . As the forward rate of HIF-1α shuttling from cytoplasm into nucleus increases , more HIF-1 transcription factor complex is formed , which promotes the synthesis of various molecules including VEGF that facilitates cellular adaptation to hypoxia by improving angiogenesis and TTP that feeds back to inhibit HIF-1α production ( Fig 4E ) . Consistent with our prediction , Ahluwalia et al . overexpressed importin-α , a nuclear importer of HIF-1α , in GMECs ( gastric mucosal endothelial cells ) of aging rats and observed a significant increase in the binding of HIF to the VEGF gene promoter region [58] . Smaller forward rate of HIF-1α nuclear import leads to a lower HIF-1α baseline in normoxia and reduces the overall HIF-1α level in hypoxia , since the majority of the stabilized HIF-1α accumulates within the cytoplasm ( Fig 4F ) , where HIF-1α is unable to dimerize with HIF-1β and is more susceptible to degradation . This suggests that cells with impaired HIF-1α nuclear import are correlated with reduced HIF-1 transcription activity and poorer pro-angiogenic adaptation in hypoxia , which is consistent with the finding that reduced importin-α level in senescent GMECs hindered the induction of VEGF expression and angiogenesis in response to hypoxia [58] . We are interested in the mechanistic interactions between let-7 , AGO1 and miR-15a and their roles in the control of subsequent VEGF mRNA release , since AGO1-associated miRs ( e . g . miR-15a ) that are capable of silencing VEGF were shown to be less abundant following hypoxia treatment [21] . To better understand this key feature in our model , we approach the analysis in two steps by looking at the direct interactions between let-7 and AGO1 that influence free form VEGF mRNA level at different O2 tensions , as well as the downregulation of miR-15a availability as a result of upstream let-7/AGO1 control . A number of in vitro and in vivo studies have been performed to characterize the therapeutic values of different miRs in treating cancer and cardiovascular disease; in some of these the goal was to uncover the entire regulatory events that give rise to the miR dysregulation [20 , 65 , 66] . Abnormal profiles of AGOs that disrupt the balance between pro- and anti-angiogenic miR expression are among the essential reasons behind the aberrant disease-related angiogenic activities of ECs [67 , 68] . Deriving and testing potential miR-based therapeutics in different diseases in silico , given the substantial analysis performed to understand our proposed pathway , are of crucial significance to future research in the field . We performed modular sensitivity analysis on four key species in the pathway: cytoplasmic HIF-1α , free form AGO1 , free form let-7 and VEGF . The analysis was accomplished in MATLAB SimBiology toolbox ( see Methods ) , assuming 2% hypoxia as O2 initial condition . Fig 8A–8D display the local sensitivity of the four species with respect to different set of selected kinetic parameters , in which direct production and degradation rates of each species were excluded since their contributions are too evident to produce valuable insights . Not surprisingly , HIF-1α is very sensitive to its affinity with PHD2-O2-Fe-DG or FIH-O2-Fe-DG complexes which will subsequently mark it for hydroxylation ( Fig 8A ) . Also , the idea we proposed in previous results that the strength of association between let-7 and AGO1 controls the expression of both molecules , was again corroborated by sensitivity analysis ( Fig 8B and 8C ) . It is noteworthy that in the sensitivity analysis for VEGF , varying TTP synthesis was as influential as varying the VEGF silencing efficiency of miR-15a RISC directly ( Fig 8D ) . Since the major conclusions of this work relate closely to the qualitative dynamics of AGO1 and VEGF , the more influential parameters identified from modular sensitivity analysis were subjected to additional evaluations ( S7 Fig ) . Motivated by the sensitivity analysis , we suggested that overexpressing TTP could be a potential anti-angiogenic therapy since it inhibits VEGF synthesis by decreasing HIF-1 signal and directly destabilizing VEGF mRNA . Since TTP suppression in cells has been associated with pro-tumorigenic phenotypes , stimulation of its expression might be an effective way to shrink the synthesis of pro-angiogenic ( e . g VEGF ) and pro-inflammatory cytokines in tumor ( Fig 8E ) [75 , 76] . In addition to the sensitivity analysis , we tried to look for core reactions and parameters that are responsible for the radical differences between system behaviors in hypoxia and in normoxia . HIF-1α and AGO1 are both master coordinators for a series of signaling events in the model , and their time course total expressions have been experimentally measured [21] . Increasing the affinity between oxygen and PHD2-Fe-DG or FIH-Fe-DG complex significantly reduces the relative HIF induction in hypoxia ( Fig 8F ) . A reason for that is the extra amount of PHD2/FIH-Fe-DG-O2 that has been stabilized in normoxia due to enhanced O2 binding , which speeds up HIF hydroxylation in hypoxia ( S8 Fig ) . A higher rate of HIF-1α translocation also strongly influences HIF induction because it temporarily pushes more HIF-1α into the nucleus and protects it from degradation , but this would later lead to a big drop in total HIF-1α because of HIF-induced TTP production ( Fig 8F ) . For AGO1 , its downregulation is necessary to induce sufficient VEGF desuppression . Like let-7 , in the model AGO1 is prevented from degradation when associated with miRs . Since more let-7-AGO1 is formed because of either increased let-7 or stronger binding , the extra amount of AGO1 that is stabilized elevates its baseline level in hypoxia; furthermore , higher level of let-7-AGO1 reduces AGO1 level in normoxia due to translational repression ( Fig 8G ) .
In this study , we presented the first mass-action based computational model of miRs in a comprehensive , whole-cell signaling network that is closely related to angiogenesis . Our model is also the first one to have described in details the regulatory process that controls miR biogenesis , such as Dicer cleavage , AGO1 loading and p-body localization , and considered these reactions as an essential module of the model framework [11 , 42] . The reason why these elements are important in the formulation of our model is more than the fact that they represent the real biological process; for example , AGO1 is commonly considered as part of miRISC for universal miR-mediated silencing activities , but in our pathway of interest it also happens to be a key factor . Challenges follow as always when researchers map signal transduction pathways , and we aim to make a model that contains enough biochemical and biophysical details to address different experimental findings while its complexity remains manageable . With a few Michaelis-Menten or Hill kinetics and more than 80% of the reactions based on simple first and second order kinetics , the model is equipped with both modular flexibility and solid biochemical background . Since most miR research only focuses on the influence of one specific miR , we decided to take a novel approach and investigate the miR-dependent regulation of miRs in hypoxia coordinated by miR-processing molecules AGO1 and Dicer , as the core theory of our model has been experimentally validated by the work by Chen et al [21] . However , current experimental data on miR signaling in the control of angiogenesis is very thin , which leaves a large space for our model to be further validated and refined . A fundamental goal of this work is to guide and stimulate more miR research that would investigate not only the function of an individual miR but also the interconnections between abnormal profiles associated with a group of miRs in vascular diseases . By the same reasoning , topics such as the time course dynamics of molecules participating in p-body configuration during miR-induced mRNA silencing or the characterization of individual miR activities via AGO1- versus AGO2-mediated regulation may also bear research values in the miR field , and these data , if available from future studies , would then significantly complement the accuracy and reliability of our model . Overall , the extensive simulations performed in this study identified and reasoned that overexpressing AGO1 , TTP or antagonizing let-7 are effective strategies to suppress VEGF production in tumor , and that miR-15a antagonists alone , compared to other proposed strategies , could most potently enhance VEGF synthesis in simulated PAD conditions . Our model will advance the current understanding of how different miRs are regulated to affect angiogenesis when cells are actively adapting to hypoxia; it will also provide valuable insights into the future research in the pathophysiology of cancer and ischemic cardiovascular disease , including PAD , as well as the development of miR-based therapeutics that target other related pathways . Sensitivity analysis reveals that the degree of HIF-1α stabilization is firmly controlled by the binding between O2 and hydroxylase enzymes ( e . g . PHD2 , FIH ) . The stronger O2 binds hydroxylase enzyme , the less HIF-1α is sustained both in normoxia and hypoxia . For tumor this change could be disadvantageous , since cells in hypoxia would fail to accumulate enough HIF-1α in order to trigger adequate pro-angiogenic adaptations . On the other side , inhibiting the interaction between O2 and hydroxylase enzyme increases HIF-1α signals and the expression of its targets , which potentially benefits patients with ischemic arterial disease [77] . We also pinpointed the molecule TTP , which destabilizes mRNA of multiple signaling molecules including its own , and we demonstrated in silico that it holds therapeutic value in tumor for its anti-angiogenic property [75 , 78] . Although TTP overexpression seems more direct than miR-based therapies in terms of its mechanism to inhibit VEGF production , its mRNA-destabilizing activity is repressed upon phosphorylation by the mitogen-activated protein kinase [79 , 80] . TTP is linked to miR as a direct target of miR-29a validated in cancer epithelial cells , which adds another layer of complexity to the delicate regulation of VEGF expression in hypoxic environments [81] . Speaking of potential model applications in the context of disease pathology , since the prominence of AGO1 and let-7 profiles in cancer has already been validated , the next step is to connect our model with miR features in vascular disease [21] . In the limited literature that studies miR dysregulation in PAD , circulating let-7 and miR-15a are shown to be downregulated in patients with PAD compared with healthy controls [82] . Interestingly , these miRs happened to be included in our model for their importance in the modulation of VEGF synthesis in hypoxia , and this motivates us to hypothesize the following mechanisms that could bridge the gap between miR data and PAD clinical symptoms , given the evidence that baseline VEGF levels in skeletal muscle of PAD patients were similar compared to healthy subjects [83] . In PAD patients , prolonged exposure to ischemia might turn on some epigenetic switch such as the LIN28/let-7 feedback loop that potently silences let-7 expression , so let-7 abundance is markedly reduced even in cell that has high HIF-1 level in response to hypoxic stimulus; as a result , AGO1 expression is less inhibited in hypoxia which leads to impaired VEGF desuppression and synthesis [84] . Therefore , the increase in tissue VEGF level might be insufficient to initiate an ischemic response to upregulate angiogenesis and re-establish perfusion in PAD patients . Since PAD prevalence increases significantly with age , another possible explanation about PAD-related miR feature looks at the loss-of-function of HIF-1 transcription factor in aging endothelial cells; nuclear import of HIF-1α is decreased in aging GMECs compared to young GMECs , and this severely blunts the induction of HIF-1 target genes including VEGF and let-7 in hypoxia [58 , 85] . However , the mechanisms discussed above , including the model simulation of miR-based treatment for PAD , focused only on the regulation of pro-angiogenic cytokines , while pathophysiology of PAD is in fact affected by dysregulation of both pro- ( e . g . VEGF ) and anti-angiogenic factors ( e . g . TSP-1 , angiostatin ) [86–88] . So far , efficacy of VEGF delivery strategies in PAD and CAD clinical trials is still below expectation , and systems biology studies that can quantitatively model multiscale cytokine interaction and angiogenesis will likely advance future drug designs for ischemic vascular disease [89] . Another limitation is that the model assumes a constantly low oxygen availability in ischemic tissues , which is more relevant to the clinical manifestation of patients with advanced-stage PAD since they tend to develop critical limb ischemia ( CLI ) and experience continuous leg pain , whereas early-stage PAD is often associated with intermittent claudication , suggesting that hypoxic exposure may be less prominent in this case [90–92] . Currently , the model describes only the dynamics of let-7 and miR-15a , but its setup allows future extensions to include more miRs and their target molecules . Ghosh et al . identified miR-424 as another HRM which targets CUL2 ( cullin 2 ) in ECs; since CUL2 is essential in the assembly of the E3 ubiquitin ligase complex , miR-424 helps to stabilize HIF-1α and consequently increase the transcription of HIF-1 targets in hypoxia [93] . A few other hypoxia-responsive miRs , such as miR-155 and miR-210 , have also been shown to negatively control HIF-1α stabilization in hypoxia by directly targeting its mRNA , and a basic but well-supported model describing the interactions between HIF-1α and miR-155 has been developed [32 , 94] . In addition , members of miR-17/92 cluster are shown to target HIF-1α and TSP-1 , and their biogenesis is repressed in hypoxia with a strong dependency on Dicer expression [95–97] . There is also evidence suggesting that AGO2 , in response to hypoxia , is differentially regulated in tumor and smooth muscle cells , which in turn modulates the maturation of various miRs and influences downstream gene expressions [98 , 99] . With all the experimental evidence , the model can be further enriched to depict a more comprehensive signal transduction network that starts with oxygen sensing , consists of diverse miR regulations , and ends at the secretion of pro- ( e . g . , VEGF ) and anti-angiogenic ( e . g . , TSP-1 ) factors , with a better predicting power in terms of capturing both the general dynamics as a result of coordinated miR regulation and the specific profile changes of key intermediate species . So far , the model is formulated mostly based on validated knowledge in ECs , however , experimental evidence suggests that HIF-let-7-AGO-VEGF pathway is present in other cell/tissue types , including muscle and tumor , and it plays a fundamental role in the angiogenic adaptations of these cells in response to hypoxia [21] . This implies a possibility of making a model that predicts the proliferative and migratory behavior of individual cells based on cytokine signals ( VEGF , TSP-1 ) released from all different cell types within a small population . In summary , our model is the first computational study that investigates miR control in hypoxia-induced angiogenesis and a pioneer in the field of miR mechanistic pathway models . To extend experiment-based mechanistic miR modeling to the next level , future computational studies could combine our work with state-of-the-art VEGF models and incorporate agent-based modeling techniques to simulate tissue-level proliferation and angiogenesis , within a bulk tissue from tumor or skeletal muscle , under different physiological conditions .
We constructed the model which produces simulations for this study based on ordinary differential equations ( ODE ) with a total of 47 species , 91 kinetic parameters and 57 reactions ( Fig 2 ) . All reactions , with description and kinetic parameter values ( S1 Table ) , and initial condition for each species ( S2 Table ) are available in the supporting information . Although the model does not focus on intracellular trafficking , for several species ( e . g . HIF-1α ) the model allows transportation and distinguishes them by cellular locations—in cytoplasm or in nucleus , since their core properties change along with the physical locations . Transcription activation and inhibition , along with Dicer processing are modeled as Hill-type or Michaelis-Menten kinetics , and we included turnover mechanisms for major species in order to capture more accurate long-term responses . 47 out of 57 reactions are modeled as first or second order biochemical reactions , and most interactions are based on experimental data from literature ( S1 Table ) . All the data are put together into a computational model using MATLAB SimBiology toolbox ( MathWorks , Natick , MA ) . Simulations are performed using the ode15s method , which is a stiff ODE solver provided in MATLAB . Since we are interested in the hypoxic response , the initial conditions of all species are their normoxic steady state levels which are obtained by a long enough simulation at 21% O2 . For the simulation of different mRNA treatments , overexpression is equivalent to increasing its initial condition; silencing is modeled as the binding of siRNA with mRNA to form a compound in which mRNA is unusable . For miR treatments , overexpression by miR mimics increases the initial condition of the corresponding precursor miR , while silencing is reflected as the association of miR antagonist with miRISC to form a compound that cannot function . In the model validation section , to quantify the Western blot data from literature , we used ImageJ software ( NIH ) to perform densitometry analysis according to the blot analysis protocol . In the simulations that evaluate different treatments against PAD , the pathological setting of PAD is simplified and modeled as an impaired let-7 induction by hypoxia ( value of kp21 in S1 Table is adjusted to 0 . 464 μM ) according to the findings by Stather et al [74 , 82] . Given the fact that time delays in processes such as transcription and translation are sometimes critical in determining the behaviors of certain biological systems , we showed in S9 Fig that a system built strictly with ODEs could still exhibit these necessary delays as a result of sufficient parameter tuning [100–102] . Although there has been little data that quantifies real-time transcription delays , our model predicts a delay in the range of 10 to 20 minutes for the transcription of VEGF and TTP ( S9A and S9B Fig ) which is very close to the experimental measurements by Ota et al [103] . S9D Fig compares the behavior of TTP in the original ODE model with the new behavior if TTP transcription/translation are modeled using delayed differential equations ( DDEs ) in MATLAB Simulink . Despite the large increase in its translation rate and a switch of modeling approach , the general time-course of TTP protein remains similar ( S9D Fig ) . Therefore , we suggest that the important time delays in biological events are not overlooked by our ODE-based model given that we have done extensive parameter tuning and optimization , and the system’s core dynamics would be largely unaffected if the more explicit , DDE approach is taken instead of ODE . Due to the novelty of our model and the limited literature in miR modeling , we only take a few reaction parameters and initial conditions from published models on oxygen sensing [27] . To estimate the initial conditions for miRs , we compare published experimental data and assume that miRs are present in the order of 103 to 104 copies per cell in normoxia; the molar concentration is then computed using a 1 pL cell volume [104 , 105] . Given the relatively low levels of miRs within the cell , some previous computational studies had employed the stochastic approach to model miR/protein regulatory networks [106 , 107] . However , given the scope and complexity of our model , we choose to take the deterministic ODE approach after careful consideration . Mac Gabhann et al . observed that stochastic and deterministic simulations , respectively , of VEGF binding to its receptors would generate consistent results , although typical VEGF concentrations in vivo are in the picomolar range [108] . Similarly , Giamperi et al . found that the outcomes from stochastic and deterministic approaches in modeling a miR/protein toggle switch achieve a high degree of agreement [107] . The results of both studies corroborated our assumption that the deterministic approach we used captures the important average behavior of the system which would be of equal essence in the stochastic case . The decay rates of mRNAs ( 1 . 2e-3 min-1 ) , miRNAs ( 1e-4 min-1 ) , proteins ( 2 . 5e-4 min-1 ) , the rate of translation per mRNA ( 3 min-1 ) , the normoxic level of various mRNAs ( 2 . 8e-5 μM ) and proteins ( 0 . 08 μM ) are estimated in a way that the values are within ±2 orders of magnitude compared to the global median value ( normalized by 1 pL cell volume and shown in the brackets ) found by large-scale quantification studies [109–111] . The rest of the reaction rates and initial conditions are either fitted or estimated based on empirical conjectures concluded by previous research ( all the sources are summarized in S1 and S2 Tables ) . The optimization routine is carried out using the Levenberg-Marquardt algorithm within the lsqnonlin function available in MATLAB . Since EC time course data of miR signaling in hypoxia are very limited , we optimize the parameters by minimizing the sum of squared errors between normalized simulation profiles and two experimental datasets [21] . The optimized parameter set produces biologically justifiable simulations regarding the absolute concentrations of each species , and it also predicts time course VEGF expressions that fit well to other Western blot data performed in different cell types ( see Results ) . Local sensitivity analysis was performed using the methods supplied in MATLAB SimBiology toolbox , which employs the “complex-step approximation” to compute time-dependent sensitivity of a species A with respect to a parameter X . For A , each sensitivity computed for a parameter was subjected to non-dimensionalization and integrated over the simulation time; then , the value was normalized to the sum of all integrals ( each integral was calculated for a certain parameter ) , and the local sensitivity of A in hypoxia was visualized in a pie chart . To evaluate the impact of adjusting local parameters on the relative difference between species expressions in normoxia and in hypoxia , a script was written in MATLAB to repeatedly run the following events in order: vary kinetic parameters , simulate the model to obtain new steady states in normoxia , set the new steady states as model initial conditions for hypoxia and simulate the model in hypoxia to calculate relative expressions . | Cells living in a hypoxic environment secrete signals to stimulate new blood vessel growth , a process termed angiogenesis , to acquire more oxygen and nutrients . Hypoxia-inducible factor 1 ( HIF-1 ) accumulates in hypoxia and expedites the release of pro-angiogenic cytokines such as vascular endothelial growth factor ( VEGF ) , a prime inducer of angiogenesis . The intermediate signaling events connecting HIF-1 and VEGF are tightly controlled by microRNAs ( miRs ) , which are endogenous , non-coding RNA molecules and powerful regulators in cancer and cardiovascular disease . Given the importance of angiogenesis in tumor development and post-ischemia reperfusion , it holds great basic research and therapeutic value to investigate how miRs modulate intracellular VEGF synthesis to control angiogenesis in hypoxia . We present a computational model that details the interactions between miRs and other key molecules which make up different hierarchies in HIF-miR-VEGF pathway . Based on simulation analysis , new potential therapies are introduced and tested in silico , from which the strategies that most effectively reduce VEGF synthesis in cancer , or enhance VEGF release in ischemic vascular disease are identified . We conclude that in hypoxia different miRs work consonantly to fine-tune the cellular adaptations; when a master miR alters its expression , dynamics of other miRs vary accordingly which together contribute to aberrant RNA/protein profiles observed in the pathophysiology of multiple diseases . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2015 | Computational Model of MicroRNA Control of HIF-VEGF Pathway: Insights into the Pathophysiology of Ischemic Vascular Disease and Cancer |
Clinical diagnosis and follow up of cystic echinococcosis ( CE ) are based on imaging complemented by serology . Several immunodiagnostic tests are commercially available , but the development of new tools is still needed to overcome the lack of standardization of the target antigen , generally consisting of a crude extract of Echinococcus granulosus hydatid cyst fluid . In a previous work , we described a chromatographic method for the preparation of a highly enriched Antigen 5 fraction from hydatid cyst fluid . The high reactivity of patient sera against this preparation prompted us to evaluate further this antigen for the serodiagnosis of CE on a larger cohort of samples . A total of 327 sera from CE patients with heterogeneous conditions for cyst stage , cyst number , organ localization , drug therapy , and surgical intervention , together with 253 sera from healthy controls , were first analyzed by an ELISA based on the Ag5 preparation in two different experimental setups and , in parallel , by a commercial ELISA routinely used in clinical laboratories for CE serodiagnosis . The Ag5 ELISAs revealed different sensitivity ( 88 . 3% vs 95 . 3% ) without significant differences in specificity ( 94 . 1% vs 92 . 5% ) , for the two setups , respectively . Moreover , possible relationships between the Ag5 ELISA absorbance results and clinical variables were investigated . Chi squared test , bivariate logistic regression and multiple regression analyses highlighted differences in the serology reactivity according to pharmacological treatment , cyst activity , and cyst number . The two Ag5 ELISAs revealed different performances depending on the setup . The good diagnostic sensitivity and the high reliability of the Ag5 preparation method make this antigen a promising candidate for the serodiagnosis of CE . Further studies will be needed to evaluate the ability of our test to provide useful information on specific CE clinical traits .
Cystic echinococcosis ( CE ) is a neglected zoonotic disease caused by the larval form of the tapeworm Echinococcus granulosus complex . The definitive hosts are dogs and other canids , while sheep and other livestock are the natural intermediate hosts; humans are occasional intermediate hosts . Intermediate hosts can be infected by ingestion of food and water contaminated with the parasite eggs eliminated with the feces of infected dogs . The early phase of infection is generally asymptomatic . Small , well encapsulated , viable cysts or old cysts with pseudosolid content typically do not induce major pathology , and patients may remain asymptomatic for years or even permanently . This is likely the reason why almost 50% of CE patients recorded in the Italian Hospital Discharge Records have been diagnosed accidentally during investigations for other diseases , and 57% of cases are people over 60 years old [1] . CE has many important economic effects , the most evident and tangible of which is the cost of expensive medical treatment for human cases; moreover , there is also a strong negative impact on the economy due to the large diffusion of CE among livestock [2–3] . Currently , CE can be treated according to four different approaches: surgery , percutaneous techniques , chemotherapy with benzimidazoles , and with a “watch and wait” approach for inactive cysts . Unfortunately , 20%–40% of the patients respond only temporarily to chemotherapy , and revert to their previous stage ( mainly CE2 and CE3b ) after the end of treatment [4] . The most affected regions include the Mediterranean , Eastern Europe , parts of South America , parts of Africa , and Central Asia/Western China . In Italy , the average annual incidence rate of hospital cases ( AIh ) between 2001 and 2012 was 1 . 6/105 inhabitants [1] . At present , no marker of cyst viability and therapy efficacy exists , and serology may remain positive for years even after successful therapy . As a consequence , long-term follow-up with imaging is required for the clinical management of patients . It is therefore important to invest in innovative technologies that facilitate the monitoring and control of this infectious disease in humans and in farm animals . Currently , CE diagnosis in humans is mostly based on imaging techniques [5] , and the clinical approach is based on the WHO international classification of ultrasound images according to the stage of the cyst: CE1 , CE2 , CE3b ( active ) , CE3a ( transitional ) , and CE4 and CE5 ( inactive ) [6–8] . Ultrasonography ( US ) , due to the relatively low cost and size of the equipment , is easily transportable in remote resource-poor areas , provides a useful tool for screening , clinical diagnosis , and cyst monitoring [9–11] . However , serology has an important role in supporting the diagnosis of CE , since serological tests are generally cheap , quick , and require less trained and specialized personnel for result interpretation . This is particularly true in areas where US expertise in the diagnosis of CE is scant and/or not easily accessible , and when lesions do not show pathognomonic signs of a parasitic origin , such as young CE1 cysts or inactive CE4–CE5 cysts . Unfortunately , these stages are also those with a broader differential diagnosis ( e . g . with simple cysts , neoplastic lesions ) whose serology results are also difficult to interpret [12–13] . Commercially available immunoassays are mostly based on hydatid cyst fluid ( HCF ) , collected from infected animals . However , the heterogeneity of this preparation negatively impacts on the sensitivity and specificity of the tests . Many purified native , recombinant , or synthetic antigen preparations have been tested in the last decade , although with controversial results [14–16] . This is most likely due to the poor inter-laboratory reproducibility of antigenic preparations that often rely on outdated methodologies , improperly defined as “purifications” . This adds to the use of different panels of sera , as well as to the lack of clinical characterization and appropriate classification/grouping of sera , used for validation [17–19] . Among HCF antigens , Antigen 5 ( Ag5 ) and Antigen B ( AgB ) are the most abundant and immunogenic proteins , whose role in the life cycle of the cestode has been assessed only partially [20] . In a recent work [21] , our research group reported a straightforward , robust and reproducible chromatography method that enables the preparation of a HCF fraction highly enriched in native Ag5 . This highly enriched antigen demonstrated a strong reactivity , both in western blotting and ELISA formats , when tested on a limited panel of sera from CE patients , encouraging a more extensive examination of its diagnostic potential . Here we present a large-scale study ( 327 cases and 253 controls ) aimed to evaluate the diagnostic accuracy of two different Ag5 ELISA setups compared with that of a commercially available ELISA widely employed for routine diagnostic . We also investigated the association between readings of the Ag5-based ELISAs with selected clinical variables of the patients . When the three assays were compared for sensitivity and specificity , the Ag5 ELISA had significantly higher sensitivity . Moreover , the performances of both Ag5 ELISA setups were statistically associated with clinical variables known to influence serology results . These data support the use of this Ag5-based preparation in highly sensitive diagnostic tests and prompt further investigation on its use in the follow-up of CE patients .
A written informed consent to use of leftover serum after routine serology for research was obtained from patients at the time of sample collection . The study was approved by the ethics committees of IRCCS San Matteo Hospital Foundation , Pavia , Italy , Prot N . 20150004877 , for sera from CE patients , and of the local health authority of Sassari ( ASL N . 1 , Sassari ) , Prot N . 1123/L , for sera from healthy controls . A prospective study was performed on sera from patients with hepatic and extra-hepatic CE , collected at the Department of Infectious Diseases of the IRCCS San Matteo Hospital Foundation , Pavia , Italy , where the WHO Collaborating Centre for Clinical Management of Cystic Echinococcosis is based . Sera from healthy subjects , collected at the Sassari Hospital Blood Donor Center , Sassari , Italy , were used as a control group . At the time of serum collection , all patients with abdominal cysts were diagnosed by US and CE cysts were classified according to the World Health Organization–Informal Working Group on Echinococcosis ( WHO-IWGE ) standardized US classification , by a clinician with long standing experience in the US diagnosis of CE , as part of the routine diagnostic procedures . This classification groups cysts in six stages based on a biological/dynamic approach: active ( CE1 and CE2 ) , transitional ( CE3a and CE3b ) and inactive ( CE4 and CE5 ) cysts . However , CE3b are actually biologically active [8] . On the other hand , from a serological point of view , CE3 cysts of both stages often show comparable results , indicating that biological activity at the time of serum collection may not immediately influence serological responses in these stages [13] . Therefore , for the purpose of analysis , sera from patients with active ( CE1 , CE2 and CE3b ) and transitional ( CE3a ) cysts were grouped together . Patients having multiple cysts in different stages were assigned to the group of the most active cyst , independently from its hepatic or extra-hepatic localization . HCF crude samples were collected in two different Sardinian slaughterhouses ( CE/IT2383M , Tula , Sassari and CE/IT2078M , Lula , Nuoro ) . Fluid was aspirated from liver and lung cysts found in infected sheep . The hydatid fluid was centrifuged at 1000 g at 4°C and the supernatant stored at -80°C . Enriched Ag5 was obtained as described previously [21] . Briefly , after desalting and concentration , aliquots of sheep HCF were fractionated by Fast Protein Liquid Chromatography ( FPLC ) on a Superdex-200 column ( 10/300 GL , GE Healthcare , Uppsala , Sweden ) . The fractions of interest were pooled and their protein content was evaluated by tandem mass spectrometry on a Q-TOF hybrid mass spectrometer equipped with a nano lock Z-spray source and coupled on-line with a nanoAcquity chromatography system ( Waters , Manchester , UK ) to verify the quality of the preparation . Sera were tested in duplicate in the parasitology diagnostic laboratory of the IRCCS San Matteo Hospital Foundation , Pavia , Italy , by laboratory personnel with long standing experience in diagnostic parasitology , using a commercial ELISA test ( RIDASCREEN Echinococcus IgG , R-Biopharm , Darmstadt , Germany ) , for the detection of Echinococcus specific total IgG , according to manufacturer’s instructions . Tests were read at 450nm in a spectrophotometer , and a Sample Index ( SI ) was calculated and interpreted for each serum according to manufacturer’s instructions; ELISA was considered positive for SI >1 . 1 , negative for SI <0 . 9 , and border line for 0 . 9 ≤SI ≤1 . 1 . However , for statistical evaluations , borderline results were classified together with negatives . Readers were blind to the results of the other tests . Sera were tested with the Ag5 ELISAs based on our Ag5 enriched preparation [21] , following two alternative setups , A and B , in the laboratory of Porto Conte Ricerche , Alghero , Italy . Briefly , for setup A , microplates ( Nunc-Maxisorp Immunoplate , Waltham , Massachusetts , USA ) were coated with 100 μL/well of a 100 ng/mL antigen solution in phosphate buffered saline ( PBS ) . After blocking and washings , sera were added at 1:500 dilution in 2% bovine serum albumin in PBS-0 . 05% tween-20 ( BSA in PBS-0 . 05%T ) and incubated at 37°C for 1 hour . For setup B , microplates were coated with 100 μL/well of a 50 ng/mL antigen solution in PBS , and sera were added at 1:200 dilution . In both cases , secondary antibody ( horseradish peroxidase conjugated anti-human IgG , Sigma-Aldrich , St . Louis , MO , USA ) was diluted 1:100 , 000 in 2% BSA in PBS-0 . 05%T and incubated at 37°C for 1 hour . Finally , the substrate ( 3 , 3' , 5 , 5'-Tetramethylbenzidine Liquid Substrate , Supersensitive , Sigma ) was added . The absorbance was read at 620 nm after 1 hour incubation using a Tecan Sunrise ( Tecan Group , Ltd . , Männedorf , Switzerland ) microplate reader . All sera were tested in duplicate . In order to compare results obtained from different plates , a Sample Ratio ( SR ) was calculated according to the following formula: SR=Sample mean−Negative control meanPositive control mean−Negative control mean where the negative control was a pool of fifteen healthy donors and the positive control was the Working Standard Anti-Echinococcus Serum , Human ( NIBSC , Potters Bar , England ) . Reproducibility among different Ag5 batches was evaluated for both setup A and setup B on three different , independent preparations . Ten sera ( high , medium and low positive control samples ) were tested in duplicate and results were evaluated by calculating the mean coefficient of variation ( CV ) among the three tests , for each setup . All measurements were carried out in parallel by two experienced research fellows . Readers were blind to the results of the other tests . Data analysis was performed with MedCalc Statistical Software version 15 . 2 . 2 ( MedCalc Software bvba , Ostend , Belgium; http://www . medcalc . org; 2015 ) . A receiver-operator characteristic analysis ( ROC ) [22–23] was performed to determine a cut-off value for each Ag5 based test . The standard error and the area under the curve were calculated according to DeLong et al . [24] . Levels of sensitivity were plotted against levels of 100 minus specificity at each cut-off point on a ROC curve . Threshold values used were those associated with the highest Youden index J [25] . In order to calculate the best ELISA cut-off values , and to improve sensitivity on active-transitional cysts , that are generally seropositive with HCF-based tests ( while patients with CE4 and CE5 cysts and post-surgical patients have most commonly negative or low results ) , ROC curves were built by using SR values from patients with CE1 , CE2 , CE3a and CE3b as positive group ( 171 sera ) and healthy controls as negative group ( 253 sera ) . The area under the ROC curve ( AUC ) was used to define the antigen discriminatory power ( between subjects with active-transitional cysts and subjects with inactive cysts or without the disease ) . A p-value <0 . 05 was considered statistically significant . McNemar test was performed , on all 580 sera , to compare the sensitivities of the two in-house Ag5 setups and the commercial assays . Differences in SR or SI values between groups were analyzed by Kruskal-Wallis test , for the three ELISAs , independently; when more than two groups were analyzed , after Kruskal-Wallis test , pairwise multiple comparisons were evaluated by Conover test with Bonferroni correction [26–27] . Inter-rater agreement test was used to evaluate the agreement between the gold standard ( US ) and the in-house ELISAs , and the results were expressed by Kappa ( K ) statistic , with 95% confidence interval [28] . Moreover , three statistical analyses were performed on the results of sera from CE patients to assess the effect of the cyst stage , the number of cysts and the previous treatment with albendazole ( ABZ ) , potentially affecting the assay performance of Ag5 ELISAs [13 , 18] . While cyst localization is an important variable and its influence on ELISA results is potentially interesting , it was not evaluated given the low number of patients with extra hepatic cysts . Chi squared test was applied to compare the sensitivity of the two Ag5 ELISA setups within subgroups of patients , classified according to their clinical variables . Then , the same variables were further evaluated by a bivariate logistic regression , to consider their influence on test performances and calculate an odd ratio ( OR ) for each pair of variables . Finally , all the examined clinical variables were concurrently analyzed by a multiple regression , to evaluate their relationship on the diagnostic result . A p-value <0 . 05 was considered statistically significant .
A total of 580 blood sera were collected from June 2008 to June 2012 , including 283 from patients with CE cysts , 44 during follow-up after surgery , and 253 from healthy subjects . As summarized in Table 1 , 295 sera were collected from patients with hepatic cysts ( 90 . 2% ) , one serum from a pulmonary case ( 0 . 3% ) , and the remaining 31 sera ( 9 . 5% ) were from patients with other localizations ( including peritoneum , kidney , and leg ) ; single cysts were found in 146 patients ( 44 . 6% ) , whilst 2 or more cysts were found in the other 137 subjects . The reproducibility among Ag5 lots was assessed in both setups , demonstrating the reliable performances of our Ag5 preparation . Specifically , setup A had a CV of 9 . 7% , and setup B had a CV of 11 . 3% . All sera were analyzed by the commercial assay RIDASCREEN and by the two experimental Ag5 ELISA setups , from November 2011 to September 2012 . Then , Ag5 ELISAs were evaluated by ROC curves ( Fig 1 and Table 2 ) to define optimal cut-off values for data analysis . The areas under the ROC curve ( AUC ) were 0 . 962 and 0 . 978 for setup A and setup B , respectively . Serological results and their statistical significance are summarized in Tables 3 and S1 . At the best cut off value ( 0 . 261 and 0 . 120 for setup A and setup B , respectively ) , the two Ag5 ELISA setups showed different sensitivity . In particular , Ag5 ELISA setup B revealed an overall sensitivity higher than both Ag5 setup A and RIDASCREEN test , whilst Ag5 setup A showed similar sensitivity to the commercial kit . More in detail , Ag5 setup B results displayed statistically significant differences when CE3b , CE4 and post surgery patients were examined . These differences persisted when grouping patients as active-transitional and inactive . Concerning the control sera , a higher number of donors tested positive in Ag5 setup B ( 7 . 5% ) , followed by Ag5 setup A ( 5 . 9% ) and RIDASCREEN ( 1 . 6% ) . Hence , the Ag5 ELISA , especially in setup B , revealed a higher sensitivity and a lower specificity than the commercial RIDASCREEN test . The comparison among the Ag5 ELISAs described in this work and the commercial ELISA is also plotted in Fig 2 . Although it should be noted that the ELISA values are not directly comparable due to the difference in OD normalization , Kruskal-Wallis test on SR or SI values confirmed that all the three ELISAs were able to discriminate between patients and healthy controls ( Fig 2A , 2B and 2C ) ; statistically different results were also obtained with the three ELISAs , when patients were grouped taking into account the active-transitional versus the inactive stages of CE ( Fig 2D , 2E and 2F ) . Finally , none of the three methods was able to completely discriminate among any of the CE groups and post surgery follow-up patients ( Fig 2G , 2H and 2I ) ; however , pairwise comparisons of the subgroups highlighted some differences . Both CE1 and CE2 cysts were different from CE5 in the three assays; CE3a and CE3b were always comparable , but behaved differently in the three tests . The inactive cysts showed important differences in all the tests: CE4 revealed SR or SI values statistically different from CE3a , CE3b and CE5 in Ag5 setup A and RIDASCREEN ELISAs , whilst it differed only from CE5 in Ag5 setup B . CE5 , on the contrary , performed differently from all the other groups ( except for post surgery patients ) in all the ELISAs . Finally , post surgery patients showed dissimilar behavior in all the tests , although with a higher divergence from active-transitional groups for RIDASCREEN results . Further , the Ag5 setup A and the commercial kit , provided a wider range of antibody levels , whilst for Ag5 setup B results were concentrated in a narrower range . The agreement between the gold standard method ( US ) and our Ag5 ELISAs was evaluated by inter-rater agreement ( Kappa ) test . When considering patients with CE1 , CE2 , CE3a and CE3b , kappa ( K ) value was 0 . 828 , with a standard error of 0 . 0279 ( setup A ) , or 0 . 869 , with a standard error of 0 . 0243 ( setup B ) , confirming the excellent agreement between the imaging diagnosis and the ELISA results . When CE4 and CE5 patients were also included in the test , this agreement was poorer , with a K value of 0 . 718 , with a standard error of 0 . 0295 ( setup A ) , or 0 . 795 , with a standard error of 0 . 0262 ( setup B ) . This is not surprising , since these patients , due to the inactivity of the cysts , are often negative to ELISA . In addition to the cyst stage , other major clinical variables such as the number of cysts and the previous ABZ treatment were taken into account using Chi squared test to assess the ability of the Ag5 ELISAs to discriminate among CE patients with different clinical traits . Results are summarized in Table 4 . For both Ag5 ELISA setups , this test showed a statistical significance ( p-value <0 . 05 ) for the stratification of patients in terms of single vs multiple cysts , active-transitional vs inactive cysts and for current or past ABZ treatment vs no ABZ . More in detail , pharmacologically treated patients gave positive results to both Ag5 setups more frequently than untreated patients; further , patients in the active or transitional stage had higher positivity rates than patients in the inactive stages; finally , patients with more than one cyst were positive to Ag5 ELISAs more frequently than patients with one cyst . All the above-mentioned clinical variables were evaluated by a bivariate logistic regression . The statistical significance persisted , with fairly high values of odds ratios , when grouping patients according to active-transitional vs inactive cysts and for ABZ treatment vs no ABZ , but it became only borderline significant for Ag5 setup A , when single vs multiple cysts were considered . The interdependence of test results from the pharmacological treatment and the activity of the cysts was confirmed for both Ag5 ELISA setups by multiple regression; the effect of the number of cysts , instead , was only borderline significant for setup A ( p-value = 0 . 055 ) , remaining significant for setup B .
CE is a public health and economic issue , concerning both humans and farm animals , and requires an early and unambiguous diagnosis . Imaging techniques remain the most reliable method for an accurate diagnosis . Serological tests are required for diagnosis confirmation in doubtful cases , but their current sensitivity and specificity are unsatisfactory , while their value in the monitoring of patients during follow-up is very limited . The development of robust and stage-specific serological tests is therefore still needed . Currently , commercially available serological kits are based on western blotting , hemagglutination , and ELISA , and mostly use HCF as target antigen , a complex mixture of host and parasite electrolytes , proteins , nitrogenous waste products , carbohydrates and lipids . Its composition is known to vary , often significantly , from cyst to cyst [21 , 29–30] . As a consequence , sensitivity and specificity are very heterogeneous across tests . Technological improvements have provided increasingly reliable antigens and tools [14 , 16 , 18 , 31–35] , but their performances are still suboptimal and their production is often expensive or patented , limiting their use in the most affected regions , which are often developing countries that cannot afford the appropriate facilities [17] . Ag5 and AgB are reported to be the most abundant and immunogenic proteins in the cyst fluid [20] . After an initial growing interest in the use of Ag5 for diagnostics , the focus has moved mostly towards AgB , and its subunits . However , it is reported that a proportion of CE patients with active cysts do not develop a detectable humoral response against AgB [21 , 36] . Ag5 cross-reactivity issues , as well as low sensitivity and specificity , have been discussed in many papers [16 , 37–38] , and they are probably the main reason for the decline in Ag5 use for CE immunodiagnosis . Part of the cross-reactivity was associated with the presence of phosphorylcholine bound to the Ag5 38 kDa subunit [39–40] . On the other hand , Ag5 protein shares 96 . 7% and 85 . 5% identity with the homologous sequences of Taenia solium and Echinococcus multilocularis , respectively , and it is inevitable that the same epitopes are present on these proteins . However , interestingly , Ahn and coworkers [30] showed that Ag5 seems to be immunoreactive in every stage of the pathology , as opposed to AgB , whose proteoforms revealed a reduced antibody capturing activity in CE1 , CE4 and CE5 stages . Further studies using sera from patients with other relevant parasitoses are needed to assess the behavior of our Ag5-based ELISAs and to evaluate the value of Ag5-based assays for patient follow-up . Concerning the low sensitivity and specificity reported in previous works , it must be underlined that all the experimental studies concerning Ag5 date back to decades ago , when the analytical techniques themselves suffered from low sensitivity . Therefore , it is likely that the low diagnostic performance reported so far for tests based on Ag5 could be explained with the high heterogeneity of the antigen preparations used at that time . Our results show that Ag5 is a sensitive antigen and further studies using sera from patients with non-CE solid lesions are warranted to evaluate and optimize the cut-off value of the ELISAs when higher sensitivity is needed in the differential diagnosis of inactive cysts . In a previous work [21] , a chromatographic method for the reproducible preparation of native Ag5 from different HCF sources was described , enabling the production of a protein fraction highly enriched in Ag5 , as verified by mass spectrometry . Preliminary ELISA experiments on a limited panel of CE sera revealed the high reactivity of this antigen . Therefore , we were prompted to evaluate the diagnostic performance of this antigen preparation on a substantial number of CE patients and healthy control sera . ROC curves generated by both Ag5 ELISA setups using CE1 , CE2 , CE3a and CE3b sera demonstrated its good sensitivity and specificity . When comparing the diagnostic accuracy of these Ag5-based ELISAs with a commercial kit routinely used in clinical laboratories , the excellent performances for Ag5 setup B outperformed those of the commercial test . Nevertheless , specificity was higher for the commercial kit . When we evaluated the effect of clinical variables on Ag5 ELISA results , we found that patients with more than one cyst , and/or in the active or transitional stage , and/or who received drug therapy , were positive to Ag5 test more frequently than the other patients . The bivariate logistic regression and the multiple regression both highlighted an effect due to the pharmacological treatment and to the cyst activity , while the number of cysts maintained a statistical significance only when setup B was used , confirming the importance of these variables as reported in other previous works [13 , 18 , 35] . The biological mechanism at the basis of the influence of ABZ treatment on serology results was attributed to high levels of ABZ in cyst fluid causing the germinal laminated membranes to become more permeable , inducing a leakage of their antigenic contents in the blood stream . In turn , the leakage of parasite antigens triggers and sustains increased concentration of circulating antibodies[41–42] . Concerning the effect of the number of cysts and the stage of the disease , no experimental study has demonstrated the biological mechanisms underlying these serology patterns , however it is likely that the loss of cyst wall integrity during the evolution of the cyst and the presence of a large antigenic mass when multiple cysts are present may explain the observed serology behavior . In conclusion , the described serological assay , combining robustness , sensitivity , and easiness of execution , with the low cost , high reproducibility and rapidity of the Ag5 preparation method , makes this antigen a promising candidate for the serodiagnosis of CE especially in the setup B . Moreover , to our knowledge , this is the first report on the influence of the pharmacological treatment , the cyst stage , and the number of cysts on the results of an Ag5-based ELISA test . | Cystic echinococcosis is a neglected disease caused by the larval stage of the tapeworm Echinococcus granulosus complex affecting both humans and livestock . The disease is considered one of the world’s major zoonoses , and represents a public health problem . Clinical diagnosis and follow-up is mainly based on imaging , while serology should complement imaging-based diagnosis when imaging features are unclear . However , current commercial immunoassays lack satisfactory sensitivity and specificity . They are mostly based on crude antigen preparations of E . granulosus hydatid cyst fluid , a heterogeneous mixture containing molecules of both parasite and host origin , thus standardization is also an issue . Ag5 is one of the most immunogenic proteins present in the hydatid cyst fluid . In a previous work , we described a method enabling the preparation of a highly enriched Ag5 fraction . Here , we present the evaluation of the diagnostic performances of this preparation in two ELISA setups , using a large number of human sera . The influence of several clinical variables on the performance of the tests was also assessed . The results obtained by the Ag5 ELISAs , combined with the robustness of the Ag5 preparation method , make this antigen a promising candidate for the serodiagnosis of CE . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"invertebrates",
"medicine",
"and",
"health",
"sciences",
"enzyme-linked",
"immunoassays",
"pathology",
"and",
"laboratory",
"medicine",
"tropical",
"diseases",
"parasitic",
"diseases",
"animals",
"echinococcus",
"surgical",
"and",
"invasive",
"medical",
"procedures",
"pharmaceutics",
"neglected",
"tropical",
"diseases",
"immunologic",
"techniques",
"research",
"and",
"analysis",
"methods",
"echinococcosis",
"serology",
"serodiagnosis",
"immunoassays",
"flatworms",
"helminth",
"infections",
"diagnostic",
"medicine",
"biology",
"and",
"life",
"sciences",
"drug",
"therapy",
"organisms"
] | 2016 | Diagnostic Accuracy of Antigen 5-Based ELISAs for Human Cystic Echinococcosis |
Most biological traits and common diseases have a strong but complex genetic basis , controlled by large numbers of genetic variants with small contributions to a trait or disease risk . The effect-size of most genetic variants is not absolute and is instead dependent upon multiple factors such as the age and genetic background of an organism . In order to understand the mechanistic basis of these changes , we characterized heritable trait differences between two domesticated strains of C . elegans . We previously identified a major effect locus , caused in part by a mutation in a component of the NURF chromatin remodeling complex , that regulates reproductive output in an age-dependent manner . The effect-size of this locus changes from positive to negative over the course of an animal’s reproductive lifespan . Here , we use a previously published macroscale model of the egg-laying rate in C . elegans to show that time-dependent effect-size is explained by an unequal use of sperm combined with negative feedback between sperm and ovulation rate . We validate key predictions of this model with controlled mating experiments and quantification of oogenesis and sperm use . Incorporation of this model into QTL mapping allows us to identify and partition new QTLs into specific aspects of the egg-laying process . Finally , we show how epistasis between two genetic variants is predicted by this modeling as a consequence of the unequal use of sperm . This work demonstrates how modeling of multicellular communication systems can improve our ability to predict and understand the role of genetic variation on a complex phenotype . Negative autoregulatory feedback loops , common in transcriptional regulation , could play an important role in modifying genetic architecture in other traits .
Most biological traits have a strong heritable , or genetic , component . There is a general interest to understand the genetic basis of these traits , often by the identification of quantitative trait nucleotides ( QTNs ) underlying heritable variation segregating within a population . Two decades of biological trait studies in humans and other model organisms indicates the genetic basis of most biological traits is incredibly complex–dozens , hundreds , or even thousands of genes are involved , often with non-linear effects . Work in model organisms demonstrates that genetic epistasis [1] ( i . e . biological epistasis [2] or compositional epistasis [3] ) between two loci is ubiquitous; it is observed in fungi [4–6] , plants [7–10] , insects [11 , 12] , nematodes [13–15] , birds [16 , 17] , and mammals [18–21] . Environment and age are also relevant covariates , influencing the effect and onset of the QTN on the phenotype . For example , a recent survey of natural variation in C . elegans gene expression identified > 900 eQTLs with time-dependent dynamics [22] . The details of these interactions are important for predicting an individual genetic variants effect on fitness . While the role of statistical epistasis ( i . e . the deviation from a linear model in a sampled population ) is debated [23] , the predicted effect size of variants with biological epistasis is dependent on their allele frequency in the mapping population [24] . As the frequency of these genetic variants change ( e . g . due to positive selection ) , their linear effect will also change . Age-dependent genetic variants also play an essential role in theories of antagonistic pleiotropy , which proposes that pleiotropic genes can have opposite effects on fitness at different ages [25] . Identification of QTNs can help us understand how epistasis and age-dependence arise in natural populations . GWAS and QTL mapping , two common quantitative genetics techniques , are usually unable to identify these genetic variants . GWAS , which can narrow down causative genetic variants to small regions , are typically underpowered to identify statistically significant epistatic interactions due to low natural allele frequencies and a large number of obligatory statistical tests [26] . QTL mapping , on the other hand , has increased power to identify interacting QTLs due to equal allele frequencies but identifies large regions in linkage disequilibrium containing thousands of potential variants [24] . Interactions between genetic variants segregating within a population are inherently difficult to study , and epistasis studies typically focus on laboratory induced loss-of-function mutations through mutagenesis or RNAi . Due to the filtering effect of natural selection , the mechanisms that underlie these types of epistasis might not apply to natural populations [27] . To better understand how naturally occurring genetic variants impact a trait , we are studying two C . elegans strains , N2 and LSJ2 , derived from an individual hermaphrodite isolated in 1951 . The two strains were then separated into distinct cultures of either solid or liquid media sometime between 1957 and 1958 ( Fig 1A ) [28 , 29] . N2 was cultured for ~15 years on agar plates while LSJ2 was cultured for ~50 years in liquid culture . We previously identified 94 new mutations fixed in the N2 lineage and 188 new mutations fixed in the LSJ2 lineage with next-generation sequencing [30] . Despite this low level of genetic diversity , a large number of phenotypic differences distinguish the two strains . A total of five QTNs have been identified in these strains , providing empirical evidence for theories of genetic targets of evolution [31] linking variation in neuropeptide receptor activity to changes in social behavior [32] , variation in sensory gene deployment with specific chemosensory responses [28 , 30 , 33] , and variation in an acetyltransferase as the source of cryptic genetic variation affecting organ development [34] . Most recently , we found that LSJ2 and N2 evolved different life-history strategies and mapped these changes to a single genetic change in a chromatin remodeling factor called nurf-1 [35] . This genetic change is an example of antagonistic pleiotropy: decreases in the reproductive rate at early time points caused by the LSJ2 allele of this gene are accompanied by extended lifespan , increased survival to a panel of drugs , and increased reproductive rate later in life . To determine how these tradeoffs arise , we examined the reproductive differences between the N2 and LSJ2 strains at five different time points spanning their reproductive lifespan . Our goals for this study were to identify examples of complex genetic architecture and to understand their molecular and cellular causes .
We previously performed QTL mapping on the reproductive rate with 94 recombinant inbred strains ( RILs ) generated between LSJ2 and CX12311 [35] . The CX12311 strain derives the majority ( >99% ) of its DNA from N2 except for a small amount of DNA backcrossed from the CB4856 wild strain near the npr-1 and glb-5 genes ( Fig 1B ) [30] . Novel mutations in these two genes became fixed in the N2 lineage and result in pleiotropic effects on a large number of phenotypes . Use of the CX12311 strain allows us to avoid studying their effects . To examine the role age plays on reproduction , we chose five time points that span the reproductive lifespan of the CX12311 animals . The egg-laying rate was quantified by counting the number of eggs laid by six animals for six hours on agar plates seeded with E . coli bacteria ( Fig 1C ) . We previously identified a major effect QTL centered over the nurf-1 gene responsible for ~50% of the observed phenotypic variation [35] . To study how the animal’s age affected the effect-size of this locus , we first segregated the 94 RIL strains based on their genotype at nurf-1 ( Fig 1D–top panel ) . At the first three time points , RIL strains with the N2 genotype laid more eggs than RIL strains with the LSJ2 genotype , however , at the fourth and fifth time point , this relationship flipped—animals with the LSJ2 allele of nurf-1 laid more eggs than animals with the N2 allele . To visualize this effect more clearly , we plotted the effect-size of the nurf-1 locus at all five time points ( Fig 1D–bottom panel ) by subtracting the egg-laying rate of the strains with each genotype . The effect-size of the LSJ2 allele of nurf-1 was negative for the first three time points and positive for the last two time points . This result is a clear example of age-dependence , i . e . the prediction of the effect of the nurf-1 locus on the egg-laying rate requires knowledge of both the nurf-1 genotype as well as the current animal’s age . To verify these observations , we next assayed a near isogenic line ( NIL ) constructed by backcrossing the region surrounding nurf-1 from LSJ2 into the CX12311 strain ( Fig 1B ) along with the CX12311 and LSJ2 parental strains . The CX12311 strain ( containing the N2 allele of the nurf-1 locus ) laid more eggs than the NILnurf-1 strain for the first three time points but fewer eggs at the fourth and fifth time points , again resulting in a time-dependent positive to negative effect size ( Fig 1E–top and bottom panel ) . Good qualitative agreement was observed between the effect size of nurf-1 in both the RIL and NIL strains ( Fig 1D and 1E–bottom panel ) . Due to additional segregating variants in the RIL strains , we do not expect these two calculations to be identical . In both experimental designs , the intersection of the two lines was due to a decrease in the egg-laying rate in the strain containing N2 nurf-1 as opposed to an increase in egg-laying of the LSJ2 nurf-1 strains . The LSJ2 strain was statistically indistinguishable at the first three time points from the NIL strain . However , at the last two time points , LSJ2 laid additional eggs resulting in a larger effect-size at these two points . These results demonstrate a correlation between the positive and negative effects of the nurf-1 locus on egg-laying and the animal’s age . The rate eggs are laid on an agar plate is dependent on multiple factors: size and rate of mitosis of the germline progenitor pool , speed of meiosis/differentiation of these cells , maturation and growth of oocytes , ovulation and fertilization of the primary oocyte to produce an egg , and the rate of active expulsion of an egg via a vulva valve motor program ( Fig 2A ) . To characterize which egg-laying rate factor might be affected by nurf-1 , we first characterized the number of fertilized eggs in each strain via DIC microscopy . If the rate fertilized eggs were laid was affected in LSJ2 and NILnurf-1 strains but the rate of production of fertilized eggs remained the same , we would expect LSJ2 and NILnurf-1 to contain more eggs in their uterus than the CX12311 strain . We measured the number of fertilized eggs in each of these three strains at 24 and 48 hours after the L4 stage ( Fig 2B ) . Contrary to our expectations , we find both LSJ2 and NILnurf-1 have significantly fewer fertilized eggs than the CX12311 strain . We conclude fertilized egg production must be affected in the LSJ2 and NILnurf-1 strain and the lower number of unlaid fertilized eggs in these strains is potentially a consequence of the reduced rate of fertilized egg production . We next measured the number of large oocytes undergoing oogenesis in these three strains ( the region marked “Oogenesis” in Fig 2A ) . We hoped to distinguish between two possible mechanisms modifying the rate of fertilized egg production: 1 ) the rate of production of mature oocytes is decreased in LSJ2/NILnurf-1 strains or 2 ) the rate of fertilization of a mature oocyte is decreased in the LSJ2/NILnurf-1 strains . In the former case , we reasoned the number of large oocytes undergoing oogenesis would be lower in LSJ2 and NILnurf-1 due to a decrease in production . In the latter case , we reasoned the number of large mature oocytes would increase over time if oocyte production was unaffected but the fertilization rate was lower . However , the difference between the number of large oocytes at 24 and 48 hours was not statistically significant between any of the three strains ( Fig 2C ) . We believe this result may indicate the presence of a homeostatic mechanism for constant oocyte maturation independent of oocyte production or fertilization rates . Finally , we measured the production rate of new progeny germ cells from progenitor germline stem cells via mitosis . The progeny germ cells lie in the mitotic zone ( Fig 2A ) and provide the source of meiotic germ cells , which will later differentiate into oocytes . There is not a one-to-one relationship between the rate of mitosis and oocyte production due to cannibalization of a subset of these cells , but it is thought there is a relationship between the size of the progenitor pool and the subsequent rate of animal reproduction [36] . Progenitor cells can be distinguished from cells in the mitotic zone based on nuclear morphology . We observed a significant difference between CX12311 and NILnurf-1 progenitor cell numbers at the 24-hour time point , suggesting this could be the cause of the egg-laying rate differences ( Fig 2D ) . However , the NILnurf-1 strain was significantly different from LSJ2 at this time point as well . This difference disappeared at the 48-hour time point when each strain had a similar average number of progenitor cells . We then measured the number of cells undergoing mitosis in the mitotic zone to test for potential differences in the ratio of dividing cells . We used an antibody to Ser10 phosphorylation in histone H3 , which is correlated with chromosome condensation in mitosis ( Fig 2E ) [37 , 38] . We observed similar results to the germline progenitor pool suggesting the ratio of progenitor cells undergoing mitosis was the same in all three strains . The NILnurf-1 was different from both parental strains at the 24-hour time point , and all three strains were indistinguishable from each other at 48 hours . From these observations , we conclude nurf-1 likely modulates the early germline proliferation rate and LSJ2 contains additional genetic variation that can suppress this effect . These proliferation differences could account for a fraction of egg-laying disparities , but we believe it does not provide a comprehensive explanation , as this would indicate different functional mechanisms between LSJ2 and NILnurf-1 . The identification of age-dependent QTLs through genetic mapping approaches has rarely led to a mechanistic understanding of how an animal’s age can influence the effect of a genetic variant . To explore possible mechanisms for nurf-1 age-dependence , we leveraged extant C . elegans egg-laying genetics literature . A recently described mechanism indicates sperm can regulate the rate of egg-laying . Before the onset of oogenesis , C . elegans hermaphrodites produce and store a limited number of sperm ( ~300 ) in an organ known as the spermatheca [39] . Sperm secrete a hormone called major sperm protein ( MSP ) that binds to and activates ephrin receptors on oocytes and gonad sheath cells thereby stimulating both oocyte maturation and gonad sheath contraction leading to ovulation ( Fig 3A ) [40 , 41] . The effect of MSP is dose-dependent and the limited number of sperm stored in the spermatheca decrease over time with each fertilization event . The decreased numbers of stored sperm , in turn , lead to decreased ovulation and oocyte maturation with increasing hermaphrodite age [40 , 41] . This process suggests a possible mechanism linking hermaphrodite age to egg-laying rate . Animals with N2 nurf-1 lay more eggs during the first three time points compared to LSJ2 nurf-1 animals and will consequently have less sperm at the fourth and fifth time point . This reduction in sperm will lower sperm hormone concentrations at the primary oocyte , which counterbalances the positive effect of the N2 nurf-1 allele on egg-laying during the transition between the third and fourth time point . We tested this hypothesis more formally using a previously published macroscale model [42] , which stipulates the egg-laying rate is proportional to the product of the number of oocytes and the number of sperm ( Fig 3B ) . This model is defined by four time-independent parameters: ko , which specifies how rapidly oocytes are created , kc , which specifies a carrying capacity of the gonad , kf , which defines the fertilization rate , and S0 , which specifies the number of self-sperm created by the animal . In this report , we excluded the kc parameter , because the carrying capacity of all three strains was similar ( Fig 2C ) , and our attempts to fit this parameter resulted in negative , non-physiological values . Moreover , due to the nature of the equations , the kf and ko parameters end up having a similar effect on the egg-laying rate . To prevent the noise caused by their tight correlation , we fit a single ko value for all of the strains . This ko value does not distinguish between the number of molecular and cellular processes described above that influence the speed of mature oocyte production . We fit individual ko and S0 parameters to each of the replicates in Fig 1E ( Fig 3C ) . This model could recapitulate the rise and fall of the egg-laying rate–the rate rose as oocytes were generated and fell as the number of sperm decreased ( Fig 3D–top panel ) . A significant increase in S0 was observed in the LSJ2 strain compared to the CX12311 and NILnurf-1 strain ( Fig 3C–top panel ) . This result qualitatively agrees with our previous observation LSJ2 animals lay more eggs over the course of their lifetime than CX12311 animals [35] . While the predicted S0 value of CX12311 was in good quantitative agreement with our previously measured fecundity ( 263 vs . 256 ) , the predicted value of S0 for LSJ2 was significantly higher than the measured fecundity ( 434 vs . 319 ) . This difference is most likely due to the inability of the model to fully account for the genetic changes in the LSJ2 strain—a possibility supported by significantly higher average residuals observed for LSJ2 compared to CX12311 ( 2 . 0 vs . 0 . 72 ) . The fitted ko parameters matched our expectations—CX12311 had a higher ko value than either the LSJ2 or NILnurf-1 strain ( Fig 3C—bottom panel ) . Furthermore , the effect size calculated from the modeling experiments ( Fig 3D–bottom panel ) qualitatively matches the effect size we experimentally measured ( Fig 1E–bottom panel ) . Our laboratory data analysis reveals a constant change in oocyte generation rate and a time-dependent effect on egg-laying rate resulting in sign-switching at later time points . Our model explicitly predicts the reduction in CX12311 egg-laying rates at later time points is not due to changes in reproductive capacity but rather to decreases in sperm number . To test this prediction , we measured the number of fertilized eggs and large oocytes in CX12311 animals at 66 hours ( Fig 4A ) . In line with our expectations , we observed a reduction in the number of fertilized eggs ( ~10 ) contained in the CX12311 animals compared to earlier time points ( Fig 2B ) . The reduction in fertilized eggs indicates the decrease in egg-laying rate cannot be explained by a decrease in the expulsion rate of fertilized eggs from the uterus . We also observed a sizable number of large oocytes ( ~22 ) in CX12311 animals , which is an increase of large oocytes compared to the previous two time points ( Figs 4A and 2C ) . This increase in large oocytes suggests the change in CX12311 egg-laying rate is not due to changes in oocyte maturation rate because mature oocytes are available for fertilization . These observations are consistent with our hypothesis that decreased sperm and MSP are responsible for the decreased rates of egg-laying in CX12311 at later time points . Our modeling predicts the CX12311 strain will have less sperm later in life due to their increased rate of fertilization at earlier time points ( Fig 4B ) . We counted the number of sperm cells present in the spermatheca in CX12311 and NILnurf-1 animals using DAPI staining combined with fluorescence microscopy . At both 48 and 66 hours , more sperm were present in the NILnurf-1 animals compared to CX12311 in a manner consistent with the modeling ( Fig 4C ) . We took advantage of C . elegans’ androdioecious mating system for the final test of our model . To increase the sperm available to hermaphrodites at later time points , we mated CX12311 and NILnurf-1 hermaphrodites to CX12311 males , which results in the transfer of ~1000 sperm to each hermaphrodite . After mating young adult animals , we separated the hermaphrodites from males and measured the egg-laying rate at two time points late in life . Consistent with our predictions , increasing sperm number prevented sign-switching from occurring at both times in mated animals ( Fig 4D ) . This finding indicates a decrease in sperm number is the primary reason for reduced CX12311 egg-laying rates at later time points . We next investigated whether additional QTLs affected egg-laying rate differences between LSJ2 and CX12311 . The presence of a major effect QTL ( such as the nurf-1 locus ) can mask the effects of smaller QTLs , so we performed additional scans using the genotype of nurf-1 as an additive or interacting covariate ( Fig 5A ) [43] . We identified five significant genome-wide QTLs: one QTL on the center of chromosome I , one QTL on the center of chromosome II , one large QTL on the center and right arm of chromosome IV , one QTL on the right arm of chromosome V , and one QTL in the center of the X chromosome ( S1–S5 Figs ) . The identification of the QTL on chromosome I was expected , as it contains a previously described missense mutation in nath-10 known to affect egg-laying [34] . However , the other four QTLs do not contain any genetic variants associated with egg-laying . In order to understand how the five modifier QTLs regulate the egg-laying rate at the five mapping time points , we segregated the 94 RIL lines based upon their nurf-1 and modifier QTL genotype ( Fig 5B ) . By segregating both genotypes , we can visually determine if any non-linear interactions ( i . e . epistasis ) exist between nurf-1 and the five modifier loci . Additive interactions result in two lines with identical slopes while non-linear interactions result in lines with different slopes . Visual inspection of these effect-size graphs indicates the presence of both additive and non-linear effects . In order to formalize this analysis , we used ANOVA to determine ( i ) if the modifier QTL had a significant effect at a particular time point , ( ii ) whether there was a significant interaction term between nurf-1 and the modifier variant , and ( iii ) the total amount of variance the modifier QTL explained at a particular time point ( Fig 5B ) . We considered two types of epistasis: positive and negative . When the modifier variant effect is the same direction in both nurf-1 genotypes ( i . e . the slope is positive for both lines but with different magnitudes ) , this is positive epistasis . When the modifier variant effect has a different direction in the two nurf-1 genotypes ( i . e . one slope is positive and the other slope is negative ) , this is negative epistasis . Our analysis indicates the effect of these modifier QTLs is similarly complex to independent nurf-1 QTL mapping effects with each QTL exhibiting age dependence and non-linear interactions with the nurf-1 locus . This mapping suggests egg-laying differences between N2 and LSJ2 are multigenic , involve extensive epistatic interactions , and are highly age-dependent . Further inspection of the age-dependence and epistasis of modifier QTLs with nurf-1 reveals several interesting trends ( Fig 5B ) . Epistasis is most likely to occur at the first and last two time points ( 6 out of 8 significant effects ) and less likely to be observed at the second and third time point ( 0 out of 7 significant effects ) . While statistically significant , it is difficult to interpret epistasis at the first and last time points because the egg-laying rate of some genetic backgrounds converges on zero . On the other hand , the negative epistasis observed in three of the modifier QTLs at the 60–66-hour time point is particularly intriguing . We observed effect-size switching for nurf-1 at the same time point and investigated whether the two features could be related . Additional modeling was implemented to determine if epistasis could arise through the unequal use of sperm . We modeled a modifier QTL by assuming it changed the oocyte generation rate ( ko ) in an additive fashion ( S6A Fig ) and calculated the egg-laying rate predicted by this model for the four possible genotypic combinations of nurf-1 and the modifier QTL ( S6B Fig ) . These values were plotted at three time points similar to Fig 5B ( S6C Fig ) . This analysis demonstrates the potential for modifier QTLs to create negative epistasis at a time point when negative epistasis is also observed in our QTL mapping data . How does negative epistasis arise ? One possible reason can be deduced from our observation regarding the sign-switching of the effect size of the modifier QTL . The modifier QTL switches signs due to changes in the oocyte generation rate ( ko ) in two genetic backgrounds . However , the exact time this occurs is also dependent on the genetic background of the nurf-1 locus . The modifier has a positive effect in one nurf-1 background but switches to a negative effect in the other nurf-1 background and creates an intersection at two different time points corresponding to the N2 and LSJ2 modifiers . The period between the two intersections is when negative epistasis is observed . After the sign-switching occurs for both genotypes , the modifier allele again has the same direction of effect in both nurf-1 backgrounds . We refer to this time window as the sign epistasis zone ( S6B Fig ) . Finally , we decided to test whether we could use the macroscale modeling of the egg-laying process to improve our QTL mapping . Instead of performing QTL mapping on each of the five time points , we used this data to estimate a ko and S0 for each RIL strain . We assumed , as before , that kf is the same for all RIL strains and excluded kc . The distribution of the fitted ko and S0 parameters are plotted in Fig 6A and 6B . We next performed QTL mapping to identify genetic regions that influence these rates . As expected , we reidentified loci from the previous analysis . The QTL surrounding nurf-1 was identified as a regulator of both the oocyte generation rate ( ko ) and the number of self-sperm ( S0 ) . The modifier QTLs on II and X regulated the oocyte generation rate but not the number of sperm number , while the QTL on V had an effect on sperm number but not the oocyte generation rate . Interestingly , this analysis also identified a new QTL on chromosome III as a regulator of sperm number ( Fig 6B ) , suggesting the process of egg-laying modeling not only helps us understand the effect of modifier QTLs but also facilitates the identification of new QTL loci . In order to test these results , we utilized a NIL strain isogenic for the region surrounding the QTL on V . We crossed the NILV strain to the NILnurf-1 strain to create the double NILnurf-1;V strain ( Fig 6C ) . This strain combination allowed us to test the effect of the modifier QTL with both nurf-1 genotypes . For each of these strains , we measured the total amount of progeny produced . Surprisingly , the NILV QTL increased the brood size in both nurf-1 backgrounds but had a larger effect size with the LSJ2 nurf-1 genotype ( 260 vs . 300 ) ; indicating epistasis does exist between QTLV and QTLnurf-1 ( Fig 6D ) . This result qualitatively agrees with the QTL mapping data ( Fig 6B–right panel ) , but discrepancies were observed between the magnitudes of the total brood sizes , suggesting additional changes to the model may be beneficial .
Our results identify four novel genetic loci that regulate the reproductive rate and/or fecundity . Locus variation arose and fixed following separation of the N2 and LSJ2 lineages . Polygenic traits are common in natural populations , but the speed of polygenic trait evolution in laboratory conditions is surprising . Based upon minimum generation time , we estimate a maximum of 3900 generations separate the LSJ2 and N2 strains . We previously identified one of the causative genetic variants as a 60 bp deletion in nurf-1 , which encodes a component of the NURF chromatin-remodeling factor [35] . This deletion arose and fixed in the LSJ2 lineage . Using competition experiments , we demonstrated this genetic variant was advantageous in the LSJ2 growth conditions , suggesting it was fixed by selection . The additional modifier variants could also be advantageous , which could explain their rapid fixation . However , due to a small effective population size and limited outcrossing , genetic draft and genetic drift are also applicable evolutionary forces for these strains . Additional work is needed to identify the responsible genetic variants , determine their lineage of origin , and test their fitness in the relevant conditions to determine if fixation was caused by selection , genetic drift , or genetic draft . Unfortunately , these experiments are not trivial . Despite the inherent advantages of this system compared to wild strains , each QTL still contains a handful to dozens of potential causal genetic variants . The small effect-size of these variants also requires a large number of animals to be tested to obtain the necessary statistical power to distinguish between strains with and without the variant . High-throughput and automated analyses of reproductive output would greatly aid this work . The different effects of the nurf-1 locus on the reproductive rate at early and late time points are an example of antagonistic pleiotropy; in the CX12311 strain , the fitness benefits of an increased reproductive rate at early time points will be counterbalanced by the fitness costs of decreased reproductive rate at later time points . Antagonistic pleiotropy can explain many aspects of aging , where selection on early life traits leads to deleterious consequences later in life [25 , 44] . In the laboratory conditions experienced by N2 , the fitness costs of the decreased late-age fecundity in the CX12311 strain is likely attenuated by the fact that intensity of selection declines with age [45] and the short period of time food is available ( 3 days ) before it is exhausted . In contrast , in the liquid , axenic conditions experienced by LSJ2 , the animals were grown for ~1 month between transfers . In these conditions , the fitness costs of the changes in reproduction are likely attenuated compared to other phenotypes controlled by nurf-1 , including lifespan , stress survival , and dauer formation [35] . We had previously shown that the nurf-1 deletion fixed in the LSJ2 lineage has a positive effect on relative fitness in the liquid , axenic conditions but a negative effect on relative fitness on agar plates [35] . While the laboratory conditions studied here will not be found in C . elegans natural environment , these results demonstrate how an antagonistically pleiotropic gene could lead to different fitness effects in different environments . C . elegans populations are globally distributed and inhabit many different niches [46 , 47] , thereby providing many opportunities for balancing selection to act upon antagonistically pleiotropic genes . A number of genetic regions appear to be under ancient balancing selection in C . elegans [15 , 48–51] and some of these regions could arise from the maintenance of naturally occurring variation in antagonistically-pleiotropic genes . The study of natural traits from a number of phyla reveals numerous examples of biological epistasis [6 , 11 , 13 , 52] . Age also acts as an important covariate for genetic variation . The presence of epistasis and age-dependence obscures the relationship between genotype and phenotype due to the diminished effects of genetic variants when mapping populations are not appropriately segregated by age or genetic background . Our results provide a framework to understand how age-dependence also arises through emergent properties of a cellular network , which we believe to be the major scientific contribution of this work . Our work indicates the complex seesaw of effects nurf-1 has on reproductive output is explained using two major considerations: ( 1 ) a hormonally mediated negative feedback loop linking sperm with oocyte maturation , and ( 2 ) the effect nurf-1 has on the rate of oogenesis . Consequently , the molecular details of how nurf-1 modifies protein and cellular function are unnecessary to explain its age-dependence . Our modeling experiments also demonstrate how sign epistasis could arise in an age-dependent manner strictly through age-independent changes in the oocyte production rate . The origin of genetic epistasis is often thought of in terms of biochemical properties of proteins , through physical interactions between two proteins , or through multiple changes to a parallel or linear signaling pathway [53–55] . However , these mechanisms are typically identified through analyses of laboratory-derived mutations or genetic perturbations with a strong negative effect on fitness . Our study provides evidence for unique epistatic mechanisms derived from natural variation . This is the second example we have identified in C . elegans of what we refer to as cellular epistasis [15]–i . e . the non-linear interactions are an emergent property of the functions of cellular networks as opposed to properties of molecular or biophysical interactions between proteins . It will be interesting to see how often cellular epistasis is responsible for genetic epistasis in natural traits . While the C . elegans reproductive system is a special case , any negative feedback loops acting on a measurable trait could cause similar age-dependent changes . For example , negative autoregulatory feedback loops are common in transcription factors networks . Any genetic variants ( either cis or trans ) that increase the rate of transcription of one of these transcription factors will initially appear to have a positive effect-size on the mRNA levels of the gene . However , as the amount of protein product increases , the transcription factor will turn off the expression of its mRNA . Since the amount of protein product will be higher in the strain with the higher initial expression , transcription will turn off sooner in this strain and the genetic variant will soon appear to have a negative effect on expression . Our work provides an example of how multidisciplinary studies can be used tackle the genetic basis of complex traits–we leveraged quantitative genetics , detailed knowledge of the egg-laying process in C . elegans , and an existing macroscale model of egg-laying to make our conclusions .
Strains were cultivated on agar plates seeded with E . coli strain OP50 at 20°C [56] . Strains used in this study are: N2 , LSJ2 , CX12311 kyIR1 ( V , CB4856>N2 ) ; qgIR1 ( X , CB4856>N2 ) . NIL strains used for this study are: PTM66 ( NILnurf-1 ) kyIR87 ( II , LSJ2>N2 ) ; kyIR1 ( V , CB4856>N2 ) ; qgIR1 ( X , CB4856>N2 ) , PTM75 ( NILV ) kahIR2 ( V , LSJ2>N2 ) , kyIR1 ( V , CB4856>N2 ) ; qgIR1 ( X , CB4856>N2 ) , PTM84 ( NILnurf-1 , V ) kahIR6 ( IV , LSJ2>N2 ) ; kyIR87 ( II , LSJ2>N2 ) ; kyIR1 ( V , CB4856>N2 ) ; qgIR1 ( X , CB4856>N2 ) RIL strains used in this study are sequentially: CX12312 –CX12327 , CX12346 –CX12377 , CX12381 –CX12388 , CX12414 –CX12437 , and CX12495-CX12510 Egg-laying assays were performed as previously described [35] . For mating experiments in Fig 4D , twelve N2 or CX12311 males were placed on experimental plates with six L4 hermaphrodites of interest for the first time point , and mated hermaphrodites were then transferred to experimental plates without males . Successful mating events were validated via the observation of males in the F1 generation of mating plates . Gonad immunofluorescence was performed on adult animals from each strain 24 and 48 hours after the L4 larval stage . Gonads were dissected in M9 containing 1% Tween and 1 mM levamisole and fixed in 2% paraformaldehyde via freeze cracking . Fixed gonads were stained with a 1:50 conjugated H3KS10p Alexa Fluor 488 antibody ( Millipore Cat . 06-570-AF488 ) specific to mitotically dividing cells and 1 . 5 mg/mL of DAPI in Vectashield mounting medium ( Vector Laboratories Cat . H-1200 ) [37 , 38] . Mitotic germline cells and germline stem cells were imaged and scored between the transition zone and the distal tip cell on an Olympus IX73 inverted microscope with a 100x/1 . 40 UPlanSApo objective ( Olympus ) and a Hamamatsu Orca-flash4 . 0 digital camera . The distal tip cell ( Fig 2A ) is located at the tip of the gonad and releases a mitosis-promoting factor . The gonad transition zone represents the location where germline cells initiate meiosis . Embryo/oocytes/sperm/germline stem cells were quantified by fixing whole animals in 95% ethanol and staining nuclei with 1 . 5 mg/mL DAPI in Vectashield mounting medium . Embryos were scored using DIC on a 10x/0 . 30 UPlanFL N objective ( Olympus ) . DAPI stained oocytes were imaged and scored between the spermatheca and posterior gonadal arm with a 40x/1 . 3 PlanApo objective ( Olympus ) . Sperm and germline stem cells were identified using z-stacking to capture multiple planes within each spermatheca or progenitor zone respectively . The transition zone was identified using morphological criteria based upon the crescent shape of the nucleus . All images and scoring were processed in ImageJ . Significant differences between means were determined using a Mann-Whitney U test , which is a nonparametric test of the null hypothesis . For simplicity , a Bonferroni correction was used to modify the type I error rate to account for multiple testing . For each figure , we listed the uncorrected p-value and the number of comparisons used for the Bonferroni correction in S1 Table . Both the non-parametric test and the Bonferroni correction should be conservative approximations of the true p-value . For the analysis of epistasis in Fig 5 , we used ANOVA to calculate an F-value and associated p-value using the fitqtl function in R/qtl . We calculated a single position for each of the five modifier QTLs to use for all five time points . All significant QTLs were simultaneously fit together for each of the five time points considering both their linear effect and their interaction with the nurf-1 genotype ( i . e . y = QI +QII + Qnurf + QIV + QV + QX + QI*Qnurf + QII*Qnurf + QIV*Qnurf + QV*Qnurf + QX*Qnurf ) . The F-value was calculated by dropping a single QTL at a time . Since most of the parameters were significant before multiple comparison testing , we used the Benjamini-Hochberg procedure to control for the false discovery rate . R/qtl was used to perform a one-dimensional scan using marker regression on the 192 markers [43] . The significance threshold ( p = 0 . 05 ) was determined using 1000 permutation tests . To identify modifier QTLs , the nurf-1 marker was used as an additive and interactive covariate for additional one-dimensional scans , assuming a normal model . The significance threshold ( p = 0 . 05 ) for these two tests was determined using 1000 permutation tests . We adapted a previously published model [42] , which simulates the egg-laying rate using the following set of equations: E˙=kfOS O˙=ko−E˙ S˙=−E˙ where E is fertilized eggs , O is oocytes , S is sperm , kf is the rate of oocyte fertilization , ko is the rate of ovulation , and kc is the carrying capacity of the uterus . A dot indicates a time-derivative . We solved these ODEs numerically using a Dormand-Prince explicit solver or using an analytical solution ( below ) . To calculate best fits , we assume the LSJ2 , CX12311 , RILs and NILnurf-1 strains have unique values of ko , but share the kf parameters . These parameters are estimated using a Levenberg-Marquardt non-linear least squares algorithm . The two fits are subtracted to calculate effect-size . We used Mathematica to analytically solve the ODE equations: S=e−kf* ( kot22−S0t ) C2+πkf2ko*ekf*S022ko*erf ( 2kfko*kot−S0 ) O=kot−S0+e−kf* ( kot22−S0t ) C2+πkf2ko*ekf*S022ko*erf ( 2kfko*kot−S0 ) C2=1S0−πkf2ko*S02ekf2koerf ( −S0*kf2ko ) We used these equations to estimate the parameters using a Levenberg-Marquardt non-linear least squares algorithm . The PTM75 NIL strain was generated from the CX12361 RIL strain by backcrossing the chromosomal region of interest into a balancer strain containing a CX12311 background along with the oxTi710 fluorescent miniMos insertion near the QTL of interest . Males heterozygous for the fluorescent marker ( as determined by fluorescence intensity ) were crossed to hermaphrodites of the balancer strain for ten generations . On the 11th generation , animals without the fluorescent marker were isolated . Genotyping of one to three markers within the candidate regions was used to confirm the successful introgression of LSJ2 DNA into CX12311 . | Complex traits are influenced by the individual effects of genetic variants in addition to the interactions of the variants with the environment , age , and each other . While complex genetic architectures are ubiquitous in natural traits , little is known about the causal mechanisms that create their complex genetic architectures . Here we identify an example of age-dependent genetic architecture controlling the rate and timing of reproduction in the hermaphroditic nematode C . elegans . We use computational modeling to demonstrate how age-dependent genetic architecture can arise as a consequence of two factors: hormonal feedback on oocytes mediated by major sperm protein ( MSP ) released by sperm stored in the spermatheca and life history differences in sperm use caused by genetic variants . Our work also suggests how antagonistic pleiotropy can emerge from multicellular feedback systems . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"invertebrates",
"medicine",
"and",
"health",
"sciences",
"reproductive",
"system",
"gonads",
"quantitative",
"trait",
"loci",
"caenorhabditis",
"animals",
"germ",
"cells",
"epistasis",
"animal",
"models",
"oocytes",
"caenorhabditis",
"elegans",
"model",
"organisms",
"experimental",
"organism",
"systems",
"molecular",
"biology",
"techniques",
"sperm",
"research",
"and",
"analysis",
"methods",
"gene",
"mapping",
"animal",
"cells",
"molecular",
"biology",
"genetic",
"loci",
"cell",
"biology",
"ova",
"anatomy",
"heredity",
"genetics",
"nematoda",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"organisms",
"genital",
"anatomy"
] | 2017 | Modeling of a negative feedback mechanism explains antagonistic pleiotropy in reproduction in domesticated Caenorhabditis elegans strains |
Mechanism-based chemical kinetic models are increasingly being used to describe biological signaling . Such models serve to encapsulate current understanding of pathways and to enable insight into complex biological processes . One challenge in model development is that , with limited experimental data , multiple models can be consistent with known mechanisms and existing data . Here , we address the problem of model ambiguity by providing a method for designing dynamic stimuli that , in stimulus–response experiments , distinguish among parameterized models with different topologies , i . e . , reaction mechanisms , in which only some of the species can be measured . We develop the approach by presenting two formulations of a model-based controller that is used to design the dynamic stimulus . In both formulations , an input signal is designed for each candidate model and parameterization so as to drive the model outputs through a target trajectory . The quality of a model is then assessed by the ability of the corresponding controller , informed by that model , to drive the experimental system . We evaluated our method on models of antibody–ligand binding , mitogen-activated protein kinase ( MAPK ) phosphorylation and de-phosphorylation , and larger models of the epidermal growth factor receptor ( EGFR ) pathway . For each of these systems , the controller informed by the correct model is the most successful at designing a stimulus to produce the desired behavior . Using these stimuli we were able to distinguish between models with subtle mechanistic differences or where input and outputs were multiple reactions removed from the model differences . An advantage of this method of model discrimination is that it does not require novel reagents , or altered measurement techniques; the only change to the experiment is the time course of stimulation . Taken together , these results provide a strong basis for using designed input stimuli as a tool for the development of cell signaling models .
One goal of systems biology is to develop detailed models of complex biological systems that quantitatively capture known mechanisms and behaviors , and also make useful predictions . Such models serve as a basis for understanding , for the design of experiments , and for the development of clinical intervention . In support of this goal , there has been a strong push to build mechanistically correct kinetic models , often based on systems of ordinary differential equations ( ODEs ) , that are capable of recapitulating the dynamic behavior of a signaling network . These models hold the promise of connecting biological and medical research to a class of computational analysis and design tools that could revolutionize how we understand biological processes and develop clinical therapies [1 , 2] . One type of experiment for model validation involves stimulating a system with a step change in the input ( typically by adding a high concentration of ligand ) and then measuring the change of network readouts ( the concentrations or activities of various downstream species ) as a function of time . Candidate models are fit to the data and the best model is selected based on criteria such as the quality of the fit , the simplicity of the model , and other factors . While it is tempting to select a simple model consistent with the known biochemical mechanisms that fits all available data , future experimentation may prove this choice incorrect . Rather , it may be preferable to collect “all” models consistent with known mechanisms and data , and to design follow-on experiments capable of distinguishing among the model candidates . In support of this less-biased approach , here we develop an approach for designing these follow-on experiments using dynamic stimuli . While the step-response experiment is attractive for its ease of implementation , dynamic stimuli have the potential to uncover more subtle system dynamics and to improve model selection in the cases where step-response experiments are not sufficiently discriminating . One example that illustrates the use of a dynamic stimulus to distinguish between two models is the work by Smith-Gill and co-workers on the detailed mechanism of antibody–antigen binding [3] . Initial step-response experiments were compatible with either a one-step or two-step binding mechanism , in which the ligand and antibody first come together in a loose encounter complex before forming a fully bound complex . To resolve this ambiguity , the authors applied a series of rectangular pulses of ligand concentration to their system . The resulting binding curves produced by this dynamic stimulus were inconsistent with the one-step model but were consistent with a two-step model and suggested the existence of an encounter complex , even though such a complex could not be measured directly by the assay . These results show that time varying inputs have the potential to distinguish closely related models of biochemical systems . For the relatively simple antibody–antigen system , an appropriate dynamic input was deduced intuitively . However this sort of intuitive design is difficult , especially in the case of more complex cell signaling pathway models , which may be described by hundreds or thousands of differential equations . An automated approach that could design experiments to test these complex systems has the potential to expand the scope of model selection experiments . Previous work in designing dynamic stimuli for the purpose of model discrimination in systems biology has focused on choosing input trajectories that maximize the expected difference in the output trajectories of competing models [4–10] . In addition to model discrimination , a rich literature exists on experimental design in systems biology for the purpose of estimating model parameters [2 , 11–13] . These optimization approaches for model discrimination have been applied to small biological systems , but the nonlinearity of the models combined with the presence of many local minima has thus far limited their application [8] . There is a need to extend these methods to design experiments that may not be optimal but are capable of discriminating between large pathway models . Instead of trying to design an input signal that maximizes the predicted difference between two model readouts , we recast the problem as a control problem ( Figure 1 ) . We choose a target trajectory , and then challenge a model-based controller to drive the system to follow the target trajectory . The extent to which the controller based upon a given model is able to drive the physical system is a measure of the fitness of that model . We demonstrate our methodology by applying it to the epidermal growth factor receptor ( EGFR ) pathway . This pathway has been extensively studied and modeled [14–18] . EGFR and its family members ( Erb2 , Erb3 , and Erb4 ) are known to mediate cell–cell interactions in organogenesis and adult tissues [19] . Overexpression of EGFR family members is a marker of certain types of cancer , including head , neck , breast , bladder , and kidney [20] . Because of their clinical importance , the EGFRs themselves , as well as various downstream proteins , are targets of therapeutic intervention [21 , 22] . Despite clinical interest in the EGFR pathway and over 40 y of intense study , there is still much about the pathway that is not known . For example , in three recent studies [23–25] , a number of proteins that changed phosphorylation state in response to EGF stimulation were found that were not previously known to be part of the pathway; in addition , many of the known pathway proteins are not part of any computational model [26] . The ordinary differential equation model of Hornberg et al . is a widely used mechanistic model of EGFR signaling [16] . This model is a refinement of earlier models of the pathway [17 , 18 , 27] . It describes signal transduction initiated at the cell surface by EGF binding to EGFR , leading eventually to the dual phosphorylation of ERK as the most downstream outcome , which then participates in a negative feedback to the top of the pathway . The elementary molecular processes modeled include bimolecular association and dissociation , phosphorylation and de-phosphorylation , synthesis and degradation , as well as endocytosis and trafficking all described with mass-action kinetics . The model contains 103 chemical species , 148 reactions , 97 independent reaction rates , and 103 initial conditions . We applied our computational methods initially to a small portion of the EGFR model for development and demonstration purposes , and then to the full model . In both cases , we formulated a set of closely related models that exhibit similar step-response behavior . We built a controller capable of controlling each candidate model and asked the controller to drive the system output ( doubly phosphorylated ERK ) to a predetermined value . Finally , by applying these designed inputs based on the reference and perturbed models , we showed that it is possible to discriminate between the various model alternatives .
In this work , we consider mass-action kinetic models consisting of zeroth- , first- , and second-order reactions described by ordinary differential equations . In the equations below , k signifies a rate constant; A , B , and C represent species or concentrations of species , depending on the context; and ∅︀ is the empty set or nothing . Zeroth-order reaction: First-order reaction: Second-order reaction: Large systems of reactions of this form can be represented compactly using Equation 4 . The state vector x describes the chemical species concentrations that are free to evolve in time according to the kinetics of the system . The input vector u represents the chemical species concentrations controlled by the experimenter . Matrices A1 and B1 represent first-order reactions , matrices A2 and B2 represent second-order reactions , and k represents constitutive ( zeroth-order ) reactions . The symbol ⊗ denotes the Kronecker product ( also known as the matrix direct product ) [28] . For vectors , this operator generates a vector of all quadratic products . The output of the model y is a linear combination of the state variables represented by the matrix C . A controller was developed to solve for the input signal u ( t ) that best achieves a particular objective in the output . We formulate this objective as a cost function G ( u ) that measures the distance between the model output and the desired output . Here , G ( u ) is the sum of squares error between y ( u , t ) , the model output for a given input u ( t ) , and ydesign ( t ) , the target output the controller is trying to match . T is the length of time of the experiment . The control problem is then to find an input function u ( t ) that minimizes G ( u ) . Equation 6 depends on models of the form of Equation 4 , which are nonlinear and potentially high order . This prevents us from solving the minimization problem directly . To address this issue , we implement two different approximations . The first is based on controlling a model formed from successive linearizations of Equation 4 ( henceforth referred to as the tangent linear controller ) , and the second is based on a local search of the input space ( henceforth referred to as the dynamic optimization controller ) [29] . A first-order approximation to Equation 4 at time t was computed by taking the Taylor series expansion about the current value of the state and input vectors ( xt and ut ) . Equation 7 is a linear differential equation with state variable Δx and time varying forcing term Δu , which has both numerical and analytical solutions . However , this approximation would tend to diverge from the solution to Equation 4 with increasing Δt , the time beyond the linearization point t , and ( Δx , Δu ) , the distance from the linearization point ( xt , ut ) . To mitigate this problem the true system ( Equation 4 ) was propagated , and successive linearizations were applied to improve the controller performance . Effectively , the linearization point is allowed to slide along with the exact simulation . Operationally , each time step was solved in three stages . First , the current state of the nonlinear simulation was used to derive a linear approximation about the current time point . Second , the linear system was solved to get the best input Δu . The linear system was solved numerically by discretizing the input as a series of scaled and shifted boxcar functions [30] of width τ . Numerical integration with the MATLAB routine ode15s [31] was used to compute the system response to a unit boxcar input . The output of a linear time invariant system can be expressed as a linear combination of scaled and shifted impulse response functions . Thus , solving for the input was achieved by computing the weights to apply to the input pulses that gave the optimal output . This was solved as a linear system of equations with box constraints on the input to limit the maximum and minimum concentration using the MATLAB routine lsqlin . Third , the computed input signal was applied to the full nonlinear system for a short time step τ . The process was then repeated for the next time interval . Effectively , each step the algorithm solves for an input signal Δu that is piecewise constant . The width of the intervals τ as well as the number of intervals is a parameter of the optimization and should be chosen based on the accuracy of the linear system . In this controller formulation , rather than exactly solving the tangent linear system , we solved the full nonlinear problem iteratively using a gradient optimization method . Application of this method requires computation of the sensitivities of the least squares objective function ( Equation 6 ) with respect to the input parameterization p . An efficient way to compute this quantity is to first solve for the adjoint sensitivities λ [29] . For the dynamical system ( Equation 4 ) and the objective function ( Equation 6 ) , the adjoint equations are given by Equation 8 . Here , λ* indicates the conjugate transpose . We use piecewise linear input functions described by parameters pi , which are the input function value at Ti , u ( Ti ) ; u ( t ) is then linearly interpolated between the control points at Ti . For these piecewise linear input signals the ith component of the gradient is given by: The adjoint equations were solved in MATLAB using ode15s [31] and the optimization was implemented using fmincon configured to use Quasi-Newton [32] with BFGS [33 , 34] in the MATLAB Optimization Toolbox Version 3 . 1 . 1 . Thus far the input signals have been unconstrained , except by the choice of the discretization . However , in practice it may be desirable to restrict the space of input signals to those that could be feasibly achieved by a given experimental setup . For example , in many experimental setups it is easy to add material but difficult to take material away . Likewise , there may be a maximum and minimum concentration for the input signals , or a maximum rate of change for the input signal . We implemented these experimental constraints as linear inequality constraints of the form of Equation 10 . The matrix A and the vector b are passed as arguments to lsqlin in the case of the tangent linear controller , or to fmincon in the case of the dynamic optimization controller . An example of a linear constraint that might be applied is that the input increase monotonically . In this case , A and b are given by Equation 11 . We based our model of EGFR signaling on that of Hornberg et al . [16] , which itself is a refinement of earlier work [17 , 18 , 27] . The model contains 103 chemical species , and 148 elementary reactions; these reactions are of the type given by Equations 1 , 2 , and 3 and may be reversible . The model is parameterized by 97 distinct reaction rate values and 103 initial conditions . The details of this model are given in Dataset S3 . Here we also introduced a modified model of EGFR signaling , which contained six additional production/degradation reactions of the form of Equation 12 , where X is one of {GAP , GRB2 , SOS , RAS-GDP , SHC , or GRB2-SOS} . The degradation rate kdeg was set such that the steady-state value of the species was the same as the steady-state value in the unmodified model computed using Equation 13 . In addition to the protein synthesis and degradation reactions , a GAP-catalyzed turnover of RAS-GTP was implemented . The rate constants ( kon , koff , and kcat ) are 5 × 10−7 cell molecules−1 s−1 , 0 . 4 s−1 , and 0 . 023 s−1 , respectively . The rate constants kon and koff are taken from the analogous reaction where GAP is part of the receptor complex and the kcat was fit so that the half–life of RAS-GTP in the absence of EGF matched literature values [35] . Finally , a first-order turnover of internalized SOS was implemented with a rate constant of 10−7 s−1 based on the turnover rate of EGFR . This augmented model has the additional property that if the input is removed ( set to zero ) it will return to its initial condition . The mitogen activated protein kinase cascade is a signaling motif found repeated throughout biology [27] . In each step of the cascade a substrate is multiply phosphorylated by a kinase , which in turn is the input to the next layer in the cascade . The off signal , present in each layer , is a phosphatase that removes the phosphate groups . Despite knowing all of the species involved , the detailed mechanism of the enzymatic steps had been difficult to determine [27] . In particular , it was unclear if the kinase acted in two distinct enzymatic steps , whereby it released the substrate between phosphorylation steps ( distributive mechanism ) or if it performed both phosphorylation steps before releasing the substrate ( processive mechanism ) . A MAP kinase cascade consisting or RAF , MEK , and ERK is contained in the Hornberg EGFR pathway model . We extracted a tier of this cascade consisting of a single kinase , phosphatase , and substrate . The four reversible bimolecular reactions representing the phosphorylation of ERK by doubly phosphorylated MEK ( MEKpp ) and the de-phosphorylation by a phosphatase were used as the basis of a new model . The model contains a distributive dual phosphorylation step catalyzed by MEKpp and a distributive dual de-phosphorylation step catalyzed by a phosphatase . MEKpp is the system input; doubly phosphorylated ERK ( ERKpp ) is the output . In addition to this basic model , three alternative models were constructed that differed in their mechanism of phosphorylation and de-phosphorylation ( processive or distributive ) . The set of four models ( distributive-kinase/distributive-phosphatase , processive-kinase/distributive-phosphatase , distributive-kinase/processive-phosphatase , processive-kinase/processive-phosphatase ) represents all possible combinations of processive and distributive phosphorylation and de-phosphorylation mechanisms . The alternative models , which contain some rate parameters not included in the distributive-kinase/distributive-phosphatase base model , were parameterized by fitting the parameters to the step response of the double distributive model , which included both a step-up and a step-down experiment . The details of these four models are given in Dataset S2 .
The dynamic optimization controller was applied to design input stimuli for each of the two alternative antibody binding reactions studied by Smith-Gill and co-workers [3] . For both the one-step and the two-step model ( Dataset S1 ) , the objective applied was to produce a constant output of antibody–ligand complex from time zero onwards . In the experiment performed by Smith-Gill and co-workers the measurement was a change in mass due to ligand binding as measured by surface plasmon resonance . While the fully bound complex is more stable than the postulated encounter complex , both have the same mass and would produce the same output signal . Therefore , in the case of the two-step model , the output is the sum of the encounter and fully bound complexes , whereas in the one-step model it is simply the fully bound complex . The basis set for the input was a 50-point piecewise-linear function with linear spacing . In the two-step model points were distributed evenly over the entire interval . In the one-step model points were placed evenly from 500 s to 600 s to accommodate the sharp transition . The results are shown in Figure 2 . Both controllers designed an input signal that starts at high concentration to form complex quickly and then drops to a lower concentration to keep the complex from overshooting the desired value . However , the controller for the one-step model drops abruptly while the controller for the two-step model drops more gradually . The desired outputs were not recovered when the stimuli from the wrong models were applied . When the one-step input was applied to the two-step system , the output produced an undershoot followed by an overshoot . When the input designed for the two-step model was applied to the one-step system , the complex concentration also produced an overshoot , but one that persisted . In both cases , accounting for the presence or absence of the encounter complex was critical for controlling the output correctly . It is interesting to note that this method allows for the selection of both the more complex model ( if it is correct ) as well as the simpler model . This is not possible using standard a posteriori metrics , such as least squares , which will always favor the more complex model . While there are methods that try to correct for this bias [36] , properly accounting for model complexity in large nonlinear systems remains an open problem [37] . Comparing our results to the Smith-Gill pulse method ( Figure 2B ) , it is clear that both computational experiments permit the two models to be distinguished in favor of the two-step method . However , for larger and more complex cases , it is unclear whether intuitive approaches or square pulse inputs will be sufficient to design distinguishing experiments . Another feature of the simulations is that the designed pulse produces a level output that does not require fine time resolution to accurately measure . This can be a significant advantage for more complex experimental systems , such as cell signaling measurements , where limitations on experimental observations are even more severe , whether in terms of numbers of species , time points or other factors . Mitogen-activated protein kinase cascades have been extensively studied experimentally and modeled computationally . While many variants exist , the canonical pathway consists of three layers of kinases and phosphatases . For each layer , the kinase activates the downstream kinase by dual-phosphorylation and the phosphatase deactivates the downstream kinase by removing the phosphate groups . Knowing the general structure of this pathway , it was still difficult to determine the details of the enzymatic steps . In particular , it was unknown if the kinase acted in a processive mechanism ( adding both phosphate groups in a single step ) , or if it acted in a distributive mechanism ( adding the phosphates in two distinct enzymatic steps ) . The difficulty arose from the fact that , without measuring all of the phosphorylation forms , both mechanisms could fit the step response data . The issue was eventually resolved by devising an experiment that could separate all of the phosphorylation forms [27] . Here we show that , in principle , the mechanisms could have been distinguished using our method , without adding additional measurements . To address this problem we generated four candidate models of a MAPK dual phosphorylation reaction . All four models contained forward phosphorylation and reverse de-phosphorylation steps , but differed in the detailed mechanisms . For both the forward and the reverse reactions we considered a processive ( one-step ) and a distributive ( two-step ) mechanism ( Figure 3A ) . Taking all combinations of distributive and processive reactions produced four models . For each model the free kinase concentration was the input variable and the concentration of doubly phosphorylated substrate was the output . For each of the four models , a stimulus was developed using the tangent linear controller . The objective was to drive the output to a fixed value that remained constant with time . Each of the four designed signals was used to stimulate each of the four models , and the resulting 16 experiments are shown in Figure 3C . Along the diagonal , one can see that the input signal derived from the correct model was able to effectively control the system . However , looking at each off-diagonal entry shows that inputs from each wrong model did a poor job controlling each system . In any real experiment , there is only one true system , which corresponds to performing the experiments from a single row of the figure . As with the antibody models , the algorithm was able to find a set of signals that distinguished amongst multiple models . It is worth noting that these solutions were generated automatically from the candidate models and did not require explicit user supervision . A popular ordinary differential equation model of the EGFR pathway is that of Hornberg and co-workers [16] . This model consists of 103 differential equations and includes ligand binding , receptor dimerization and activation , adaptor protein binding , trafficking of the receptor complex , and activation of the MAPK cascade terminating with ERK dual phosphorylation ( Figure 4 ) . This model was built as a set of successive refinements of earlier models [17 , 18 , 27] , with each refinement adding a new level of detail to the model . In its most recent formulation an additional negative feedback loop was added whereby activated ERK phosphorylates SOS and deactivates it . This model has been shown to agree with time course data collected in cell based assays as well as literature values for parameters measured in vitro [18] . We compare the original Hornberg model to a version with additional changes . We continue this model evolution by modifying the Hornberg model so that , when the input ( EGF ) is removed , the model returns to its initial conditions . This reset behavior is observed experimentally . Cells cultured in media containing EGF but switched to serum and EGF free media 12 h before stimulation , are able to respond to a dose of EGF added to the media [23] . This indicates that after EGF has been removed , the pathway returns to an EGF responsive state . In the Hornberg model , the dominant mechanism for desensitization and adaptation of the pathway to EGF is endocytosis and degradation of the receptor complex . Opposing this process are constitutive production and degradation reactions for the receptor , which allow the receptor level to return back to steady state after stimulation . This same process degrades other proteins in the receptor complex GAP , GRB2 , SOS , and RAS , but the current model does not contain synthesis terms for these proteins . As a result , prolonged stimulation depletes these proteins and prevents the activation of RAF , MEK , and ERK . We added production and degradation reactions analogous to the reactions for the receptor for all of the proteins in the receptor complex . Rate constants were chosen such that the steady-state levels in the absence of stimulation were the same as the initial conditions for the model and the exponential time constant for the approach to steady state was the same as for EGFR . The second modification to the model was in the RAS-GDP/RAS-GTP cycle . In the Hornberg model , activated receptor is needed to catalyze the recycling of RAS-GTP* ( a molecule of RAS-GTP that has already activated a molecule of RAF ) that is waiting to be recycled to RAS-GDP . If EGF is removed , RAS can be trapped in the RAS-GTP* form , preventing the system from returning to steady state . We addressed this by adding an additional enzymatic step to recycle RAS-GTP* back to RAS-GDP catalyzed by GAP and parameterized using literature rate constants [35] . With the addition of these new reactions , the modified model returns to its initial conditions after stimulation . For the remaining model parameters ( the parameters shared with the original model ) we fit the modified model to the original using data from a simulated step-response experiment ( Figure 5A ) constraining them to be within 10% of their original value . Despite the introduction of these new mechanisms and the tight constraints on the parameters , the step responses of the six molecular species modeling those presented in the original paper [18] ( Figure 5B ) are very similar in the original model ( blue curves ) and the modified model ( red curves ) . The largest difference is in the SHC* time course , which has a very similar shape and varies by at most 11% . While significant , this difference would be very difficult to detect in a standard biological experiment . As such , the modified model is a reasonable alternative to the original model , and it would be hard to reject either mechanism using the step-response data alone . From this starting point we used our methodology to design an experiment that could distinguish between the current model and the modified model of the EGFR pathway . For each model we tasked the dynamic optimization controller with driving the concentration of doubly phosphorylated ERK to a constant level of 104 molecules per cell . The input basis set was 25 points linearly spaced over the interval . To model the experimental condition where it is easy to add EGF to the dish of cells but difficult to remove , we implemented a monotonicity constraint . Figure 5B shows the inputs designed for each of the two models applied to each system with the resulting ERKpp time courses . Due to the negative feedback loops , both models required a steadily increasing concentration of EGF to maintain a constant level of ERKpp . However , the original model was much more difficult to control; as time progressed increasingly high doses of EGFR were required to maintain a constant output . The modified model required a much gentler increase in EGF concentration to maintain its level and was able to keep the concentration of ERKpp high to the end of the time period . Trial calculations showed that this result was robust to order of magnitude changes in the new rate parameters introduced in the modified model . Applying these two signals in an experiment could be used to distinguish between these two models , as demonstrated by the simulations .
The most common stimulus-response protocol involves applying a step change in one or more input concentrations and following the evolution of one or more downstream molecules . For a linear system , this type of experiment can provide enough information to fully identify the system [38] . However , even simple biochemical systems are nonlinear , and as such there is no a priori reason to believe that a step-response experiment will be sufficient to uncover the relevant dynamics of the system and allow for the selection of a unique model . As a result , it is often possible , if not probable , that multiple mechanisms fit the same set of step-response data . We have shown here that using dynamic stimulation can improve stimulus-response experiments . Even in the context of complex pathways with limited numbers of inputs and outputs , experiments can be designed that are capable of distinguishing amongst alternative mechanisms . Moreover , for the EGFR pathway studied , the differences detected were in the middle of the pathway , far from the location of the stimulus or the readouts . One possible explanation for the results presented here is that we have stimulated the systems with high-frequency signals , and it is this fact that allows for model discrimination . While the high-frequency content almost certainly plays a part , the fact that differences between models are observed at low frequency distinguishes our results from other standard test signals . For example , in linear systems it is common to use random or pseudorandom signals to discriminate among models . Figure S1 shows such an experiment . While the signal is discriminating , the observed differences are high frequency and would be difficult to distinguish in a standard biological assay , which is usually sampled sparsely in time . Formulating experimental design as a control problem yielded a relatively straightforward numerical solution , which allowed us to apply our method to large pathway models . While the method does not yield optimal experiments , in the sense of maximizing the least squares error between model ouputs , the results are still of practical benefit and appear sufficient to distinguish amongst model candidates . In the systems studied here , the designed inputs were able to substantially increase the differences observed between competing models when compared to the corresponding step-response experiment . By prescribing the target output trajectory , it should be possible to tailor the experiments to the available measurement methods , thereby achieving the most benefit from existing assays . It is worth noting that in all of the examples presented here , the target function was a constant output concentration . This was chosen for simplicity rather than for any special property of these targets . The problem of the best target function is an interesting one but is beyond the scope of this work . However , in Figure S2 we show calculations for the antibody–ligand system using other simple target functions , lines of constant slope , and find that designed inputs based on these signals have similar discriminating power . In each of the cases presented here , the dynamic stimuli allowed us to select the correct mechanism from a set of plausible candidates . However , it is possible that for a particular system and set of constraints , the algorithms presented here may fail to find a signal that is sufficiently discriminating . In this case a different choice of target function or a more sophisticated optimization approach may yield better results . However , it is worth noting that in the systems studied here both methods were able to find very good solutions in all cases . In general , the tangent linear controller was more computationally efficient and yielded smoother signals , whereas the dynamic optimization controller was slower but did not require tuning of parameters such as τ . One potential limitation of our method comes from our reliance on parameterized models . The accuracy of the parameterizations will affect the quality of the predictions made by the controllers and thus the ability to distinguish between models . To demonstrate this , we generated 100 different parameterizations of the one-step and two-step antibody models and then applied the control signals designed using the nominal parameter set ( Figure S3 ) . The parameter variation resulted in output trajectories that were quantitatively different from the predicted output trajectories . However , the overall shape of the output trajectories was preserved . All of the results presented here were in simulation . In practice , experimental error and measurement noise will make it more difficult to distinguish between models . As a result , one may only be able to effectively discard some candidate models , and reduce the pool of hypotheses . However , these experimental challenges also motivate our method , as it has the potential to increase the experimental observability of model differences when compared to a more traditional experiment , such as a step response . Moreover , the fact that potential mechanisms can be evaluated without having to resort to additional inputs or outputs is especially valuable in laboratory experiments , where adding additional inputs or outputs may require significant effort , such as developing new experimental reagents . | A major focus of systems biology is the development of mechanism-based models of cell signaling pathways . These models hold the promise of encapsulating our understanding of complex biological processes while also predicting new behavior . However , as these models become more complex , it can be difficult to distinguish between model alternatives . One means of improved model discrimination involves making measurements of additional components in the biological system to provide more detailed data . Here we present an alternative , which is to apply a time-varying input while monitoring the same network components . This new method was able to discriminate among models with subtle mechanistic differences . A particular advantage is that for many cases , time-varying input stimulation is fairly easy to apply experimentally , whereas measuring additional network components can involve the creation of new reagents or measurement assays . Thus , we believe that the application of time-varying input stimulation will become a powerful tool in the field of systems biology as the community places increased emphasis on the development of quantitative , mechanistic , and predictive models of biological network behavior . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"biochemistry",
"mammals",
"computational",
"biology",
"cell",
"biology"
] | 2008 | Stimulus Design for Model Selection and Validation in Cell Signaling |
An early and yet indispensable step in the alphaherpesvirus infection is the engagement of host receptors by the viral envelope glycoprotein D ( gD ) . Of the thus-far identified gD receptors , nectin-1 is likely the most effective in terms of its wide usage by multiple alphaherpesviruses for cell entry . The molecular basis of nectin-1 recognition by the gD protein is therefore an interesting scientific question in the alphaherpesvirus field . Previous studies focused on the herpes simplex virus ( HSV ) of the Simplexvirus genus , for which both the free gD structure and the gD/nectin-1 complex structure were reported at high resolutions . The structural and functional features of other alphaherpesviral gDs , however , remain poorly characterized . In the current study , we systematically studied the characteristics of nectin-1 binding by the gD of a Varicellovirus genus member , the pseudorabies virus ( PRV ) . We first showed that PRV infects host cells via both human and swine nectin-1 , and that its gD exhibits similar binding affinities for nectin-1 of the two species . Furthermore , we demonstrated that removal of the PRV gD membrane-proximal residues could significantly increase its affinity for the receptor binding . The structures of PRV gD in the free and the nectin-1-bound states were then solved , revealing a similar overall 3D fold as well as a homologous nectin-1 binding mode to its HSV counterpart . However , several unique features were observed at the binding interface of PRV gD , enabling the viral ligand to utilize different gD residues ( from those of HSV ) for nectin-1 engagement . These observed binding characteristics were further verified by the mutagenesis study using the key-residue mutants of nectin-1 . The structural and functional data obtained in this study , therefore , provide the basis of receptor recognition by PRV gD .
There are three major subfamilies , Alpha- , Beta- and Gamma-herpesvirinae , in the Herpesviridae family [1] . The three subfamilies differ in their host range capacities . In contrast to beta and gamma herpesviruses which exhibit restricted or limited cell-type tropism , the alphaherpesviruses have a much broader host range and are able to infect a wide variety of cell types [2 , 3] . For example , the representative alphaherpesvirus , herpes simplex virus ( HSV ) , shows low species specificity and is able to infect human and non-human cells alike [4] . The capability of HSV to infect most human cell types is recognized as an important contributing factor to its high prevalence in the world populations [5] . Pseudorabies virus ( PRV ) , another member of the Alphaherpesvirinae subfamily , is reported to infect both farming ( e . g . pigs , sheep , etc ) and pet ( e . g . cats ) animals [6] . Herpes B virus , an alphaherpesvirus that causes mild or asymptomatic infections in macaques , can cross the species barriers and lead to fatal diseases in humans [7] . Alphaherpesviruses contain multiple surface glycoproteins ( e . g . more than 11 in HSV ) in the virion envelope [8] . An early step in the infection of the alphaherpesvirus is the engagement of host receptors by the virus glycoprotein D ( gD ) [5] . The broad tropism of alphaherpesvirus is , therefore , at least partially the result of its capacity to recognize and bind multiple cell surface molecules via gD [8] . Thus far , six gD receptors have been identified . These include 3-O-sulfonated-heparan sulfate ( 3-O-S-HS ) [9]; the herpes virus entry mediator A ( HveA , also known as HVEM ) , a TNF receptor-related protein [10]; and three immunoglobulin superfamily members: HveB ( PRR2 , nectin-2 ) [11] , HveC ( PRR1 , nectin-1 ) and HveD ( PRV , CD155 ) [12] . Of these renowned host molecules , nectin-1 serves as a broadly used receptor mediating the entry of all of the commonly tested viruses , including HSV type 1 ( HSV-1 ) and type 2 ( HSV-2 ) , PRV , and bovine herpes virus type 1 ( BHV-1 ) [12–14] . In addition , nectin-1 was also identified as the primary receptor for the HSV-1 infection of rat and mouse sensory neurons [15] . Nectin-1 is an immunoglobulin ( Ig ) -like cell adhesion molecule [16] . It plays important roles in organizing the intercellular junctions by self homodimerization and/or by heterodimerization with other nectin and nectin-like molecules ( e . g . nectin-3 , nectin-4 , nectin-like 1 , etc . ) [17] . Nectin-1 is highly conserved in mammalian animals . The swine and human nectin-1 share 96% amino acid identity , and both homologs are able to mediate the entry of HSV-1 , HSV-2 , PRV and BHV-1 [13 , 18] . It is demonstrated that the gD proteins of these alphaherpesviruses bind to nectin-1 with nanomolar affinities [13 , 19] . The detailed binding mode between HSV gD and human nectin-1 has been successfully illustrated in several previous studies [20–22] . The gD molecule is composed of a V-set Ig-like ( or IgV-like ) core and long N- and C-terminal extensions , It utilizes both terminal-extension elements to interact with nectin-1 . The receptor , with three Ig-like domains arranged into a rod-shaped structure , projects its membrane-distal IgV domain for gD engagement[21–23] . It is notable that the whole gD footprint in nectin-1 overlaps extensively with the nectin-1 dimerization interface , which explains a novel mechanism of exploiting the host cell-adhesion functions by HSV for virus spread and infection [21 , 22] . In addition to tethering virus particles onto cell surface and facilitating the viral spread , gD is also believed to play a key role in triggering the membrane fusion cascade , thereby leading to the entry of alphaherpesviruses [24–26] . Several previous studies demonstrated that gD binding to its receptors could displace the C-terminal pro-fusion domain ( PFD ) to activate the fusion executor composed of glycoproteins B , H and L ( or gB , gH and gL ) [24 , 25] . Interestingly , the structures of HSV gD in the receptor bound forms ( with nectin-1 and HVEM , respectively ) and a dimeric free gD structure indeed reveal that receptor engagement would displace the gD C-terminal loop [20–22 , 25 , 27] . These structural investigations pave the way for understanding the basis of gD recognition of multiple cellular receptors and thereby of the broad cell tropism of alphaherpesviruses . Nevertheless , the previous studies are almost exclusively based on HSV gDs . The structural and functional features of other alphaherpesviral gDs remain poorly characterized . These are interesting scientific issues awaiting answers in the herpesvirus field , because gD homologs of the prevalent alphaherpesviruses ( e . g . HSV , PRV , BHV-1 , etc . ) only share very low ( 22–33% ) amino acid sequence identities [19] . To further delineate the receptor recognition basis of alphaherpesviruses , we set out to investigate the detailed interactions between PRV gD and nectin-1 . Despite the significant homology between PRV and HSV , the two viruses belong to different genera ( PRV in the Varicellovirus genus and HSV in the Simplexvirus genus , respectively [1] ) . In the current study , we show that PRV infects host cells via both human and swine nectin-1 , but not with human HVEM . PRV gD engages the human and swine receptors with similar affinities . We further demonstrate that a gD variant lacking the C-terminal membrane-proximal residues exhibited much higher affinity for nectin-1 , which we believe represents an inspiring evidence that the PRV gD C-terminal loop likely also plays a role in triggering the virus/cell membrane fusion as observed for HSV PFD . The structures of PRV gD in the unbound and the nectin-1-bound forms were then solved at high resolutions . The PRV protein exhibits overall similar 3D fold and nectin-1 binding mode to its HSV counterpart but utilizes different gD residues ( from those of HSV ) for nectin-1 engagement . Finally , we also conducted a mutagenesis study using the interface-residue mutants of nectin-1 , which in combination with the structural observations , provide the basis of receptor recognition by PRV gD .
Of the thus-far identified gD receptors , nectin-1 has been shown to mediate the cell entry of multiple alphaherpesviruses [5 , 8 , 12] . Several previous studies demonstrated that both human ( HU-nectin-1 ) and swine nectin-1 ( SW-nectin-1 ) serve as a cellular receptor for PRV [13 , 28] . As a step towards understanding the basis of receptor recognition by the virus , we first reconfirmed the nectin-1 mediated viral entry using the PRV vaccine strain Barth K61 ( 28 ) in CHO-K1 cells . CHO-K1 cell lacks any of the known alphaherpesvirus receptors and is therefore resistant to PRV infection [29] . Transient expression of HU- and SW-nectin-1 in CHO-K1 , however , suffices the cells for PRV entry . In the presence of either the human or the swine receptors , PRV infection with significant cytopathic effects was observed; whereas HU-HVEM failed to promote the entry of the virus ( Fig 1A ) . This is consistent with previous reports showing that nectin-1 , but not HVEM or 3-O-S-HS , could mediate the PRV infection [8] . To quantitatively compare the cell entry mediated by HU- and SW-nectin-1 , we further set up a cell based fusion assay as previously applied in the HSV studies [22] . With CHO-K1 cells , remarkable cell fusion could be observed when the nectin-1-expressing cells are mixed with cells expressing HSV gD along with the viral fusion executor of gB , gH and gL . By replacing the HSV glycoproteins with the PRV homologs , significant increase in the luciferase activity was recorded , which is highly indicative of fusion of the cells . On the whole , the cell fusion mediated by HU- and SW-nectin-1 is quantitatively equivalent ( Fig 1B ) , indicating similar capacities of gD engagement by the two receptors . To gain further insight into the PRV-gD/nectin-1 interaction , we set out to characterize their binding features using the real-time surface plasmon resonance ( SRP ) assays . Previous studies have shown that , by truncation of the C-terminal trans-membrane and cytoplasmic domains , the entire ectodomain of PRV gD could be yielded as a recombinant protein in the soluble form [13] . Following the same strategy , we successfully expressed and purified a truncated PRV gD protein spanning residues 1–337 ( hereafter referred to as gD337 ) using a baculovirus expression system ( Fig 2A and 2B ) . Noted that the residue numberings for HSV gD in previous structural studies are based on the mature protein [21 , 22] , the numberings for PRV gD amino acids in the current study are therefore also based on the mature protein ( unless otherwise specified ) to facilitate sequence comparison ( Fig 2A ) . The ecto-domain proteins of both HU- and SW-nectin-1 prepared from E . coli were individually immobilized on the chip and tested for the interaction with gD337 . As expected , the gD protein potently interacts with both receptors , exhibiting typical slow-on/slow-off binding kinetics ( Fig 2C ) . The equilibrium dissociation constants ( Kd ) were determined to be 191 nM with HU-nectin-1 ( Kon , 1 . 41 × 105 M-1s-1; Koff , 2 . 7 × 10−2 s-1 ) and 301 nM with SW-nectin-1 ( Kon , 1 . 46 × 105 M-1s-1; Koff , 4 . 4 × 10−2 s-1 ) , respectively . Therefore , PRV gD recognizes the human and swine receptors with essentially the same binding affinity . This result is consistent with our observation in the cell fusion assay which shows that nectin-1 of the two species mediates gD-dependent CHO-K1 fusion with similar efficiencies shown in the previous section . Furthermore , the determined Kd values are also in good accordance with a previous study reporting an affinity of approximate 130 nM for the PRV-gD/HU-nectin-1 interactions [13] . Inspired by the studies on HSV gD which revealed an approximate 2-digit fold affinity difference in receptor binding between a long and a short ( lacking about 20 residues at the membrane-proximal region ) gD form [20 , 25] , we further constructed a shorter PRV gD variant without the membrane-proximal loop . This new construct spans the gD amino acids 1–284 ( hereafter referred to as gD284 ) ( Fig 2A ) . The resultant protein was similarly prepared via the baculovirus expression system and purified to homogeneity ( Fig 2B ) . When tested using SPR , this short form of PRV gD exhibited significantly increased binding avidity to nectin-1 . The calculated Kds of gD284 to the human and swine receptors were 16 . 1 nM ( Kon , 7 . 25 × 105 M-1s-1; Koff , 1 . 17 × 10−2 s-1 ) and 18 . 4 nM ( Kon , 1 . 15 × 106 M-1s-1; Koff , 2 . 11 × 10−2 s-1 ) , respectively ( Fig 2C ) . These values represent an approximate 12–16 fold enhancement in affinity in comparison to gD337 . It is notable that gD284 , with the enhanced receptor binding capacity , exhibits a similar affinity for nectin-1 to HSV-1 gD285—a representative short form of HSV gD [21 , 22] . According to the previous studies , the observed affinity-difference between the long and short forms of HSV gD stem mainly from the changes in their kinetic association rates ( up to 40-fold in Kon ) [25] . The enhanced affinity for nectin-1 by PRV gD284 ( relative to gD337 ) , however , arises from both the increase of the association rate ( by approximate 5–8 fold in Kon ) and the decrease of its dissociation rate ( by approximate 2-fold in Koff ) . To probe into the structural features of PRV gD , both gD337 and gD284 were subjected to intensified crystallization screenings . We finally managed to collect a 1 . 5-Å resolution data set from the gD337 crystals . The solved structure was refined to Rwork = 0 . 169 and Rfree = 0 . 192 ( Table 1 ) , and contains 244 amino acids spanning from P7 to P250 . The terminal residues ( A1 to V6 in the N-terminus and R251 to S337 in the C-terminus ) , however , were untraceable . We believe these two parts , especially the large C-terminal region , are flexible and most likely disordered loops , which might have undergone unexpected proteolytic digestions during crystallization . The overall PRV gD structure is expectedly composed of an IgV-like core and the long N- and C-terminal extensions ( Fig 3A ) , as observed in its HSV homologs [20–22] . The core-domain , which spans from D38 to V164 , contains a nine-stranded ( A' , B , C , C' , C'' , D , E , F , and G ) β-barrel in the center . These nine strands are topologically arranged in a typical IgV manner . In contrast to the canonical IgV fold , however , the barrel C'' strand is kinked in the middle , leading to two strand halves connected with a distorted loop . Similar kinked C'' strand has also been observed in both HSV-1 and HSV-2 gD structures , which likely represents a unique feature of the alphaherpesviral gDs . In addition to the compact central barrel , the core-domain also contains two α-helices ( α1 and α1' ) , locating in between the BC and the C''D strands , respectively ( Fig 3A ) . The IgV-like core of PRV gD is further wrapped by the N- and C-terminal extensions . The former , extending from P7 to A37 , is largely a loop structure . It also encompasses a small β-strand ( str2 ) which is aligned in parallel with the first half of strand C'' . The latter extension , with residues A165-P250 , is much more extended . It structurally folds into four α-helices ( α2' , α2 , α3' and α3 ) and one small β-strand ( str3 ) , covering about half of the central-core outer surface ( Fig 3A ) . In total , three disulfide bonds were formed in the PRV gD structure . One ( C100 with C109 ) is observed in the IgV-like core , connecting the C' and C'' strands; while the other two ( C49 with C172 and C88 with C188 , respectively ) are located between the central core and the C-terminal extension , tying the extension loops to the core components . These disulfide bridges are also conserved in HSV gDs ( Fig 3C ) . As expected , the structure of PRV gD is quite similar to those of its HSV homologs . Superimposition of our structure with previously reported HSV gD structures revealed well-aligned core domain and terminal extensions ( Fig 3B ) . Despite of the low sequence identity , a majority of the secondary structure elements , including the core-strands and most of the extension helices , were parallel-formed in PRV and HSV gDs . The PRV structure , however , contains an extra α2' helix in the C-terminal extension . In addition , while both PRV and HSV gDs encompass a core α1' helix , their steric positions are quite different in the structures . In PRV gD , helix α1' is located between strands C'' and D; whereas the HSV-gD α1' helix lines before strand C , directly following the α1 helix ( Fig 3B and 3C ) . Further structural variations between PRV and HSV gDs were observed for the equivalent α3 helices and their terminal loops . The former showed a small variance in the conformation , while the latter exhibited large orientation differences ( Fig 3B ) . Facilitated by the structure-based sequence alignment , it is notable that HSV gD encompasses a much longer N-terminal loop ( or N-loop ) than PRV gD ( Fig 3C ) . Noted that HSV gD reconstitutes its N-loop into a hairpin structure for HVEM binding [27] , the lack of an N-loop of sufficient size in PRV gD therefore coincides well with its incompetence of utilizing HVEM . Inspired by the high affinity between gD284 and nectin-1 , we prepared the complex of this short form of PRV gD bound with the IgV-domain protein of the SW-nectin-1 , and obtained a complex crystal that diffracts to 2 . 7 Å resolution . The complex structure , with an Rwork of 0 . 238 and an Rfree of 0 . 266 ( Table 1 ) , contains gD residues P9-D241 and the nectin-1 amino acids D37-M143 . Although a shortened gD form was used for the complex crystallization , a large fraction of the C-terminal residues were still untraceable in the structure . It is notable that similar proportion of the gD residues were successfully traced in both the free gD and the gD/nectin-1 complex structures , despite that two different gD forms ( gD337 and gD284 , respectively ) were individually used in the crystallization experiments . As expected , the gD molecule in the receptor-bound state similarly folds as an IgV-like core wrapped by the extensive terminal extensions . It recognizes SW-nectin-1 mainly through the extension elements ( including str2 and its flanking loops in the N-terminal extension , and the α2' , α3' , α3 helices , and the α3'/α3 intervening loop in the C-terminal extension ) and the exposed α1' helix of the core domain ( Fig 4A and 4B ) . These gD components were exquisitely positioned over the binding interface , engaging exclusively the V-domain CC'C''FG sheet of the receptor ( Fig 4A and 4B ) . In comparison to the free PRV gD structure , its N-loop was clearly reoriented upon receptor binding , extending to the vicinity of helix α3 in the complex structure ( Fig 4B ) . Apart from this N-loop re-orientation , engagement of SW-nectin-1 does not induce other significant conformational changes in the viral ligand . Overall , the nectin-1 binding mode of PRV gD resembles its HSV counterparts . All the three viral ligands utilize the terminal extensions to contact the CC'C''FG sheet of the receptor IgV domain . A detailed superimposition between the current structure and the previously reported HSV-gD/nectin-1 complex structures , however , revealed an obvious difference in their steric position ( relative to the receptor ) for the bound gD proteins . With well-aligned nectin-1 molecules , a shift-distance of about 3 . 5 Å was observed between the PRV and HSV gDs ( Fig 4C ) . Furthermore , the receptor CC' loop also showed quite different conformations when bound to the viral ligands of PRV and HSV ( Fig 4C ) . Noted that PRV gD contains an extra α2' helix which is projected outwards for engagement of the receptor , the SW-nectin-1 CC' loop therefore adopted an alternative conformation to accommodate the bulged helix . Apart from the aforementioned differences , the PRV and HSV gDs exhibited quite similar overall-receptor-binding mode for engagement of nectin-1 ( Fig 4C ) . On the whole , extended surface areas of about 1118 . 1 Å2 in PRV gD and 1170 . 9 Å2 in SW-nectin-1 were buried upon complex formation . We therefore scrutinized this binding interface to delineate the amino acid interaction details between the two binding entities . Along the nectin-1 CC'C''FG sheet , the major receptor-engagement components of gD include its N-loop , the α1' helix in the IgV-core , and helices α2' , α3' , α3 and the α3'/α3 loop in its C-terminal extension ( Fig 5A ) . In the N-loop , two aromatic residues F11 and W22 were projected outwards for receptor binding . They were found to be packed against nectin-1 amino acids K61 , Q64 , I80 , N82 , M85-S88 , and F129 , thereby providing important hydrophobic interactions ( Fig 5B ) . It should be noted that F11 of gD is located in the N-loop region that was shown to undergo large conformational changes after receptor binding ( Fig 4B ) . We believe the subsequent interactions of F11 with nectin-1 K61 and F129 are the major forces stabilizing the observed N-loop orientation in the complex structure . As for the α1' and α2' helices , each component presents one amino acid ( T119 in α1' and Y183 in α2' ) towards the receptor , contacting mainly the nectin-1 residues G73-K75 , I123 , E125 , N133 , and E135 via multiple Van der Waals ( vdw ) interactions . In addition , the α1'-residue T119 also forms a weak H-bond with nectin-1 E125 ( Fig 5C ) . Further PRV-gD/SW-nectin-1 interactions were contributed by the α3' , α3 helices and their intervening loop , which position multiple residues , including M201-R202 , P206-Y208 , V216 , and Y219 , to interact with nectin-1 amino acids T63-Q64 , Q76-N77 , I80 , S88 , L90-A91 , A127-P130 and N133 ( Fig 5D ) . In addition to providing extensive vdw contacts , these amino acids further stick the viral ligand to the receptor by creating an inter-molecule H-bond network . In total , five strong and two weak H-bonds were observed to form , arising from PRV gD M201 and R202 interacting with SW-nectin-1 Q76 , N77 and A91 , gD P206 with nectin-1 Q64 , and gD Y208 with nectin-1 T63 and A127 , respectively ( Fig 5D ) . It is notable that the identified interface residues in the receptor are absolutely conserved between HU- and SW-nectin-1 ( Fig 5E ) , explaining the similar affinities of the two receptors for PRV gD binding . Owing to the low sequence identity , PRV and HSV gDs utilize different amino acids for nectin-1 engagement . Comparison of their footprints in nectin-1 , however , reveals essentially the same surface patch in the receptor . For both PRV and HSV gDs , the nectin-1 residues locating within a distance of 4 . 5 Å from the bound ligand were selected and compared in detail . The subsequent interface-residue list revealed 31 amino acids interacting with PRV gD , 22 with HSV-1 gD and 26 with HSV-2 gD , respectively . Of these residues , 20 were utilized for contacting all the three viral ligands , contributing more than 90% of the total inter-molecule contacts with either ligand ( Table 2 ) . Though the vdw contacts conferred by each of these amino acids were different between HSV and PRV gDs ( Table 2 ) , these shared 20 nectin-1 residues dominate the binding to gDs of both viruses . To further confirm the binding features observed in our complex structure , the key interface residues in nectin-1 , including N77 , M85 and F129 ( which contribute more than 20 hydrophobic vdw contacts each ( Table 2 ) and/or the side-chain H-bonds ( Fig 5D ) , were individually mutated and tested for their interactions with PRV gD337 . In the context that HU- and SW-nectin-1 are equally competent in binding with PRV gD ( Fig 2B ) and their interface residues contacting gD are absolutely conserved ( Fig 5E ) , we therefore utilized the human homolog for the test in our mutagenesis study . In consistent with the structural observation that F129 provides the maximum amount of intermolecule contacts ( 45 contacts , Table 2 ) , the F129A mutation almost completely abrogated the gD/nectin-1 binding ( Kd > 40 μM ) ( Fig 6A ) and therefore the gD/nectin-1 dependent cell fusion ( Fig 6B ) . The functional indispensability of nectin-1 F129 in PRV-gD engagement was also demonstrated in a previous study[30] . In contrast to F129 with 45 vdw contacts for PRV-gD , the intermolecule binding contributed by nectin-1 N77 ( 29 contacts ) and M85 ( 20 contacts ) were only about half of that for F129 ( Table 2 ) . Accordingly , substitution of the two residues with alanine decreased , but not abolished , the gD/nectin-1 interaction . The affinities of the nectin-1 N77A and M85A for gD337 were determined to be 3 . 4 μM and 2 . 8 μM , respectively ( Fig 6A ) , and the two mutants remain competent in the gD-mediated cell fusion , though with decreased efficiencies ( Fig 6B ) . It is notable that nectin-1 N77 and M85 interact more extensively with HSV-gD ( e . g . 42 and 33 contacts , respectively with HSV-2 gD ) than with PRV-gD ( 29 and 20 contacts , respectively ) ( Table 2 ) , echoing a previous study reporting their important roles in binding the HSV ligand but much compromised contributions to the engagement of the PRV protein [31] . These mutagenesis data coincide well with our structural observation , which conversely verified the PRV-gD/nectin-1 binding mode observed in the complex structure . It is also noteworthy that nectin-1 N77A and M85A engage PRV-gD with a fast-on/fast-off mode ( Fig 6A ) , forming contrast to the wild type protein which shows both slow association and dissociation rates . Similar phenomena of altered Kon/Koff rates have also been observed in other virus-ligand/receptor binding pairs when their interface residues are mutated [22] .
Of the thus-far identified alphaherpesviral gD receptors , nectin-1 is likely the most effective in terms of its wide usage by different viruses . HSV-1 , HSV-2 , PRV and BHV-1 are all reported to utilize nectin-1 for cell entry [8 , 12 , 13 , 30 , 31] . The molecular basis of nectin-1 recognition by the envelope gD proteins of these viruses , therefore , is an interesting but yet an unresolved issue . Previous studies focused on HSV , a member of the Simplexvirus genus in the Alphaherpesvirinae subfamily . The structures of both the free HSV gD and its complex with nectin-1 are reported at high resolutions [6 , 21 , 22 , 25] . Nevertheless , the structural and functional features of other alphaherpesviral gDs remains poorly understood . In this study , we have reported the first structure of gD derived from a Varicellovirus member of the alphaherpesviruses , the PRV . Despite of its low sequence identity to the HSV homologs ( ~ 22% ) , PRV gD reserves the canonical gD features , including an IgV-like core with a kinked C” strand and the surface-exposed N- and C-terminal extensions . We further solved the complex structure of PRV gD bound with SW-nectin-1 , which revealed a similar nectin-1 binding mode as observed for HSV gD . Nevertheless , several unique features at the PRV-gD/nectin-1 binding interface ( e . g . a bulged α2' helix of PRV gD interacts with an adjusted CC’ loop in nectin-1 , an about 3 . 5-Å shift for PRV gD ( relative to HSV gD ) when bound to the receptor ) suffice the PRV ligand to recognize nectin-1 using quite different gD residues from those of the HSV homologs . These structural observations therefore provide a systematic view on the receptor binding mechanism of a second alphaherpesvirus and yet the first in the Varicellovirus genus . It is interesting that both in the free PRV gD structure and in the PRV-gD/nectin-1 complex structure , a large C-terminal region of the gD molecule are untraceable . We noted that it has been a long time ( over four months ) before the proteins ( gD337 for the free structure and gD284 for the complex structure ) were successfully crystallized . Taking into account that large disordered parts of a protein are likely perturbing crystal formation , it may indicate a possible proteolysis of both the gD337 and the gD284 C-terminal regions during crystallization , leading to similar proportion of the gD residues ( P7-P250 in the free gD structure and P9-D241 in the complex structure ) being successfully traced in the free gD and the nectin-1 bound structures . We therefore believe that the re-orientation of the gD N-loop before and after nectin-1 binding is more likely the result of receptor engagement than arising from the differences in the gD forms used for crystallization . A flexible N-loop which experiences rearrangement or is structurally stabilized upon receptor binding has also been observed for HSV gDs [22 , 26] . With an overall similar structure , PRV and HSV gDs all engage the nectin-1 CC'C''FG sheet for receptor recognition . Along the sheet , a total of 20 residues ( out of 31 for PRV gD , 22 for HSV-1 gD and 26 for HSV-2 gD ) were recognized by all the three viral ligands ( summarized in Table 2 ) . It is notable that these amino acids contribute the majority of the inter-molecule interactions and therefore constitute a conserved central contact interface dominating gD-binding , whereas the remaining nectin-1 residues provided additional and supplementary contacts for engagement of gD . It is interesting that the PRV and HSV gDs , with only 22% sequence identity , select essentially the same contact interface in nectin-1 for recognition and engagement . We believe this interface patch likely has evolved as a major ligand-binding entity , dominating the nectin-1 interaction not only with PRV and HSV , but also with other alphaherpesviruses ( e . g . BHV-1 ) . It is also noteworthy that it is not rare phenomena that different viruses could recognize the same host surface molecule as the cellular receptor . It would therefore be interesting to investigate if other viruses similarly recognize extensively overlapped surface patches during receptor engagement , as observed for the PRV/HSV pair . To our knowledge , the Middle East respiratory syndrome coronavirus and the related bat-derived HKU4 coronavirus represent another example of such case . Both viruses recognize human CD26 via the viral spike protein , and they were shown to bind to the same propeller elements of the receptor [32 , 33] . The PRV gD footprint in nectin-1 also coincides with the receptor dimerization interface . According to the previous studies , formation of the homo- and/or hetero-dimers is the basis for nectin-1 to exert its cell adhesion functions [34] . PRV should therefore compete against the receptor dimerization by gD engagement , via which jeopardizing the host junction architecture . It should be noted that the affinity of PRV gD for HU-nectin-1 is determined to be 191 nM , which is about 100 fold higher than that calculated for nectin-1 self-interactions [23] . This should confer the binding with PRV gD an astonishing priority during the viral infection over nectin-1 dimerization , therefore exploiting its cell adhesion functions . In HSV , the membrane-proximal loop of the gD ectodomain is proposed as the PFD which would interact with gB/gH/gL to trigger membrane fusion [19] . According to a previously reported dimeric structure of HSV gD , the viral ligand prelocates its PFD in a position that will preclude the binding of gD with its receptors [25] . The gD/receptor engagement therefore would only occur when this prelocated loop is displaced , exposing the otherwise locked receptor binding site . Consistent with this structural observation , a short HSV gD protein lacking PFD is about 100-fold more competent than the long gD form in terms of receptor binding [20 , 25] . In the current study , we demonstrated that the affinity of PRV gD284 ( a short gD variant without the membrane-proximal loop ) for nectin-1 is also dramatically ( ~12–16 folds ) higher than that of gD337 ( a long gD form containing the whole ectodomain ) . We believe this phenomenon echoes what has been observed with HSV-gD PFD , indicating a role of the PRV-gD membrane-proximal loop in membrane fusion . It is also noteworthy that the enhanced receptor binding affinity for the short variant of HSV gD stems mainly from a significantly increased binding on-rate ( ~ 40 fold in Kon ) , falling in a good accordance with the prelocation of PFD in gD [25] . In contrast , the changes in Kon between PRV gD284 and gD337 is only about 5–8 folds . This argues against a prelocation of the membrane-proximal loop in PRV gD and may indicate a novel mechanism of the loop interfering with nectin-1 binding . These are interesting issues that are worth of studying in the future . In contrast to HSV , both previous studies and the functional data in this report demonstrated that PRV can not use HVEM as a cellular receptor [12 , 35] . It is notable that the HVEM binding site in HSV gD is allocated to an N-terminal hairpin which was re-constituted by the long protein N-loop upon HVEM engagement . An N-loop of sufficient size is therefore a prerequisite for the gD ligand to interact with HVEM [27] . Facilitated by the structure-based sequence alignment between PRV and HSV gDs , it is clear that the N-loop of the PRV protein is only about half size of that in HSV gD . We believe this dramatically shortened N-loop in the PRV ligand can not support HVEM-binding , representing the structural basis of its inertness in HVEM recognition . The PRV infection can cause pseudorabies ( also known as Aujeszky's disease ) in pigs and other animals [6] . Although the disease can be prophylactically controlled with the live attenuated virus vaccine , it remains a serious problem and potential threat to the swine industry of many countries , leading to heavy economic losses each year . With an indispensable role in the viral infection , PRV gD represents a good vaccine candidate [36–38] . The strategy reported in the current study of preparing recombinant PRV-gD proteins ( gD337 and gD284 ) with both structural integrity ( correctly folded as demonstrated with the free and the nectin-1-bound structures ) and functional competency ( able to engage the receptor as shown with the in vitro SPR data ) might facilitate the development of a gD-based subunit vaccine in the future .
CHO-K1 cells ( a Chinese hamster ovary cell line which is an already-existing collection in the laboratory ) were maintained in Dulbecco's minimal essential medium ( DMEM ) supplemented with 10% fatal bovine serum ( FBS ) , 100 U/ml penicillin , and 100 mg/ml streptomycin . For the PRV infection experiment , cells were seeded in the 6 well plate with 70% confluence . After 24h , cells were transfected with pcDNA4 . 0-HU-nectin-1 , pcDNA4 . 0-HU-HVEM or pcDNA4 . 0-SW-nectin-1 expression plasmids , separately . After 24 h , cells were incubated with PRV Bartha-K61 solution ( 100 TCID50/ml PRV in DMEM ) at 37°C for 3 h , after which 2% FBS was added to the medium . Cells were then photographed under microscope 72 h post infection . DNAs encoding the ectodomains of the human and swine nectin-1 ( amino acids 30–335 ) were obtained by PCR using the primer pairs of human-Nectin1-F ( 5'- CGCGGATCCGTCCCAGGTGGTCCAGGTGAAC -3' ) / human-Nectin1-R ( 5'-AAGGAAAAAAGCGGCCGCTTCTGTGATATTGACCTC-3' ) and porcine-Nectin1-F ( 5'-CGCGGATCCGACCCAGGTGGTCCAGGTGAACG-3' ) / porcine-Nectin1-R ( 5'-AAGGAAAAAAGCGGCCGCCTCTGTGATATTGACCTCCACC-3' ) , respectively . The amplified fragments were then individually cloned into the prokaryotic expression vector pET21b ( Invitrogen ) with the BamHI and NotI sites . The proteins were then prepared following the reported methods [21] . In brief , the proteins were expressed as inclusion bodies in E . coli BL21 ( DE3 ) and the inclusion bodies were prepared in a buffer composed of 100 mM Tris-HCl , pH 9 . 0 , 400 mM L-Arginine , 2 mM EDTA , 5 mM reduced glutathione and 1 mM oxidized glutathione . Then , the refolded proteins were subjected to gel-filtration on a Hiload 16/60 Superdex 200 column ( GE Healthcare ) . By estimation of their MWs , the monomer peaks could be obtained . The resultant monomeric proteins were carefully collected , concentrated , and then applied for kinetic studies using surface plasmon resonance . The DNAs coding for PRV ( Becker strain ) gD ectodomain of residues 1–284 ( gD284 ) and 1–337 ( gD337 ) were amplified by PCR with the same forward primer of PRV-gD-F ( 5'-CCGGAATTCAGGACGTGGACGCCGTGCC-3' ) pairing with different reverse primers of PRV-gD284-R ( 5'-CCGCTCGAGTTAGTGGTGATGGTGGTGGTGCCGCGTCGCCGGCTCGGGCAG-3' ) and PRV-gD337-R ( 5'-CCGCTCGAGTTAGTGGTGATGGTGGTGGTGGCGGTGGCGCGAGACGCC-3' ) , respectively . The coding sequence of the swine nectin-1 IgV domain ( amino acids 37–143 ) was amplified using primers pNectin-1-IgV-F ( 5'-CCGGAATTCAGGACTCCATGTATGGTTTCATCGGC-3' ) and pNectin-1-IgV-R ( 5'-CCGCTCGAGTTAGTGGTGATGGTGGTGGTGCATCACAGTGAGGTTGAGCT-3' ) . The resultant DNA fragments that contain a C-terminal 6×his coding sequence were then individually cloned , via EcoRI and XhoI sites , into a previously modified pFastBac1 vector , which has been engineered to incorporate a baculovirus gp67 signal sequence at the N-terminus [39–41] . The pFastBac1 construct for HSV-1 ( KOS strain ) gD ( residues 1–285 , gD285 ) , which is an already-existing collection in the lab , has been described previously ( 21 ) . All the proteins were expressed with the Bac-to-Bac baculovirus expression system ( Invitrogen ) . The recombinant baculovirus was used to infect Hi5 cells ( a Trichoplusia ni insect cell line which is an already-existing collection in the laboratory ) for the production of the soluble proteins . The proteins were purified from the Hi5 cell supernatants first by nickel affinity chromatography ( GE Healthcare ) and then by gel-filtration chromatography using a Superdex 200 column ( GE Healthcare ) . The mutant proteins of human nectin-1 ectodomain were prepared as previously described [21] . The binding kinetics between the soluble gD and nectin-1 was analysed at 25°C on a BIAcore 3000 machine with CM5 chips ( GE Healthcare ) . HBS-EP buffer ( 10 mM HEPES , pH7 . 4 , 150 mM NaCl , 3 mM EDTA , 0 . 005% Tween 20 ) was used for all measurements . For SPR measurements , both gD and nectin-1 proteins were purified by gel filtration using Superdex 200 column ( GE Healthcare ) . We used the blank channel as negative control . About 1 , 200 response units of nectin-1 were immobilized on the chip . When the data collection was finished in each cycle , the sensor surface was regenerated with 10 mM NaOH . A serial of concentrations up to 2 , 500 nM were designed for the experiment . The kinetic analysis was performed by SPR using the BIAcore 3000 system . Gradient concentrations of PRV gD284 and PRV gD337 were flowed at 30 μL/min over nectin-1 ( WT or mutant ) immobilized at about 1 , 200 response units , and tested for binding at 25°C . The running buffer is composed of 10 mM HEPES , pH7 . 4 , 150 mM NaCl , 3 mM EDTA , 0 . 005% Tween 20 . All the crystals were obtained , with the hang-drop vapor-diffusion method , by initial screening with the commercial Hampton Research kits and then by condition optimizations . Free PRV gD crystal was finally obtained by mixing 1 μl of the concentrated gD337 protein at 10 mg/ml with 1 μL reservoir solution consisting of 0 . 1 M ammonium acetate , 0 . 1 M bis-tris pH 5 . 5 , 17% w/v polyethylene glycol 10 , 000 . To obtain the crystal of the PRV gD/SW-nectin-1 IgV complex , two proteins ( gD284 and SW-nectin-1 IgV ) were separately purified by gel-filtration chromatography , mixed at 1:1 molar ratio , and incubated on ice for 2 h . The mixture was further purified by gel-filtration chromatography using a Superdex 200 column ( GE Healthcare ) and the complex peak was carefully collected and then the PRV gD/SW-nectin-1 IgV complex was concentrated to 5 mg/mL . Diffractable crystals were finally obtained by mixing 1 μL of the protein complex with 1 μL reservoir solution consisting of 0 . 2 M sodium chloride , 0 . 1 M sodium acetate pH 5 . 5 , 20% w/v PEG 10 , 000 , followed by incubation at 4°C for about 4 months . The diffraction datasets were collected at beamline BL19U1 of the Shanghai Synchrotron Radiation Facility ( SSRF ) using the synchrotron radiation at 100K . The cryoprotectant solution is composed of 15% V/V glycerol and 85% V/V reservoir solution . Data were then processed with HKL2000 [42] for indexing , integration and scaling . Via molecular replacement using the PHASER [43] program in the CCP4 suite [44] , the free PRV gD337 and the PRV gD284/ SW-nectin-1 IgV complex structure were solved with HSV2 gD ( PDB code 4MYV ) and HSV-1 gD285/ human nectin-1 complex structure ( PDB code 3U82 ) as the search models . The coordinates and the related structural factors have been deposited into the Protein Data Bank with the PDB codes of 5X5V for the free PRV gD structure and 5X5W for the PRV-gD/nectin-1 complex structure . The fusion mediated by HSV gB/gD/gH/gL and the receptors has been validated in various cell types [24 , 29] . In this study , we set up a cell-based fusion system using CHO-K1 cells as previously reported [29] . In brief , the genes of PRV gB , gD , gH , and gL and nectin-1 were cloned into the pcDNA4 . 0-myc-his vector to yield the respective plasmids for protein expression in mammalian cells . The T7 polymerase and T7 luciferase expression plasmids were constructed previously in our lab . The expressing plasmids for PRV gB , gD , gH , gL and the T7 luciferase or the plasmids for nectin-1 and the T7 polymerase were separately co-transfected into CHO-K1 cells using Lipo2000 ( Invitrogen ) according to the manufacturer's instructions . After 24 h of transfection , the cells expressing gB/gD/gH/gL and nectin-1 were mixed and incubated for cell fusion . The luciferase activity was tested using a luciferase assay system kit ( Promega ) . | Both herpes simplex virus ( HSV ) and pseudorabies virus ( PRV ) recognize nectin-1 as the cellular receptor . They utilize the envelope glycoprotein D ( gD ) on the virion surface to interact with nectin-1 , initiating the virus infection . Although the molecular basis of nectin-1 binding by HSV gD has been successfully elucidated with high resolution structures , the atomic features of PRV gD interacting with the same receptor remains uncharacterized . Here , we show that PRV gD exhibits nano-molar affinities for both human and swine nectin-1 , and deletion of the membrane-proximal loop in the gD ectodomain dramatically increases its receptor-binding avidity . We further solved the free and the nectin-1-bound PRV gD structures . The free gD structure reveals a canonical fold of an IgV-like core wrapped by the N- and C-terminal extensions as observed in the HSV homologs . The N-terminus of PRV gD , however , is shorter than that of HSV gDs . The solved complex structure demonstrates that PRV gD exhibits a homologous receptor-binding mode to the HSV counterpart . Nevertheless , several unique features at the PRV gD binding interface suffice the viral ligand to engage nectin-1 with a series of residues differing from the HSV amino acids . These observations not only delineated the molecular basis of PRV engaging nectin-1 but also enriched our knowledge on the receptor-binding mechanism of the Alphaherpesvirinae subfamily . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"cell",
"physiology",
"herpes",
"simplex",
"virus",
"medicine",
"and",
"health",
"sciences",
"crystal",
"structure",
"pathology",
"and",
"laboratory",
"medicine",
"viral",
"transmission",
"and",
"infection",
"pathogens",
"condensed",
"matter",
"physics",
"microbiology",
"vertebrates",
"viral",
"structure",
"animals",
"mammals",
"viruses",
"cell",
"signaling",
"membrane",
"receptor",
"signaling",
"dna",
"viruses",
"protein",
"structure",
"crystallography",
"herpesviruses",
"swine",
"viral",
"entry",
"solid",
"state",
"physics",
"proteins",
"medical",
"microbiology",
"microbial",
"pathogens",
"molecular",
"biology",
"physics",
"biochemistry",
"signal",
"transduction",
"macromolecular",
"structure",
"analysis",
"cell",
"biology",
"virology",
"viral",
"pathogens",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"amniotes",
"organisms",
"cell",
"fusion"
] | 2017 | Structural basis of nectin-1 recognition by pseudorabies virus glycoprotein D |
Biofilm formation on implanted medical devices is a major source of lethal invasive infection by Candida albicans . Filamentous growth of this fungus is tied to biofilm formation because many filamentation-associated genes are required for surface adherence . Cell cycle or cell growth defects can induce filamentation , but we have limited information about the coupling between filamentation and filamentation-associated gene expression after cell cycle/cell growth inhibition . Here we identified the CDK activating protein kinase Cak1 as a determinant of filamentation and filamentation-associated gene expression through a screen of mutations that diminish expression of protein kinase-related genes implicated in cell cycle/cell growth control . A cak1 diminished expression ( DX ) strain displays filamentous growth and expresses filamentation-associated genes in the absence of typical inducing signals . In a wild-type background , expression of filamentation-associated genes depends upon the transcription factors Bcr1 , Brg1 , Efg1 , Tec1 , and Ume6 . In the cak1 DX background , the dependence of filamentation-associated gene expression on each transcription factor is substantially relieved . The unexpected bypass of filamentation-associated gene expression activators has the functional consequence of enabling biofilm formation in the absence of Bcr1 , Brg1 , Tec1 , Ume6 , or in the absence of both Brg1 and Ume6 . It also enables filamentous cell morphogenesis , though not biofilm formation , in the absence of Efg1 . Because these transcription factors are known to have shared target genes , we suggest that cell cycle/cell growth limitation leads to activation of several transcription factors , thus relieving dependence on any one .
Candida albicans is an invasive fungal pathogen that causes lethal infections in approximately 400 , 000 people per year worldwide [1] . Susceptibility to infection can be caused by a weakened immune system or presence of an implanted medical device , which provides a niche for biofilm formation . Limitations of the antifungal armamentarium lead to a 40% mortality rate among infected patients despite therapy [2–4]; hence the development of new therapeutics is of the utmost importance [5] . The goal of our study is to tie essential genes , which are candidate drug targets [6–9] , to biological processes . Such connections may be useful to develop screens for growth inhibitors . Several approaches have been used with C . albicans to reduce expression of an essential gene in order to assess its biological function [10–17] . Most approaches have used promoters with activity that can be regulated by presence of a nutrient or small molecule ( Tet-off [12–14] , MET3 [15] , PCK1 [16] , and MAL2 [17] ) . In fact , a collection of GRACE strains is now available in which 2 , 356 different genes have been placed under control of a doxycycline-repressible promoter [12 , 18] . This approach allows growth under a permissive condition in which the gene of interest is expressed at high levels , and then allows functional assays after expression of the gene is reduced . Several previous studies of C . albicans have shown that cell cycle or cell growth inhibition induces polarized growth , yielding elongated cells that resemble naturally occurring filamentous cells such as hyphae or pseudohyphae ( reviewed in [19 , 20]; see [18] for a recent study ) . Filamentous growth is of particular significance for C . albicans because it is required for invasive infection and for biofilm formation . Many filamentous cell functions are tied to their expression of a distinctive set of filamentation-associated genes whose products have direct roles in adherence and pathogenicity [21 , 22] . It is not clear to what extent cell cycle/cell growth inhibition induces filamentation-associated gene expression , as illustrated by the foundational study of Bachewich et al . [23] . They used genome-wide profiling to establish that depletion of the essential protein kinase Cdc5 , an M-phase regulator , induced both filamentous growth and several filamentation-associated genes that included ECE1 and RBT1 . However , they found that filamentous growth and filamentation-associated gene expression could be uncoupled . Specifically , treatment with the DNA synthesis inhibitor hydroxyurea induced filamentous growth but had little effect on filamentation-associated gene expression . A more extreme case of uncoupling comes from work of Bastidas et al . [24] who studied cell growth control by the Tor1 kinase . They observed that the Tor1 inhibitor rapamycin blocked filamentous growth yet induced filamentation-associated genes that included ECE1 and RBT1 . Therefore , the extent of coupling between filamentation and filamentation-associated gene expression after cell cycle/cell growth inhibition is uncertain . In addition , the identities of transcriptional regulators that mediate the effects of cell cycle/cell growth inhibition on filamentation-associated genes remain largely undefined . We focus here on a set of protein kinase genes that have been implicated in cell cycle or cell growth control . We chose to examine protein kinases because they are among the most druggable eukaryotic protein targets [25 , 26] . In fact , our prior study of C . albicans protein kinase-related gene insertion mutants revealed that many non-essential protein kinases may be useful antimicrobial targets [27]: 35 of 80 viable mutants were defective in virulence-associated traits that included stress tolerance , biofilm formation , or filamentation . However , because DNA insertions were used to create the mutations in that study , we were unable to recover mutations in essential protein kinase genes . Here we have engineered constitutive diminished expression of 15 protein kinase and protein kinase-related genes . Our findings extend the connection between cell cycle/cell growth inhibition and filamentation to include filamentation-associated gene expression . We find that the impact of a partial defect in cell cycle regulator Cak1 is sufficient to promote biofilm formation and filamentation-associated gene expression in the absence of major transcriptional activators of these processes .
We chose 18 protein kinase genes that we had been unable to disrupt previously [27] for functional analysis ( Table 1 ) . Half of the genes had been included in a screen of doxycycline-repressible GRACE strains for altered hyphal morphogenesis [12]; the other half were not represented among the GRACE strains ( Table 1 ) . In addition , we included as an internal control an essential protein kinase-related gene , CLN3 , that had been analyzed previously through conditional expression approaches [28 , 29] . To create a set of strains with decreased expression of each gene , we deleted one allele and replaced the 5' region of the second allele with a weakly expressed promoter . We were able to introduce the PGA5 promoter in front of most genes to create DX1 alleles ( S1 Table ) . However , for three genes , only fusions to the promoters of PGA42 ( DX2 alleles ) or ORF19 . 7606 ( DX3 alleles ) were recovered ( S1 Table ) . We refer to a strain of genotype yfg1-DX1/yfg1Δ , yfg1-DX2/yfg1Δ , or yfg1-DX3/yfg1Δ as a yfg1 DX strain . Of the 19 genes chosen ( 18 protein kinase genes and CLN3 ) , we recovered DX strains for 15 genes ( Table 1 ) . We used nanostring RNA measurements to determine whether the DX strains had reduced expression of the targeted genes ( S1 Table ) . Under yeast growth conditions ( 30° in YPD medium ) , the strains displayed 7 . 3 ± 8 . 1% of the wild-type expression levels for the 12 genes with DX1 alleles . Under hyphal growth conditions ( 37° in YPD + serum medium ) , the strains displayed 8 . 5 ± 10 . 8% of the wild-type expression levels for the 12 genes with DX1 alleles . Too few genes were represented by DX2 or DX3 alleles to draw general conclusions . These results indicate that DX strains engineered by promoter replacement with the PGA5 promoter have reduced expression of several different targeted genes under both yeast and hyphal growth conditions . Several DX strains grew as filamentous forms under conditions that supported nonfilamentous growth of the wild-type strain ( Fig 1 , Table 1 ) . The cln3 , cak1 , kin28 , gin4 , dbf2 , and ctk1 mutants had the most extreme phenotypes; the snf1 and mck1 mutants had prominent though less severe phenotypes . Many of the remaining mutants had mild or heterogeneous phenotypes ( summarized in Table 1 ) . Filamentous cells of the cln3 and cak1 DX strains resembled true hyphae in having parallel cell walls along the filament ( Fig 1 ) . For many of the other mutants , filament diameters varied and lengths were stunted ( see especially kin28 , gin4 , dbf2 , ctk1 , and snf1 ) . Our observations with the cln3 DX strain agreed with previous studies of hyperpolarized growth after repression of CLN3 RNA [28 , 29] . Filamentous growth of wild-type C . albicans is accompanied by high-level expression of filamentation-associated genes [19 , 38] . Therefore , we sought to determine whether filamentous growth of DX strains observed under noninducing conditions was accompanied by filamentation-associated gene expression . We measured expression of 181 genes in each DX mutant , including five of the eight core filamentation genes [38]: HWP1 , ECE1 , ALS3 , IHD1 , and RBT5 . Many of the mutants display elevated expression of these core filamentation genes ( Fig 1 , Table 1 ) . In general , the extent of core filamentation gene up-regulation correlated with the degree of hyperpolarized growth: cln3 , cak1 and kin28 DX strains expressed the highest levels of HWP1 , ECE1 and ALS3 , followed by gin4 and ctk1 DX strains . These results argue that growth limitation imposed by reduced expression of four protein kinase genes , or the essential protein kinase-related gene CLN3 , activates both morphogenesis and gene expression arms of the filamentous growth program . Several DX strains provided exceptions to correlation between filamentous morphology and filamentation gene expression . This group included dbf2 , ipl1 , snf1 , and ire1 DX mutants . These strains displayed filamentous morphology and seemed to have slightly elevated core filamentation gene expression , but the difference from the wild type was not statistically significant ( Table 1 ) . We believe that the poor correlation between morphology and gene expression in this group reflects the small number of core filamentation genes . In fact , a large number of other genes that were up-regulated in this set of DX mutants ( as well as the cln3 , cak1 , kin28 , gin4 , and ctk1 DX mutants ) under noninducing conditions ( YPD , 30° ) were also up-regulated in the wild-type strain under filamentation-inducing conditions ( YPD + 10% serum , 37° ) ( S1 Fig; S2 Table ) . An additional complication was that several of these DX mutants had heterogeneous morphology ( Fig 1 ) , so the gene expression profile of the population may not reflect the gene expression profile of the rare filamentous cells . In summary , there is a good correlation between filamentation and core filamentation gene expression for DX mutants with the most severe phenotypes , and an uncertain correlation for DX mutants with weak or heterogeneous phenotypes . We sought to verify that each defined mutation was the cause of each mutant strain's phenotype . We took two approaches to establish a genotype-phenotype correlation: independent isolate characterization and complementation . For the first approach , we chose seven genes and characterized 3–5 five mutant isolates of each: cln3 , cak1 , ctk1 , kin28 , snf1 , sak1 and mck1 . For the 5 genes that caused hyperfilamentous morphology , we observed that all independent isolates were hyperfilamentous , as indicated by wrinkled colony morphology [39] . In addition , independent DX isolates for most genes showed highly reproducible expression alterations for 181 genes assayed ( Fig 2 , S3 Table ) . The one exception was that independent snf1 DX strains had a more variable gene expression phenotype than the other DX strains . It is possible that the snf1 DX strains acquire second-site suppressor mutations that have impact on their gene expression phenotype . These results indicate that independent DX isolates for six of the seven genes tested yield reproducible phenotypes , and support the argument that the defined DX mutations cause the mutant phenotypes in those cases . The second approach to test the genotype-phenotype relationship was complementation . We introduced a wild-type copy of the respective affected gene into each of six DX mutants: cak1 , cln3 , kin28 , ctk1 , snf1 , and sak1 . Complementation was assessed through cell morphology , Cek1 phosphorylation , and gene expression profiling . Five of the DX mutants in this group were hyperfilamentous , and we observed that the complemented cln3 , kin28 , cak1 , ctk1 , and snf1 DX strains had nonfilamentous morphology when grown at 30° ( Fig 3A ) , similar to the wild-type strain ( Fig 1 ) . The sak1 DX strain had nonfilamentous morphology under these conditions , so complementation could not be assessed by morphology . These results indicate that reduced expression of CLN3 , KIN28 , CAK1 , CTK1 , or SNF1 causes hyperfilamentation . Complementation was assessed through levels of phosphorylated MAP Kinase Cek1 as well . Activity of Cek1 has been implicated in both filamentation and growth ( [40–42]; reviewed in [43] ) . Immunoblot analysis revealed that the ctk1 , kin28 , snf1 , and sak1 DX mutants had increased levels of phosphorylated Cek1 compared to the wild type ( Fig 3B ) . Complementation of these DX mutants reduced levels of phosphorylated Cek1 ( Fig 3B ) . These results indicate that reduced expression of CTK1 , KIN28 , SNF1 , or SAK1 causes increased accumulation of phosphorylated Cek1 . Complementation was also assessed through assays of RNA levels for 181 genes ( Fig 3C , S3 Table ) . We observed that the most prominent gene expression changes were almost completely reversed by complementation for most of the DX mutants tested . Specifically , the mean fold-change in expression for the 20 most greatly affected genes , compared to the wild-type strain , was reduced in the complemented strains from 9 . 4 to 1 . 2 ( cak1 ) , 10 . 1 to 1 . 5 ( kin28 ) , 8 . 4 to 2 . 0 ( ctk1 ) , 6 . 3 to 1 . 5 ( snf1 ) , and 2 . 9 to 1 . 1 ( sak1 ) . We observed partial complementation of the cln3 DX mutant: introduction of wild-type CLN3 reduced the mean fold-change in expression for the 20 most greatly affected genes from 12 . 6 to 3 . 1 . The level of CLN3 RNA in the complemented strain was ~20% of the wild-type level ( S3 Table ) , which may be the reason for partial complementation . Our analysis supports the idea that reduced expression of CAK1 , CLN3 , KIN28 , CTK1 , SNF1 , and SAK1 causes the major gene expression changes observed in the respective DX strains . Biofilms of C . albicans include both filamentous cells and yeast cells [44 , 45] . We sought to determine whether hyperfilamentous DX strains showed altered biofilm formation . For this analysis we focused on the cak1 DX strain because it had a prominent phenotype . The wild-type , cak1 DX , and complemented strains all formed biofilms of similar depth , as shown in side-view projections of confocal microscopy images ( Fig 4A ) . However , apical views revealed defects in the cak1 DX biofilm ( Fig 4B ) . Specifically , cells in the upper regions ( above 180 μm from the substrate ) displayed aberrant morphology . In addition , cell staining in these regions was uneven , which may reflect cell wall defects . Biofilm growth is accompanied by accumulation of extracellular matrix , a complex mixture of protein , carbohydrate , and other components [46] . We observed that cak1 DX biofilm matrix had significantly reduced levels of protein ( Fig 4C ) . Levels of one of the best characterized matrix carbohydrate components , soluble β-1 , 3 glucan , also accumulated at significantly lower levels in the cak1 DX biofilm than in the wild-type or complemented strain biofilms . These results indicate that diminished Cak1 expression does not prevent biofilm formation , but Cak1 is required for normal biofilm cell morphology and matrix production . Results above indicate that several growth-regulatory protein kinases function as negative regulators of filamentation and filamentation-associated gene expression . We sought to determine whether this role of the protein kinases may be mediated by known transcriptional activators of filamentation genes . We again focused on CAK1 for the analysis . We chose five major activators of filamentation genes: Bcr1 , Brg1 , Efg1 , Tec1 , and Ume6 . All are required for filamentation-associated gene expression [47–51] . We verified that a panel of newly made mutants affecting these genes had reduced expression of core filamentation genes ( Fig 5A , S4 Table ) . Under these growth conditions ( YPD medium at 37° , which is sufficient for induction of the filamentation pathway [52 , 53] ) , filamentous cell morphogenesis was severely defective in the efg1Δ/Δ mutant , partially defective in the brg1Δ/Δ and tec1Δ/Δ mutants , and largely similar to the wild-type strain in the bcr1Δ/Δ and ume6Δ/Δ mutants ( Fig 5B ) . All of the mutants were also defective in biofilm formation ( Fig 5C ) . Quantitative measurement of gene expression in single and double mutant strains revealed that each filamentation activator was required for the full gene expression response to reduced CAK1 levels ( Fig 5A ) . We used filamentation-inducing growth conditions , in which core filamentation gene expression is similar in the cak1 DX and wild-type strains . In the cak1 DX strain background , expression of most core filamentation genes assayed was substantially reduced by any filamentation activator deletion mutation . For example , levels of ECE1 RNA were reduced between 5- and 50-fold by deletion of any single filamentation activator gene in the cak1 DX background . However , mutation of any particular filamentation activator had less gene expression impact in the cak1 DX background than in the wild-type CAK1 background . For example , levels of ECE1 RNA were increased between 10- and 50-fold by a cak1 DX alteration in any filamentation activator gene deletion background . Therefore , although the expression of filamentation-associated genes remains largely dependent upon these filamentation activators regardless of the state of CAK1 , reduced CAK1 expression causes increased residual filamentation-associated gene expression in the absence of major filamentation activators . A broader view of gene expression levels supports the idea that the cak1 DX background is epistatic to filamentation activator defects . This point is illustrated by the fact that the cak1 DX strain groups together with the cak1 DX-filamentation activator double mutants by unsupervised hierarchical clustering ( Fig 5D ) . Inspection of the data indicates that the cak1 DX strain has highly pleiotropic gene expression effects , and many of those effects are insensitive to deletion of filamentation activators . Therefore , our gene expression data indicate that reduced CAK1 expression can overcome much of the impact of defects in filamentation activators . Biological impact of the filamentation activator mutations was muted when examined in a cak1 DX background . For example , biofilm formation ability was restored for all activator mutants except the efg1Δ/Δ mutant ( Fig 5C ) . Even though the efg1Δ/Δ mutant remained biofilm-defective , its filamentation ability was greatly improved in the cak1 DX background ( Fig 5B ) . These phenotypic assays indicate that reduced CAK1 expression can bypass the need for Bcr1 , Brg1 , Tec1 , and Ume6 in biofilm formation , and can overcome the dependence of filamentation on Efg1 . These biological results argue that the residual expression of filamentation-associated genes that is observed in filamentation activator mutants in the cak1 DX background has substantial phenotypic impact . The epistasis tests above show that diminished CAK1 expression can override filamentation/biofilm activator defects , based upon several phenotypic assays . One simple interpretation of these outcomes is that Cak1 acts downstream of the filamentation/biofilm activators [54 , 55] . However , each filamentation/biofilm activator's target genes overlap considerably with those of the other filamentation/biofilm activators [48] . This point led us to consider an alternative explanation . Specifically , we considered a model in which diminished CAK1 expression may stimulate the functions of several filamentation/biofilm activators . A deletion mutation that removes only one activator might be bypassed through the effects on the other activators . We tested this model through analysis of mutant strains with deletions of two filamentation/biofilm activator genes , BRG1 and UME6 . Recent studies indicate that Brg1 promotes initiation of filamentation , whereas Ume6 promotes maintenance of filamentation [51 , 56] . In addition , our results show that BRG1 and UME6 RNA levels are increased in the cak1 DX background at 30° ( S2 and S3 Tables ) . We observed that the brg1Δ/Δ ume6Δ/Δ cak1 DX strain had some biofilm formation ability ( Fig 6A ) , and could produce filamentous cells at both 37° and 30° ( Fig 6B and 6C ) . When a wild type copy of CAK1 was introduced into the strain , biofilm formation was eliminated ( Fig 6A ) and cell morphology ( Fig 6B and 6C ) more closely resembled that of the brg1Δ/Δ mutant strain ( Fig 5B ) . Quantitative measurement of gene expression in the brg1Δ/Δ ume6Δ/Δ cak1 DX strain revealed a significant increase in expression of core filamentation genes HWP1 , ECE1 , and ALS3 over that observed in the CAK1-complemented strain ( Fig 6D ) , though the magnitude of expression was greatly reduced compared to the brg1Δ/Δ cak1 DX strain and the ume6Δ/Δ cak1 DX strain ( Fig 5A ) . The fact that the cak1 DX phenotypes are less pronounced in a brg1Δ/Δ ume6Δ/Δ double mutant background than in either the brg1Δ/Δ or ume6Δ/Δ single mutant backgrounds is consistent with the hypothesis that both Brg1 and Ume6 contribute to the cak1 DX phenotypes .
Most studies of C . albicans essential genes have employed fusions to regulated promoters [12–17] . The GRACE strain collection is an exceptionally useful resource built upon such a platform [12] . The function of a gene is assessed after it is transcriptionally repressed . This approach is extremely useful , but has some limitations . For example , repression is occasionally incomplete [12] . In addition , it can take many generations to dilute out pre-existing gene product after the regulated promoter has been shut off ( see references [59] and [60] for examples ) . Finally , growth under permissive conditions , when the regulated promoter is active , can result in overexpression of the targeted gene [61 , 62] and novel phenotypic consequences . Our approach here was to reduce gene expression constitutively , a concept that has proven extremely useful for uniform analysis of S . cerevisiae essential genes [63] . This approach has the advantage of enabling phenotypic assays under diverse conditions where controlled manipulation of drug concentrations or carbon sources may not be feasible . Also , constitutive down-regulation relieves concern about a protracted time for dilution of pre-existing gene product . Our approach has the disadvantage of potentially selecting for secondary suppressor mutations , and such suppressors may have been the cause of heterogeneity of our snf1 DX strains . A second disadvantage is some genes seemed to be affected very little by our promoter replacement . These and other considerations lead us to suggest that no single approach is ideal for analysis of all genes under all circumstances . In fact , it seems advantageous to have several approaches available for gene function analysis . A comparison of DX strain cell morphology with results of prior studies illustrates this point ( Table 1 ) . In most cases ( 7 genes ) , the DX and prior depletion or deletion strains have the same cell morphology , thus providing validation of results . In two cases–CDC7 and CDC28 –the prior conditional depletion strains were filamentous and the DX strains grew as yeast . It seems reasonable that the prior conditional depletion strains had lower gene expression levels than the DX strains . In fact , these two particular DX strains expressed the targeted genes at higher levels than most other DX strains ( S1 Table ) . In the case of CAK1 , the DX strain was filamentous and the prior conditional depletion strain grew as yeast . We validated the cak1 DX strain phenotype by both complementation and consistency of independent isolates . Thus it seems reasonable that the DX strain had lower gene expression levels than the prior conditional depletion strain . These examples of gene-specific considerations illustrate that it is useful to have several approaches for functional analysis . Several previous reports show that mutation of cell cycle or cell growth regulatory genes can induce filamentous growth under conditions that are noninducing for wild type strains ( reviewed in [19 , 20] ) . A recent genome-scale survey of GRACE strains extended the list of such mutants considerably [18] , and our study brings the number of protein kinase-related genes that control cell cycle/cell growth and filamentation from nine to thirteen . Our results together with prior studies indicate that many protein kinases with cell cycle or cell growth regulatory functions are negative regulators of filamentous growth in that their inactivation induces filamentation . Although the connection between cell cycle/cell growth impairment and filamentous growth is well established , the mechanisms that mediate this relationship are uncertain . Our findings shed light on this connection . Specifically , we found that several major transcriptional activators of filamentation-associated genes are required for the full gene expression response to diminished CAK1 expression . This result argues that the filamentation response to cell cycle/cell growth impairment probably does not reflect rogue activity of one particular activator , but instead results from activity of signaling pathways that normally govern filamentous growth . One unexpected outcome of our analysis is that there is substantial expression of filamentation-associated genes in response to diminished CAK1 levels even in the absence of any key transcriptional activator . The net impact is functional , as revealed through biological assays for filamentation and biofilm formation in vitro . If we interpret the results in the framework of a linear pathway model [54 , 55] , they suggest that Cak1 may act downstream of the filamentation/biofilm activators to repress target gene expression . However , the overlap among targets of these activators has been well documented [48 , 51 , 56] . We thus considered the model that diminished CAK1 expression may stimulate several filamentation/biofilm activators , and that each can compensate partially for the absence of any other . The mechanism of activation would have to be post-transcriptional , because at 37° the cak1 DX background has little impact on activator RNA levels ( S4 Table ) ; several post-transcriptional control mechanisms for filamentation/biofilm activators are known [64–67] . The model that Cak1 has impact on several filamentation/biofilm activators is consistent with our experimental test , which showed that the cak1 DX phenotype is more strongly affected by absence of Brg1 and Ume6 together than it is by absence of either Brg1 or Ume6 alone . Our results thus argue that the cell cycle impairment by reduced CAK1 expression results from stimulation of multiple filamentation/biofilm activators . The relationship between filamentation-associated gene expression levels and biofilm formation ability seems puzzling . For example , expression of filamentation-associated genes is not dramatically higher in the brg1Δ/Δ ume6Δ/Δ cak1 DX strain , which is biofilm-competent , as compared to the CAK1 complemented brg1Δ/Δ ume6Δ/Δ cak1 DX strain , which is biofilm-defective ( Fig 6D; S5 Table ) . Perhaps there is a precise threshold for adhesin gene expression levels that can support biofilm formation . This threshold may be reduced in the brg1Δ/Δ ume6Δ/Δ cak1 DX strain because it expresses lower levels of YWP1 than the CAK1 complemented brg1Δ/Δ ume6Δ/Δ cak1 DX strain ( S5 Table ) ; YWP1 specifies an anti-adhesin [68] . A second possibility is that reduced CAK1 activity may cause increased expression of alternative biofilm adhesins . We note that several cell wall protein genes are up-regulated in cak1 DX strains ( Fig 5D ) ; these genes may specify such alternative adhesins . Our results together with prior studies indicate that it is common for essential protein kinases to function as negative regulators of filamentation and filamentation-associated genes ( [18 , 31 , 32 , 35] , reviewed in [20] ) . Other essential genes are also negative regulators of filamentation [18] and one well studied example , HSP90 , exerts its effects through activities of essential protein kinases [37 , 69] . There are a few growth inhibitors that activate filamentation or filamentation-associated genes at sub-lethal concentrations [24 , 70–72] . It seems possible that such drugs target some of the essential genes whose inactivation causes increased filamentation , as has been shown for inhibitors of Tor1 and Hsp90 [24 , 72] . How may the inverse relationship between cell cycle/cell growth and filamentation benefit C . albicans ? A rationale originally proposed for S . cerevisiae pseudohyphal growth—that filamentous growth allows escape from an unfavorable environment [73]—seems applicable to C . albicans as well . A second rationale views this relationship in the context of biofilm formation . The cells at the base of a biofilm are critical for adherence of the entire structure to the substrate . Those cells are likely to be nutrient-limited as well , because they are surrounded by other cells with better access to the outside environment . Filamentous growth forms express key biofilm adhesin genes ALS3 and HWP1 [38] . Therefore , induction of filamentation by growth limitation would have the effect of reinforcing biofilm-substrate attachment .
C . albicans strains were grown at 30°C in YPD or 37°C in YPD , YPD + 10%FBS or in RPMI-1640 . Transformants were selected on synthetic medium ( 2% dextrose , 1 . 7% Difco yeast nitrogen base with ammonium sulfate and auxotrophic supplements ) . Based on data from previous experiments , we searched for genes whose expression was unmodulated between planktonic 30° growth in YPD , 37° growth in hyphal inducing spider media , and in kidney infection models . We then looked among those genes for the ones whose expression was low from nanostring data analysis ( normalized expression values below 500 counts ) and we selected three of varying expression levels: PGA5 ( ORF19 . 3693 ) , PGA42 ( ORF19 . 2907 ) , and ORF19 . 7606 ( S6 Table ) . Heterozygotes were constructed in the BWP17 background by replacing one allele of the gene of interest with the URA3 marker by homologous recombination . PCR was performed using plasmid pGEMURA as template; PCR primers had homology to 100 bp upstream of the ATG or to 100 bp downstream of the stop codon followed by 18–19 bp homology to adaptor sequences on the plasmid surrounding the selective marker . This PCR product was transformed into the C . albicans strain . The promoter sequences were inserted downstream of ARG4 in pRSARG4ΔSpe , oriented in the same direction as ARG4 , by digesting the plasmid with NotI to direct integration of the PCR product carrying promoter sequences by homologous recombination in S . cerevisiae . The promoter sequences to PGA5 , PGA42 , and ORF19 . 7606 were inserted into the plasmid . The promoter for PGA5 included only 400 bp upstream of the start codon ( due to the end of an upstream gene located 411 bp upstream of the PGA5 ORF ) and the promoters for PGA42 and ORF19 . 7606 each included 950 bp upstream of their respective start codons . In order to be able to use the same primers to amplify each of the promoters from their respective plasmids , the primers used to amplify sequences for the PGA42 and the ORF19 . 7606 promoters for insertion into the plasmid included homology to the 30 nucleotides directly upstream of the start codon of PGA5 as a sequence adaptor substituting for those equivalently located nucleotides of PGA42 and ORF19 . 7606 . Thus the forward and reverse primers for amplifying promoter sequences from pRSARG-DX1 ( PGA5 promoter ) , pRSARG-DX2 ( PGA42 promoter ) , or pRSARG-DX3 ( ORF19 . 7606 promoter ) were the same for each of the promoters for any specific gene . The primers for amplifying promoter sequences to be transplaced in front of specific genes were designed to delete somewhere between 50 and 500 bp of the native promoter region of the gene . The general design of the primers used for amplifying the promoter sequences to be transplaced in front of genes is as follows: Forward primer: 5’ [100 bp upstream sequence of GENE X within 500 bp of ATG] [GTGTGGAATTGTGAGCGGATA ( the reverse complement of bp 3960–3980 pRSARG4ΔSpe -this sequence is upstream of ARG4 ) ]3’ Reverse primer: 5’[reverse complement of 1st 100 nts of GENE X ORF] [GATGGATTAAGATGATTGATTGTGATGATT ( the reverse complement of the 30 bp adaptor sequence for promoters from 30 bp upstream of PGA5 orf to 1 bp upstream of PGA5 orf ) ] 3’ Primer pairs were designed to check the specific alleles present in any given transformant: an upstream F check and an ORF rev check to check for the presence of a wild type allele of a gene; the upstream F check and a URA3 rev check to check for the URA3 marked allele; and an adaptor seq F check and the ORF rev check to check for the promotor transplacement . Strains used are listed in S7 Table . When complementing DX mutants , the WT allele was amplified from genomic DNA ( SC5314 ) , including 1000 bp upstream and 250 bp downstream of the ORF ( shorter distances were used when there were additional genes located within this region ) . Complementation primers were approximately 80 bp in length and were comprised of a sequence to direct in vivo recombination into plasmid pDDB78 ( CAATTTCACACAGGAAACAGCTATGACCATGATTACGCCAAGCT for the forward primer and GTCGACCATATGGGAGAGCTCCCAACGCGTTGGATGCATAG for the reverse primer ) followed by a 45mer gene specific sequence . The complementing PCR product was co-transformed along with EcoRI/NotI digested pDDB78 into the S . cerevisiae BJ2698 strain ( his1 ) . The resulting complementing plasmid was amplified in E . coli and digested with NruI to direct insertion to the his1 locus of the DX mutant strains . NruI digested pDDB78 was inserted into mutant strains to obtain a marker-matched prototrophic mutant . When DX mutations were in conjunction with a homozygous deletion of another gene , an unmarked deletion was first constructed of the other gene using the URA3 mini-blaster protocol [74] which utilizes a recyclable cassette , and then this strain was used as the parent for the construction of the DX mutant . His+ prototrophs were made by inserting NruI digested pDDB78 into the mutant strains . An unmarked deletion of BRG1 was first constructed using the URA3 mini-blaster protocol [74] . Then UME6 was deleted using a transient CRISPR-Cas9 system [75 , 76] utilizing a NAT1 marker . This strain was used as the parent for the construction of the cak1 DX mutant . His+ prototrophs or CAK1 complemented strains were constructed as described above . Overnight cultures of cells were diluted into 50 mls fresh medium at specified OD . Cultures were allowed to grow with shaking and cells were then harvested by vacuum filtration and quickly frozen at -80°C until RNA extraction . RNA extractions were performed using Qiagen RNeasy Mini Kit ( cat#74104 ) with the following modifications . Cells were resuspended from the membranes with 2 x 900 μl ice cold water washes with vortexing for 30 sec after each addition . 1 . 5 mls of cell suspension was added to prechilled microfuge tubes and centrifuged at top speed for 30 sec . 600 μl RLT + 1%BMSH was added to resuspend the cells and this was added to a fresh 2 ml screw cap tube containing 300 μl Zirconia beads ( Ambion , Fisher Scientific ) and 600 μl phenol:chloroform:isoamyl alcohol 25:24:1 , vortexed on a mini-beadbeater ( Biospec Products ) for 3 min , and centrifuged at 14000 rpm for 5 min at 4°C . 550 μl of the aqueous phase was transferred to a new microfuge tube , an equal volume of 70% ethanol added , and RNA isolation proceeded as the manufacturer’s instructions . NanoString analysis is a sensitive method for analyzing gene expression of 100–800 targets at a time . For our analysis , 166 targets associated with environmental response pathways , including cell wall stress , osmotic and oxidative stress conditions , hypha development and biofilm formation , along with 15 targets corresponding to the mutants constructed for this work , were selected for analysis . For each nanoString assay , 125–300 ng of Candida RNA extracted from an in vitro liquid culture was added to a nanoString codeset mix and incubated at 65°C overnight ( 16–18 hours ) . The reaction mixes were loaded on the nanoString nCounter Prep Station for binding and washing , and the resultant cartridge was transferred to the nanoString nCounter digital analyzer for scanning and data collection . A total of 600 fields were captured per sample . To calculate gene expression ratios among different samples , we normalized adjusted raw counts by total counts of 181 genes after background subtraction [77] . The heat maps were generated using Multiexperimental Viewer 4 . 9 . 0 [78] . Preliminary determination of expression levels of the DX alleles was carried out on single his- isolates . Statistical significance was assayed by t-test . Strains were grown overnight and diluted into fresh 30° or 37° YPD at an OD600 of 0 . 1 . Cultures were grown for 4 hr and cells pelleted . 0 . 5 mls of media was retained in the tube , and cells transferred to microfuge tubes . Cells were pelleted and the supernatant removed . One ml 4% formaldehyde/PBS was added and the tubes were vigorously vortexed for 15 min . Cells were washed one time with PBS , resuspended in 100 μl PBS , and then vortexed . Calcofluor White ( 5 . 5mg/ml in 50% DMSO ) was added to 0 . 2 mg/ml , cells were vortexed and then incubated for 15 min with an occasional flicking . Cells were stored at 4° in the dark until visualization . Cells were diluted 1:2 with PBS and pipetted onto slides coated with concanavalin A . Cells were visualized with a Zeiss Axio Observer Z . 1 fluorescence microscope and a 20x objective . The strains being analyzed were grown overnight in YPD at 30°C . The overnight cultures were used to inoculate wells with 2 mls of fresh YPD media at an OD600 of 0 . 5 on silicone squares ( Bentec Medical Inc . ) that were pretreated with fetal bovine serum ( FBS ) . The cells were allowed to adhere to the silicone for 90 min in an incubator-shaker at 37°C and 60 rpm . Following the adherence , the squares were washed in PBS to remove any nonadherent cells and placed in wells containing 2 mls of new YPD media . Biofilm imaging is adapted from [79] with modifications . After 48 hours , the biofilms were fixed using 4% formaldehyde and 1 . 5% glutaraldehyde in 1xPBS on an orbital mixer for 1 hour . After fixation the specimens were washed in 1xPBS . The fixed biofilms were stained with concanavalin A , Alexa Fluor 594 Conjugate ( Life Technologies ) at a concentration of 25 μg/ml in PBS for two days on an orbital mixer . The fixed and stained biofilms on the silicone squares were then transferred to glass scintillation vials . To dehydrate the samples , 2 mls of methanol were added to the samples and allowed to infiltrate on an orbital mixer for 20 min . The methanol was aspirated out and 2 mls of methanol were briefly added . After the 100% methanol addition , a 50:50 mixture of methanol and methyl salicylate was added . The 50:50 mixture was aspirated out and replaced with 100% methyl salicylate . The vials were gently agitated until the samples were completely cleared through the matching of refractive index . In order to image these samples using an inverted confocal microscope and avoiding the use of plastic , a cover glass was cemented to the bottom of a black-anodized aluminum stage insert . A silicone ring ( thickness 300μm ) and a small amount of methyl salicylate were added to this constructed well and the biofilm was placed on the ring with the apical side facing down , using the surface tension between the methyl salicylate and the silicone square to hold the silicon in place . The biofilms were imaged using a slit-scan confocal optical unit on a Zeiss Axiovert 200 microscope . A 40x 0 . 85-numerical aperture oil immersion objective was used in order to provide enough working distance to focus through the full thickness of the biofilms . Optical sections were collected in several series of 130 planes with a total sum of 500 planes ( Fig 4 ) or 557 planes ( Fig 6A ) at 0 . 9 μm step-size . The stacks were concatenated and processed using FIJI software [80] . The images were processed using the Background Subtract plugin and the final images were obtained using a resliced , maximum intensity Z-projection . The apical view projections were obtained using the Temporal Color-code plugin and the Ice lookup table . Protein extraction and immunoblots were performed as described [81] . Proteins were extracted from cells grown to mid-log phase ( ~5 hr ) at 30°C in YPD media . Protein samples were separated on 10% SDS polyacrylamide gels ( SDS-PAGE ) and transferred to nitrocellulose membranes ( Protran BA85 , VWR International Inc . , Bridgeport NJ ) . Membranes were blocked in immunoblot buffer ( 5% nonfat dry milk , 10mM Tris-HCl [pH 8] , 150mM NaCl and 0 . 05% Tween 20 ) for 16h at 4°C . Cdc2 p34 antibody that recognizes the PSTAIRE motif was used as a loading control ( Santa Cruz Biotechnology , Santa Cruz , CA; #sc-53 ) . P~Cek1 was detected by p42/p44 antibodies ( Cell Signaling Technology , Danvers , MA; #4370 ) . Secondary antibodies included goat α-mouse IgG–HRP ( Bio-Rad Laboratories , Hercules , CA; #170–6516 ) , goat α-rabbit IgG-HRP ( Jackson ImmunoResearch Laboratories , Inc . , West Grove , PA; #111-035-144 ) , donkey α-goat IgG-HRP ( Santa Cruz Biotechnology , Santa Cruz , CA; #sc-2020 ) . Ponceau S ( Sigma , St . Louis , MO; #P7170 ) was used to confirm protein levels . WesternBright MCF fluorescent Western blotting kit from Advansta Inc . ( Menlo Park , CA; LPS #K-12045-D20 ) was used for detection . Densitometry was performed on immunoblots to calculate the levels of P~Cek1 in selected DX mutants . After background subtraction , band intensity of P~Cek1 and PSTAIRE were measured for each sample by ImageJ ( https://imagej . nih . gov/ij/ ) . The ratio of P~Cek1/PSTAIRE was set to 1 . 0 for wild type and calculated for other samples to measure the relative difference in P~Cek1 levels . Extracellular matrix was collected from two roller bottles , as published previously [82] . Briefly , cells from an overnight culture were used to create an inoculum of 106 cells/ml in RPMI-MOPS . One ml inoculum was added to each bottle , and biofilms were incubated at 37°C for 48 hr at 50 rpm , with a media exchange at 24 hr . The biofilms were removed from the bottles using a spatula and suspended in water , then sonicated for 20 min . These were centrifuged at 4 , 000 rpm for 20 min at 4°C , separating the soluble matrix from the cell pellet ( biomass ) . For normalization of subsequent ELISA data , the biomass of each biofilm from the group was collected , lyophilized , and weighed . Biofilm matrix protein was measured using a BCA protein assay ( Thermo-scientific ) , slightly modified from the manufacturer’s instructions . Briefly , 100μl of isolated matrix samples in triplicate were incubated with the BCA reagents 30min at 37°C . ODs were read at 562nm and protein concentrations calculated from a BSA standard curve . Raw data generated by the BCA assay was normalized by relative biofilm biomass then presented as μg/mg . The total carbohydrate contents of each sample was measured as hexoses by the phenol-sulfuric acid method and normalized by biomass . T-tests were used to determine statistical significance for both BCA and phenol-sulfuric acid assays . Matrix samples were analyzed by ELISA using a monoclonal antibody against β-1 , 3 Glucan ( Australia Biosupplies ) as previously described [46 , 83] . This assay used laminarin ( Sigma ) as standard curve , at least two biological replicates were performed , and the mean of three technical replicates from one representative assay was calculated . Once normalized by biomass , these values were presented as μg/mg . T-tests were used to determine statistical significance . | The ability of the pathogen Candida albicans to grow on surfaces as biofilms is a determinant of infection ability , because biofilms on implanted medical devices seed infections . Biofilm formation by this organism requires growth in the form of filamentous cells and the expression of filamentation-associated genes . Inhibition of cell proliferation can induce filamentous cell formation , as we find here for strains that express greatly reduced levels of the cell cycle regulator Cak1 . Surprisingly , biofilm formation occurs independently of many central biofilm regulatory genes when Cak1 levels are reduced . This response to proliferation inhibition may reflect the activation of numerous biofilm regulators , thus relieving the dependence on any one regulator . The stimulation of biofilm formation by proliferation inhibition , a property of many bacterial pathogens as well , may contribute to the limited effectiveness of antimicrobials against biofilms . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"biofilms",
"medicine",
"and",
"health",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"enzymes",
"pathogens",
"cell",
"cycle",
"and",
"cell",
"division",
"cell",
"processes",
"rna",
"extraction",
"microbiology",
"enzymology",
"mutation",
"fungi",
"model",
"organisms",
"fungal",
"pathogens",
"extraction",
"techniques",
"research",
"and",
"analysis",
"methods",
"mycology",
"mutant",
"strains",
"proteins",
"medical",
"microbiology",
"gene",
"expression",
"microbial",
"pathogens",
"protein",
"kinases",
"yeast",
"biochemistry",
"candida",
"cell",
"biology",
"phenotypes",
"genetics",
"biology",
"and",
"life",
"sciences",
"yeast",
"and",
"fungal",
"models",
"organisms",
"candida",
"albicans"
] | 2016 | Bypass of Candida albicans Filamentation/Biofilm Regulators through Diminished Expression of Protein Kinase Cak1 |
Hepatitis B virus ( HBV ) infection is a common problem in the world , especially in China . More than 60–80% of hepatocellular carcinoma ( HCC ) cases can be attributed to HBV infection in high HBV prevalent regions . Although traditional Sanger sequencing has been extensively used to investigate HBV sequences , NGS is becoming more commonly used . Further , it is unknown whether word pattern frequencies of HBV reads by Next Generation Sequencing ( NGS ) can be used to investigate HBV genotypes and predict HCC status . In this study , we used NGS to sequence the pre-S region of the HBV sequence of 94 HCC patients and 45 chronic HBV ( CHB ) infected individuals . Word pattern frequencies among the sequence data of all individuals were calculated and compared using the Manhattan distance . The individuals were grouped using principal coordinate analysis ( PCoA ) and hierarchical clustering . Word pattern frequencies were also used to build prediction models for HCC status using both K-nearest neighbors ( KNN ) and support vector machine ( SVM ) . We showed the extremely high power of analyzing HBV sequences using word patterns . Our key findings include that the first principal coordinate of the PCoA analysis was highly associated with the fraction of genotype B ( or C ) sequences and the second principal coordinate was significantly associated with the probability of having HCC . Hierarchical clustering first groups the individuals according to their major genotypes followed by their HCC status . Using cross-validation , high area under the receiver operational characteristic curve ( AUC ) of around 0 . 88 for KNN and 0 . 92 for SVM were obtained . In the independent data set of 46 HCC patients and 31 CHB individuals , a good AUC score of 0 . 77 was obtained using SVM . It was further shown that 3000 reads for each individual can yield stable prediction results for SVM . Thus , another key finding is that word patterns can be used to predict HCC status with high accuracy . Therefore , our study shows clearly that word pattern frequencies of HBV sequences contain much information about the composition of different HBV genotypes and the HCC status of an individual .
The hepatitis B virus ( HBV ) is a DNA virus infecting around 257 million people worldwide ( http://www . who . int/mediacentre/factsheets/fs204/en/ ) and can cause liver diseases and hepatocellular carcinoma ( HCC ) , one of the most common types of liver cancer [1 , 2] . About 500 , 000 HBV patients die each year worldwide from HBV related complications and about 10% of the HBV infected individuals will have HCC during their life time [3] . However , the understanding of the differences of HBV compositions based on next generation sequencing ( NGS ) technologies between chronic hepatitis B ( CHB ) and HBV related HCC is limited . The HBV sequences are currently divided into 10 HBV genotypes , A to J , with genome wide differences of 8% , and 35 subgenotypes using genome wide differences of 4% [3–5] . HBV genotypes have been shown to be associated with geographical locations [6 , 7] . In China , the most common genotypes are B and C [8 , 9] . Besides , some individuals can be infected by viruses of multiple genotypes and there can be some recombinations among the different genotypes . Different genotypes have varied effects on disease severity , course and likelihood of complications , response to treatment and possibly vaccination [10 , 11] . It has been shown that genotype C is associated with more disease complications and higher chance of HCC transition than genotype B [12] . Due to the high mutation rate of the HBV and the possibility of multiple HBV infections , there are high inter- and intra- patient HBV geneticdiversities . Previous studies revealed that basal core promoter ( BCP ) A1762T/G1764A mutations were strongly associated with the occurrence of HCC [13–16] . Truncated large surface proteins due to deletions in the pre-S gene were observed to accumulate in the endoplasmic reticulum ( ER ) , resulting in ER stress and hepatocarcinogenesis [17 , 18] . It was also shown that some pre-S deletions or mutations were risk factors for the development of liver cirrhosis and HCC [19–22] . Meta-analysis studies indicated that pre-S deletion mutations and BCP double mutations were associated with HCC risk [13 , 23–25] . Several studies have found that combination of mutations in the HBV genome could predict HCC occurrence more accurately than individual mutations [26–28] . Traditionally , only the dominant genotypes and haplotypes within the patients were investigated due to the technological limitations of Sanger sequencing that are usually time consuming and economically expensive to sequence a large number of sequences within individuals . With the development of high-throughput NGS technologies , it is now possible to investigate the HBV genetic diversity within individuals carefully and to develop more sophisticated and robust prediction models for predicting HCC . In this study , we aim to explore the diversity of HBV pre-S sequences within HCC and CHB patients , to identify their differences , and to establish prediction models for HCC with machine learning methods based on word pattern frequencies . In detail , we first carried out a large scale HBV pre-S region study of 94 HCC patients and 45 chronic HBV infected individuals . The heterogeneity of HBV composition and the HBV genotype fraction in individuals were investigated . We used a novel alignment-free method based on word pattern frequencies to cluster the individuals and investigated the cluster distributions of HCC patients and CHB individuals . We further applied K-nearest neighbors ( KNN ) and support vector machine ( SVM ) approaches to predict HCC status based on word counts and the predictive model was validated using an independent data set consisting of 46 HCC patients and 31 CHB individuals . The key novelties of this study are the use of word patterns for the analysis of HBV sequences to cluster HBV infected individuals and to predict HCC status . Our study clearly showed the surprising high power of word patterns for clustering HBV genotypes and predicting HCC status .
We genotyped each sequence in the NGS data using STAR [49] and calculated the fraction of genotypes B and C sequences for every individual as described in the “Materials and methods” section . The fraction of recombinants in 95% of the individuals ( 132/139 ) was less than 5% and most of the reads were of genotype B or C ( S1 Supplementary material S1 Fig ) . Therefore , we ignored the recombinant reads and the reads of other genotypes and concentrated on the reads of genotype B or C in all the individuals . The histograms of the fractions of genotype B sequences among the 94 HCC patients and 45 CHB individuals are given in Fig 1 ( A ) . It can be seen from the figure that most individuals have both genotypes B and C sequences for both HCC and CHB individuals . The fraction of genotype B sequences among HCC patients has a tendency to be lower than that for the CHB individuals , consistent with previous observations that genotype C individuals are more likely to have HCC than genotype B individuals [29] . About 70% of the HCC patients have genotype B fraction less than 30% and only about 50% of the CHB patients have genotype B fraction less than 30% . While about 37% of the CHB individuals have genotype B fraction at least 70% , only about 5% of the HCC patients have genotype B fraction at least 70% . Based on our data , we further investigated the relationship between having HCC and the fraction of genotype B in an individual . It can be shown that the probability of having HCC for given genotype B fraction increases with the ratio of fraction of individuals having the given genotype B fraction among HCC patients over that of CHB patients . Therefore , we binned both the HCC and CHB individuals according to the genotype B fraction . For each bin , we calculated the fractions HCC and CHB individuals and then calculated their ratio as shown in Fig 1 ( B ) . When the number of occurrences in a bin was small , the estimated fraction was not reliable . Thus , we required that the fractions for both HCC and CHB in each bin to be at least 5% . If either the HCC fraction or the CHB fraction in an interval was smaller than 5% , we merged it with the later intervals until both fractions were above 5% . Therefore , we merged the bins 0 . 3~0 . 4 , 0 . 4~0 . 5 , and 0 . 5~0 . 6 into one bin when we calculated the ratio of the fractions . Similarly , we merged the bins 0 . 7~0 . 8 , 0 . 8~0 . 9 and 0 . 9~1 . 0 to form another bin . As we can see from Fig 1B that this fraction is higher than 1 . 0 when the fraction of genotype B sequences is less than 0 . 6 , while it is much less than 1 when the fraction of genotype B sequences is above 0 . 6 . To see how genotyping method would affect the results , we also used another genotyping program , jpHMM [30] , to genotype the reads . The histogram of the fraction of recombinant reads for the 139 individuals is shown in FigS2a in the S1 Supplementary material . The fraction of genotype B using jpHMM is highly associated with that based on STAR ( Pearson correlation coefficient = 0 . 9968 and p-value = 1 . 0e-151 ) as shown in FigS2b ) in the S1 Supplementary material . FigS3 in S1 Supplementary material shows a similar figure as Fig 1 when jpHMM was used for genotyping . Again we see that the probability of having HCC increases with the fraction of genotype C sequences based on jpHMM . Based on the word pattern frequencies of the NGS reads from the HBV pre-S region for the individuals , we used Manhattan distance to calculate the dissimilarity between any pair of individuals . We then used principal coordinate analysis ( PCoA ) to project the individuals onto two-dimensional Euclidean space . Fig 2A and 2B show the PCoA results for the 94 HCC patients and 45 CHB individuals using word length k = 6 and k = 8 , respectively . To see the relationship between the PCoA results and the fraction of genotype B or C in the NGS data of the HBV pre-S sequences , we colored the points corresponding to the individuals according to the fractions of B and C genotypes with red indicating 100% genotype B and blue indicating 100% genotype C with intermediate color in between based on the STAR genotyping results . We also downloaded the HBV genotypes B and C reference sequences from NCBI ( accession number of genotype B: D00329 , AB073846 , AB602818; genotype C: X04615 , AY123041 , AB014381 ) and used the pre-S region to serve as references . We counted the occurrences of word patterns of these sequences , calculated their dissimilarity with the 139 samples , and plotted the 141 samples in the PCoA figure . We have several observations from Fig 2 . First , the fraction of genotype B sequences in each individual is highly associated with the values of the first principal coordinate . From left to right of the figures , the fraction of genotype B sequences increases with the first coordinate . To see this pattern more clearly , we plotted Fig 2C and 2D that show the relationship between the first principal coordinate and the fraction of genotype B using k = 6 and k = 8 , respectively . The Pearson correlation coefficient ( PCC ) between the fraction of genotype B sequences and the first principal coordinate is as high as 0 . 97 when k = 6 and k = 8 . Second , the HCC tumor samples are distributed more broadly on the PCoA plots and are more diverse than the CHB individuals . The second principal coordinate seems to be associated with the HCC status with high second PCoA coordinate indicating high probability of HCC . Although the second principal coordinates for most of the CHB individuals are at similar levels as for the reference genotypes B and C sequences , many HCC samples have much higher second principal coordinate . To see the pattern more clearly , we divided the second coordinate into 5 bins: < −0 . 15; −0 . 15~−0 . 1; −0 . 1~−0 . 05; −0 . 05~0; > 0 . In each bin , we calculated the fractions of CHB and HCC individuals in the bin . We also calculated their ratio and plot the relationship between the ratio and the second coordinate in Fig 2E and 2F . It can be seen that when the second coordinate is smaller than -0 . 1 , the fraction of CHB individuals dominates and with the increase of second coordinate , the fraction of HCC individuals increases . When the second coordinate is bigger than 0 , there are no CHB individuals . On the other hand , some of the HCC patients and CHB individuals mix together in the principal coordinate plots and there is no clear separation for HCC patients and CHB individuals . The above conclusions are consistent for both k = 6 and k = 8 . Fig 2 shows that the first principal coordinate is highly associated with the fractions of genotype B ( C ) when intuitively choosing k = 6 and k = 8 . Therefore , we chose the word length k to maximize the correlation . Table 1 shows the Pearson and Spearman correlations between the first principal coordinate and the fraction of genotype B sequences for word length k ranging from k = 2 to k = 8 . Both the Spearman and the Pearson correlation coefficients increase with word length k . When k ≥ 6 , the PCC becomes stable . Note that for k = 6 the correlation is already very high and considering computational efficiency , we use k = 6 to show our results on the training data in the rest of the paper . In addition to the PCoA plots , we also grouped the individuals using hierarchical clustering with UPGMA ( Un-weighted Pair Group Method with Arithmetic Mean ) to calculate the distance between two clusters . We used the distance matrix calculated from Manhattan distance with k = 6 and input it into the software Mega ( http://www . megasoftware . net/ ) . Fig 3 shows the clustering results and the genotypes are analyzed using STAR . The corresponding results using jpHMM are given in S1 Supplementary material Fig S5 . The individuals are generally divided into two main clusters . Cluster I contains 44 individuals , 38 of them with dominant genotype B and cluster II contains 95 individuals , 94 of them with dominant genotype C . The overlaps between the two clusters and groups of individuals with genotypes B or C are given in Table 2 . The clusters are significantly associated with the dominant individual genotypes ( p-value = 2 . 2e-16 , χ2-test ) . Six individuals ( HCC1 , HCC13 , HCC83 , HCC84 , HCC88 , and HCC102 ) out of 101 ( 76HCC+25CHB ) with dominant genotype C belong to cluster I . Their corresponding fractions of genotype B are 0 . 49 , 0 . 49 , 0 . 18 , 0 . 27 , 0 . 14 , 0 . 29 , respectively . On the other hand , only one individual ( CHB60 ) out of 38 ( 18HCC+20CHB ) with dominant genotype B belong to the second cluster and its fraction of genotype B is 0 . 59 . We can see that the mis-clustered individuals are highly mixed , and their secondary genotypes also have relatively high fraction . The normalized fractions of genotypes B and C sequences of all individuals using STAR and jpHMM are given in the S1 Table . Within cluster I , there is a small sub-cluster Ia that is dominated by CHB individuals . On the other hand , the HCC patients and CHB individuals are not clearly separated in cluster I . Within cluster II , a small cluster IIa is dominated by CHB individuals and the HCC patients are generally far away from this group . The results from the hierarchical clustering of the individuals are consistent with the observations based on PCoA results . We noticed 11 CHB patients within the large cluster IIb that contains mostly HCC patients . Therefore , we checked the meta-data to see if these 11 individuals had high risk factors for HCC including liver cirrhosis , advanced age , male sex , etc . Six out of the 11 CHB patients in cluster IIb had meta-data available . Five patients ( CHB46 , CHB48 , CHB50 , CHB60 , CHB91 ) are male and one is over 60 . Patient CHB55 is female , who has liver cirrhosis and was over 60 years old . Thus , our meta-data do show that these patients have more risk factors . We also colored the points in the PCoA plots corresponding to the individuals according to the fractions of B and C genotypes with red indicating 100% genotype B and blue indicating 100% genotype C with intermediate color in between based on the jpHMM genotyping results , and the corresponding figure is shown as FigS4 in the S1 Supplementary material . Similar observations as based on STAR genotyping were obtained . Table S2 in the S1 Supplementary material shows again that the first principal coordinate is highly associated with the fraction of B genotypes in an individual , consistent with the results using the STAR genotyping tool . We used two methods , K-nearest neighbors and support vector machine ( SVM ) , to predict HCC status based on the word pattern frequency vector of the HBV pre-S region of the samples . The prediction results based on KNN are given in Table 3 . It can be seen from the table that the cross validation results measured by AUC are roughly the same with different word length k and the AUCs center around 0 . 88 . For the independent test data , the AUC increases slightly with the word length from 0 . 62 for k = 2 to 0 . 67 when k is between 6 and 8 . The AUC values of SVM using cross validation and testing set and corresponding parameter C using different word length k are shown in Table 4 . We observe from the table that the prediction accuracy measured by AUC with cross-validation increases slightly with word length from 0 . 86 when k = 2 to 0 . 93 when k = 7 . On the other hand , the AUC for the independent data decreases with word length from 0 . 77 when k = 3 to 0 . 70 when k = 8 . When k = 2 , the AUC is only 0 . 65 . The good performance of the SVM model when k = 3 may be due to the relatively small number of learning samples such that the derived SVM model with small number of word patterns is more stable .
Several recent studies have clearly shown the advantage of NGS over traditional Sanger sequencing in detecting rare HBV sequence mutations [15] and for the prediction of anti-virus therapy response [31 , 32] . In this study , we used high throughput sequencing to investigate composition of HBV sequences in a large number of both CHB and HCC individuals , to compare differences of genetic composition between them , and to predict HCC status using novel word pattern based approaches . Several interesting results were obtained . First , we showed that there was extensive heterogeneity of HBV composition among the individuals based on the NGS data . Almost all the individuals contain some marked fractions of both genotype B and genotype C HBV sequences in Chinese individuals infected with HBV . Previous studies have shown the existence of co-infection of different genotypes of HBV [33–35] and inter-genotype HBV co-infection is the prerequisite of HBV recombination incidence that have been reported broadly [36–38] . Our results highlight the importance of using NGS to study the distribution of different genotypes within individuals . Second , we used a novel word pattern based approach to cluster the individual samples and investigated the cluster distributions of HCC patients and CHB individuals . Alignment-free sequence comparison based on word counts has been widely used in studying the relationships among sequences or NGS data as reviewed in [39 , 40] . However , this approach has not been used for the analysis of HBV data . In this paper , we used alignment-free sequence comparison methods based on word counts to study the relationship among the individuals . We used a dissimilarity matrix based on Manhattan distance between the word frequencies of the NGS data to cluster all the individuals . We showed that there was a strong correlation between the clustering and the fractions of genotypes ( B or C ) of individuals . This observation was surprising and proved the effectiveness of the alignment-free method on classification based on sequence dissimilarity . Third , since the second coordinate of PCoA was remarkably correlated with the probability of having HCC , we further applied K-nearest neighbors ( KNN ) and support vector machine ( SVM ) approaches to classify HCC or CHB individuals based on word counts . Using cross-validation , we achieved a high area under the receiver operational characteristic curve ( AUC ) of around 0 . 88 for KNN and 0 . 92 for SVM for word length from 4 to 8 . Fourth , we validated the prediction models on an independent set of 46 HCC patients and 31 CHB individuals . The AUC for the independent set was around 0 . 70 when word length is from 6 to 8 for SVM and 0 . 67 for KNN . Surprisingly , the AUC for SVM was 0 . 77 when word length is 3 . The good result of k = 3 may be explained by the appropriate number of features compared with the number of individuals . The results showed the usefulness of our prediction models for separating HCC patients from CHB individuals . Numerous studies have revealed the divergence in pre-S region between CHB and HCC patients and deletions in pre-S was one of the most noticeable characteristic of HCC patients [41–44] . In addition , fewer studies also found that several nucleotide mutations were also associated with incidence of HCC [19 , 45 , 46] . Nevertheless , we have succeeded in the establishment of predictive model for HCC via the word pattern frequencies of the pre-S gene following the NGS . The superior performances in both the cross validation and independent cohort validation are also indicative of the advantages of NGS compared with Sanger sequencing . Finally , we showed that the HCC status can be effectively predicted based on word pattern frequencies using support vector machine and that prediction accuracy increases with the number of reads and becomes stable at about 3000 reads per individual . To our knowledge , this is the first study focusing on the implication of the number of reads on model effectiveness trained on NGS data . With the development of NGS technology , investigators are interested in appropriate number of reads and our study provides guidelines for designing of NGS studies . Despite these significant results , our study has several limitations . First , the numbers of HCC and CHB individuals , although large compared to previous studies , were still not very large and more individuals are needed to further confirm the applicability of our word pattern based method for investigating HBV infected individuals . Second , the AUC values for the independent test data using both KNN and SVM were much smaller than the corresponding mean AUC values for cross-validation . Potential explanations for the lower AUC value for the independent test data is that the independent samples may come from populations different from that in the training data . Potential experimental variations from the testing data may also decrease the prediction accuracy . Third , we concentrated on the HBV pre-S region in this study and other regions may have different properties . Further studies for other regions or even the whole genome are needed . Fourth , we investigated Chinese HCC and CHB individuals with dominant B and C genotypes . The applicability of our results to other ethnic groups or population samples needs to be further investigated . In conclusion , our study showed the applicability of word pattern based methods to investigate the diversity of HBV sequences , to compare HBV communities among different individuals , and for the prediction of HCC status . Further studies are needed to extend the results to much larger genomic regions over large number of individuals .
HBV were divided into ten major genotypes A to J with the dominant genotype B or C in China . Merged pre-S region sequences were genotyped with HBV STAR software [49] that is one of the most widely used software tools for HBV genotyping [50–52] . It is based on a statistically defined , position-specific scoring model ( PSSM ) [53] . Even though our sequence reads are relatively short compared to the whole genome , it has been shown that any 300 bps sequence segment of the polymerase N-terminal domain containing pre-S is reliable for sequencing-based HBV genotyping [54] . STAR [49] uses all the known HBV sequences with known genotypes to construct a PSSM for each genotype A to H ( I and J are not well understood ) and then scores each read with respect to each genotype to have eight scores . We further transformed the scores into Z scores as in [49] . As recommended in [49] , if the maximum score of a read was above 2 . 0 , we predicted the genotype of the read as the one yielding the highest Z score . If the maximum score was below 2 . 0 , STAR uses a slide window of 150bps to find the genotype for each window . We considered the reads with Z score below 2 . 0 and having windows with distinct genotypes as recombinant reads . Consistent with the fact that the dominant HBV genotypes are B and C in China , over 95% of the reads are of the two genotypes or recombinants of B and C for all the samples with some small fractions of genotype A . The fraction of recombinant reads for 95% of the samples ( 132/139 ) was less than 5% , and only 3 samples had the fraction of recombinant reads above 20% . Therefore , we ignored the fractions of other genotypes and recombinant reads , normalized the fractions of B and C to sum to 1 , and calculated the fraction of genotypes B and C , respectively , for each sample . In addition to STAR , we also used another program jpHMM [30] for the identification of recombinant reads in NGS reads to see how different programs will affect our results . jpHMM uses a jumping hidden Markov model to identify recombinant reads between different genotypes . For each read , it identifies regions corresponding to a particular genotype . We defined a read to be a non-recombinant if a consecutive region of at least 400bps belongs to the same genotype while only at most 57bps belong to different genotypes . The details were given in the S1 Supplementary material section 2 . For each individual , we counted the number of occurrences of any word pattern of length k ( also called k-tuples , k-mers , k-grams ) in the NGS data . The relative frequency of a word of length k was its count divided by the total counts of all the words of length k for the individual . The distance between any pair of individuals was measured by the Manhattan distance between their corresponding frequency vectors . We constructed a distance matrix of all samples from the training set to see how the individuals cluster together . We chose the Manhattan distance because previous studies showed that it gave better clustering results than Euclidean distance for the clustering of genome sequences in many applications [55] . For different values of k , we used principal coordinate analysis ( PCoA ) to project the data onto two-dimensional space to see how the individuals group together . The basic idea of PCoA was to represent the data in the low dimensional space so that the distances between the samples in the low dimensional space are as close as possible to their true distances . In addition , we hierarchically clustered the individuals based on their word pattern frequencies . We used UPGMA to calculate the distance between any two clusters as the average of all the pairwise distances between the pairs of individuals from both clusters . We investigated the optimal approaches for predicting HCC status from the word pattern frequencies . Based on the PCoA and hierarchical clustering results , it can be seen that if the word pattern frequency vector of an individual is similar to others having HCC status , the individual is more likely to have HCC . Therefore , we first used the K-nearest neighbors ( KNN ) algorithm to predict HCC status , where K is the number of neighbors used for prediction . In KNN , an individual is predicted as having HCC if the fraction of HCC individuals among the top K most similar individuals according to word pattern frequency is above a threshold . We also used supporting vector machine ( SVM ) to predict HCC status using word pattern frequencies as features . For SVM , we had several kernel functions and parameters to choose from . We used linear kernel only here because for most cases it can work well and it has only one parameter C . For the parameter C , we used cross validation within the training set to choose C yielding the highest AUC ( area under the receiver operational characteristic curve ) value and used the parameter to construct a model for predicting the testing set . | HBV infection can lead to many liver complications including hepatocellular carcinoma ( HCC ) , one of the most common liver cancers in China . High-throughput sequencing technologies have recently been used to study the genotype sequence compositions of HBV infected individuals and to distinguish chronic HBV ( CHB ) infection from HCC . We used NGS to sequence the pre-S region of a large number of CHB and HCC individuals and designed novel word pattern based approaches to analyze the data . We have several surprising key findings . First , most HBV infected individuals contained mixtures of genotypes B and C sequences . Second , multi-dimensional scaling ( MDS ) analysis of the data showed that the first principal coordinate was closely associated with the fraction of genotype B ( or C ) sequences and the second principal coordinate was highly associated with the probability of HCC . Third , we also designed K-nearest neighbors ( KNN ) and support vector machine ( SVM ) based classifiers for CHB and HCC with high prediction accuracy . The results were validated in an independent data set . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"sequencing",
"techniques",
"biogeography",
"medicine",
"and",
"health",
"sciences",
"ecology",
"and",
"environmental",
"sciences",
"linguistics",
"pathology",
"and",
"laboratory",
"medicine",
"pathogens",
"carcinomas",
"population",
"genetics",
"microbiology",
"cancers",
"and",
"neoplasms",
"social",
"sciences",
"gastrointestinal",
"tumors",
"hepatitis",
"b",
"virus",
"liver",
"diseases",
"viruses",
"oncology",
"next-generation",
"sequencing",
"mathematics",
"forecasting",
"statistics",
"(mathematics)",
"artificial",
"intelligence",
"gastroenterology",
"and",
"hepatology",
"genome",
"analysis",
"molecular",
"biology",
"techniques",
"population",
"biology",
"genotyping",
"research",
"and",
"analysis",
"methods",
"computer",
"and",
"information",
"sciences",
"geography",
"genomics",
"medical",
"microbiology",
"mathematical",
"and",
"statistical",
"techniques",
"support",
"vector",
"machines",
"microbial",
"pathogens",
"hepatitis",
"viruses",
"phylogeography",
"molecular",
"biology",
"hepatocellular",
"carcinoma",
"machine",
"learning",
"viral",
"pathogens",
"earth",
"sciences",
"transcriptome",
"analysis",
"genetics",
"biology",
"and",
"life",
"sciences",
"semantics",
"physical",
"sciences",
"computational",
"biology",
"evolutionary",
"biology",
"dna",
"sequencing",
"statistical",
"methods",
"organisms"
] | 2018 | Deep sequencing of HBV pre-S region reveals high heterogeneity of HBV genotypes and associations of word pattern frequencies with HCC |
Much effort and interest have focused on assessing the importance of natural selection , particularly positive natural selection , in shaping the human genome . Although scans for positive selection have identified candidate loci that may be associated with positive selection in humans , such scans do not indicate whether adaptation is frequent in general in humans . Studies based on the reasoning of the MacDonald–Kreitman test , which , in principle , can be used to evaluate the extent of positive selection , suggested that adaptation is detectable in the human genome but that it is less common than in Drosophila or Escherichia coli . Both positive and purifying natural selection at functional sites should affect levels and patterns of polymorphism at linked nonfunctional sites . Here , we search for these effects by analyzing patterns of neutral polymorphism in humans in relation to the rates of recombination , functional density , and functional divergence with chimpanzees . We find that the levels of neutral polymorphism are lower in the regions of lower recombination and in the regions of higher functional density or divergence . These correlations persist after controlling for the variation in GC content , density of simple repeats , selective constraint , mutation rate , and depth of sequencing coverage . We argue that these results are most plausibly explained by the effects of natural selection at functional sites—either recurrent selective sweeps or background selection—on the levels of linked neutral polymorphism . Natural selection at both coding and regulatory sites appears to affect linked neutral polymorphism , reducing neutral polymorphism by 6% genome-wide and by 11% in the gene-rich half of the human genome . These findings suggest that the effects of natural selection at linked sites cannot be ignored in the study of neutral human polymorphism .
The neutral theory of molecular evolution [1] postulates that adaptive substitutions occur so rarely that they can be safely ignored in most studies in population genetics or molecular evolution . This view has dominated the field of molecular evolution for the past 40 years . However , the past 4–6 years have seen a strong challenge to this view . This challenge comes not only from numerous studies detailing specific cases of molecular adaptation in a number of organisms ( for example , see [2]–[8] ) but also , and most compellingly , from a number of studies that indicate that adaptation might be common on the genomic scale [9]–[17] . High rates of adaptation on the genomic scale have been inferred from the excess of substitutions in functional regions relative to neutral expectations . The neutral expectations are derived from the polymorphism data at functional and putatively neutral sites and the divergence at the neutral sites using the reasoning of the McDonald-Kreitman ( MK ) test [18] . The excess in the number of substitutions at functional sites over this expectation can be used to estimate the number of adaptive substitutions [10] , [19] . McDonald-Kreitman approaches can be modified to account for the presence of deleterious polymorphisms in the sample and the effects of demographic processes on polymorphism [10] , [20] , [21] . The approach can also be extended to estimate rates of adaptation in regulatory regions [12] , [22] . McDonald-Kreitman analysis indicates that adaptive evolution in functional regions might be common in a range of organisms . In Drosophila , it has been estimated that from 30 to 60% of amino acid substitutions and ∼20% of substitutions in non-coding regions are adaptive [10] , [11] , [16] , [23]–[25] . The rate appears similarly high in E . coli ( >56% of amino acid substitutions are adaptive ) [26] but not in Arabidopsis ( 0–5% of amino acid substitutions are adaptive ) [27] and yeast [28] . In humans , McDonald-Kreitman-based estimates have varied from zero to ∼35% of all amino acid substitutions being adaptive [15] , [29]–[33] . A recent estimate by Boyko et al [21] used information from the allele spectra of nonsynonymous and synonymous SNPs in human genes and the divergence with chimpanzee orthologs to estimate that ∼10% of amino acid substitutions between humans and chimpanzees have been fixed by positive selection . Thus , some of these studies suggest that adaptation might be fairly common in humans , although probably substantially less common than in Drosophila or E . coli . McDonald-Kreitman approaches are very powerful at detecting positive selection , however , they can be misleading for a variety of reasons [15] , [34] , [35] . For example , if the strength of purifying selection over the evolutionary period separating two species has been different than it is in the present , McDonald-Kreitman-based approaches can either over- or underestimate the rate of adaptive evolution . As these estimates do not provide consistent answers about the prevalence of adaptation in humans and because they can be misleading under plausible demographic scenarios , reaching more reliable conclusions about the importance of adaptations in humans requires the investigation of other signatures of positive selection . An adaptive substitution reduces the level of polymorphism at neutral sites in its vicinity in a phenomenon known as a selective sweep [36] . The width of the region in which the polymorphism is reduced is inversely proportional to the local recombination rate and directly proportional to the selection coefficient associated with the adaptive substitution [37]–[39] . The reduction of polymorphism is transient and the levels of polymorphism are expected to recover within roughly Ne generations [40] . In addition to the reduction of the level of polymorphism , recurrent selective sweeps may also generate other signatures such as ( i ) an overabundance of low-frequency alleles [41] , [42] , ( ii ) a greater proportion of high-frequency derived alleles [43] , [44] , ( iii ) unusual haplotype structures [45] , [46] . A number of these expectations have been used to define signatures of positive selection for genome-wide scans for recent adaptation in humans: i . e . , the detection of candidate regions that are likely to be experiencing a selective sweep at present or that have experienced one recently . For example , Nielsen et al . [30] and Kelley et al [47] used the deviation of the allele frequency spectrum from its background characteristics to detect candidate regions that may have experienced a sweep; several other methods have used summaries of haplotype structure and their deviation from the background to detect candidate regions that are undergoing a selective sweep [45] , [46] . Genomic scans for positive selection are primarily used to choose candidate regions for future investigation , but their application to the quantification of positive selection or even the establishment of its prevalence is problematic . To quantify the extent of positive selection based on the deviations of these signatures from the background requires a prior expectation about the likelihood of observing them under neutrality . These expectations , however , may be sensitive to the effects of non-equilibrium demography [44] , [48]–[50] . As a result , it is difficult to generate robust a priori expectations for these statistics under neutrality . Therefore , scans for positive selection do not , by themselves , provide reliable quantification of the extent of positive selection in humans or establish that positive selection is prevalent in humans . To evaluate whether selective sweeps are common in the human genome , we require signatures that are unlikely to be generated by demography alone . The effects of recurrent selective sweeps ( RSS ) should be stronger in the regions of lower recombination and in regions of more frequent and selectively consequential adaptation . In Drosophila , for example , the level of neutral polymorphism is positively correlated with the recombination rate [51]–[53] and negatively with the rate and number of nonsynonymous substitutions in a region [9] , [13] , [14] . These correlations , which are expected under models of RSS but should not be generated by demography alone , support the notion of high rates of adaptation in these taxa . Despite several compelling examples of adaptations , clear genome-wide signatures of RSS have been difficult to detect in humans . A relationship of diversity and recombination has been reported , but was attributed primarily to an association between recombination and mutation processes rather than to the effects of selected at linked sites [54]–[57] , with the possible exception of telomeric and centromeric regions [58] . In turn , the relationships between levels of polymorphism and functional divergence have not yet been examined . If the recent MK estimates of the rate of adaptive evolution are correct and approximately 10% of amino acid substitutions are adaptive [21] , we should expect to see a substantial number of recent selective sweeps in the polymorphism data . Indeed , ∼7×104 amino acid differences between human and chimpanzee proteins [31] have accumulated over the past ∼14 million years . If 10% of these have been adaptive , then we can estimate that ∼7×103 adaptive amino acid substitutions have taken place over ∼14 million years . Assuming a constant rate of adaptation , this translates into ∼100 adaptive amino acid substitutions that occurred during the past Ne generations ( Ne = ∼2×105 years ) [59] . Moreover , if regulatory adaptations are common as well , then hundreds of recent selective sweeps should be detectable in the human polymorphism data . With these considerations in mind , we analyze genomic patterns of nucleotide polymorphism , recombination , functional density and functional divergence in humans using two independent , genome-wide SNP datasets . Consistent with the expectations of positive selection , we detect a positive correlation between levels of neutral polymorphism and recombination rate and a negative correlation between levels of nucleotide polymorphism and both functional density and functional divergence . These correlations remain intact after controlling for a number of possible covariates . The evidence is consistent with positive selection in both regulatory and protein-coding regions . We consider alternative explanations for these findings and argue that , in addition to recurrent selective sweep , only background selection ( BS ) ( loss of neutral variants due to hitchhiking with linked deleterious mutations ) can possibly generate most of these patterns . Hitchhiking of neutral polymorphisms with linked selected variants—either due to recurrent positive selection or background selection or possibly both—appears to be a substantial force determining levels of neutral polymorphism in the human genome .
To study the effects of RSS , we separate the genomic sequences into two mutually exclusive sets of sequences: “functional” ( genic and regulatory ) and “nonfunctional” . Both sets of sequences are taken only from the internal parts of autosomes; specifically , we remove all sequences located within 10 Mbp of a telomere or a centromere . We further remove all sequences that cannot be aligned with the chimpanzee genome [31] . The functional set is composed of several types of sequences ( see Material and Methods ) . First , it contains all the genic regions , specifically those that ( i ) encode exons or are located within 1 kb of any predicted exon and ( ii ) are located within 5 kb from the starting and ending position of transcripts of protein-coding genes . Because many functional , noncoding sequences are located far from genes in the human genome [60]–[62] , we also take all the sequences that can be aligned between primates and zebrafish; sequences that can be aligned over such large evolutionary distances are very unlikely to be unconstrained [63] ( see Materials and Methods ) . The nonfunctional set contains all other sequences except for the repetitive sequences that are filtered out using RepeatMasker [64] . We remove repetitive regions because both alignment and SNP discovery are more problematic in such regions [65] . Hereafter , we will refer to the sequences in the primarily nonfunctional set ( totaling ∼1 , 080 Mbp ) as “neutral” sequences for brevity . We use two SNP datasets: ( i ) ∼1 . 2 million Perlegen [66] “A” SNPs discovered using Perlegen chip technology [67] in a panel of 71 individuals of mixed ancestry [68] and ( ii ) ∼2 . 0 million SNPs discovered in the diploid sequence of James Watson [69] ( see Materials and Methods ) . In the remainder of the paper , we show the results derived from the analysis of the Perlegen dataset . The results derived from the analysis of the Watson SNPs are shown in the Supplementary Materials . All of the conclusions in the paper are supported by the analysis of either dataset . We measure the level of neutral nucleotide variation in a genomic window using the number of SNPs within the neutral regions divided by the total number of neutral sites ( θneu ) in a window ( see Materials and Methods ) . This measure is proportional to the conventional Watterson's θ [70] . In the remainder of the paper , all measurements are carried out over 400 kb windows . We have also carried out all of the analyses with two other window sizes , 200 and 600 kb; none of the conclusions change depending on the window size ( Table S1 , S2 and S3 , Figures S7 , S8 , S9 , and S10 ) . The level of neutral polymorphism ( θneu ) depends both on the average time to coalescence within a particular genomic region and on the local constraint and mutation rate . For the purposes of detecting signatures of RSS , variation in constraint and mutation rate generates noise . We assess variability in constraint and mutation rate by measuring divergence per neutral site ( dneu ) within the neutral regions between the human and chimpanzee genomes ( see Materials and Methods ) . We detect a positive correlation between dneu and θneu ( Table 1 ) , confirming that , as expected , constraint and/or mutation rate vary across the human genome . We control for the variation in neutral mutation rate either by carrying out partial correlations with dneu or by using a normalized measure of neutral variation , , where #SNPneu stands for the number of SNPs found in the neutral regions and Dneu stands for the number of divergent sites within neutral regions between humans and chimpanzee genomes . Pneu and θneu also correlate significantly with repeat density ( RD ) and GC content ( GC ) ( Table 1 , Table S1 ) . Finally , in the case of the Watson data , we further carry out controls for the depth of sequence coverage ( Table S4 ) . The overall effect of RSS on the regional levels of neutral polymorphism should depend on ( i ) the regional rate of recombination , ( ii ) the number of recent sweeps ( the rate of RSS ) , and ( iii ) the strength of positive natural selection associated with a typical adaptive substitution ( the strength of RSS ) . The levels of neutral polymorphism across the genome should correlate positively with the rate of recombination and negatively with the rate and the strength of RSS . We take estimates of recombination rate from Myers et al . [71] , who used a statistical approach to infer recombination rates from linkage disequilibrium data in humans; these rates have been shown to be highly reliable by comparison to pedigree data [72] . The levels of neutral polymorphism measured by both θneu and Pneu increase with the recombination rate ( Figures 1 , S1 , and S2 ) . The correlation remains when we control for possible confounders such GC content ( GC ) , repeat density ( RD ) , and divergence at neutral sites ( dneu ) separately ( Table 2S ) or together ( Pearson r ( θneu , RR|GC , RD , dneu ) = 0 . 254 , Pearson r ( Pneu , RR|GC , RD ) = 0 . 209 , P<0 . 001 in both cases ) . Under a model of RSS regions experiencing more frequent or stronger selective sweeps should show lower levels of neutral polymorphism . Because positive selection should be more prevalent in regions of greater functional density , RSS is expected to generate a negative correlation between the degree of functional density and the level of neutral polymorphism . We measure functional density in two complementary ways . First , in each 400 kb window , we count the number of protein-coding codons ( FDn ) as a proxy of protein-coding density . In addition , we count the number of nongenic sites that can be aligned between primates and zebrafish ( FDx ) as a proxy of the number of conserved noncoding sites ( CNRs ) ( see Materials and Methods for details ) . Consistent with the predictions under RSS , there are strongly negative correlations between either measure of functional density ( FDn , FDx ) and measures of neutral variability ( Figures 2 , S3 , S4 , and Tables 2 , S3 ) . After controlling for GC content ( GC ) , recombination rate ( RR ) , repeat density ( RD ) , and divergence at putatively neutral sites ( dneu ) ( in the case of θneu ) the correlations become substantially weaker but do remain statistically significant ( Tables 2 , 3S ) . The correlations between FDn and both θneu and Pneu remain significant after we control for FDx; and similarly , the correlations between FDx and both θneu and Pneu are still significant when we control for FDn ( Tables 2 , 3S ) . The number of differences between humans and chimpanzee genomes at functional regions is likely to be a more direct proxy of the rate of positive selection than the functional density . Consistent with the expectations of RSS , we detect lower levels of θneu ( Pneu ) in regions of higher Dn ( the count of divergent amino acid coding sites ) or Dx ( the count of divergent sites within conserved noncoding regions ) ( Tables 3 , S3 , Figures 3 , S5 , S6 ) . These correlations remain significant when we control for GC content ( GC ) , recombination rate ( RR ) , repeat density ( RD ) , and functional density ( FDn , FDx , or both ) ( Tables 3 , S4 ) . The correlations between either Dn or Dx and either of the two measures of neutral variation ( θneu or Pneu ) remain statistically significant when we control/correct for the other measure of functional divergence ( i . e . control for Dn in the case of correlations of neutral diversity with Dx and , similarly , control for Dx in the case of correlations of neutral diversity with Dn ) ( Tables 3 , S4 ) .
All SNP datasets suffer from ascertainment biases during the SNP discovery phase that can systematically under- or overestimate numbers of SNPs in particular genomic regions or at particular types of sites . We address this concern by using two very different SNP datasets that are likely to have different ascertainment biases: ( i ) the high quality ( type A ) SNPs from the Perlegen dataset [66] and ( ii ) SNPs discovered in the sequenced diploid genome of James Watson [69] . The type A SNPs were discovered using Perlegen oligo hybridization chip technology in a panel of 71 individuals of mixed ancestry [66] . This set is biased against SNPs located in repetitive regions , given that it is difficult to design uniquely hybridizing oligonucleotides in such repetitive regions [66] . The diploid genome of James Watson was sequenced using the 454 technology and does not suffer from the same technological problems as the Perlegen oligonucleotide chip hybridization technology . We obtain very similar results using both datasets , which argues that it is unlikely that specific ascertainment biases are responsible for the observed patterns . In addition , we also used the density of the repeats , GC content and functional density as variables in our statistical analyses and showed that all of the signatures of genetic hitchhiking in our data are robust to statistical controls for these variables . The depth of coverage in the Watson sequencing data also does not noticeably affect any of the detected correlations ( Table S4 ) . The demographic history of human populations in general , and specifically of the populations that have been used for SNP discovery and SNP typing in the Perlegen data , is very complex . Bottlenecks , quick population growth and complex patterns of admixture ( for example in the African–American population ) are expected to perturb levels of neutral polymorphism across the genome . Collectively , we will denote these forces as “demography” . The effects of recent demography undoubtedly generate much variation in neutral polymorphism; however , the correlations that we observe are likely to be weakened and unlikely to be generated by the demographic processes alone . For instance , the lower levels of neutral polymorphism in the regions that have large numbers of the protein-coding ( Dn ) and functional noncoding ( Dx ) differences are hard to explain by demography; demographic events cannot easily affect the longer-term rates of functional divergence that have been accumulating for ∼10–14 million years between chimpanzees and humans [73] . On the other hand , it is clear that demography needs to be taken into account in order to use the detected signatures to evaluate the strength of hitchhiking in the human genome . Some of the variation in levels of polymorphism in the sequences that we use to measure levels of neutral polymorphism could be due to the variability in the rates of mutation and levels of selective constraint . We measure levels of neutral variation in the sequences that are less likely to be under selective constraint: they are noncoding , located far from exons , and cannot be aligned with distantly related species such as zebrafish . Nevertheless , some residual variation in constraint is likely to remain . Indeed , the positive correlation between our measures of the levels of neutral polymorphism ( θneu ) and divergence ( dneu ) ( Table 1 and 1S ) suggests that mutation rates and/or levels of constraint vary systematically in these regions . It is therefore important to control for the variation in the levels of selective constraint and mutation rate; we do so by using the levels of divergence ( dneu ) as a variable in partial correlation analyses or by using the measure Pneu ( ) . The levels of neutral variation correlate strongly with recombination rate , functional density and functional divergence after controlling for neutral divergence suggesting that these correlations are not due to the variation in mutation rate or constraint Partial correlations may not remove all of the effects of the variation in mutation rate and constraint , however . The variation in selective constraint among neutral regions should have a stronger effect on the levels of neutral divergence ( dneu ) than on the levels of neutral polymorphism ( θneu ) because deleterious mutations have a greater chance of segregating in the population than to become fixed . This implies that if the negative correlation between θneu and levels of functional density were entirely due to the variation in selective constraint ( specifically higher remaining constraint in regions of higher functional density ) , then controlling for divergence ( dneu ) should make the partial correlation between neutral polymorphism ( θneu ) and functional density positive . Yet we see the opposite: the correlations between Pneu and functional density and the partial correlation between θneu and functional density with respect to dneu both remain strongly negative . This suggests that the variation in selective constraint is unlikely to generate the correlations between levels of neutral variation and recombination rate , functional density and functional divergence that we see in this study . On the other hand , variability in mutation rates might contribute to some of the observed patterns . Specifically , the positive correlation between neutral diversity and rates of recombination could be due to the mutagenic effects of recombination . Because rates of recombination at local scales ( although not necessarily at the 200–600 kb scales relevant to this study ) evolve fast [55] , [56] , [74]–[77] , mutagenic effects of recombination should have more pronounced effects on the levels of polymorphism than on the levels of divergence . If so , controlling for neutral divergence ( dneu ) may not entirely account for the higher mutation rates produced by recent recombination [55] . Mutagenic effects of recombination are expected to affect levels of polymorphism proportionately to the rate of recombination in the area , whereas hitchhiking ( RSS or BS ) is expected to affect levels polymorphism in regions of very low recombination much more substantially [78] . We observe a mostly linear effect of recombination on divergence ( dneu ) suggestive of the mutagenic effect of recombination and further arguing that the regional recombination rates at the level of our analysis ( 200 to 600 kb ) do not evolve as fast as the location of recombination hotspots . In contrast , the effect of recombination on the levels of polymorphism ( θneu and Pneu ) is curvilinear , with most of the effect limited to the regions of the lowest recombination rates ( Figure 1 and S1 ) . Indeed , when we split the data by the median value of recombination rate ( RR = 1 . 040 cM/MBp ) , the correlation between the levels of neutral divergence ( dneu ) and recombination rate ( RR ) for the two halves of the data are of similar strength ( r ( dneu , RR|RR<1 . 040 ) = 0 . 197 and r ( dneu , RR|RR>1 . 040 ) = 0 . 220 ) . However , the correlations between recombination rate and levels of polymorphism ( θneu or Pneu ) are much stronger in the low recombination regions than in the high recombination regions ( ( r ( θneu , RR|RR<1 . 040 ) = 0 . 249 versus r ( θneu , RR|RR>1 . 040 ) = 0 . 045; r ( Pneu , RR|RR<1 . 040 ) = 0 . 194 versus r ( Pneu , RR|RR>1 . 040 ) = −0 . 0241 ) . These considerations suggest that most of the positive correlation between recombination rates and levels of neutral polymorphism , and especially the reduction at lower recombination rates , is caused by some form of hitchhiking . These results are consistent with the findings of Hellman et al [58] who detected lower levels of polymorphism in the areas of low recombination close to centromeres and telomeres . Note that in our study we explicitly excluded telomeric and centromeric regions ( see Materials and Methods ) , making our findings complementary to those of Hellman et al [58] . Background selection ( BS ) is the process of hitchhiking of neutral or weakly deleterious polymorphism with linked strongly deleterious polymorphisms [79]–[82] . BS should be more efficacious and lead to lower levels of neutral polymorphism in regions of lower recombination . It is thus quite possible that the positive correlation between neutral polymorphism and recombination rate is due in part to BS . In addition , BS should be stronger in the more constrained genomic regions because such regions should experience higher rates of deleterious mutation ( e . g . [58] ) . Therefore BS is likely to contribute to the negative correlation between levels of neutral polymorphism and functional density as well . Because regions of higher functional density also exhibit higher rates of functional divergence ( Tables 1 and S1 ) , BS could contribute to the negative correlation between levels of neutral polymorphism and functional divergence as well . It is less clear whether BS could generate the negative correlation between the levels of neutral polymorphism and functional divergence after controlling for levels of functional density ( Tables 3 , S4 , Figure S6 ) . Two regions of equal functional density can differ in the strength of BS if they differ in the rate of deleterious mutations in the functional sequences . The higher level of deleterious mutations should lead to stronger BS and therefore lower levels of polymorphism in the linked neutral sequences . At the same time , the higher rate of deleterious mutations is likely to come at the expense of neutral mutations at functional sites and thus should lead to lower levels of protein and regulatory divergence . The reduction of neutral mutation rate in the regions of higher deleterious mutation should lead to a positive correlation between levels of neutral polymorphism and functional divergence after controlling for functional density—the opposite of what is seen . On the other hand , the increase in the rate of fixation of weakly deleterious mutations , also expected in the regions of stronger BS , counteracts the reduction of the rate of functional divergence due to the reduction of neutral mutation rate . The combined effect is difficult to estimate given that we do not have information about the distribution of the rates of mutations of different selective effects along the genome . There is another pattern we observed that is not naturally predicted by BS . The correlations between functional density ( FDn or FDx ) and neutral polymorphism weaken very substantially and in some cases become nonsignificant when we control for functional divergence at replacement ( Dn ) and conserved noncoding sites ( Dx ) ( Table 2 ) . Functional density is likely to be a better of proxy of regional constraint than functional divergence . If BS is indeed the dominant force in the generation of the observed patterns , we might have expected correlations between neutral polymorphisms with FDn and/or FDx to be the most robust . Without a better understanding of the distribution of selective effects and rates of new mutations , we cannot reject the possibility that BS contributes substantially to all of the detected patterns . It appears , however , that only specific distributions of selective effects of new mutations would generate all of the observed patterns . Whether such a distribution exists in principle and whether the distribution of selective effects of human mutation satisfies these requirements in fact remains to be determined . The arguments above suggest strongly that some form of hitchhiking , either BS or RSS , needs to be invoked to explain the results presented in this paper . These results also suggest that natural selection at both coding and regulatory sites affect linked neutral polymorphism . This is because the measures of the rate of functional evolution at coding and regulatory sites appear to influence levels of neutral polymorphism independently of each other . Specifically , divergence at coding sites and divergence at regulatory sites correlate negatively with the levels of neutral polymorphism after controlling for each other and for the variation in levels of functional divergence ( Table 3 , S3 ) . To the extent that this is due to recurrent adaptation selection at both coding and regulatory sites , this would echo results of McDonald-Kreitman analyses of adaptation in Drosophila [12] . Levels of neutral polymorphism correlate stronger with divergence at coding than at non-coding regions , possibly implying that either a higher proportion of nonsynonymous changes are adaptive compared to changes in regulatory regions or that the nonsynonymous adaptations have higher selective coefficients . It is also possible and even likely that Dx is a noisier measure than Dn due to greater difficulties in identification of regulatory regions and the noise in estimating Dx due to misalignments . This pattern may also be due to different rates or distributions of the selective effects of deleterious mutations located in coding and regulatory regions , leading to varying effects of BS on linked neutral polymorphism and functional divergence . These results can also be used to assess the importance of hitchhiking ( either RSS or BS ) in affecting patterns of neutral polymorphism . The levels of neutral polymorphism appear to be ∼50% lower in the regions of high Dn or Dx ( Figures 3 , S5 ) relative to the regions of zero functional divergence ( Dn or Dx = 0 ) . If we assume that this effect is entirely due to hitchhiking , then by using the observed correlation between θneu and Dn , we estimate that the levels of polymorphism genome-wide are reduced by 6% genome-wide ( Materials and Methods ) . This reduction is much more pronounced in the more gene-rich regions . For instance , in the 50% of the most gene-rich regions ( regions that have greater than the median density of codons ( FDn ) ) , the neutral polymorphism is reduced by 11% , while in the regions that contain 50% of the genes ( regions that have greater than the mean density of codons ( FDn ) ) , the neutral polymorphism is reduced by 13% . It is clear that hitchhiking has left a significant imprint on the patterns and levels of neutral variability in the human genome and that the effects of natural selection at linked sites cannot be ignored in the analysis of polymorphism data in humans . The challenge for the future is to use these signatures to answer a number of outstanding questions . What are the selective effects and genomic distributions of adaptive and deleterious changes responsible for RSS and BS ? What is the biological nature of these changes ? What is the relative importance of RSS and BS ? Can we estimate parameters of adaptive evolution in the presence of BS ? The availability of whole genome sequences in a large number of humans may provide the necessary data to answer these questions . What is needed now are the models and tools to harness these data to provide a cogent picture of the effects of natural selection on human genome and human evolution .
All analyses have been carried out using two SNP data sets—Perlegen data [66] and Watson data [69] . Perlegen data were downloaded from http://genome . perlegen . com . These data were annotated based on the NCBI build 35 of the human genome sequence . We updated all the genomic positions of the SNPs to match the latest NCBI build 36 , according to the rs number of SNPs in the dbSNP build 127 . During the processing , 1 , 361 SNPs were discarded because they could not be uniquely mapped to the human genome . Perlegen data contain three classes of SNPs: ( A ) array-based genomic resequencing , ( B ) reliable external SNP collections , and ( C ) unvalidated , lower confidence sources ( see Supplementary text of [66] ) . We excluded class B and C SNPs and retained 1 , 235 , 057 class A SNPs located on autosomes for our analysis . The Watson data were downloaded from http://jimwatsonsequence . cshl . edu/ . The genome of James Watson was sequenced at 6× coverage using 454 Life Sciences Technology [69] and matched to the human genome project's published reference sequence [83] . In the Watson DNA sequence , heterozygous sites , in which each site was sequenced multiple times and both forms of the base were found in the diploid genome , were ascertained as SNPs . Homozygous sites of Watson's DNA sequence that have been sequenced multiple times and that differ from the reference sequence of the human genome were also ascertained as SNPs . In total our Watson dataset consisted of 2 , 020 , 767 SNPs . Whole-genome alignments of human ( H ) , chimpanzee ( C ) , and zebrafish ( Z ) sequences were obtained from the Ensembl compara database [84] through the Ensembl Application Program Interfaces ( APIs ) . We defined the “neutral” genomic regions of the human genome if the regions were: ( 1 ) H-C aligned , ( 2 ) not H-C-Z aligned , ( 3 ) located at least 5 kb away from the starting and ending position of transcripts of protein-coding genes and at least 1 kb away from any exons , ( 4 ) located on autosomes at least 10 Mbp away from the boundaries of centromeres and the ends of telomeres , ( 5 ) not located in the simple repetitive regions of the human genome . The chromosomal coordinates of exons , transcripts and simple repeats were obtained from the finished and annotated human chromosome sequence from the Ensembl database ( build 36 ) . Neutral divergence was assessed from H-C alignments . The accuracy of estimation of neutral divergence may be influenced by the misaligned sequences . Indeed , we discovered some short ( 2 kb on average ) neutral genomic regions having extremely high levels of divergent sites , which may result from misalignments ( data not shown ) . To minimize the possible influence of misalignments , we only counted “isolated” substitutions that are flanked by two monomorphic positions on each side ( i . e . no substitutions or SNPs were mapped to these sites ) . We denoted the number of isolated substitutions between human and chimpanzee sequences as Dneu , and the number of isolated substitutions per neutral site , dneu . To measure neutral polymorphism , we counted the number of SNPs in neutral regions and denoted the number of SNPs per site as θneu . Alternatively , we measured neutral polymorphism with . Data manipulation was done using Matlab functions based on PGEToolbox [85] and MBEToolbox [86] . We used four metrics as proxies of the rate of adaptive evolution for a given region in the human genome . Functional density was measured using FDn , the number of codons , and FDx , the number of aligned bases in the H-C-Z three-way alignments . Functional divergence was measured using Dn , the number of codons involved in nonsynonymous substitutions between H-C orthologous gene pairs , and Dx , the number of H-C substitutions in H-C-Z alignments that are located in noncoding human genomic regions . For each pair of genes , the amino-acid sequences were extracted and aligned using CLUSTALW [87] with the default parameters . The corresponding nucleotide sequence alignments were derived by substituting the respective coding sequences from the protein sequences . The synonymous substitution rate ( Ks ) was then estimated by the maximum-likelihood method implemented in the CODEML program of PAML [88] . Insertions and deletions within alignments were discarded . Poorly aligned orthologous pairs , as indicated by Ks>5 , were excluded . The codons containing nonsynonymous substitutions were mapped back onto the human genome and positions were recorded . For simplicity we counted the numbers of codons causing amino-acid changes instead of the numbers of single nucleotide replacement substitutions . In calculation of Dx , we excluded “tri-allelic” sites where the bases of H , C and Z all differ from each other . We used 400 kb ( as well as 200 and 600 kb ) sliding window with a step of 100 kb to scan along the human genome . For each window , two measures of neutral polymorphism ( θneu and Pneu ) and four proxies of the rate adaptive evolution ( FDn , FDx , Dn , and Dx ) were estimated . To reduce noise arising from small sample size , we also discarded the windows with Dneu<500 and the ones with the total amount of “neutral” sequence less than 2 kb . 22 , 553 400 kb windows have been used for the correlation analysis . Spearman rank correlation or Kendall's correlation coefficients have been calculated in all cases . To visualize correlations between variables , we used scatter plots with regression lines superimposed . We also pooled the data points of neutral polymorphism by the values of the proxy of adaptation under consideration ( e . g . Dn ) . To do this , we ranked all the data points of the neutral polymorphism by the values of the proxy and then pooled them into 100 bins such that each bin had equal size ( i . e . , 1% ) of the data points . We then computed average values of the proxies of adaptation and the average value of neutral polymorphism for each bin , and superimposed them onto the scatter plots . To control for confounding variables , we calculated Spearman partial correlation coefficients between variables X and Y controlling for Z , using the function partialcorr in the Matlab statistic toolbox . Recombination rate estimated by using the coalescent method of [89] were downloaded from http://hapmap . org/downloads/recombination/ . The density of simple repeats was computed as the proportion of bases of simple repeats in the given region . Chromosomal coordinates of simple repeats in the human genome , identified by RepeatMasker [64] , were obtained from the UCSC genome browser [90] . We also calculated the partial correlation coefficients between variables X and Y by calculating the correlation between the two sets of residuals formed by two linear models X∼Z and Y∼Z ( see also [91] ) where Z stands for either one or a series of variables . The distribution of Dn , Dx , FDn , and FDx values is approximately exponential , which is a problem in a least squares linear model framework in controlling for a third variable , Z . The linear model used to regress out Z is sensitive to the highly non-normal distribution of variables , and the residuals will be highly non-normal , making the results difficult to interpret . Therefore , we quantile-normalized values , replacing the original estimates with their theoretical quantiles based on a normal distribution . Then , we fitted linear models , using as the response variable quantile-normalized Dn , Dx , FDn , or FDx , and using as the predictor variables various combinations of recombination rate , GC content , and the density of simple repeats . Estimation of the effect of hitchhiking on the level of neutral polymorphism was calculated using the regression between θneu on Dn , using the formulawhere q is the reduction of polymorphism due to hitchhiking , i is a window count for the subsets of windows used in the analysis ( e . g . FDn>median ( FDn ) ) , b is the intercept of the regression of θneu on Dn . | There is much reported evidence for positive selection at specific loci in the human genome . Additional papers based on comparisons between the genomes of humans and chimpanzees have also suggested that adaptive evolution may be quite common . At the same time , it has been surprisingly hard to find unambiguous evidence that either positive or negative ( background ) selection is affecting genome-wide patterns of variation at neutral sites . Here , we evaluate the prevalence of positive or background selection by using two genome-wide datasets of human polymorphism . We document that levels of neutral polymorphism are substantially lower in the regions of ( i ) higher density of genes and/or regulatory regions , ( ii ) higher protein or regulatory divergence , and ( iii ) lower recombination . These patterns are robust to a number of possible confounding factors and suggest that effects of selection at linked sites cannot be ignored in the study of the human genome . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"evolutionary",
"biology/human",
"evolution",
"computational",
"biology/genomics",
"evolutionary",
"biology/genomics"
] | 2009 | Pervasive Hitchhiking at Coding and Regulatory Sites in Humans |
Onchocerciasis ( river blindness ) is a parasitic disease transmitted by blackflies . Symptoms include severe itching , skin lesions , and vision impairment including blindness . More than 99% of all cases are concentrated in sub-Saharan Africa . Fortunately , vector control and community-directed treatment with ivermectin have significantly decreased morbidity , and the treatment goal is shifting from control to elimination in Africa . We estimated financial resources and societal opportunity costs associated with scaling up community-directed treatment with ivermectin and implementing surveillance and response systems in endemic African regions for alternative treatment goals—control , elimination , and eradication . We used a micro-costing approach that allows adjustment for time-variant resource utilization and for the heterogeneity in the demographic , epidemiological , and political situation . The elimination and eradication scenarios , which include scaling up treatments to hypo-endemic and operationally challenging areas at the latest by 2021 and implementing intensive surveillance , would allow savings of $1 . 5 billion and $1 . 6 billion over 2013–2045 as compared to the control scenario . Although the elimination and eradication scenarios would require higher surveillance costs ( $215 million and $242 million ) than the control scenario ( $47 million ) , intensive surveillance would enable treatments to be safely stopped earlier , thereby saving unnecessary costs for prolonged treatments as in the control scenario lacking such surveillance and response systems . The elimination and eradication of onchocerciasis are predicted to allow substantial cost-savings in the long run . To realize cost-savings , policymakers should keep empowering community volunteers , and pharmaceutical companies would need to continue drug donation . To sustain high surveillance costs required for elimination and eradication , endemic countries would need to enhance their domestic funding capacity . Societal and political will would be critical to sustaining all of these efforts in the long term .
To reduce disease prevalence to a locally acceptable level ( i . e . , microfilaria prevalence≤40% or community microfilarial load≤5mf/s [3] ) , all endemic African countries implement annual CDTi in hyper- and meso-endemic areas , and after at least 25-years of CDTi , conduct epidemiological surveillance to confirm that CDTi can be safely stopped ( former OCP projects having implemented regular surveillance continue their surveillance strategies ) . To reduce the incidence of infection to zero in a defined area , all endemic African countries except those with epidemiological and political challenges implement annual or biannual CDTi , and conduct regular active epidemiological and entomological surveillance to evaluate epidemiological trends , to decide a proper time to stop CDTi , and to detect and respond to possible recrudescence . To reduce the incidence of infection to zero in Africa , which would lead to global eradication , all endemic African countries implement not only annual or biannual CDTi but also locally tailored treatment strategies to deliver sustainable treatments to areas with operational challenges , and implement regular active epidemiological and entomological surveillance to evaluate epidemiological trends , to decide a proper time to stop CDTi , and to detect and respond to possible recrudescence . We estimated financial resources and societal opportunity costs for endemic African countries ( Table 1 ) associated with the control , elimination , and eradication scenarios to support policymakers’ and donors’ informed decisions and provide a basis for further economic evaluation of the elimination and eradication of onchocerciasis .
We obtained 2012 budgets from APOC , approved for onchocerciasis CDTi , that cover 67 of all 112 ongoing projects ( as of November 2013 ) in sub-Saharan Africa . All budget documents include information on the unit cost and the unit quantity of each resource , demography , available human resources ( community health workers , community volunteers ) , and funding from the ministry of health , APOC , and non-governmental organizations . These data were used as the main sources to estimate financial costs . To estimate economic costs , agriculture value added per worker was used as an opportunity cost of community volunteers’ unpaid time [17] , considering most volunteers are farmers in remote rural areas [18] . The opportunity cost of donated ivermectin was $1 . 5054 per treatment ( three 3mg-tablets ) , based on Merck’s suggested drug price of $1 . 5 per treatment before the donation was decided [19] and on the insurance and freight cost of $0 . 0018 per tablet [20] . For projects with missing unit costs , we used the national average if relevant unit costs were available; otherwise , the regional average ( Table 2 ) across available national averages for endemic African countries . For the countries that did not have agriculture value added per worker , we used the regional average for sub-Saharan Africa [17] . For projects with missing data for the determinants of unit quantities ( e . g . , the ratio of health workers over population , the ratio of volunteers over population ) , we used the national average if relevant data were available; otherwise , the regional average across available national averages for endemic African countries . Unit costs and the determinants of unit quantities at the country and regional levels are included in S1 Text . We conducted sensitivity analysis to assess the robustness of results to parametric uncertainties . The parameters included either cost items for which unit costs were missing for more than one third of total projects or total countries with available budgets . Also the parameters included the time-variant determinants of unit quantities: population living in endemic areas , the number of required treatments ( determined by population , treatment coverage linked to required treatment duration , and possible delay in starting and ending treatments ) , the number of required community volunteers ( determined by population and the ratio of community volunteers over population ) , and the number of required community health workers ( determined by population and the ratio of community health workers over population ) . We conducted one-way sensitivity analysis to examine the impact of parameters related to CDTi performance , the cost items with high uncertainty , and discount rates on total costs . We conducted multivariate probabilistic sensitivity analysis ( PSA ) to examine the joint effects of uncertainties about all selected variables on total costs . For PSA , we attached statistical distributions to the selected cost items and the determinants of unit quantities , and fitted to relevant data . S1 Text describes the methodological details of the sensitivity analysis .
Total financial and economic costs would be concentrated in the early stage during which treatments are scaled up to remaining endemic areas , and decrease as the treatment phase nears the end ( Fig 2 ) . In endemic African regions , total financial and economic costs over the period 2013–2045 would be $4 . 3 billion ( 95% central range from multivariate PSA: $3 . 9 billion[bn]–$5 . 0bn ) for the control scenario , $2 . 9 billion ( $2 . 6bn–$3 . 4bn ) for the elimination scenario , and $2 . 7 billion ( $2 . 4bn–$3 . 2bn ) for the eradication scenario . That is , switching from control to elimination and eradication would lead to cost-savings of $1 . 5 billion ( $1 . 0bn–$1 . 9bn ) and $1 . 6 billion ( $1 . 2bn–$2 . 1bn ) , respectively ( S1 Fig ) . The eradication scenario would lead to cost-savings of $144 million ( -$25 million[M]–$462M ) as compared to the elimination scenario . Unit financial and economic cost per treatment for the control scenario would decrease from $2 . 5 to $0 . 9 over 2013–2045 . For the elimination scenario , it would decrease from $2 . 5 to $1 . 3 until 2035 , and increase to $1 . 6 afterwards . For the eradication scenario , it would decrease from $2 . 5 to $1 . 5 over 2013–2030 , and increase to $3 . 9 afterwards until the end of the treatment phase in endemic African regions ( Fig 3 ) . Total financial costs over the period 2013–2045 would be $640 million ( $572M–$711M ) for the control scenario , $650 million ( $574M–$751M ) for the elimination scenario , and $649 million ( $566M –$745M ) for the eradication scenario ( Fig 4 ) . That is , the total financial costs associated with the elimination and eradication scenarios are slightly lower than those associated with the control scenario; however , these cost differences are not robust to sensitivity analysis ( S2 Fig ) . The main difference between scenarios is the proportion of surveillance costs in total costs . Total surveillance costs over 2013–2045 would increase from 7% ( $47M ) of total financial costs under the control scenario to 33% ( $215M ) and 37% ( $242M ) under the elimination and eradication scenarios , respectively ( Fig 4 ) . Unit financial cost per treatment for the control scenario would decrease from $0 . 4 to $0 . 1 over 2013–2045 . For the elimination scenario , it would stay between $0 . 4 and $0 . 5 until 2035 , and increase to $0 . 9 afterwards . For the eradication scenario , it would stay between $0 . 4 and $0 . 5 until 2030 , and increase to $3 . 1 as the treatment phase nears the end in endemic African regions ( Fig 3 ) . Economic costs would be six times higher than financial costs under the control scenario and three times higher under the elimination and eradiation scenarios . Total economic costs over 2013–2045 would be $3 . 7 billion ( $3 . 3bn–$4 . 3bn ) for the control scenario , $2 . 2 billion ( $2 . 0bn–$2 . 7bn ) for the elimination scenario , and $2 . 1 billion ( $1 . 8bn–$2 . 5bn ) for the eradication scenario ( Fig 5 ) . That is , the total economic costs associated with the elimination and eradication scenarios are lower than those associated with the control scenario by $1 . 5 billion ( $1 . 1bn–$1 . 9bn ) and $1 . 6 billion ( $1 . 2bn–$2 . 1bn ) , respectively ( S3 Fig ) . Donated ivermectin and community volunteers would account for 75% and 25% of the total economic costs over 2013–2045 in all scenarios . One-way sensitivity analysis ( Fig 6 ) shows that , among the parameters related to CDTi performance , the delay in ending CDTi ( after the infection levels reach the threshold for stopping CDTi ) is the most influential parameter , leading total costs to increase by $2 billion ( undiscounted ) over 2013–2045 in all scenarios . Among the cost items with high uncertainty ( based on the number of missing data ) , the most influential one is the salary top-ups for stabilizing new projects in the elimination and eradication scenarios , leading total costs ( undiscounted ) to range from $3 . 807 billion to $3 . 847 billion , and from $3 . 460 billion to $3 . 498 billion , respectively . Increasing the discount rate from 0% to 6% would decrease total costs over 2013–2045 by 46% from $6 . 1 billion to $3 . 3 billion for the control scenario , by 39% from $3 . 8 billion to $2 . 3 billion for the elimination scenario , and by 35% from $3 . 5 billion to $2 . 2 billion for the eradication scenario .
The elimination and eradication scenarios are predicted to generate substantial cost-savings in the long run compared to the control scenario . The main factors contributing to cost-savings are the reduction in economic costs of community volunteers and donated ivermectin due to a shorter treatment phase as a result of regular active surveillance . This finding implies that the saved volunteers’ time and ivermectin can be used for other health programs . Willing volunteers and their well-established networks , which have enabled successful implementation of CDTi in Africa , could contribute to improving access to primary health care in remote rural areas with insufficient human resources . In addition , the saved ivermectin drugs could be used for other disease programs , for example , anti-LF mass drug administration . To realize these possibilities , policymakers would need to keep empowering community volunteers through training and societal or economic appreciation . Also , pharmaceutical companies’ continuous commitment to donating drugs would be needed . The main operational difference between the elimination/eradication scenarios and the control scenario is regular active surveillance . Our analysis shows that the cumulative financial costs for surveillance over 2013–2045 in the elimination and eradication scenarios would be five times higher those in the control scenario . This implies that endemic countries would need to improve their domestic funding capacity to sustain high surveillance costs to achieve elimination , as the post-treatment surveillance period could last beyond 2045 [9] and external funding would be temporary . The development and operationalization of new affordable and effective diagnostic tools , for example , OV-16 ( ELISA and Rapid Test ) and the DEC patch test under development [27 , 28] , might lead to the savings of surveillance costs . The financial unit cost per treatment in the elimination and eradication scenarios would increase by factors of respective two and eight as the regional intervention phase nears the end . This increase is driven by the reduction in the number of people in need of treatment and steady or increasing costs for surveillance and capital goods . Additionally , in the last stage , the majority of people in need of treatment are expected to live in areas with epidemiological and political challenges [9] . This implies that , in the last mile towards elimination and eradication , political , financial , and societal commitment across a whole spectrum of stakeholders will be essential to meet high unit costs and to deliver treatments in challenging areas [29] . Studies based on social choice theory and game theory [30–33] show that the elimination and eradication of infectious diseases are public goods that can only be achieved through the coordinated efforts of multiple countries . These studies suggest that high benefit-cost ratios associated with elimination and/or eradication could incentivize endemic countries to pursue elimination and/or eradication and global donors to finance endemic countries lacking the financial capacity . Equity and social justice arguments for elimination and eradication [34 , 35] could also complement and strengthen the economic rationality . The role of global stakeholders can play a decisive role to overcome national challenges . A study by Shaffer suggests that , to prevent potential holdout problems caused by unwilling or unable countries , which could hinder elimination and eradication , the centralized efforts led by international organizations would be necessary [36] . In line with this , it has been argued that the explicit inclusion of NTDs elimination in the Sustainable Development Goals ( SDGs ) of the United Nations ( UN ) [37 , 38] would further motivate the commitment of national and global policymakers and donors . Societal commitment at local level will be also essential , because delivering treatments to operationally challenging areas would require successful drug administration by community volunteers and communities’ compliance to treatments . To promote such commitment by communities , endemic countries’ continuous investments in enhancing the operational capacity of community volunteers and in mobilizing communities will be needed . The uncertainty analysis showed that the delay in ending CDTi would have the highest impact among those related to CDTi performance on total costs . Thus , planning to move towards the post-treatment phase , along with regular monitoring and evaluation to decide the proper time of stopping treatments , would be important to avoid the delay in ending CDTi . The uncertainty analysis also showed that the salary top-ups for stabilizing new projects would have the most influence of all cost items on total costs . Many new projects are in potential hypo-endemic areas where parasitological surveys are still needed to confirm endemicity [39] . This suggests that complete epidemiological mapping should be a priority to choose areas to start new projects and to predict required human resources for those projects . The results presented in this study should be interpreted considering the limitations of the approach and data used . To calculate financial costs for projects without available budgets , we relied on national or regional average unit costs which might only approximately represent the actual costs in those projects . For economic costs , we assumed agriculture value added per worker as an opportunity cost of community volunteers’ unpaid time . However , other studies used different proxies such as national minimum wage and GNI per capita [24 , 40] . We did not use national minimum wage , as it was unavailable for 11 of 28 endemic countries [41] . We did not use GNI per capita , as it does not represent the income level in remote rural areas . In the opportunity cost of donated ivermectin , we did not include tax deduction provided to donating manufacturers [19] , as the relevant detailed information is proprietary and unavailable . There were some other factors that could affect resource utilization , but were not included in the analysis . We assumed no recrudescence , because it was difficult to predict when recrudescence would happen . If that were to happen , costs would increase because the treatment phase would have to be restarted . We did not consider the potential impact of new diagnostic and treatment tools , because it was difficult to predict when they would be developed and operationalized . If new effective and affordable tools are operationalized , the strategies of treatment and surveillance could change , thereby influencing costs . We assumed no unexpected political unrest that could interrupt interventions and would increase costs to restart the interventions . Despite these limitations , to our knowledge and based on literature review ( see S1 Text ) , our study is the most up-to-date cost analysis of potential regional elimination strategies in Africa . National and global policymakers and donors could use our cost analysis to make informed policy decisions and to predict the funding needs for implementing elimination programs in Africa . Our cost estimates could also be used by policymakers and researchers to compare costs and potential benefits associated with potential elimination strategies in Africa . | River blindness ( onchocerciasis ) is a parasitic disease transmitted by blackflies . Symptoms include severe itching , skin lesions , and vision impairment including blindness . More than 99% of all cases are concentrated in sub-Saharan Africa . Fortunately , vector control and community-directed treatment with ivermectin have significantly decreased morbidity , and the treatment goal is shifting from control to elimination in Africa . To inform policymakers’ and donors’ decisions , we estimated financial resources and societal opportunity costs associated with alternative treatment goals—control , elimination , and eradication . We found that rapid scale-up of ivermectin treatment for elimination and eradication would result in substantial cost-savings in the long term as compared to staying in a control mode , because regular active surveillance would allow treatments to end earlier , thereby saving the economic costs of community volunteers and donated ivermectin . To realize cost-savings , policymakers should keep empowering community volunteers , and pharmaceutical companies would need to continue drug donation . To sustain high surveillance costs required for elimination and eradication , endemic countries would need to enhance their domestic funding capacity . Societal and political will would be critical to sustaining all of these efforts . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [] | 2015 | Financial and Economic Costs of the Elimination and Eradication of Onchocerciasis (River Blindness) in Africa |
Genome amplification and cellular senescence are commonly associated with pathological processes . While physiological roles for polyploidization and senescence have been described in mouse development , controversy exists over their significance in humans . Here , we describe tetraploidization and senescence as phenomena of normal human placenta development . During pregnancy , placental extravillous trophoblasts ( EVTs ) invade the pregnant endometrium , termed decidua , to establish an adapted microenvironment required for the developing embryo . This process is critically dependent on continuous cell proliferation and differentiation , which is thought to follow the classical model of cell cycle arrest prior to terminal differentiation . Strikingly , flow cytometry and DNAseq revealed that EVT formation is accompanied with a genome-wide polyploidization , independent of mitotic cycles . DNA replication in these cells was analysed by a fluorescent cell-cycle indicator reporter system , cell cycle marker expression and EdU incorporation . Upon invasion into the decidua , EVTs widely lose their replicative potential and enter a senescent state characterized by high senescence-associated ( SA ) β-galactosidase activity , induction of a SA secretory phenotype as well as typical metabolic alterations . Furthermore , we show that the shift from endocycle-dependent genome amplification to growth arrest is disturbed in androgenic complete hydatidiform moles ( CHM ) , a hyperplastic pregnancy disorder associated with increased risk of developing choriocarinoma . Senescence is decreased in CHM-EVTs , accompanied by exacerbated endoreduplication and hyperploidy . We propose induction of cellular senescence as a ploidy-limiting mechanism during normal human placentation and unravel a link between excessive polyploidization and reduced senescence in CHM .
It is commonly assumed that trophoblasts of the human placenta exit their cell cycle as they develop into differentiated subtypes . This hypothesis follows the assumption that quiescence is a consequence of terminal differentiation in tissues . In detail , the villous epithelium of the human placenta hosts progenitor cells that either fuse to form multinucleated , hormone-secreting syncytiotrophoblasts or differentiate into invasive extravillous trophoblasts ( EVTs ) . In placental anchoring villi , the latter undergo a multi-step differentiation process that starts at the epidermal growth factor-positive ( EGFR+ ) proximal cell column ( CC ) , characterized by high proliferative activity . These cells further differentiate into non-dividing human leukocyte antigen G-positive ( HLA-G+ ) distal CC trophoblasts and invade the endometrial epithelium of the pregnant uterus , termed decidua . Remarkably , EVTs fulfil a great variety of functions including vascular remodelling [1 , 2] , interaction with immune cells [3 , 4] as well as defense against pathogens [5] and they undergo an epithelial to mesenchymal-like transition [6] . In rodents , invasive trophoblastic giant cells ( TGCs ) , the functional equivalent to human EVTs , contain a highly polyploid DNA content [7 , 8] . Differentiation of mouse TGCs is characterized by the omission of mitosis and induction of endoreduplication by undergoing multiple rounds of S- and G-phases [9] . A single polyploid TGC nucleus contains up to 1000 copies of the genome [10] . Functional studies showed that oscillating levels of KIP2p57 ( from now on referred to as p57 ) in G-phase as well as Cyclin E and A in S-phase are necessary for DNA polyploidization in rodent TGCs [11 , 12] . Well in line , genomic deletion of Cyclin E or p57 prevents trophoblast endoreduplication in mice [13–15] . While it was assumed for years that polyploidization in TGCs results in a linear amplification of the entire genome , recent data showed that endoreduplication in TGCs leads to under- and overreplication of specific genomic regions [16 , 17] . In contrast to rodents , our knowledge about endoreduplication in human trophoblasts is scarce . For instance , very little data exist to suggest an increased DNA content in human EVTs [18–20] . Moreover , it is not clear at which stage of differentiation the proposed increase in DNA copy numbers occurs as the currently accepted model suggests that generation of trophoblasts , expressing the prime EVT marker HLA-G , is accompanied with terminal growth arrest of these cells [21] . In addition , whether DNA amplification in human EVTs affects the entire genome or is characterized by copy number variations ( CNVs ) of specific chromosomes and/or genes is currently unknown . As mentioned above , different to rodent invasive trophoblasts , HLA-G+ trophoblasts are thought to enter a growth arrested status prior to terminal differentiation . Commonly , end-differentiated cells in tissues undergo cellular quiescence by entering G0 [22] . In vitro , entry into the G0-phase can be provoked by nutrient starvation , cell contact inhibition or signals that induce differentiation [22] . Cellular quiescence is reversible as noticed for instance in endothelial cells during wound healing and angiogenesis or in activated stem cells [23] . In addition , G0-phase cells are usually defined as metabolically and transcriptionally less active than dividing or senescent cells reflecting an end-point differentiational program which dictates cells to execute highly specific functions [24] . In contrast to quiescence or G0 , senescent cells are in an irreversible growth arrest [25] . This status is believed to be induced by various stimuli of which telomeric shortening and stress are the best characterized triggers [22] . Although senescent cells do not differ from quiescent cells by their DNA content , the former population shows an array of specific markers or phenotypical characteristics . The most commonly used markers to identify senescent cells include beta-galactosidase ( βG ) activity [26] together with the induction of cyclin-dependent kinase inhibitors ( CDKNs ) such as p16 [27] , p21 [28] and p57 [29 , 30] . Senescence is accompanied by high transcriptional and metabolic activity [31] as well as induction of a so-called senesecence-associated secretory phenotype ( SASP ) [32] . In addition , when cells senesce they often increase in size , show a tetraploid phenotype and , different to quiescent cells , remain in a G1-arrest [24] . Apart from its induction at the end of the cellular replicative lifespan , senescence is thought to be triggered by oncogenes , DNA damage and oxidative stress suggesting senescence as part of a pathological signature [24] . This view has recently been challenged by demonstrating that cellular senescence has its origin in embryonic development in mice and likely is involved in tissue remodelling [33 , 34] . The aim of this study was to firstly characterize the cell cycle status and genomic amplification during the different stages of EVT differentiation . Secondly , we asked whether terminal differentiation in EVTs is associated with induction of senescence . Finally , we studied cases of complete hydatidiform moles ( CHM ) addressing the question whether the hyperplastic phenotype in CHM would affect ploidy , cell cycle and senescence in CHM-EVTs .
First we performed immune fluorescence ( IF ) -based phenotypical analyses of EGFR+ and HLA-G+ cell column trophoblasts . Interestingly , nuclei of HLA-G+ CCTs appear to be bigger than those of EGFR+ cells ( S1A Fig ) . Magnetic bead-assisted isolation of these two populations revealed that HLA-G+ trophoblasts show a bigger nuclear diameter when compared to non-invasive EGFR+ trophoblasts ( S1B and S1C Fig ) . To further study this phenomenon we performed Confocal Laser Scanning Microscopy ( CLSM ) -assisted measurement of nuclear volumes in 3-D reconstructed nuclei of cell column trophoblasts ( Fig 1A ) . Again , these analyses revealed a significantly increased nuclear volume in distal CCTs when compared to proximal CCTs , suggesting a positive correlation between enhanced nuclear volume and EVT differentiation ( Fig 1B ) . Moreover , flow cytometric ( FC ) analysis of DAPI content in EGFR+ and HLA-G+ CCTs as well as in HLA-G+ EVTs revealed that HLA-G+ trophoblasts show a predominant 4N status ( Fig 1C ) . A small but significant subpopulation containing a DNA content beyond 4N was detectable in both HLA-G+ CCTs and EVTs ( Fig 1D ) . In parallel , we isolated HLA-G+ EVTs from first trimester placental and deciudal tissues , which were pre-selected for an embryonic male 46 , XY karyotype . Again , the majority of EVTs showed a tetraploid , mononuclear XXYY phenotype ( Fig 1E ) . Figure S1D shows positive embryonic male and female nuclei and the graph in S1E Fig indicates the percentage of male and female cells after positive selection for HLA-G+ cells . Finally , we subjected EGFR+ and HLA-G+ CCTs to DNA deep sequencing ( DNAseq ) . Significant CNVs were not found when comparing human polyploid HLA-G cells and diploid EGFR cells ( S1F Fig , upper two plots ) . For comparison , mouse polyploid TGCs display a large amount of significant CNVs when compared to diploid embryonic cells ( S1E Fig , lower plot ) . Altogether , these studies suggest a predominant tetraploid status in invasive , HLA-G+ human EVTs with no signs for CNVs . Next , we addressed the question whether increased nuclear volume and DNA content observed in HLA-G+ trophoblasts correlate with an active endo-cycle in these cells . First , we transduced outgrowing first trimester placental explants with a BacMam FUCCI reporter system , labelling cells expressing Cdt1-RFP in G1/S phase ( red ) or geminin-GFP in G2/S/M phase ( green ) . Cells in S phase express both reporters and thus appear in yellow . Interestingly , we found distal CCTs labelled in red , green or yellow indicative for an active cell cycle ( Fig 2A ) . We also noticed some positive nuclei in detaching , migratory EVTs ( S2A Fig ) . To further study this phenomenon , we determined the expression pattern of various cell cycle markers in first trimester placental anchoring villi harbouring both cell column trophoblasts and invading decidual EVTs . As expected pCCTs express high levels of Ki67 and mitosis-associated cyclin B , both of which were absent from invasive EVTs ( Fig 2B ) . Interestingly , EVTs induce the endocycle-associated cell cycle regulators cyclin E and p57 . Of note , p57 was the only CDKN tested with a clear expression profile in dCCTs and EVT whereas p21 , p27 and p16 were generally absent or weakly expressed ( S2D–S2F Fig ) . Further analyses revealed that mitosis-specific expression of phospho-histone H3 ( pH3 ) and Aurora B are restricted to EGFR+ trophoblastic subpopulations including vCTBs ( S2B Fig ) and pCCTs ( Fig 2C ) . A consecutive tissue section of Fig 2C showing localisation of pH3+/EGFR+ pCCs is presented in S2C Fig . Interestingly , HLA-G+ dCCTs and EVTs show reciprocal expression of the G1/S-phase cyclin A and p57 ( Fig 2D and 2E ) . Double-positive HLA-G and cyclin A trophoblasts at the proximal and distal ends of the CC are shown in S2G Fig . In contrast , cyclin E+ nuclei of dCCTs and EVTs were also positive for p57 , suggesting suppression of S-phase entry in these cells ( S2H and S2I Fig ) . Evaluation of cyclin A and p57 expression during EVT differentiation revealed that approximately 36% of pCCT , 12% of dCCTs and 0 . 2% of EVTs are Cyclin A+ and p57- ( Fig 2F ) . To confirm de novo DNA synthesis in dCCTs and EVTs , we incubated floating explants of placental villi as well as of decidua basalis tissues with EdU for 18 hrs to study DNA incorporation in situ ( Fig 2G and 2H ) . The degree of EdU incorporation was 55% in proximal HLA-G- pCCTs , 20% in HLA-G+ dCCTs and 0 . 5% in HLA-G+ EVTs ( Fig 2I ) . These data suggest that formation of HLA-G+ dCCTs is accompanied by the induction of endocycles , which markedly decline in decidual EVTs as noticed by strong induction of p57 , downregulation of proliferation markers and lowered EdU incorporation . Although previous studies have shown signs for cellular senescence in syncytiotrophoblasts [35] no study has evaluated senescence-associated ( SA ) markers in EVTs . First , SA beta-galactosidase ( βG ) activity was determined in first and third trimester decidual basalis frozen tissue sections revealing intense SAβG signals in EVTs ( Fig 3A ) . In contrast , CCTs showed very low SAβG activity ( S3A Fig ) . Decidua stromal cells have recently been shown to exhibit SAβG activity and thus decidua parietalis tissue sections served as a positive control ( S3B Fig ) [36] . Strongest SAβG activity was found in third trimester EVTs ( Fig 3B ) . Moreover , the majority of first trimester , decidual EVTs showed SAßG at a strong or moderate level , suggesting a global induction of a senescent phenotype in invasive trophoblasts ( Fig 3C ) . Well in line , differentiated EVTs showed intense SAβG signals in vitro while early cultures of non-differentiated EGFR+ trophoblasts where mostly negative for SAβG ( Fig 3D ) . In parallel , we stained paraffin-embedded tissue sections from the same patients with an antibody against βG ( Fig 3E and 3F ) . Generally , βG protein expression intensity in first and third trimester tissue sections correlated well with SAβG activity in EVTs and CCTs ( Fig 3 and S3C Fig ) . Quantification of both SAβG activity and βG protein expression revealed a significant induction in term EVTs when compared to first trimester sections ( Fig 3G ) . Further IF stainings of cryo and paraffin embedded decidual tissue sections confirmed that SAβG activity and βG protein expression widely co-localize in EVTs ( Figs 3H and S3E ) . Well in line , βG protein and the lysosomal marker cathepsin A ( CTSA ) are co-expressed in SAβG+ , decidual EVTs ( Fig 3I ) . Altogether , these data demonstrate induction of SAβG activity in invasive EVTs and confirm that SAβG activity likely reflects accumulation of lysosomal βG protein as previously suggested [37–39] . Additionally , SA-associated metabolic and secretory phenotypes were analysed in isolated HLA-G+ and EGFR+ trophoblasts . First , electron microscopy-assisted analysis revealed pronounced glycogen storages within EVTs ( S3D Fig ) . Using gas chromatography we determined cellular contents of triglycerides and fatty acid species in isolated vCTBs and EVTs . Firstly , EVTs display a trend towards increased triglyceride levels ( S3F Fig ) . Analysis of total fatty acids revealed a significant increase in total fatty acids as well as different fatty acids species in these cells ( Fig 4A ) . Additionally , transcripts of members of the SA secretory phenotype ( SASP ) were significantly upregulated in EVTs when compared to villous cytotrophoblasts ( S3G Fig ) . Besides well-studied genes such as fibronectin , MMP2/3 or IGFBP3 we noticed induction of the two pro-inflammatory cytokines interleukin ( IL ) -8 and -6 . Subsequent luminex-based measurement confirmed secretion of both IL-6 and IL-8 by cultivated EVTs ( Fig 4B ) . In addition , EVT-associated expression of IL-6 was confirmed by IF stainings of first trimester decidua basalis tissues ( Fig 4C ) . Similar analysis revealed induction of phosphorylated H2A histone family , member X ( γH2AX ) ( Fig 4D ) , a well described marker for cellular senescence in non-malignant tissue [40] . To further study possible regulators of EVT-associated senescence we performed siRNA-mediated knockdown of CDKN1C and/or CCNE1 , encoding p57 and cyclin E , respectively ( Fig 4E ) . Strikingly , double knockdown of CDKN1C and CCNE1 , significantly reduced expression of SAβG in cultivated , primary EVTs ( Fig 4F and 4G ) . In summary , these data suggest that EVT invasion into the decidua is accompanied by the induction of cellular senescence . CHM is classified as a hypertrophic disease characterized by hyperproliferative villous cytotrophoblasts and cell column trophoblasts [41] . Therefore , we were interested to investigate whether EVTs are also affected by this hyperplastic condition . First we determined the nuclear size of CHM-EVTs and noticed a markedly increased nuclear volume in these cells ( Fig 5A ) . When compared to age-matched healthy EVTs , the nuclear volume of CHM-EVTs was approximately 10-fold higher ( Fig 5B ) indicative for exacerbated polyploidization . Although , all cell types at the placental villus lack p57 expression we noticed induction of p57 in CHM-EVTs ( S4A and S4B Fig ) . Triple-stainings revealed that the ratio between KRT7+/cyclin A+/p57- and KRT7+/cyclin A-/p57+ was strongly shifted towards an endocycling phenotype in CHM placentas when compared to healthy controls ( Fig 5C and 5D ) . Well in line , CHM-EVTs frequently express pRB ( S4C Fig ) . Based on our previous finding we determined βG protein expression to analyse senescence in CHM-EVTs . These analyses revealed significantly reduced levels of βG when compared to age-matched healthy control sections ( Fig 5E and 5F ) . Finally , EVT-associated IL-6 and cyclin E expression was diminished in cases of CHM ( Fig 5G–5I ) . Altogether , these data reveal exacerbated polyploidization , markedly induced cell cycle marker expression and reduced signs for senescence in CHM-EVTs .
EVT differentiation is a multi-step process , which involves massive cell proliferation of EGFR+ pCCTs and formation of non-dividing HLA-G+ dCCTs that invade the maternal uterus upon contact with the decidua . Their equivalent trophoblast subtype in rodents , TGCs are well-characterized for their highly polyploid genome harboring regions with under- or–over-replicated domains [16 , 17] . However , genomic alterations in human EVTs have been poorly elucidated . Some previous data described the whole genome content in human trophoblasts . Zybina et al . suggested polyploidization in human EVTs by measuring diameters of DAPI-stained nuclei in villous , decidual and myometrial trophoblasts [20] . They reported an up to 18N genomic status in myometrial EVTs . To more deeply characterize the genome of EVTs we performed various different methods including CLSM-guided 3D-construction of trophoblastic nuclei demonstrating that EVT differentiation is associated with a significant increase in nuclear volume . Subsequent determination of DNA content revealed a predominant tetraploid status in HLA-G+ CCTs and EVTs with no signs for CNVs . This finding contrasts cytogenetic analyses , including our own , reporting higher ploidy numbers , aneuploidies and gene amplification in human EVTs [18 , 19] . However , FISH-based analyses of genomic and in particular centromeric regions in trophoblasts may be difficult to interpret as polyploid cells often show delayed or asymmetrical chromosomal separation , polyteny or diplochromosomal ploidy . Indeed , polyteny has been demonstrated in mouse trophoblasts [42 , 43] . Asymmetrical chromosomal separation has been confirmed in polyploid human megakaryocytes [44] and is a common phenomenon in tetraploid cells [45] . Finally , genomic reduplication was suggested to result in diplochromosomes characterized by tightly associated quartets of centromeres and chromosome arms [46] . To circumvent this , we determined DNA content of isolated trophoblasts by FACS analysis and screened for CNVs by performing whole-genome sequencing ( WGS ) in combination with BIC-seq analysis . Here , the vast majority of HLA-G+ CCTs and EVTs showed a 4N status . Different to mouse TGCs , analysis of WGS data revealed no sings for CNVs in isolated HLA-G+ trophoblasts , suggesting that human EVTs fully replicate their entire genome . However , on the basis of our experiments we cannot exclude the presence of a small EVT subpopulation yet showing signs for CNVs and/or higher polyploidy rates . Therefore , further studies using single cell DNAseq approaches are required to address this uncertainty . Moreover , these studies led us to propose that HLA-G+ trophoblasts undergo active replication as these cells contain more DNA than EGFR+ trophoblasts . Indeed , a fluorescence-based reporter system indicated cycling trophoblasts at the distal end of outgrowing explant cultures . While , proximal EGFR+ CCTs showed proliferation marker expression including cyclin B , Ki67 , pH3 and AuroraB we could not detect these cell division proteins in HLA-G+ CCTs . Of note , Ki67 expression has recently been shown to exert important functions during mitosis [47] . Instead , HLA-G+ CCTs expressed G/S-phase markers such as cyclin E and cyclin A along with strong induction of the cyclin-dependent kinase p57 . Interestingly , cyclin E and p57 were shown to be essential for mouse TGCs polyploidization [13–15] . In mice , TGC endocycles display oscillating p57 and cyclin E/cyclin A levels of expression [11 , 12] . Moreover , degradation of p57 is triggered by CCNA/Cdk2-dependent phosphorylation facilitating resetting of DNA replication in G1/S . Our data show that indeed cyclin A and p57 are reciprocally expressed in human HLA-G+ CCTs and EVTs suggesting that a CCNA+/p57- expression profile is indicative for an active S-phase in these cells . Whether the human EVT-associated endocycle is indeed controlled by oscillating G/S phases as shown in TGCs or also involves other cell cycle phases needs to be addressed in future studies . Evaluation of CCNA and p57 co-expression as well as EdU incorporation during EVT differentiation revealed DNA replication in the distal cell column that sharply drops in EVTs . Given that we could not detect any signs for cell division in dCCTs we assume that S-phase activity in these cells finally results in a predominant tetraploid-phenotype in human EVTs . We detected only a very small number ( approx . 1% ) of EVTs that incorporated EdU in vitro or expressed cyclin A . In agreement with these data , we consistently detected a minor subpopulation containing a DNA content of > 4N in pCCTs and EVTs . Whether enhanced polyploidization in this small population of EVTs is linked with specific functionality and/or phenotype is currently under investigation . Nevertheless , the vast majority of EVTs seems to be in a growth arrested status . Interestingly , EVTs show high expression levels of cyclin E and p57 , two markers indicative for cellular senescence in non-cycling cells [29 , 30 , 48 , 49] . In addition , tetraploidy is believed to trigger cellular senescence in order to prevent excessive polyploidization , genomic instability and tumorigenesis [50 , 51] . Senescence has traditionally been attributed with pathological alterations and loss of functionality such as aging . However , two recent scientific reports demonstrated a senescence-related function in mouse development [33 , 34] . Senescent cells are non-proliferative , exhibit activation of SAβG , altered metabolic signatures and induce a SASP [24] . Surprisingly , we noticed strong induction of SAβG activity in EVTs as well as in differentiated EVTs in vitro whereas all other trophoblast subtypes showed no prominent signal , except some activity in the syncytium and in dCCTs . Moreover , we found that SAβG activity was more pronounced in EVTs at term . In this context , it might be that increased SAβG activity is a sign for loss of functionality . In addition , it is well-documented that cellular senescence induces self-targeted clearance by immune cells . For instance , oncogene-induced senescence in hepatocytes initiates their immune-mediated clearance to prevent malignancy [52] . This phenomenon is triggered by a Th1-polarized CD4+ T-cell response [53] . It is therefore tempting to speculate that while immune-mediated clearance of senescent EVTs is suppressed by the anti-inflammatory Th2-like decidual environment , senescence may protect from systemic spread of intact EVTs . In addition to SAβGal activity and protein levels we also characterized the specific senescence-associated phenotype of EVTs . Senescent cells often show a global change in their metabolism . This includes enhanced glycogen storage [54] , induction of fatty acid synthesis [55] as well as secretion of inflammatory cytokines , chemokines , extracellular matrix ( ECM ) -associated factors and other signaling molecules including interleukin ( IL ) -6 , IL-8 , fibronectin , vascular endothelial growth factor and matrix metalloproteinases [32 , 56 , 57] . While an EVT-specific increase in glycogen and lipid content has been demonstrated [58] , we show for the first time that EVTs formation is associated with an induction of fatty acid content . Since senescence cells increase in size [25] and in particular enhance their secretory activity , increased global fatty acid synthesis is likely reflected by a greater need for membrane synthesis . High levels and pronounced induction was observed for stearic acid ( C18:0 ) , oleic acid ( C18:1 ) arachidonic acid ( C20:4 ) and docosahexaenoic acid ( C22:6 ) . Oleic , arachidonic and docosahaexaenoic acid may also impact on EVT function as they were shown to induce tube formation in HTR-8 cells and/or survival [59 , 60] . Arachidonic acid is the precursor of eicosanoids including prostaglandins and leukotrienes . Interestingly , prostaglandins induce uterine vasodilation [61] or promote Th2 and regulatory T-cell responses [62] . Leukotriene B4 is a well described leukocyte chemoattractant including recruitment of T-cells [63] and neutrophils [64] . The latter have recently been suggested to show a pro-angiogenic phenotype in the human deciuda [65] . We further found that EVTs secret IL-6 and -8 , two prominent members of the SASP [32] . Since both cytokines are well-described for their crucial role in neutrophil recruitment [66–68] , EVT-mediated release of IL-6 and -8 might support vascular remodelling during pregnancy . Along these lines , IL-6 has been shown to potently suppress vascular smooth muscle contraction [69] . We also detected a positive staining for γH2AX indicating DNA double strand breaks in EVTs , a well described trigger of cellular senescence [40] . In this context , it is interesting to note that overexpression of cyclin-E causes DNA damage [70 , 71] and was shown to induce cellular senescence [48 , 72] . Although , knockdown of cyclin E alone was not sufficient to alter induction of senescence in EVTs combined suppression of cyclin E and p57 significantly reduced SAβG activity in cultured EVTs . Since p57 has also been shown to induce cellular senescence in human cancer cells [29 , 30] and vascular smooth muscle cells in mice [73] we propose a role p57 and cyclin E in the induction of the EVT-associated senescent phenotype in humans . The fact that both cell cycle regulators are also expressed in non-senescent trophoblasts points towards multifaceted roles for p57 and cyclin E in placental development . Available data suggest that increased expression of cyclin E and p57 beyond physiological levels are necessary to induce cellular senescence . Similar regulatory mechanisms are likely to occur during EVT differentiation since p57 [21] is upregulated during EVT differentiation and published microarray and RNAseq data also suggest induction of cyclin E transcripts [74 , 75] . CHM placentas are characterized by a hyperplastic , androgenetic placental phenotype . Hyperproliferation noticed in vCTBs and pCCTs [41] is likely caused by a complete lack of growth restricting maternally expressed genes [76] . Indeed , cases of Beckwith-Wiedemann syndrome characterized by loss of p57 expression or mutations in CDNK1C [77] share several pathological features in placental development with CHM such as hyperplasia or excessive EVT formation [78 , 79] . Whether the hyperplastic phenotype in CHM also affects cell cycle of EVTs has not been studied so far . Interestingly , we noticed a highly enlarged nuclear volume in EVTs of CHM placentas suggesting exacerbated endocycles in these cells . In addition , we found that EVTs of CHM placentas induce p57 confirming previous reports demonstrating that haploid , androgenic mole placentas express paternally imprinted genes such as H19 [80] and CDKN1C [81] , normally expressed from the maternal allele . Despite reactivation of p57 we also found a pronounced endocycle-associated phenotype characterized by markedly elevated numbers of p57-/cyclin A+ EVTs . Unfortunately , we were not able to determine SAβG activity in CHM since endogenous enzyme activity can only be measured in fresh tissues . Nevertheless , our data show that SAβG activity strongly correlates with EVT-associated overexpression of lysosomal βG protein when compared to other trophoblast subtypes . We therefore suggest that reduced βG protein expression in CHM-EVTs is indicative for a suppressed senescent phenotype . This conclusion is further supported by significant lower levels of IL-6 , p57 and cyclin E in these cells . This suggests that the hyperplastic phenotype noticed in CHM extends to EVTs , which show excessive endocyclic activity characterized by hyperpolyploidization and suppressed senescence-induced growth arrest . As discussed above , multiple endocycles in mouse TGCs result in CNVs . It is tempting to speculate that repeated DNA replication in the absence of mitosis might also result in a gradual disorganization of the DNA methylation at CpG islands in EVTs , as it has been demonstrated in rodent TGCs . Along these lines , hyperpolyploidization in CHM-EVTs could lead to loss of epigenetic methylation marks and therefore reactivate expression of p57 and H19 . Indeed , partial escape from X chromosomal inactivation together with reduced promotor methylation of X-linked genes has been noticed in endreduplicating mouse TGCs [82] . In summary , this report is the first to show that induction of EVT differentiation is accompanied with induction of endocyclic activitiy and tetraploidization ( Fig 6A ) . We suggest a Cyclin A+/p57- expression profile as indicative marker for endoreduplicating HLA-G+ trophoblasts . However , upon invasion into the decidua EVTs exit the replicative cell cycle followed by growth arrest and induction of cellular senescence . In contrast , hyperplastic CHM-EVTs continue their endoreduplicative cell cycle likely leading to repression of senescence and as a consequence exacerbated polyploidization ( Fig 6B ) .
Placental and decidual tissues ( 6–12th week of gestation , n = 85 ) were obtained from legal , elective pregnancy terminations . CHM placentae ( n = 23 ) were obtained from the archive of the Clinical Institute of Pathology , Medical University of Vienna , Austria . Diagnosis of CHM was based on the original report including high resolution ultrasound , determination of the proliferative index and human gonadotropin beta levels as well as absence of p57 expression in placental villous cytotrophoblasts . Utilization of tissues and all experimental procedures were approved by the local Ethics Committee of the Medical University of Vienna , Austria . Methods were carried out in accordance with the approved guidelines . Written informed consent was obtained from all patients . Cytotrophoblasts ( CTBs ) were isolated by enzymatic dispersion and Percoll density gradient centrifugation ( 10–70% ( vol/vol ) ; GE Healthcare ) of pooled first trimester placentas ( n = 2–5 per isolation ) as described in [83] and plated ( 45 min ) in culture medium ( DMEM/Ham’s F-12 , 10% ( vol/vol ) FCS , 0 . 05 mg/mL gentamicin , 0 . 5 μg/ml fungizone; Gibco ) allowing for adherence of contaminating stromal cells . Nonadherent trophoblasts were collected and seeded in culture medium onto fibronectin-coated ( 20 μg/mL; Millipore ) dishes ( 2 . 5 × 105 cells per square centimetre ) . The contamination with stromal cells was routinely tested by IF with antibodies detecting cytokeratin 7 ( trophoblast cells ) and vimentin ( fibroblasts ) . Vimentin-positive cells were < 3% . CTBs were cultured for up to 72 h with media changed after 20h . EGFR+ trophoblasts were separated from HLA-G+ trophoblasts by magnetic-activated cell sorting , using EGFR-PE and HLA-G-PE antibodies , which were labelled with anti-PE micro-beads ( Miltenyi Biotec ) . Decidua basalis tissue was minced into ~3 mm3 pieces and digested under agitation in 2 mg/ml collagenase I ( Life Technologies ) and 0 . 5 mg/ml DNase I ( Sigma Aldrich ) in HBSS containing 25 mM HEPES ( Life Technologies ) for 30 min at 37°C . Dispersed cells were pooled and filtered through a 70 μm cell strainer . To label extravillous trophoblasts , cells were incubated with HLA-G-PE antibodies for 20 min at 4°C and prepared for flow cytometric analysis as described below . First trimester placental and decidual tissues ( 6th - 12th week of gestation ) were fixed in 7 . 5% ( wt/vol ) formaldehyde and embedded in paraffin . Serial sections ( 3 or 30 μm ) were deparaffinised in Xylol ( 10 min ) and rehydrated in a decreasing series of ethanol ( 100% , 90% , 70% , 0%; 1 min each step ) . Antigen retrieval was performed using 1× PT module buffer 1 ( pH 6 , Thermo Fisher Scientific ) for 35 min at 93°C using a KOS microwave histostation ( Milestone ) . Sections were blocked using 0 . 05% cold water fish skin gelatine ( Sigma-Aldrich ) for 30 min at room temperature and incubated overnight at 4°C with primary antibodies in PBS with 0 . 05% fish skin gelatine ( see antibody list ) . Secondary antibodies were incubated for 45 min at room temperature in PBS with 0 . 05% fish skin gelatine and 1 μg/ml 4′ , 6-Diamidin-2-phenylindol ( DAPI , Roche ) . Finally , tissue sections were mounted with Fluoromount-G ( Thermo Fisher Scientific ) and covered . Images were acquired with a fluorescence microscope ( Olympus BX50 equipped with Cell^P software ) or with a confocal laser scanning microscope ( LEICA , TCS SP8X , equipped with Leica LAS AF software ) . Confocal images are depicted as maximum projection of total z-stacks and brightness and contrast were adjusted in a homogenous manner using Leica LAS AF . RNA isolation , reverse transcription and qPCR analyzes were performed as described previously [83] using TaqMan Gene Expression Assays: ADAM12L ( Hs 00185774_m1 ) . Signals ( ΔCt ) were normalized to TATA-box binding protein ( TBP ) ( ABI , 4333769F ) . Protein extracts were immobilized on PVDF membranes and incubation with primary and secondary antibodies ( S1 Table ) was performed as published [41] . Signals were developed using ECL prime detection Kit ( GE Healthcare ) and visualized with FluorChemQ imaging system ( Alpha Innotech ) . Signal quantification was performed using Image J software . Freshly isolated trophoblasts ( see above ) were transfected with SMARTpoolON-TARGETplus siRNAs ( GE Dharmacon ) using Lipofectamine RNAiMAX reagent ( Invitrogen , Life Technologies ) and cultivated for 48 hrs . The following siRNAs against the indicated mRNAs were used at a concentration of 40 nM: CDKN1C ( L-003244-00-0005 ) , CCNE1 ( L-003213-00-0005 ) and non-targeting control ( D-001810-10-05 ) . After 48 h , transfected cells were subjected to western blotting . Alternatively , cells were fixed in 1x fixative solution ( Cell Signaling ) for 10 min at room temperature , washed 3 times with PBS and incubated overnight at 37°C with the β-galactosidase staining solution at pH 6 . 0 ( Cell Signaling ) . In total , 1620 images were analyzed quantitatively . P57 and CCNA1 positivity in CCTs was analyzed by evaluating 15 tissue sections per sample ( n = 10 ) . Consecutive tissue sections were stained with antibodies against EGFR+ and HLA-G+ to determine proximal and distal regions of the CC . P57 and CCNA1 expression in KRT7+ EVTs was determined by evaluating 12 tissue sections per sample ( n = 9 ) . EdU incorporation into HLA-G+ dCCTs and EVTs , respectively was analyzed by evaluating 8 tissue sections per sample ( n = 4 ) . To ensure objectivity , the quantification of all IF stainings was analyzed by two independent investigators . Image stacks were obtained using confocal laser scanning microscopy with 0 . 2 μm steps in z-axis of 30 μm tissue sections . 3D reconstruction was performed from DAPI signals using Imaris software ( Bitplane AG ) . Nuclei that had been cut and were incomplete were disregarded from analysis . Volumes of nuclei were calculated and compared . 3D reconstructions were performed over five cell columns of three different placentas . Proximal cell column trophoblasts ( CCTs ) were defined as CCTs within 30 μm from the vCTB layer , distal CCTs were defined as CCTs within 60 μm from the most distally recorded CCTs . First trimester placental villous explants ( 7th - 9th week of gestation , n = 12 ) were dissected under the microscope as mentioned in [83] and seeded onto collagen I drops ( BD Biosciences , mixed with 10× DMEM and 7 . 5% sodium bicarbonate ) . One explant per 48 well was seeded and incubated for 6 h at 37°C to allow anchorage . Explants were then mounted with 200 μl DMEM/Ham's F-12 medium and 2 . 5 μg/ml fungizone ( Invitrogen ) and incubated over night at 37°C . The next day , media was discarded and 160 μl media supplemented with 20 μl Premo geminin-GFP reagent and 20 μl Premo Cdt1-RFP reagent ( 2×10^6 particles of each per explant ) were added according to manufacturer’s protocol ( Thermo Fisher Scientific ) to those explants showing trophoblast outgrowth and incubated for an additional 24 h . Finally , media was discarded and explants were either mounted with prewarmed PBS and immediately digitally photographed using the EVOS FL Color Imaging System or fixed ( 4% PFA , 15 min , 4°C ) , permeabilized ( 0 . 1% Triton X-100 , 5 min , 4°C ) and stained ( DAPI , 10 min , room temperature ) in the dark and then digitally photographed . Isolated trophoblast were centrifuged and incubated with 0 . 5% KCL and incubated at 37°C for 20 minutes . Subsequently , cells were treated with ice-cold fixative ( one part acetic acid and three parts methanol ) and incubated for 10 minutes at -20°C . Fixed cells were air-dried on glass slides at 42°C for 15 minutes and treated with pepsin ( 350 μl 0 . 5% Pepsin with 1 ml 1N HCl ad 100 ml H2O ) for 15 min at 37°C . Slides were then washed in PBS and PBS/20mM MgCl2 each 5 minutes at room temperature . Subsequent to dehydration in a 10% formaldehyde solution , cells were incubated with a “ready to use” solution containing centromer specific probes for X- and Y-chromosomes ( DXZ1 ( green ) and DYZ3 ( red ) ( Leica Biosystems ) . Hybridization was performed using a ThermoBrite system ( Leica ) for 16 h at 37°C after 5 minutes denaturation at 75°C . Afterwards , cells were washed for two minutes with 0 . 4x SSC with 0 . 3% NP40 followed by a 1 minute wash step with 2x SSC with 0 . 1% NP40 . Slides were air-dried and covered with Vectashield Mounting Medium containing DAPI ( Vector Laboratories Burlingame , USA ) . FISH signals were detected using an Axioplane2 imaging system ( Zeiss ) equipped with “MetaSystems Isis” software version 5 . 3 . 18 . CTBs were isolated as described above and labelled with the FITC- and PE-conjugated antibodies outlined in S1 Table for 20 min at 4°C . Appropriate isotype-specific control antibodies were used accordingly . Then , cells were fixed ( 4% PFA , 15 min , 4°C ) , permeabilized ( 0 . 1% Triton X-100 , 3 min , 4°C ) and stained ( DAPI , 10 min , room temperature ) in the dark . Data were acquired on a FACScan flow cytometer ( BD Biosciences ) and analysed using FlowJo 7 . 6 . 5 software ( Tree Star , Ashland , OR ) . Doublet discrimination was performed by plotting the area ( FL-A ) of the fluorescence light pulse against the width ( FL-W ) . Lipids were isolated from cell pellets by standard Folch extraction . An aliquot of the pellet was used for cell protein determination by the Bradford assay . Triglycerides were directly analyzed by GC as described [84] . In brief , lipids were separated using a GC-2010 gas chromatograph ( Shimadzu ) equipped with a programmed temperature vaporizer injector and a ZB-5HT capillary column ( 15 m x 0 . 32 mm x 0 . 1 μm; Phenomenex ) . Trinonadecanoin ( Sigma ) was used as standard . For fatty acid analysis , FOLCH-extracts were trans-esterified using boron trifluoride-methanol solution ( Sigma ) at 80°C for 2 hrs followed by extraction with hexane . Lipids were separated on a ZB-FFAP capillary column ( 15 m x 0 . 32mm x 0 . 25 μm; Phenomenex ) using pentadecanoin ( Sigma ) as standard . Chromatograms were analyzed using GC Solutions 2 . 3 ( Shimadzu ) and values were normalized to cell protein . Multiplex bead assay kits were used according to the manufacturer’s protocol ( Millipore ) to determine the levels of IL-6 ( detection limit 3 . 1 pg/ml ) and IL-8 ( detection limit 1 . 6 pg/ml ) in culture supernatants from EVT cells . The analyses were performed using the Luminex200 IS system ( Millipore ) and the MasterPlex QT 2010 software ( MiraiBio ) . Values below the detection limit were given half the value of the detection limit and the concentration in the corresponding control CM was subtracted from the concentration measured in the cell supernatants . DNA isolated from EGRF+ and HLA-G+ human placental trophoblasts from two different donors ( 11th and 12th week of gestation ) were sent to the Macrogen Laboratory for WGS . Two libraries were made for each sample/cell type using the TruSeq Nano Kit ( Illumina ) . 150 bp paired-end read sequencing was performed on the HiSeq X Ten ( Illumina ) , resulting in approximately 30X coverage for each library ( ~60X coverage for each sample ) . Sequence data was mapped to human reference genome hg19 using Burrows-Wheeler aligner [85] . Mouse placental WGS data was retrieved from BioProject accession number PRJNA213010 [17] . To determine whether there is copy number variation ( CNV ) in polyploid placental cells , data ( sorted bam files ) were analyzed using BIC-seq [86] for paired data with lambda = 4 , following [16] . We also applied the HugeSeq pipeline that integrates four algorithms , including Pindel , CNVnator , BreakDancer and BreakSeq to discover CNVs , and we did not detect any high confidence CNVs that are unique to the polyploid placental cells [87–91] . Tissue sections ( 3 μm ) of paraffin-embedded placental and decidual tissues were obtained and stained as described above . Mean diameters of DAPI signals were determined of epithelial vCTBs and invasive EVTs . For determination of the volume a perfect sphere ( V = 4/3 × π × r3 ) was assumed for all nuclei . Images were taken using the Leica confocal microscope TSC SP8 . ImageJ software was used to measure stained area ( SAβG , βG ) in relation to the HLA-G or KRT signal by analysing staining intensities at a predefined threshold . Cyclin E and IL-6 expression in EVTs was analysed using ImageJ software , by creating a mask containing all KRT+ cells and measuring the mean intensity of the respective signal within the mask . KRT staining intensity within the mask was used as a calibrator . To quantify the percentage of SAβG+ EVTs fluorescence and bright filed images were overlaid in Photoshop ( CS6 Extended ) and HLA-G+ areas were selected using the Magic Wand tool . The bright filed channel was then isolated and pre-selected HLA-G+ areas were cropped . Intensities were classified into strong , moderate and absent as illustrated in S5 Fig . SAβG expression in isolated EVTs was determined relative to total cellular area using ImageJ software . SAβG activity was determined in first and third trimester placental and decidual ( 6th - 12th week of gestation ) cryosections using the senescence β-galactosidase staining kit ( Cell Signaling ) and by adapting published protocols [33 , 34] . Briefly , tissues were preserved in OCT compound , sectioned ( 4 μm ) and fixed in 1x fixative solution ( Cell Signaling ) for 10 min at room temperature , washed 3 times with PBS and incubated overnight at 37°C with the β-galactosidase staining solution at pH 6 . 0 . Subsequently , slides were counterstained with antibodies against HLA-G , ßG and DAPI . Alternatively , SAßG staining was performed in whole-mount first trimester placentas using the Senescence ß-Galactosidase Staining Kit . Briefly , whole-mount placentas were fixed at 4°C over night with 1X Fixative Solution , washed 3 times in PBS and incubated over night at 37°C with ß-Galactosidase Staining Solution ( pH6 ) . Placentas were subsequently dehydrated and perfused with Paraplast X-TRA ( Sigma , St . Louis , MO; USA ) using a KOS Microwave Histostation ( Milestone , Sorisole; Italy ) , then embedded in paraffin for serial sectioning and counterstained with antibodies against HLA-G , ßG and DAPI . BioProject accession number for WGS of human trophoblastic samples: PRJNA445189 Statistical analysis was performed with Student’s unpaired t-test using SPSS 18 ( SPSS Inc . ) . Gaussian distribution and equality of variances were examined with Kolmogorov–Smirnov test and Levene test , respectively . Comparisons of multiple groups were evaluated with one-way ANOVA and appropriate post hoc tests . A P-value of < 0 . 05 was considered statistically significant . | In tissues , cellular differentiation is normally associated with cell cycle arrest . However , in some cases differentiating cells continue their cyclic activity in the absence of cell division . This event , referred to as endoreduplication , leads to polyploidy , which means an increased number of chromosome sets per cell and has been demonstrated in rodent placental trophoblast giant cells ( TGCs ) . However , it is unknown whether human placental trophoblasts also endoreduplicate their genome . To study this , we focused on extravillous trophoblasts ( EVTs ) , a specific trophoblastic subtype that invades the uterus during pregnancy in order to control blood supply and nutrient transfer to the growing embryo . We show that initiation of EVT differentiation is characterized by induced endoreduplication leading to genomic tetraploidization . Different to TGCs , EVTs duplicate their genome in an even manner . Upon invasion into the uterus , EVTs stop their endoreduplicative cycle and undergo cellular senescence . We further show that EVTs of hyperplastic complete hydatidiform mole ( CHM ) placentas , a genetic pregnancy disorder , are characterized by exacerbated endoreduplication , a greater DNA content and reduced signs for senescence . In summary , we propose senescence as a ploidy limiting factor during placental development and describe its suppression in hyperploid CHM-EVTs . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Material",
"and",
"methods"
] | [
"blastocysts",
"uterus",
"medicine",
"and",
"health",
"sciences",
"reproductive",
"system",
"maternal",
"health",
"obstetrics",
"and",
"gynecology",
"cell",
"cycle",
"and",
"cell",
"division",
"cell",
"processes",
"departures",
"from",
"diploidy",
"cell",
"differentiation",
"developmental",
"biology",
"women's",
"health",
"pregnancy",
"mammalian",
"genomics",
"embryology",
"decidua",
"polyploidy",
"animal",
"genomics",
"cell",
"biology",
"trophoblasts",
"anatomy",
"genetics",
"biology",
"and",
"life",
"sciences",
"genomics",
"cyclins"
] | 2018 | Genome amplification and cellular senescence are hallmarks of human placenta development |
The incidence of dengue fever and dengue hemorrhagic fever in Brazil experienced a significant increase since the emergence of dengue virus type-3 ( DENV-3 ) at the early 2000s . Despite the major public health concerns , there have been very few studies of the molecular epidemiology and time-scale of this DENV lineage in Brazil . In this study , we investigated the origin and dispersion dynamics of DENV-3 genotype III in Brazil by examining a large number ( n = 107 ) of E gene sequences sampled between 2001 and 2009 from diverse Brazilian regions . These Brazilian sequences were combined with 457 DENV-3 genotype III E gene sequences from 29 countries around the world . Our phylogenetic analysis reveals that there have been at least four introductions of the DENV-3 genotype III in Brazil , as signified by the presence of four phylogenetically distinct lineages . Three lineages ( BR-I , BR-II , and BR-III ) were probably imported from the Lesser Antilles ( Caribbean ) , while the fourth one ( BR-IV ) was probably introduced from Colombia or Venezuela . While lineages BR-I and BR-II succeeded in getting established and disseminated in Brazil and other countries from the Southern Cone , lineages BR-III and BR-IV were only detected in one single individual each from the North region . The phylogeographic analysis indicates that DENV-3 lineages BR-I and BR-II were most likely introduced into Brazil through the Southeast and North regions around 1999 ( 95% HPD: 1998–2000 ) and 2001 ( 95% HPD: 2000–2002 ) , respectively . These findings show that importation of DENV-3 lineages from the Caribbean islands into Brazil seems to be relatively frequent . Our study further suggests that the North and Southeast Brazilian regions were the most important hubs of introduction and spread of DENV-3 lineages and deserve an intense epidemiological surveillance .
Dengue virus ( DENV ) is a member of the genus Flavivirus , family Flaviviridae , and one of the most important arboviral pathogens . The single-stranded positive-sense genomic RNA encodes one large open reading frame ( ORF ) as a polyprotein , which undergoes proteolytic processing into three structural proteins: capsid ( C ) , membrane ( M ) and envelope ( E ) ; and seven non-structural proteins: NS1 , NS2A , NS2B , NS3 , NS4A , NS4B and NS5 . DENV is transmitted to humans through the bites of infected Aedes mosquitoes , principally A . Aegypti , which is widely distributed around the tropical and subtropical regions of the world [1] . Infection with DENV causes a wide spectrum of disease manifestations , ranging from unapparent infection to severe and potentially fatal disease [2] . There are four distinct antigenic groups or serotypes of DENV ( DENV-1 to DENV-4 ) that are causing human pandemics . A number of phylogenetically distinct lineages , termed genotypes , have been also identified within each serotype , which may differs in both geographical distribution and viral virulence/transmissibility [3] , [4] , [5] . Among them , the genotype III of DENV-3 has been frequently associated with severe dengue outbreaks in Asia , Africa and Latin America [6] , [7] , [8] , [9] , [10] , [11] , [12] , [13] , [14] , [15] , [16] , [17] , [18] . DENV-3 genotype III probably emerged in the Indian sub-continent around the middle 1970s and subsequently spread to other countries from Asian , Africa and the Americas [6] , [19] . This genotype was first detected in the Americas during dengue fever/dengue hemorrhagic fever ( DF/DHF ) outbreaks in Nicaragua and Panama , in 1994 [20] , [21] . In the following years the virus spread through the region using several independent routes from Central America and Mexico to the Caribbean and South America [18] , [19] , [22] . In Brazil , millions of dengue infections have been detected all over the country since 1986 [23] . The first autochthonous case of DENV-3 ( genotype III ) was reported in December 2000 in the state of Rio de Janeiro ( Southeast region ) , from a patient with dengue fever [24] . During the summer of 2002 , the newly introduced DENV-3 serotype caused one of the largest dengue outbreaks in the state of Rio de Janeiro , infecting a susceptible population that had only experienced DENV-1 and DENV-2 epidemics . In the first half of the 2002 , the state reported 288 , 245 dengue cases , including 1 . 831 DHF cases and 91 deaths; which exceed the total number of DHF cases reported in Brazil from 1986 to the time of the epidemic [25] . Subsequent outbreaks of DENV-3 continued to be documented through the 2000s in Rio de Janeiro as well as in almost all Brazilian territory , revealing the rapid spread of this new serotype in the country . Despite the public health importance of DENV-3 genotype III in Brazil , there have been very few studies of the molecular epidemiology of this DENV genotype in the country . The first study surveyed the phylogenetic diversity of a small number ( n = 19 ) of Brazilian DENV-3 genotype III E gene sequences collected up to 2004 [16]; while two more recent studies have focused on the analysis of localized dengue outbreaks occurring in the state of Sao Paulo during 2006 [26] , [27] . This prompted us to perform a more comprehensive study to investigate the origin , evolution , and dispersion dynamics of DENV-3 genotype III in Brazil by examining a large number ( n = 107 ) of E gene sequences sampled between 2001 and 2009 from different locations within the country .
Virus isolates were derived from human serum specimens obtained from 19 Brazilian patients with confirmed dengue virus type 3 ( DENV-3 ) infection from Rio de Janeiro ( n = 12 ) , Espirito Santo ( n = 3 ) and Goias ( n = 4 ) states ( Table 1 ) . The case-patients included in this study had acute febrile illness with two or more of the following clinical manifestations: headache , retrobulbar pain , myalgia , arthralgia , rash and hemorrhage . Ethical clearance was obtained with the approval resolution number CSN196/96 from the Oswaldo Cruz Foundation Ethical Committee in Research ( CEP 274/05 ) , and all subjects provided written informed consent before participation . All samples were received refrigerated and stored at −70°C until tested . The viruses were isolated by inoculation into Aedes albopictus C6/36 cell lines [28] and the serotype was identified by indirect immunofluorescence using type-specific monoclonal antibodies [29] . Viral RNA was extracted from 140 µL of cell culture supernatant by use of the QIAamp Viral RNA Mini Kit ( QIAGEN , Valencia , CA ) , according to the manufacturer's instructions . The complete E gene ( 1479 bp in length ) was then amplified by reverse transcription-PCR ( RT-PCR ) as described previously [30] . Amplicons were directly sequenced in both directions using a BigDye Terminator Cycle Sequencing Ready Reaction kit ( Applied Biosystems , US ) , 1 µM of primers combined with 200 ng of DNA , after purification using PCR purification kit ( Qiagen , US ) . Thermocycling conditions consisted of 30 cycles of 94°C for 1 min , 60°C for 2 min and 72°C for 3 min . After purification using Centri-Sep columns ( Applied Biosystems , US ) the DNA was dried at 37°C , overnight . The pellet was resuspended in 10 µl of Hi-Di Formamide ( Applied Biosystems , US ) , heated for 2 min at 95°C and kept on ice until 10 µl was loaded on an Applied Biosystems Prism 3730 Sequencer ( Applied Biosystems , US ) . The sequences generated here were combined with all DENV-3 genotype III complete E gene sequences available at the GenBank by July 2010 , from which the country and year of isolation were available . One sequence from Mozambique ( GenBank accession FJ882575 ) previously identified as inter-genotype recombinant and two sequences from Brazil ( GenBank accession FJ898446 and FJ898447 ) , from which no information about country region was available , were excluded from the analysis . We also excluded four sequences that displayed anomalously long branches in the phylogenetic analysis: one from Brazil ( GenBank accession AY038605 ) , one from Puerto Rico ( GenBank accession EU529696 ) and two from Argentina ( GenBank accession EU052792 and EU052792 ) ( data not shown ) . This resulted in a final data set of 564 DENV-3 genotype III E sequences ( 1 , 479 nt long ) from the Americas ( n = 469 ) , Asia ( n = 88 ) , Middle-West ( n = 5 ) , South Pacific ( n = 1 ) and Africa ( n = 1 ) , covering a total of 29 countries ( Table 1 ) . Nucleotide sequences were aligned using CLUSTAL X program [31] . Alignment is available from the authors upon request . Phylogenetic analyses were performed under the GTR+I+Γ4 model of nucleotide substitution , selected using the jModeltest program [32] . A Maximum Likelihood ( ML ) phylogenetic tree was inferred for the complete data set of 566 DENV-3 genotype III E sequences with PhyML program [33] , using an online web server [34] . Heuristic tree search was performed employing the SPR branch-swapping algorithm and the reliability of the phylogenies was estimated with the approximate likelihood-ratio test ( aLRT ) based on a Shimodaira–Hasegawa-like procedure . A Bayesian phylogenetic tree was inferred for a subset of 202 DENV-3 sequences using MrBayes program [35] . Chains were run for 10×106 generations and convergence of parameters was assessed by calculating the Effective Sample Size ( ESS ) using TRACER v1 . 5 program [36] , after excluding an initial 10% for each run . All parameter estimates for each run showed ESS values >100 . The rate of nucleotide substitution per site per year ( subs . /site/year ) , the time to the most recent common ancestor ( Tmrca ) and the spatial diffusion of a given DENV-3 lineage were jointly estimated using the Bayesian Markov chain Monte Carlo ( MCMC ) statistical framework implemented in the BEAST v1 . 6 . 1 package [37] , [38] . A matrix of geographic locations was constructed based on the place of sampling for each sequence . A full model was used in which all possible reversible exchange rates between locations were equally likely ( flat prior ) [39] . Where two discrete locations were grouped together , the longitude and latitude used were those of the midpoint of the line connecting them . Where more than two locations were grouped , the latitude and longitude of the centroid of the polygon defined by them were used . Analyses were carried out with a Bayesian Skyline coalescent tree prior [40] , under the GTR+I+Γ4 model of nucleotide substitution and using a relaxed ( uncorrelated Lognormal ) [41] molecular clock model . The MCMC analysis was run for 10×107 generations and convergence of parameters ( ESS>200 ) was assessed with TRACERv1 . 5 program as described above . Uncertainty in parameter estimates was reflected in the 95% highest probability density ( HPD ) intervals . The programs TreeAnnotator v1 . 5 . 2 and FigTree v1 . 1 . 2 ( http://tree . bio . ed . ac . uk/software/figtree/ ) were used to summarize the posterior tree distribution and to visualize the annotated maximum clade credibility ( MCC ) tree , respectively .
The phylogenetic analysis of 564 DENV-3 genotype III E gene sequences sampled world-wide revealed that all American strains segregate in a monophyletic cluster ( Fig . 1 ) , suggesting a single introduction of this genotype into the continent , consistent with previous findings [6] , [19] . The only sequence of African origin included in our data set , which correspond to a virus isolated in Somalia in 1993 ( GenBank accession DQ341208 ) , branched between Asian and American strains ( Fig . 1 ) , supporting an scenario in which DENV-3 genotype III may have gone from Asia into Africa , and then into the Americas [6] , [19] . Inside the DENV-3 genotype III American cluster , strains isolated in Central America ( from 1994 to 1998 ) and Mexico ( from 1995 to 2007 ) branched close to the root of the cluster , while sequences isolated in the Caribbean ( from 1998 to 2007 ) and South America ( from 2000 to 2009 ) segregate in three different monophyletic sub-clusters ( Fig . 1 ) . This pattern support the view that DENV-3 genotype III was introduced into Central America or Mexico and from there spread to the Caribbean and South America following three major routes . In the first route the virus spread to Puerto Rico , Venezuela and Colombia , producing the lineage Caribbean/South America I , which also includes one sequence isolated in Brazil ( Fig . 2a ) . In the second route the virus disseminated into the Pacific side of South America hitting Peru , Ecuador and Colombia , and subsequently moved back to Venezuela , Cuba , Puerto Rico and Nicaragua; constituting the lineage Caribbean/South America II ( Fig . 2b ) . In the third route the virus went to the Caribbean ( Martinique , Puerto Rico , Cuba , Trinidad and Tobago , Saint Lucia and Anguilla ) and from there into the Southern Cone of South America ( Brazil , Argentina , Bolivia and Paraguay ) , originating the lineage Caribbean/South America III , which also included one sequence from Guyana and another one from Venezuela ( Fig . 2c ) . To analyze the diversity of DENV-3 in Brazil in more detail , we undertook a more rigorous Bayesian phylogenetic analysis of a subset of 202 DENV-3 genotype III sequences which combine all sequences sampled from Brazil ( n = 107 ) , along with selected non-Brazilian ‘background’ sequences ( n = 95 ) . DENV-3 Brazilian sequences were sampled from the Southeast ( n = 74 ) , North ( n = 20 ) , Central-West ( n = 8 ) and Northeast ( n = 5 ) regions ( Fig . 3 and Table S1 ) . DENV-3 non-Brazilian background sequences comprise all genotype III sequences from the lineage Caribbean/South America III ( n = 54 ) and representative sequences from Asia ( n = 10 ) , Africa ( n = 1 ) , Central America and Mexico ( n = 10 ) , lineage Caribbean/South America I ( n = 10 ) , and lineage Caribbean/South America II ( n = 10 ) . The DENV-3 genotype III Brazilian sequences analyzed were distributed in four independent lineages , revealing at least four introduction events of this DENV-3 genotype in the country ( Fig . 4 ) . Most Brazilian sequences ( n = 92; 86% ) grouped in a well supported monophyletic clade ( PP = 0 . 86 ) within the Caribbean/South America III cluster , called BR-I; which contains sequences sampled from all Brazilian regions from 2001 to 2009 , along with sequences isolated in Argentina , Bolivia and Paraguay . A minor proportion of Brazilians sequences ( n = 13; 12% ) isolated in the North region from 2003 to 2008 and Sao Paulo state ( Southeast region ) in 2006 , segregate in a second monophyletic clade ( PP = 1 ) within the Caribbean/South America III lineage , called BR-II; which also includes sequences isolated in Paraguay and Argentina . The third Brazilian lineage is represented by a single isolate sampled in the state of Roraima ( North region ) in 2002 ( GenBank accession DQ118865 ) , which formed a monophyletic cluster ( PP = 1 ) with sequences from several Caribbean Islands within the Caribbean/South America III clade . The fourth Brazilian lineage also correspond to a single sequence isolated in the North region at 2003 ( GenBank accession FJ850079 ) , that is closely related to Venezuelan and Colombian sequences from clade Caribbean/South America I . Most DENV-3 Brazilian sequences included in the present study were retrieved from Sao Paulo ( N = 58; 54% ) and Rio de Janeiro ( N = 13; 12% ) , which are the most populated states of the country . A closer inspection of those DENV-3 strains reveals a significant difference in the pattern of viral dissemination within these regions . While nearly all sequences isolated in Rio de Janeiro from 2002 to 2008 segregate in a single monophyletic cluster ( BR-RJ ) , sequences sampled in Sao Paulo split in three major independent lineages: BR-SP-I ( from 2003 to 2007 ) , BR-SP-II ( from 2006 to 2009 ) and BR-SP-III ( at 2006 ) ( Fig . 4 ) . Furthermore , sequences sampled in Sao Paulo were closely related to Brazilian sequences isolated in the Central-West ( BR-SP-I , BR-SP-II ) and North ( SP-III ) regions; while sequences from Rio de Janeiro showed a closer relationship with sequences isolated in the states of Espirito Santo ( Southeast region ) and Pernambuco ( Northeast region ) ( Fig . 4 ) . A few DENV-3 sequences from Rio de Janeiro and Sao Paulo branched outside the major clades and possibly represent viruses that did not succeed in getting established in those regions . In order to gain insight into the place and timing of introduction of major DENV-3 Brazilian lineages ( BR-I and BR-II ) , we used a Bayesian MCMC phylogeographic approach that jointly estimates the substitution rate , the Tmrca and the spatial diffusion from sampled sequences , while accommodating phylogenetic uncertainty arising from the sequence data . DENV-3 sequences from clades BR-I and BR-II were combined with Caribbean DENV-3 sequences from lineage Caribbean/South America III and with DENV-3 sequences isolated in Central America at the middle 1990s . A specific “character state” was assigned to each DENV-3 sequence based on its geographic origin , according to the following scheme: Central America ( Panama , Nicaragua , and Honduras ) , Greater Antilles ( Puerto Rico and Cuba ) , Lesser Antilles ( Anguilla , Martinique , Saint Lucia , and Trinidad and Tobago ) , South America ( Guyana and Venezuela ) , North Brazil , Southeast Brazil , Northeast Brazil , and Central-West Brazil . Analyses were performed under an equal rates model that assumes the same rate of virus movement between the eight locations . The mean evolutionary rate and Tmrca of the DENV-3 data-set were estimated at 11 . 0×10−4 subs . /site/yr ( 95% HPD: 8 . 3–13 . 8×10−4 subs . /site/yr ) and 1991 ( 95% HPD: 1988–1993 ) , respectively; which are close to those previously reported for the DENV-3 genotype III in the Americas [19] . The spatio-temporal reconstruction suggests that the Caribbean/South America III lineage likely originated in the Lesser Antilles ( PP = 75% ) at around 1997 ( 95% HPD: 1995–1999 ) , and rapidly spread to the Greater Antilles and South America ( Fig . 5 ) . The Brazilian clade BR-I was probably imported from the Caribbean islands into the Southeast region ( PP = 83% ) at around 1999 ( 95% HPD: 1997–2000 ) , while the clade BR-II probably migrated from the Caribbean islands to the North Brazilian region ( PP = 97% ) at around 2001 ( 95% HPD: 1999–2002 ) ( Fig . 5 ) .
The genotype III has established as the major lineage of DENV-3 in the Americas . The phylogenetic analysis presented here confirms that DENV-3 outbreaks occurring in the American continent since the mid-1990s are the result of a single introduction of genotype III . This analysis also suggest that viral introduction probably occurs through Central America or Mexico , and from there the virus spread to the Caribbean and South America following three major routes , giving rise to three independent evolutionary lineages ( Caribbean/South America I to III ) , consistent with previous findings [18] , [19] , [22] . According to this model , Central American countries and Mexico were the hubs of genotype III spread in the Americas , while the Caribbean region acted as a staging post between Central America/Mexico and South America . The lack of evidence of dissemination of new DENV-3 strains in the Americas reveals that despite massive human movement between continents , the establishment of new DENV-3 lineages of Asian and/or African origin in the Americas seems to be an improbable phenomenon . In the last two decades , the current lineages of DENV-3 circulating in the Americas immunized “naturally” the population , due to its wide spread in the continent . This factor may have been decisive to explain the lack of new introductions of Asian and/or African lineages . Brazil has been heavily affected by DENV-3 since the early 2000s and our study reveals that there have been at least four separate introductions of the genotype III into the country , as signified by the presence of four phylogenetically distinct lineages . Three lineages ( BR-I , BR-II , and BR-III ) belong to the Caribbean/South America III clade and were probably imported from the Caribbean islands . The fourth lineage ( BR-IV ) belong to the Caribbean/South America I clade and was probably introduced from Colombia or Venezuela; while we found no evidence of dissemination into Brazil of the Caribbean/South America II lineage that mainly hits the Pacific side of the Andes ( Peru , Ecuador and Colombia ) . Recent studies have shown the re-introduction into Brazil of a new DENV-2 lineage of the American/Asian genotype , that is closely related to DENV-2 strains circulating in some Caribbean islands ( Martinique , Cuba and the Dominican Republic ) [42] , [43] . Thus , the Caribbean region seems to be the main source of new DENV strains introduced into Brazil . The final outcome of each DENV-3 introduction into Brazil was highly variable . The most successful clade was the BR-I . This lineage comprises 86% of DENV-3 Brazilian sequences analyzed in the present study , was detected in all country regions from 2001 to 2009 , and was also disseminated to other countries from the Southern Cone ( Argentina , Bolivia and Paraguay ) . The lineage BR-II comprises 12% of Brazilian sequences and has been detected in the North region between 2003 and 2008 , in Sao Paulo state ( Southeast region ) and Paraguay in 2006 and in Argentina in 2007 . In contrast with the previous clades , the BR-III and BR-IV lineages seem to have failed to become established in Brazil , since both lineages were detected in one single individual each from the North region at 2002 and 2003 , respectively , and none of the later Brazilian isolates grouped within these clades . The detection of DENV-3 lineages BR-I and BR-II in Argentina , Bolivia and Paraguay , points to Brazil as an efficient hub of dissemination of DENV-3 in the Southern cone . Our phylogeographic analysis suggests that the most widely disseminated Brazilian DENV-3 clade ( BR-I ) probably entered into the country through the Southeast region . The state of Rio de Janeiro in the Southeast region is considered as the most important point for the introduction and dissemination of new DENV strains in Brazil as this sate was the place where the first cases of DENV-1 ( 1986 ) , DENV-2 ( 1990 ) , and DENV-3 ( 2000 ) were detected in Brazil [24] . The states of Rio de Janeiro and Sao Paulo also displayed among the largest and heavily dense urban population in the country , contain the most important national and international airports , and are highly connected to other states trough a large roadway and railway system . These conditions create an excellent milieu for introduction and rapid dissemination of new viral strains . On the other hand , lineages BR-II , BR-III , and BR-IV were probably introduced through the North Brazilian region . The close geographic proximity of the Northern states to the Caribbean region may explain the existence of a considerable diffusion of DENV-3 lineages across the northern Brazilian border . The evolutionary analysis suggests that DENV-3 lineage BR-I displayed a period of cryptic circulation for about 1–3 years before its detection by the Brazilian public surveillance network . According to our estimates the BR-I clade emerged around 1999 ( 95% HPD: 1997–2000 ) ; but its presence was first reported in Rio de Janeiro at December 2000 . In agreement with our findings , Romano et al [43] estimated that the new DENV-2 lineage detected in Rio de Janeiro at 2007 was introduced in the country at least 2–3 years earlier . Our analysis also supports that DENV-3 clades BR-I and BRII were disseminated at a very fast rate throughout Brazil . The reconstructed spatio-temporal pattern of DENV-3 dissemination indicates that lineages BR-I and BR-II emerged at around 1999 and 2001 in the Southeast and North regions , respectively; and only a few years later ( 2002–2006 ) both lineages were detected in regions located almost 5 , 000 km away . A similar finding was recently observed for DENV-4 that , after an absence of 30 years , was detected in Roraima state ( North region ) in July 2010 and only eight months later the virus was detected in several states of the Northeast and Southeast regions [44] . This study also points to the existence of a non-random pattern of DENV-3 dissemination across Brazilian regions , and further reveals significant differences in the molecular profile of DENV-3 epidemics occurring at the two most populated states of the country , Sao Paulo and Rio de Janeiro . The DENV-3 epidemics occurring in Sao Paulo state during the 2000s were seeded by the introduction and co-circulation of at least three viral strains ( BR-SP-I , BR-SP-II , and BR-SP-III ) , consistent with previous observations [26] , [27] . By contrast , the successive DENV-3 outbreaks taking place in Rio de Janeiro over the last decade were the result of the long-term persistence and in situ evolution of a single viral lineage ( BR-RJ ) . Those DENV-3 strains detected in Sao Paulo were closely related to DENV-3 strains circulating in the Central-Western and Northern Brazilian regions . While , the BR-RJ lineage circulates in the states of Espirito Santo ( Southeast region ) and Pernambuco ( Northeast region ) . Of note , despite the intense movement of people , high geographic proximity and dense viral sampling , we found no evidence of an important DENV-3 flux between Sao Paulo and Rio de Janeiro . In conclusion , our study demonstrates that there have been at least four introductions of the same DENV-3 genotype III in Brazil , although only two viral lineages seems to have become efficiently established and disseminated in the country . The Caribbean islands were the main source of DENV-3 viruses that arrived into Brazil , and the Northern and Southeastern Brazilian regions seems to be most important hubs of introduction and dissemination of such DENV-3 lineages . Our analyses also suggest that DENV-3 strains circulated for at least 1–2 years until meet favorable conditions to initiate an outbreak and to be detected by the Brazilian public surveillance system . Continuous epidemiological surveillance and dense sequencing of viral strains circulating in all Brazilian regions are of paramount importance to early detection of newly emerging DENV lineages , to understanding the patterns of DENV dissemination across country regions , and to guide the actions for dengue control programs in Brazil . | Dengue is a major health problem in the tropics and the incidence of dengue fever and dengue hemorrhagic fever in Brazil experienced a significant increase since the emergence of dengue virus type-3 ( DENV-3 ) . In this study , the authors reconstruct the spatio-temporal dispersion pattern of the DENV-3 lineage that circulates in Brazil and the Americas . The authors found that DENV-3 outbreaks occurring in the American continent since the mid-1990s are the result of a single introduction of genotype III . The Central American countries and Mexico were the hubs of genotype III spread in the Americas , while the Caribbean region acted as a staging post between Central America/Mexico and South America . The authors estimate that there have been at least four introductions of the DENV-3 genotype III in Brazil , although only two of them succeeded in getting established and disseminating through the country . The Lesser Antilles ( Caribbean ) were the main source of DENV-3 viruses that arrived into Brazil , and the North and Southeast country regions seem to be most important hubs of introduction and dissemination of DENV-3 lineages . These findings offer important information to perform more effective surveillance programs to detect introduction and dispersal of new DENV lineages in Brazil . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"virology",
"emerging",
"viral",
"diseases",
"biology",
"microbiology",
"viral",
"evolution"
] | 2012 | Origin and Evolution of Dengue Virus Type 3 in Brazil |
Dengue is an increasingly incident disease across many parts of the world . In response , an evidence-based handbook to translate research into policy and practice was developed . This handbook facilitates contingency planning as well as the development and use of early warning and response systems for dengue fever epidemics , by identifying decision-making processes that contribute to the success or failure of dengue surveillance , as well as triggers that initiate effective responses to incipient outbreaks . Available evidence was evaluated using a step-wise process that included systematic literature reviews , policymaker and stakeholder interviews , a study to assess dengue contingency planning and outbreak management in 10 countries , and a retrospective logistic regression analysis to identify alarm signals for an outbreak warning system using datasets from five dengue endemic countries . Best practices for managing a dengue outbreak are provided for key elements of a dengue contingency plan including timely contingency planning , the importance of a detailed , context-specific dengue contingency plan that clearly distinguishes between routine and outbreak interventions , surveillance systems for outbreak preparedness , outbreak definitions , alert algorithms , managerial capacity , vector control capacity , and clinical management of large caseloads . Additionally , a computer-assisted early warning system , which enables countries to identify and respond to context-specific variables that predict forthcoming dengue outbreaks , has been developed . Most countries do not have comprehensive , detailed contingency plans for dengue outbreaks . Countries tend to rely on intensified vector control as their outbreak response , with minimal focus on integrated management of clinical care , epidemiological , laboratory and vector surveillance , and risk communication . The Technical Handbook for Surveillance , Dengue Outbreak Prediction/ Detection and Outbreak Response seeks to provide countries with evidence-based best practices to justify the declaration of an outbreak and the mobilization of the resources required to implement an effective dengue contingency plan .
Responding to the rapidly increasing public health importance of dengue , the 2002 World Health Assembly Resolution WHA55 . 17 urged greater commitment to dengue among Member States and throughout the World Health Organisation ( WHO ) . One response of particular significance was the Revision of the International Health Regulations ( WHA58 . 3 ) in 2005 , where dengue was included as an example of a disease that would constitute a public health emergency of international concern . It was against this background that the World Health Organization’s Special Programme for Research and Training in Tropical Diseases ( WHO/TDR ) initiated a Dengue Scientific Working Group ( SWG ) of 60 experts from 20 countries , which met in October 2006 to review existing knowledge on dengue and establish priorities for future dengue research [1] . The research priorities identified were organized into four major research streams and those for dengue surveillance and outbreak response included the following primary recommendations: At the same time , a discussion began that was centred on the need for an evidence base to better inform policy recommendations . The WHO Dengue Guidelines for Diagnosis , Treatment , Prevention and Control [2] was followed by the WHO Handbook for Guideline Development [3] , which stressed specifically the need for high-level evidence when developing guidelines , particularly through systematic literature reviews . The importance of systematic reviews for linking research and practice was also highlighted by others [4] , with one [5] stating “policymakers need systematic reviews that are policy relevant , rigorous , and translatable to their local context , actionable , timely and well communicated” . With this in mind , WHO/TDR together with the WHO/NTD ( Department for Neglected Tropical Diseases ) and WHO Regional Offices set out to develop an evidence-based handbook [6] for early dengue outbreak detection and response . The project was financially supported by a grant from the European Commission ( grant number m281803 ) to the IDAMS network ( www . idams . eu ) within the 7th Framework Programme and by TDR/WHO . Accordingly , this handbook is not intended to be a direct implementation guideline but a framework for developing a national plan , requiring local adaptations to acknowledge fine-scale programme components . The latter point takes into account that contingency response planning requires consideration and incorporation of numerous contextual details such as recognition of the structure of the health and vector control services , available infrastructure and budget , human resources , willingness of staff to cooperate , and many others . Here we present an outline of the handbook , summarizing the main components of a national contingency plan for dengue outbreaks and indicating the key elements that are evidence-based and those that require further research efforts .
The development of this evidence-based handbook for dengue contingency planning used a step-wise approach . The first step established an overview by identifying knowledge gaps and commissioning new systematic literature reviews covering the following topic areas: a ) dengue vector control [7–16] b ) outbreak response [17]; c ) dengue disease surveillance [18 , 19] and dengue vector surveillance [20]; and d ) economic aspects [21] . In a second step , mixed ( qualitative and quantitative ) research methods were used to identify a ) factors leading to the success or failure of national dengue control programmes , b ) decision-making that resulted in the declaration of a state of emergency , c ) stakeholders`perceptions of their contingency plans , and d ) gaps regarding the practical application of contingency plans . These studies were conducted in Bolivia , Brazil , Cambodia , Indonesia and Thailand [22] and were complemented by a comparative analysis of dengue contingency plans from 13 countries [23] . Finally , a multi-country study was conducted that assessed dengue contingency planning and outbreak management in 10 countries [24] . The country selection process varied from study to study based on the dengue burden , information available for the information searched , willingness to participate or a history of recent dengue outbreaks , where appropriate . In the third step , a retrospective analysis of the predictive ability of variables to warn of forthcoming outbreaks was conducted . Epidemiological and meteorological variables were analysed using datasets from Brazil , Dominican Republic , Malaysia , Mexico and Vietnam [25] . These were selected based on dengue endemicity , dengue burden and those countries with a recent history of dengue outbreaks . In common with the existing scientific literature , the model identified a number of variables that could be used to predict dengue outbreaks with sufficient sensitivity and relatively few false alarms . This model is currently being evaluated in a prospective feasibility and cost-effectiveness study in Brazil , Malaysia and Mexico , as part of an evaluation of a staged response system , designed to gradually implement timely interventions in response to weak or stronger alert signals . In a last step , we developed a computer-assisted early warning system designed to run on a wide variety of platforms such as Microsoft Excel , STATA , R and SPSS . Such software was developed to build capacity in countries that currently lack the resources to implement predictive dengue technologies . A user-guide was prepared to describe and explain the early warning system , how to use it to identify potential alarm signals at the district level , and how programme managers might use these indicators to provide timely evidence-based alerts to subsequent dengue outbreaks . These developments can equip regional epidemiologists with the technical capacity to rapidly obtain the information required to formulate timely outbreak response . NB: A formal assessment of quality of evidence of the included literature was not performed in this paper—this article describes the developmental process of the handbook . The material used for the development of the handbook , however , included the highest available evidence for each subsection: a ) Guidelines and Handbooks ( 2 , 3 , 26 and 27 ) , b ) Systematic Reviews and Meta-analysis ( 7–22 ) , c ) RCTs/cRCTs ( 28 ) , d ) Cohort Studies ( 29–32 ) , e ) Mixed-Method Study Designs ( 22–24 , 33 and 34 ) , f ) Others ( primary research–non controlled and reviews-non systematic ) ( 4 , 5 , 25 , 34 , 40-67 ) , and g ) Reports ( 1 , 68–70 ) .
In a comparison of existing practices in 10 countries in Asia and Latin America [24] , outbreak response plans varied in quality and comprehensiveness , particularly regarding early response measures as well as detailed specifications of actions to be taken . Harrington et al . [23] compared 13 country contingency plans for dengue from Asia , Latin America and Australia , and one international plan by the World Health Organization . The authors found that outbreak governance was weak , in part due to a lack of clarity of the roles of stakeholders , poor surveillance contributed to delays in response , there was a lack of combining routine data with additional alerts , and the absence of triggers to initiate an early response . Frequently , an outbreak was undefined and early response mechanisms based on alert signals were neglected . Therefore it was concluded that a model contingency plan for dengue outbreak prediction , detection and response , including resource planning , training , monitoring and evaluation , could help national disease control authorities to develop their own more detailed and functional context-specific plans . Badurdeen et al . [24] also found that information on dengue was based on compulsory notification and reporting ( “passive surveillance” ) , coupled with laboratory confirmation ( in all participating Latin American countries and some Asian countries ) or by using a clinical syndromic definition . Seven countries [24] had sentinel sites with active dengue reporting , and some also had virological surveillance . Six countries had a formal definition for dengue outbreaks , distinguishing them from seasonal incident peaks . Countries collected data on a range of warning signs that could identify outbreaks early , but none had developed a systematic approach to identify and respond to the early stages of an outbreak . Through discussions at an expert meeting , suggestions were made for the development of a more standardised approach in the form of a model contingency plan , together with agreed upon outbreak definitions and country-specific risk assessment schemes , in order to initiate timely response activities [24] . Among the systematic reviews performed to date , considerable variation was observed in the number and application of outbreak definitions , and definitions have been numerous , non-standardised and inconsistently applied [24] . In order to ensure that an early warning system for dengue outbreaks is effective , efficient and timely , outbreak definitions must be able to distinguish between true outbreaks and seasonal increases in dengue . Therefore , outbreaks were defined as caseloads of an order much larger than would otherwise be expected during the respective season and/ or occurring in unexpected locations . This task is complex but has been somewhat simplified by the use of the Endemic Channel . Outbreak definitions defined using the Endemic Channel often base thresholds on 2 standard deviations ( SD ) above the mean number of historic dengue cases , which closely reflects the 1 . 96 SDs used in confidence estimates to capture 95% of the variation about the mean . However , such values are often applied across large spatial dimensions , resulting in the loss of information that may be reflective of the localised transmission dynamics inherent to dengue [25] . Considering this , models need to be parameterised according to the context [41] . In support of this evidence , Bowman et al . [25] also found that the multiplier of the standard deviation may be context-dependent and reported that 1 . 25SD could be used as an efficient multiplier . Brady et al . [34] modelled five approaches to define an outbreak using different summary statistics ( i . e . , recent mean , monthly mean , moving mean , cumulative mean , and fixed incidence threshold ) . The authors reconfirmed that outbreaks remain highly heterogeneous , in part due to location-specific transmission factors but also due to the methodologies used to define the outbreaks . In summary , outbreak definitions may need to be spatially stratified , with consideration given to available contextual data and summary statistics , and include operational perspectives to best identify the most important stages of an outbreak in order to ensure a timely response . Until consensus is reached on the most appropriate method to define outbreaks , definitions using simple approaches such as the Endemic Channel should not be discounted . Although outbreak definitions require further empirical work , they remain accessible to both programme managers and regional epidemiologists alike , and if applied at relatively fine scales offer a useful tool for outbreak detection , planning and response [25] . Syndromic surveillance [69] may contribute important data on alarm signals in early warning systems for dengue outbreaks . A number of variables that provide predictive warning have been identified and include the rate of school absenteeism [42–44] , the volume of internet-based health inquiries [45] , the malaria negative rate in fever patients [46 , 47] , non-specific laboratory requests ( as malaria negativity rates or as thrombocytes requested ) , and fever alerts or use of clinical syndromic definitions [48–51] and the proportion of virologically confirmed cases [52 , 53] . Runge-Ranzinger et al . [19] also found six studies [52 , 54–58] that showed serotype changes were positively correlated with the number of reported cases or with dengue incidence , with lag times of up to 6 months , indicating that a change in serotype may be a predictor ( alarm signal ) for dengue outbreaks . Three studies [59–61] found that data on Internet searches and event-based surveillance correlated well with the epidemic curve derived from surveillance data , suggesting that this method may be useful to predict outbreaks . Other approaches such as the use of socioeconomic indicators ( presence of water and trash collection services ) or environmental parameters ( e . g . , presence of tire repair shops , rainfall , relative humidity ) for risk assessment [62] . Modelling tools [63] also have potential , although at this stage they remain either context-dependent or under evaluation . In order to develop a dengue outbreak alert model , several potential alarm signals were evaluated retrospectively [25] . A simple approach combining the Shewhart method and Endemic Channel was used to identify alarm signals that could predict dengue outbreaks . Five country datasets were compiled by epidemiological week over the years 2007–2013 and these data were split to form a historic period ( 2007–2011 ) and evaluation period ( 2012–2013 ) . To parameterise the model , associations between alarm signals and outbreaks were analysed using logistic regression during the historic period . Thereafter , these associations were combined with alarm variable data during the evaluation period to predict dengue . Subsequently , model performance was described using sensitivity and positive predictive value ( PPV ) ( the proportion of false alarms ) . Across Mexico and Dominican Republic , an increase in probable cases predicted outbreaks of hospitalised cases with sensitivities and PPVs of 93%/ 83% and 97%/ 86% respectively . In addition , an increase in mean temperature in Mexico and Brazil predicted outbreaks of hospitalised cases , with sensitivities and PPVs of 79%/ 73% and 81%/ 46% respectively . These results were particularly promising as these variables were broadly predictive of dengue outbreaks across different countries , despite the varied surveillance systems , case definitions and localised variation in transmission potential often associated with dengue [25] . Clearly , routine surveillance can underestimate the true burden of disease , however the prediction of cases was not hindered , as the case definition remained consistent throughout the historic and evaluation periods and the systems were accurately reflecting the burden of disease . Documented effective outbreak interventions and evidence gaps were analysed in a systematic review by Pilger et al . [17] . Different strategies in the organization of outbreak response were identified , showing that control activities for a dengue outbreak need to be multi-sectoral , multidisciplinary and multilevel; they also require environmental , political , social and medical inputs for coordination so that successful activities of one sector are not weakened by the lack of commitment from another . Risk communication is a fundamental element of managing a public health threat by encouraging positive behavioural change and maintaining public trust [26] . Outbreaks can be highly charged political and social events whereby “outbreak declaration and transparency from expert to audience is surrounded by political and economic overtones” [64] . Therefore it is critical that risk communication plans are prepared prior to an event and that individuals serving as spokespersons are provided with training in public speaking and risk communication in order to proactively manage the outbreak response , along with political or other issues that may arise [26] . The logistics of outbreak response activities are challenging . It is important to assess the additional human resources that will be required , both for clinical management of cases and vector control . This includes redistribution of staff , increased staffing levels and extension of work shifts [24 , 70] . Overwork and subsequent demotivation of health staff have been identified as likely problems , often caused by increased demands by politicians and the community [7] . Therefore , staff training and preparation for an outbreak in the inter-epidemic period and supportive supervision during the outbreak can help staff cope with excessive challenges during the outbreak [17] . Investment in human resources must come prior to the outbreak , thus outbreak response planning requires a section documenting the activities to be performed in the inter-epidemic period in preparation for an outbreak , as opposed to preventative control . The contingency plan has also to include the “stopping rules” , i . e . , when and how to declare the end of the outbreak , halting the outbreak response and continuing with routine interventions . Horstick et al . [7] undertook an analysis of vector services with two methods: a systematic literature review and case studies that included stakeholder interviews and completion of questionnaires in Brazil , Guatemala , The Philippines , and Vietnam . In the systematic literature review , staffing levels , capacity building , management and organization , funding , and community engagement were found to be insufficient . The case studies confirmed most of these findings , with stakeholders reporting: 1 ) lack of personnel ( entomologists , social scientists and operational vector control staff ) ; 2 ) lack of technical expertise at decentralized levels of services; 3 ) insufficient budgets; 4 ) inadequate geographical coverage; 5 ) interventions that rely mostly on insecticides; 6 ) difficulties engaging communities; 7 ) little capacity building; and 8 ) minimal monitoring and evaluation . Stakeholders’ doubts about service effectiveness were widespread , but interventions were assumed to be potentially effective with increased resources . The authors highlighted the need for operational standards; evidence-based selection/ delivery of combinations of interventions; development/ application of monitoring and evaluation tools; and needs-driven capacity building . These recommendations are in line with those from Pilger et al . [17] , who reported that combining interventions that involved vector control ( elimination of larval habitats with community involvement; appropriate use of insecticides in and around houses ) and capacity training of medical personnel , in combination with laboratory support , were crucial for the successful control of outbreaks . For single vector control interventions , systematic reviews are available on peridomestic space spraying [12] , Bacillus thuringiensis israelensis ( BTI ) [9] , temephos [16] , copepods [13] and larvivorous fish [15] . Horstick and Runge-Ranzinger [65] found that: 1 ) vector control could be effective , but implementation and coverage remained an issue; 2 ) single interventions were probably not useful; 3 ) combinations of interventions had mixed results; 4 ) careful implementation of vector control measures may be most important; and 5 ) outbreak interventions were often applied with questionable effectiveness . A systematic review and meta-analysis found that community-based multiple interventions ( such as environmental management or clean up campaigns , refuse collection , the formation of community working groups , social mobilization strategies , water covers , and larviciding ) can signficiantly reduce vector densities [14] . Results from a cluster randomised controlled trial in Latin America [28] reported reductions in dengue cases following similar interventions . Bowman et al . [14] also reported that house screens on external doors and windows could be protective against dengue transmission , but that there was insufficient evidence from randomized controlled trials to determine whether or not insecticide space-spraying or fogging could impact dengue transmission . Best practices in vector control remain to be defined for any setting ( i . e . , which tools or methods the community should employ ) , as well as what constitutes adequate or sufficient coverage in order to impact the vector population and virus transmission . This includes operational aspects , quality of delivery and best combination of interventions for successful vector control during outbreaks . Bowman et al . [14] also found no evidence that interventions such as mosquito coils , repellents , bed nets , or mosquito traps could reduce dengue incidence . Finally , indoor residual insecticide spraying and approaches involving the use of genetically modified ( GM ) mosquitoes or the intracellular symbiont Wolbachia [66] have considerable potential for dengue vector control , but have not yet been evaluated sufficiently to draw conclusions about their effectiveness . Good clinical case management during an outbreak has been crucial in reducing the case fatality of dengue from 10–20% to less than 1% in many countries over the past two decades [67] . The training of health professionals in diagnosis and management , as well as robust laboratory facilities must be prioritized , as this will effectively dictate case management and influence mortality rates . The best ways to achieve successful training may be through hands-on training during ward rounds and case conferences [17] . The importance of emergency resources and funding for outbreak response including clinical supplies has been highlighted as an important element of preparedness and response planning [2 , 24] . Badurdeen et al . [24] found that the surge capacity of hospitals with recent dengue outbreaks varied . Hospital outbreak management plans were present in 9 of 22 participating hospitals in Latin America and 8 of 20 participating hospitals in Asia , also highlighting the need for contingency planning . Further information on triage systems , case management and referrals are available elsewhere [27] .
Preparedness planning starts in the inter-epidemic phase and success is dependent on the combination of year-round routine activities , often established in a National Dengue Prevention and Control Plan , up-scaling of routine vector control interventions and communication activities , and timely and systematically initiated additional measures during an outbreak . The proposed handbook suggests seven areas for contingency planning which can either be integrated into the existing national plan or developed as a separate add on . A summary of the recommendations for dengue surveillance , outbreak alert and response are given below in Fig 2 . With respect to timely contingency planning , it is crucial to ensure that a context-specific dengue contingency plan has detailed instructions that allow managers to distinguish between routine interventions required during inter-epidemic periods and those needed during outbreak interventions ( i . e . , up-scaling of preventive interventions before the start of the “dengue season” vs specific outbreak procedures ) . The contingency plan should ensure continuity between timely surveillance ( including multiple signals ) , outbreak alerts , and outbreak confirmation based on a clear definition , outbreak declaration , and finally implementation of contingency responses . A key first step is to identify the person/ unit/ agency/ institution responsible for specific activities , to define the roles and responsibilities of each person involved , to ensure the regulatory framework exists to support and facilitate the contingency response , and to ensure that the means and capacity exist to implement the full set of specified contingency activities . This initial planning also takes into consideration the need for human resource preparedness planning for all sectors including distribution of the plan to all stakeholders , instructions for training , and a detailed plan for monitoring and evaluation of preparedness activities and response . In order to optimize surveillance , a focus on reducing under-reporting and improving reporting timeliness should strengthen routine surveillance systems . It is important to establish a common understanding across all stakeholders on the purpose and objectives of surveillance , to improve feedback of reported data and to provide a clear—ideally electronic—data flow . Enhancement strategies such as sentinel-based reporting , staff motivation , syndromic surveillance , and monitoring additional alarm signals , e . g . , virological , serological surveillance , should be included along with use of the simplified and standardized WHO 2009 [2] case classification . With respect to laboratory support , reporting available laboratory confirmation of cases to the surveillance system is recommended along with information about the circulating serotype/genotype . The laboratory section of the national contingency plan should include details on virus isolation , PCR , NS-1 , ELISA , serological confirmation by IgM and IgG , appropriate use of rapid diagnostic tests , storage of samples , and cold chain logistics ( see WHO [2] ) . The purpose of laboratory tests , test results and their interpretation should be described and accompanied by a flowchart that visually depicts the timing of various tests and destinations of samples provided . Laboratory-specific processes of outbreak investigation and confirmation should be defined , including quality control , capacity building , prevention of stock-outs , and the role of different levels of laboratories within the national laboratory network . The outbreak definition in a national dengue contingency plan should be context-specific and based on the threshold of local historical disease data reported through the national surveillance system . For example , countries may use the Endemic Channel where the threshold is based on z standard deviations ( SD ) above the mean number of historic dengue cases ( currently often z = 2 , or according to recent evidence z = 1 . 25 , which is close to the 3rd percentile above the median ) . Efforts should be made to distinguish between standardized definitions of an outbreak and the local/ national threshold used to initiate outbreak response , considering that large spatial dimensions will result in the loss of information of localised transmission dynamics . In addition to those mentioned herein , additional predictive variables , such as meteorological variables , in particular mean daytime weekly temperature , may be of use in local contexts . It is crucial to define an alert algorithm based on different alarm signals ( epidemiological thresholds plus the use of meteorological data , syndromic surveillance data , laboratory results or perhaps entomological metrics ( although there is currently little evidence of quantifiable associations between vector indices ) ) to increase sensitivity and specificity for predicting forthcoming outbreaks . The outbreak response should be staged in accordance with the identified level of risk ( i . e . , Initial Response , Early Response , Emergency Response ) to ensure that resources are used efficiently and proportionately . From a managerial aspect , the organization of multidisciplinary response teams , details of logistic/ operational considerations , including standard operating procedures , stopping rules , monitoring and evaluation , staff training prior to an epidemic , resource mobilisation and financial management , legal framework , and recruitment of additional staff during outbreak response , are all important issues for consideration . This includes the training of management personnel in risk communication to ensure timely and appropriate communication within and without the health sector and throughout the broader population . The process of outbreak declaration and risk communication should be well defined and described , so that community engagement and stakeholder involvement contribute to a successful outbreak response at the local level . With respect to vector control interventions , the focus should be on quality and coverage of vector interventions , as these remain key issues . The involvement of communities in vector control activities , for example “search and eliminate” , increases the likelihood that expanded coverage will be achieved; notably , community-based interventions can impact vector indices , and some evidence exists for an impact on dengue incidence . House screening has demonstrated an impact on dengue incidence and may be an effective intervention against dengue where the context is appropriate . While limited evidence demonstrates a reduction in vector indices following outdoor fogging , there is no evidence yet for an impact on dengue incidence . With respect to clinical case management , timely alert of clinicians and a hospital outbreak management plan that includes planning for additional beds and staff are essential . Ensuring triage systems for case management , referrals [27] and mortality reviews will improve case management . Disease transmission control in hospitals as well as regular and timely training of hospital personnel must also be considered . While gaps in knowledge and evidence still remain , much has been accomplished over the past decade that provides a solid basis for evidence-based contingency planning . With the WHO 2009 [2] dengue case classification , improved diagnostic tests and increased national laboratory capacity , stronger national surveillance systems , and ongoing research to develop algorithms that can be used in an operational setting , countries are in a better position to create a dengue contingency plan that reflects their national and local contexts and optimizes available resources for outbreak response . | An evidence-based handbook was generated to facilitate deployment of dengue surveillance and response systems for timely and effective management of outbreaks , and to identify the factors required for success . Evidence was evaluated using literature reviews , policymaker and stakeholder interviews , assessment of dengue contingency planning and outbreak management in ten endemic countries , and a statistical analysis to identify outbreak early warning signs in five countries . Best practices for managing dengue outbreaks included timely and context-specific dengue contingency plans that distinguished between routine practices and outbreak interventions , surveillance systems , outbreak definitions , alert algorithms , and managerial , clinical and vector control capacity . A computer-assisted early warning system was developed to enable each locality to develop its own context-specific scheme . Today , most countries do not have comprehensive , detailed contingency plans for dengue outbreaks , responding simply by intensifying vector control , with minimal focus on integrated management of clinical care , epidemiological , laboratory and vector surveillance , and risk communication . To rectify this , our handbook provides countries with evidence-based best practices to justify the declaration of an outbreak and for the mobilization and management of appropriate resources required to implement a dengue contingency plan . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"medicine",
"and",
"health",
"sciences",
"tropical",
"diseases",
"global",
"health",
"neglected",
"tropical",
"diseases",
"infectious",
"disease",
"control",
"research",
"facilities",
"research",
"and",
"analysis",
"methods",
"public",
"and",
"occupational",
"health",
"infectious",
"diseases",
"agrochemicals",
"epidemiology",
"dengue",
"fever",
"research",
"assessment",
"research",
"laboratories",
"agriculture",
"insecticides",
"systematic",
"reviews",
"disease",
"surveillance",
"government",
"laboratories",
"biology",
"and",
"life",
"sciences",
"viral",
"diseases"
] | 2016 | Dengue Contingency Planning: From Research to Policy and Practice |
The neutral theory of molecular evolution predicts that the amount of neutral polymorphisms within a species will increase proportionally with the census population size ( Nc ) . However , this prediction has not been borne out in practice: while the range of Nc spans many orders of magnitude , levels of genetic diversity within species fall in a comparatively narrow range . Although theoretical arguments have invoked the increased efficacy of natural selection in larger populations to explain this discrepancy , few direct empirical tests of this hypothesis have been conducted . In this work , we provide a direct test of this hypothesis using population genomic data from a wide range of taxonomically diverse species . To do this , we relied on the fact that the impact of natural selection on linked neutral diversity depends on the local recombinational environment . In regions of relatively low recombination , selected variants affect more neutral sites through linkage , and the resulting correlation between recombination and polymorphism allows a quantitative assessment of the magnitude of the impact of selection on linked neutral diversity . By comparing whole genome polymorphism data and genetic maps using a coalescent modeling framework , we estimate the degree to which natural selection reduces linked neutral diversity for 40 species of obligately sexual eukaryotes . We then show that the magnitude of the impact of natural selection is positively correlated with Nc , based on body size and species range as proxies for census population size . These results demonstrate that natural selection removes more variation at linked neutral sites in species with large Nc than those with small Nc and provides direct empirical evidence that natural selection constrains levels of neutral genetic diversity across many species . This implies that natural selection may provide an explanation for this longstanding paradox of population genetics .
The level of neutral genetic diversity within populations is a central parameter for understanding the demographic histories of populations [1] , selective constraints [2] , the molecular basis of adaptive evolution [3] , genome-wide associations with disease [4] , and conservation genetics [5] . Consequentially , numerous empirical surveys have sought to quantify the levels of neutral nucleotide diversity within species , and considerable theory has focused on understanding and predicting the distribution of genetic variation among species . All else being equal , under simple neutral models of evolution , levels of neutral genetic diversity within species are expected to increase proportionally with the number of breeding individuals ( the census population size , Nc ) . Although this prediction is firmly established , surveys of levels of genetic variation across species have revealed little or no correlation between levels of genetic diversity and population size [6–9] . This discrepancy—first pointed out by Richard Lewontin in 1974 [6]—remains among the longest standing paradoxes of population genetics . One possible explanation for this disagreement is an inverse correlation between mutation rate and population size . This is expected if there is relatively weak selection against alleles that cause higher mutation rates [8 , 10] . Alternatively , this paradox could result from greater impact in large populations of nonequilibrium demographic perturbations such as higher variance in reproductive success [11] or population size fluctuations [12] . Indeed , one recent empirical study suggests that demographic factors play an important role in shaping levels of genetic diversity within animal populations [13] . However , none of these potential explanations is sufficient to fully account for the observed patterns of neutral diversity across species [8] . Another potential cause of this paradox is the operation of natural selection on the genome [7 , 14 , 15] . Natural selection can impact levels of neutral diversity via the adaptive fixation of beneficial mutations ( hitchhiking; HH ) [7 , 15 , 16] and/or selection against deleterious mutations ( background selection; BGS ) [17 , 18] . Both processes purge neutral variants that are linked to selected mutations , implying that if natural selection is sufficiently common in the genome , it can reduce observed levels of neutral polymorphism . Furthermore , theoretical arguments [7 , 14 , 19] suggest that , when the impact of natural selection is substantial , the dependence of neutral diversity on population size is weak or even nonexistent . Although many authors have demonstrated that natural selection could , in principle , be sufficiently common to explain Lewontin’s paradox [7 , 8 , 14–16 , 20] , few direct empirical tests of this explanation exist . One unique prediction of the hypothesis that natural selection is a primary contributor to disparity between Nc and levels of neutral genetic variation within species is that natural selection will play a greater role in shaping the distribution of neutral genetic variation in species with large Nc . To test this prediction , we relied on the fact that the impact of natural selection on linked neutral diversity depends on the local recombinational environment . In regions of relatively low recombination , selected variants affect more neutral sites through linkage , and vice versa , in regions of relatively high recombination . The resulting correlation between recombination and polymorphism [21–26] ( reviewed in depth in [27] ) allows a quantitative assessment of the magnitude of the impact of selection on linked neutral diversity ( e . g . , [22 , 23 , 26 , 28] ) . Specifically , if the effects of linked selection can explain the lack of correlation between neutral diversity and population size , we expect that species with larger population sizes will display stronger correlations between recombination and polymorphism than those with smaller population sizes and show a concurrently larger impact of natural selection on levels of neutral diversity across the genome . Although empirical studies that explore the relationship between neutral diversity and population size are relatively infrequent compared to theoretical studies on this topic , there are two interesting patterns that merit consideration here . First , the proportion of nonsynonymous substitutions that have been driven to fixation by positive selection varies widely across taxa . In humans [29] , yeast [30] , and many plant species [31] , estimates of this proportion are close to zero . In contrast , in Drosophila [32 , 33] , mice [34] , and Capsella grandiflora [35] , as well as other taxa ( reviewed in [8] ) , a large fraction of nonsynonymous substitutions are inferred to have been driven to fixation by positive selection , implying that natural selection is common in the genomes of these organisms ( which generally have large Nc ) . Second , the strength of the correlation between polymorphism and recombination varies widely among the limited number of taxa [8 , 27] that have been studied in depth . Here again , Drosophila [21 , 25 , 36] is among the taxa that shows the strongest correlation and thus the clearest evidence for natural selection , and the correlation in Drosophila is substantially larger than , for example , in humans [28] . In a related study to the work presented here , Bazin et al . [37] showed that there is no correlation between nucleotide diversity in nonrecombining mtDNA and nucleotide diversity in the nuclear genome . While this is consistent with some predictions of theoretical work on this subject , the mitochondrion has unusual patterns of replication and inheritance , and it is therefore challenging to disentangle the processes that generate diversity from those that shape its distribution across the genome . Although suggestive , the evidence accrued thus far is fragmentary , has not been analyzed in aggregate , and varies widely in quality of samples , data collection , and analyses performed [8 , 27] . It is therefore difficult to draw firm conclusions about the relative importance and prevalence of natural selection in shaping patterns of genetic variation in the genome based on existing studies . Due to rapid advances in genome sequencing technologies , whole genome polymorphism data are now available for a wide variety of species ( e . g . , [36 , 38] ) , and these data enable us to conduct a quantitative test of the natural selection hypothesis as an explanation for Lewontin's paradox . Towards this , we identified 40 species with sufficiently high quality reference genomes , linkage maps , and polymorphism data to enable a broad-scale , robust comparison of the relative strength of correlation between polymorphism and recombination rate within a single unified alignment , assembly , and analysis framework . Using these data , and reasonable proxies for Nc , we show that the effect of selection on linked nucleotide diversity is indeed strongly correlated with population size . In other words , natural selection plays a disproportionately large role in shaping patterns of genetic variation in species with large Nc , confirming the idea that natural selection is an important contributor to Lewontin’s paradox .
We identified 40 species ( 15 plants , 6 insects , 2 nematodes , 3 birds , 5 fishes , and 9 mammals ) for which a high-quality reference genome , a high-density , pedigree-based linkage map , and genome-wide resequencing data from at least two unrelated chromosomes within a population were available ( Table 1 , S1 Table , S2 Table ) . Because our model ( below ) requires that recombination has been sufficiently frequent to uncouple genealogies across large tracts of DNA on chromosomes , we required that each species have an obligatory sexual portion of its life cycle . This requirement necessarily excludes clades such as bacteria , which are predominantly clonally propagated . Nonetheless , extending this framework to bacterial taxa will be an important step towards understanding the mechanisms by which natural selection shapes patterns of variation across the tree of life . Additionally , our sampling is biased towards more commonly studied clades ( e . g . , mammals ) , but this is unavoidable in this type of analysis , and there is no reason in principle why this taxonomic bias would affect the basic conclusions we describe here , as the sampled taxa likely span a large range of census population sizes . After acquiring sequence data , we developed and implemented a bioinformatic pipeline to align , curate , and call genotype data for each species ( see S1 Fig and methods for a full description of the bioinformatics pipeline ) . We further used the available genetic maps to estimate recombination rates across the genomes . Across all species , we analyzed recombination across nearly 385 , 000 markers and aligned more than 63 , 000 , 000 , 000 short reads . This is therefore one of the largest comparative population genomics dataset that has been assembled to date . We used both simple nonparametric correlations and explicit coalescent models to test for a relationship between the impact of selection on linked neutral diversity and census size . Although correlations between recombination rate and neutral diversity are informative , the extensive literature in theoretical population genetics provides an opportunity to develop a robust modeling approach . Two primary types of selection can introduce a correlation between recombination rate and levels of nucleotide diversity: background selection ( BGS ) and hitchhiking ( HH ) . Here , we are not primarily concerned with distinguishing between the two models , and so focus on their joint effects . In addition to combining BGS and HH , we would also like to relax the assumption that these processes act uniformly across the genome . All else being equal , regions of the genome with a higher density of potential targets of selection should experience a greater reduction in neutral diversity . Starting from considerable prior theoretical work [14 , 17 , 18 , 32 , 39–41] , we develop an explicit model relating polymorphism , recombination rate , and density of functional elements in the genome . We fit both a joint model that allows for both HH and BGS , as well as models of BGS only , HH only , and a purely neutral model ( in which there is no predicted correlation between recombination or functional density and neutral diversity ) . Using these models , we estimate the proportion of neutral diversity removed by linked selection for beneficial alleles and/or against deleterious alleles ( Fig . 1 ) for each species , as well as the relative likelihood of each model . In practice , it is not feasible to determine Nc for the majority of species we studied . Instead , we used the species’ geographic range and individual body size as proxies for Nc . Size has been previously validated as a proxy for individual density in a wide variety of taxa and ecosystems ( e . g . , [42–44] ) . Under some simplifying assumptions , the product of geographic range and local density should be sufficient to roughly estimate a species census population size , and each factor is expected to independently capture some information related to species’ Nc . Specifically , we expect that range will be positively correlated with Nc , size will be negatively correlated with Nc , and Nc will be positively correlated with the impact of selection . For many of the species that we studied , it is clear that selection plays a central , even dominant , role in shaping patterns of neutral genetic diversity . Specifically , both our correlation analysis and our explicit modeling support the hypothesis that natural selection on linked sites eliminates disproportionately more neutral polymorphism in species with large Nc , and in this way , natural selection truncates the distribution of neutral genetic diversity . At a coarse scale , there is a stronger correlation between polymorphism and recombination in invertebrates ( mean partial τ after correcting for gene density = 0 . 247 ) , which likely have a large Nc on average , than in vertebrates ( median partial τ = 0 . 118 ) , which likely have a smaller Nc on average ( two-tailed permutation p = 0 . 021 ) . We observe similar patterns for herbaceous plants ( mean partial τ = 0 . 106 ) versus woody plants ( mean partial τ = −0 . 020; two-tailed permutation p = 0 . 058 ) and for medians as opposed to means ( Fig . 2 ) . When we repeat the analysis with alternate window sizes , we observe consistent effect sizes , albeit occasionally with reduced statistical support ( S3 Table ) . More generally , we tested the hypothesis that Nc is positively correlated with the impact of selection by fitting a linear model that includes body size , geographic range , kingdom , and the significant interactions among them as predictors , and uses the impact of selection estimated from our coalescent model as the response variable ( Table 2; Fig . 3 ) . Both size and range are significant predictors of the impact of selection in the expected directions ( Table 2; log10 ( size ) : coefficient = −0 . 092 , p = 0 . 0005; log10 ( range ) : coefficient = 0 . 112 , p = 0 . 0002 ) , and model as a whole explains 63 . 88% of the variation in impact of selection across species ( Table 2; overall p = 3 . 518 x 10−8 ) . This is clear evidence that more variation is removed by linked selection from the genomes of species with smaller body size and larger ranges than from the genomes of species with larger body size and smaller ranges . A number of confounding factors could potentially influence our conclusions , including variation in map or assembly quality across species , differences in overall recombination rate , and differences in genome size . To test whether these factors can explain our results , we fit a confounder-only model including two measures of genetic map quality ( density of useable markers and proportion of total markers scored as useable ) ; two measures of assembly quality ( proportion of assembly that is not gaps and proportion of total assembly assembled into chromosomes ) ; overall recombination rate; and genome size . We then compare this confounder-only model to a model that includes all confounding parameters and , in addition , includes our population size proxies ( kingdom , size , and range ) . The model with proxies for Nc both explains substantially more total variation in impact of selection ( adjusted R2 of 0 . 6359 compared to 0 . 3388 for the confounder-only model ) and is a significantly better fit to the data ( F = 7 . 7322 , df = 4 , p = 0 . 0002 ) . In order to ensure that variable sampling of chromosomes is not a source of bias ( given that the number of chromosomes sampled ranges from a minimum of 2 to a maximum of 517; S1 Table ) , we tested whether sampling depth is correlated with either size or range . In neither case do we find a correlation ( size versus sampling depth: Kendall’s τ = 0 . 022 , p = 0 . 84; range versus sampling depth: Kendall’s τ = 0 . 044 , p = 0 . 699 ) . We also find no evidence that species with only two chromosomes sampled are atypical with respect to range ( Wilcoxon Rank Sum Test , p = 0 . 944 ) or size ( Wilcoxon Rank Sum Test , p = 0 . 423 ) . Finally , we find no evidence that mean depth per individual is correlated with either size ( Kendall's τ = −0 . 044 , p = 0 . 683 ) or range ( Kendall's τ = −0 . 02 , p = 0 . 862 ) . Taken together , these results strongly suggest that the variable sampling across species , both in terms of sequencing depth and in terms of number of chromosomes sequenced , does not bias our conclusions . To get a lower bound on the proportion of variation in impact of selection explained by our parameters of interest ( range , size , kingdom , and the kingdom–size interaction ) , we fit a linear model with these parameters as predictors and the residuals of the confounder-only model as the response variable ( S4 Table , S5 Table ) . This is a conservative test , as genome size is strongly correlated with body size ( Kendall's τ = 0 . 296 , p = 0 . 007 in our dataset ) . Nonetheless , our proxies for Nc explain 34 . 05% of the remaining variation in impact of selection after accounting for all confounding parameters ( overall model p = 0 . 0008 , S4 Table ) , and 47 . 36% of the variation after accounting for all confounding parameters except genome size ( overall model p = 2 . 042 x 10−5 , S5 Table ) . For five species , our polymorphism data included individuals from domesticated populations , which could potentially affect our conclusions if selection has a different signature during domestication events than it leaves in natural populations . However , removing these five species has virtually no impact on our model fit ( overall adjusted R2 = 0 . 6281 , overall p = 6 . 094 x 10−7 , S6 Table ) , suggesting that their inclusion has not biased our results . Additionally , we obtain similar results if we fit our model ( excluding the kingdom term and its interaction with size ) to animals and plants independently ( S7 Table , S8 Table ) . Finally , varying the filtering criteria , window size , assumed deleterious mutation rate ( U ) , or population genetic modeling approach produces nearly identical results ( Fig . 3C ) , implying our primary conclusion is robust to a wide range of analysis choices . Taken together , our analysis demonstrates that the central pattern—natural selection reduces neutral diversity more strongly in species with large Nc than in species with small Nc—is consistently observed with both nonparametric model free approaches ( Fig . 2; S3 Table ) and with explicit population genetic models ( Fig . 3A , B , Table 2 ) across a wide range of possible analysis and filtering choices ( Fig . 3C , S4–S8 Tables ) . If the process of recombination is itself mutagenic , neutral processes could produce a correlation between recombination and polymorphism [21 , 25 , 27] . However , no or very weak correlations between divergence and recombination have been found in most species that have been closely studied [21 , 25] ( reviewed in [27] ) . Moreover , for those species in which a positive correlation between divergence and polymorphism has been found ( e . g . , [45 , 46] ) , it is likely at least partially the result of linked selection acting on polymorphisms present in the ancestral population [27 , 32] . Furthermore , the two species that showed the strongest correlation between polymorphism and recombination ( partial τ = 0 . 5196 for D . melanogaster , partial τ = 0 . 4637 for Drosophila pseudoobscura ) have no such correlation between recombination rate and divergence either on broad scales [21] or fine scales [25] . Finally , many authors have found strong evidence that recombination is not mutagenic in a number of other animal species ( e . g . , [28 , 47 , 48] ) , and it therefore appears a general consensus has emerged that recombination-associated mutagenesis is unlikely to influence the overall patterns we report in this work [27] . As an alternative approach to estimating the impact of natural selection on linked neutral diversity , we considered whether our proxies for Nc correlate with the strength of evidence that selection shapes patterns of neutral diversity , derived from our population genetic modeling approach . To do this , we focus on the relative likelihoods ( Akaike weights ) of four models: the BGS+HH model , the BGS-only model , the HH-only model , and the neutral model . These relative likelihoods can be interpreted as the probability that a particular model is the best model according to Akaike Information Criteria ( AIC ) , given the set of models tested and the underlying data . We initially focus on the relative likelihood of the support for a purely neutral model . Species with weak or no support for neutrality ( relative likelihood of the neutral model < 0 . 05 ) have significantly larger ranges ( p = 0 . 006 , Wilcoxon Rank Sum Test , Fig . 4A ) and significantly smaller sizes ( p = 0 . 0001 , Wilcoxon Rank Sum Test , Fig . 4B ) than species with moderate ( relative likelihood of neutral model ≥ 0 . 05 and < 0 . 90 ) or strong ( relative likelihood of neutral model ≥ 0 . 90 ) support . This pattern also holds if we compare the species with strong support for neutrality or species with moderate support for neutrality individually to species with weak or no support ( moderate versus weak: p = 0 . 0005 for size and 0 . 02 for range; strong versus weak: p = 0 . 02 for size and 0 . 02 for range , all p-values from Wilcoxcon Rank Sum Tests ) . This suggests that the evidence for non-neutral processes ( BGS and/or HH ) is significantly stronger in species with larger ranges and/or smaller sizes , consistent with our results above and with the hypothesis that natural selection explains Lewontin's paradox . Given the extensive debate on the relative importance of HH versus BGS in shaping patterns of diversity across the genome [17 , 21] , we also attempt to disentangle the relative roles of these two processes in reducing neutral diversity . This is potentially relevant to the resolution of Lewontin's paradox , as models of frequent , recurrent HH ( i . e . , genetic draft [7] ) demonstrate that recurrent HH can remove the dependence of neutral diversity on population size entirely . Thus , evidence that HH specifically is more likely to occur in species with large census sizes would be compelling evidence for a role of selection in resolving the discrepancy between population sizes and neutral diversity . However , it is crucial to note that our test does not take into account features , such as patterns of polymorphism around amino acid fixations [23 , 49] , that are particularly powerful for distinguishing HH and BGS , and thus suffers from many of the limitations of previous work relying purely on patterns of neutral diversity across the genome ( e . g . , [26 , 28 , 40 , 41] ) . With that caveat , we begin by noting that , consistent with recent work in Drosophila [49 , 50] and other organisms [26 , 28 , 48] , background selection is ubiquitous . Either the BGS-only model or the BGS+HH model has at least some support ( relative likelihood ≥ 0 . 05 ) for 95% ( 38 of 40 ) of the species we analyzed , and for 90% ( 36 of 40 ) of species one of the BGS-containing models was the best fit , as measured by AIC . Thus , it seems clear that , in most cases , BGS is a more appropriate null model for tests of natural selection than strict neutrality . To test whether species with moderate ( relative likelihood of HH or BGS+HH ≥ 0 . 05 and < 0 . 9 ) or strong ( relative likelihood of HH or BGS+HH ≥ 0 . 9 ) evidence for HH differ from species with little or no evidence for HH ( relative likelihood of HH or BGS+HH < 0 . 05 ) , we examined our proxies for Nc among these evidence classes . Species with moderate or strong evidence for HH have significantly larger ranges than species with weak or no evidence for HH ( p = 0 . 03 , Wilcoxon Rank Sum Test , median range ( weak ) = 2 , 681 , 693 sq km , median range ( moderate/strong ) = 5 , 592 , 037 sq km ) , and these species tend to have smaller sizes as well ( p = 0 . 15 , Wilcoxon Rank Sum Test , median size ( weak ) = 0 . 91 m , median size ( moderate/strong ) = 0 . 54 m ) . As a second test of this pattern , we compared whether the relative likelihood of HH was greater for species estimated to have particularly high Nc compared to species estimated to have particularly low Nc . We define the high-Nc class as those species with ranges greater than the median range , and sizes below the median size , and we define the low-Nc class as those species with ranges below the median range and sizes above the median size . The relative likelihood of HH models is greater for species in the high-Nc class than the low-Nc class ( p = 0 . 023 , Wilcoxon Rank Sum Test ) , and the proportion of species with moderate or strong evidence for HH ( either alone or in combination with BGS ) is higher in the high-Nc class than the low-Nc class ( 4/10 in high-Nc class , 0/10 in low-Nc class , p = 0 . 086 , Fisher's Exact Test ) . Despite the fact that our test is unlikely to have substantial power to distinguish BGS and HH models , we suggest that these results imply that HH in particular is a stronger force shaping genomic diversity in species with large Nc , while BGS appears to be much more pervasive . The observation that pervasive HH may predominantly occur in species with large Nc suggests that genetic draft may play a substantial role in limiting neutral diversity among the species with the largest population sizes . More data on species with very large Nc , and the application of tests specifically designed to detect HH to a wider taxonomic range , will be necessary to fully disentangle the relative roles of HH and BGS in shaping levels of neutral diversity .
On the strength of early allozyme polymorphism data , Lewontin [6] observed that in contrast with theoretical predictions of the neutral theory [51–53] , the range of neutral genetic variation among species is substantially smaller than the range of Nc among species . Because both positive selection via HH and negative selection via BGS purge linked neutral mutations , the operation of natural selection affects patterns of neutral genetic variation at linked sites across the genome . Although many authors have suggested that natural selection may play a role in truncating the distribution of genetic variation and may play a greater role than neutral genetic drift in shaping patterns of neutral nucleotide polymorphism [7 , 8 , 14 , 15] , few empirical tests of this hypothesis have been proposed or conducted . Here , we show that species with larger Nc display a stronger correlation between neutral polymorphism and recombination rate , and that natural selection removes disproportionately more linked neutral variation from species with larger populations . This indicates that natural selection plays a disproportionately large role in shaping patterns of polymorphism in the genome of species with large Nc . One important consideration when interpreting our results is that cryptic population structure can influence patterns of variation across the genome in a way that obscures the effects of selection . In the extreme case , where populations do not exchange any migrants for an extended period of time , genetic divergence is expected to accumulate at equivalent rates across the genome and would obscure the effects of linked selection . Elucidating the complex relationship between population structure and patterns of natural selection is an important and longstanding question in population genetics ( for recent work see [54 , 55] ) . Nonetheless , especially given the scope of our analysis , it is not feasible to simultaneously estimate the effects of linked selection and population structure , and there are many reasons to believe that the results presented here will be robust to potential cryptic population structure . So long as the population subdivision is not especially ancient ( in the timescale of coalescence , on the order of Ne generations ) , a correlation between recombination and polymorphism is expected to remain due to the effects of selection on linked sites in the ancestral population [27 , 32] . Additionally , if migration is sufficiently common , it is reasonable to treat data derived from samples from separate localities as a single population [56] . One straightforward assumption is that species with larger geographic ranges will have greater opportunity on average to accumulate cryptic population structure than species with small ranges , which would imply we should preferentially underestimate the effects of linked selection in species with larger ranges . If population structure is a primary determinant of patterns of nucleotide diversity in taxa that we studied , we could reasonably expect a negative correlation between species range and the effects of selection on linked sites . Given that we instead obtain the opposite effect—one consistent with the effect of selection on linked neutral sites—it is reasonable to conclude that cryptic population structure has not drastically influenced the basic results presented herein . Understanding the proximate and ultimate factors that affect the distribution of genetic variation in the genome is a central and enduring goal of population genetics and it carries important implications for a number of evolutionary processes . One implication of this work is that in species with large Nc , such as D . melanogaster , selection plays a dominant role in shaping the distribution of molecular variation in the genome . Among other things , this can affect the interpretation of demographic inferences because it indicates that even putatively neutral variants are affected by natural selection at linked sites . Furthermore , to whatever degree standing functional variation is also affected by selection on linked sites ( e . g . , [40] ) , local recombination rate in organisms with large Nc may also predict what regions of the genome will contribute the greatest adaptive responses when a population is subjected to novel selective pressures . More broadly , this work provides direct empirical evidence that the standard neutral theory may be violated across a wide range of species . Indeed , it is clear from this work that in many taxa , natural selection plays a dominant role in shaping patterns of neutral molecular variation in the genome . It is therefore essential to consider selective processes when studying the distribution of genetic diversity within and between species . Incorporating selection into standard population genetic models of evolution will be a central and important challenge for evolutionary geneticists going forward .
Reference genome versions , annotation versions , map references , and other basic information about the genetic and genomic data for species we included in our analysis is summarized in S1 Table and S2 Table , and described in more detail below . Our approach to estimating recombination rates is to first obtain sequence information and genetic map positions for markers from the literature , map markers to the genome sequence where necessary , filter duplicate and incongruent markers , and finally estimate recombination rates from the relationship between physical position and genetic position . Specific details of map construction for each species are described in S1 Text . We begin with the very general selective sweep model derived by Coop and Ralph [41] , which captures a broad variety of HH dynamics . To include the effects of BGS , we rely on the fact that to a first approximation , BGS can be thought of as reducing the effective population size and therefore increasing the rate of coalescence . This effect can be incorporated by a relatively simple modification to equation 16 of [41] . Specifically , we scale N by a BGS parameter , exp ( -G ) , in equation 16 , which then leads to a new expectation of average pairwise genetic diversity ( π ) : E[π]=θ1/exp ( −G ) +α/rbp ( 1 ) where α = 2N * Vbp * J2 , 2 ( per [41] ) and rbp is the recombination rate per base pair . This is very similar to previously published models of the joint effects of background selection and HH ( e . g . , [39] ) . To account for variation in the density of targets of selection , we build upon the approach of Rockman et al . [40] and Flowers et al . [26] , which derives from the work Hudson , Kaplan , Charlesworth , and others that originally described models of background selection in recombining genomes [17 , 18] . Specifically , we fit the following model to estimate G for each window i: Gi=ΣkU*fdi*sh2* ( sh+P|Mi−Mk| ) * ( sh+P|Mi−Mk+1| ) ( 2 ) where U is the total genomic deleterious mutation rate , fdi is the functional density of window i , sh is a compound parameter capturing both dominance and the strength of selection against deleterious mutations , Mk and Mi are the genetic positions in Morgans of window k and window i , respectively , and P is the index of panmixis , which allows us to account for the effects of selfing . We estimate functional density as the fraction of exonic coding sites in the genome that fall within the window in question . We focus on exonic coding sites as a proxy for targets of selection as they are the only functional measure that is uniformly available for all the species in our study . Because P , U , and sh are not known , we fit this BGS model with a variety of parameter combinations . As U is generally unknown , and estimating U is difficult in most cases ( e . g . , [67 , 68] ) , we fit our models with three different values: Umin , where we assume U is equal to the mutation rate times the number of exonic protein-coding bases in the genome; Uconst , where we assume that U is equal to one for all species; and Umax where we assume that U is equal the lesser of the mutation rate times fives times the number of exonic protein-coding bases in the genome or the mutation rate times the genome size . Umin and Umax are multiplied by two to convert to diploid estimates . We believe that these estimates of U should roughly span the reasonable range for most species . Umin is likely to underestimate the true deleterious mutation rate as the number of exonic protein-coding bases will typically underestimate the number of evolutionarily conserved bases in a genome . Umax assumes that 20% of conserved bases are exonic coding bases and 80% are noncoding , which we admit is a relatively arbitrary assumption , but likely close to the maximum plausible U . For P , we assume one for all vertebrates , insects , and obligate outcrossers among plants; 0 . 04 for highly selfing species , and 0 . 68 for partial selfers . These estimates correspond to selfing rates of 0% , ∼98% , and ∼50% , respectively . Estimates of selfing are available in S14 Table . For a few species of plants , we were unable to obtain reliable estimates of selfing rate ( indicated by NA in S14 Table ) , and in this case we include all estimates of P in our model selection approach below . For sh , we fit a range of values evenly spaced ( on a log scale ) between 1e-5 and 0 . 1 . Code to estimate Gi was implemented in C++ and is available from the GitHub repository associated with this manuscript . To incorporate functional density into the HH component of the model , we make the simplifying assumption that sweeps targeting selected sites outside a window will have little effect on neutral diversity within a window , and that sweeps occur uniformly within a window . Under this assumption , we can consider functional density as a scaling factor on the rate of sweeps , Vbp . Specifically , we reparameterize the rate of sweeps , Vbp , as V , the total sweeps per genome , and then consider the fraction of sweeps that occur in a particular window i as V*fdi . This results in a simple scaling of α in Equation 1 . While we note that this assumption is likely to be violated in practice , it allows us to use the homogeneous sweep model of [41] with different rates of sweeps for each window across the genome . Ultimately , of course , it would be preferable to derive a nonhomogenous sweep model that directly incorporates variation in functional density , but doing so is beyond the scope of this work . However , we believe that our simplifying is likely adequate , as the largest reduction in diversity associated with a sweep is localized to the window containing the swept site ( e . g . , [41] ) . Incorporating the effects of functional density in both BGS and HH , our final model for the expectation of neutral diversity in window i is: E[πi]=θneutral1/exp ( −Gi ) +α*fdi/rbpi ( 3 ) To obtain an estimate of the effect of selection for each species , we fit this model for estimates of Gi derived from different parameter combinations ( see above ) , using the nlsLM ( ) function from the minpack . lm package in R . In addition , we fit three simpler models: a BGS-only model ( in which α is 0 and thus the second part of the denominator is 0 ) , an HH-only model ( in which G is 0 for all i , and thus the first part of the denominator is 1 ) , and a neutral model in which both G and α are 0 , and thus the model predicts that neutral genetic diversity is equal to mean genetic diversity across the genome . Together , we refer to these four models as model set 1 . Finally , we fit a second set of models ( model set 2 ) in which we use the same approach to model background selection , but use the homogenous HH model of [41] without modification to allow for variation in functional density across the genome , and thus remove the fdi term from Equation 3 . From each model fit we estimate θneutral for all four models ( full , BGS-only , HH-only , and neutral ) and also extract the likelihood of the fit . We then compute the AIC for each parameter combination , extract the fit with the best AIC for each model , and use that AIC to estimate the Akaike weight ( relative likelihood ) of each model j as RELj=e ( AICmin−AICj ) /2 ( 4 ) which we then normalize so that the weights for all four models for a species sum to one . We focus on AIC as it provides a straightforward way to compare non-nested models . We estimate expected neutral genetic diversity in the absence of selection ( θneutral ) for each species as the parameter value obtained by the model with the best AIC . We then compute average observed genetic diversity for each species , and report the magnitude of the impact of selection on linked neutral diversity as 1 – ( observed / neutral ) . Values below zero are replaced by zero . This value can be interpreted as the proportion of neutral variation removed by selection acting on linked sites , averaged across the genome . This modeling approach has some important limitations: in particular , our approach calculates the effects of BGS and HH in windows across the genome instead of per base and we use the parameter sh instead of integrating across the distribution of fitness effects ( as is done in e . g . [48 , 50] ) . Additionally , we do not use information such as locations of amino acid fixations , as is used by [49] . We fully acknowledge that these simplifying assumptions will , to a certain extent , degrade the accuracy of our modeling approach compared to other possible approaches . We argue , however , that these assumptions are necessary for this work: more sophisticated models typically require additional data ( e . g . , the distribution of fitness effects of new mutations or the location of recent amino acid fixations ) , or significantly increased computational time ( i . e . , by computing the effects of background selection at each base instead of in windows ) . For most of the species we studied , the necessary additional data are not clearly available to fit more complex models , and the increased computational time to fit per-base models would rapidly make our analysis computationally intractable . Thus , we believe that we have made reasonable tradeoffs between modeling complexity , data availability , and taxonomic breadth . Our goal is to test whether Nc predicts the degree to which selection shapes patterns of neutral diversity , using log-transformed measures of body size and geographic range as proxies for Nc . However , many other factors could potentially influence our measure of strength of selection , including biological factors such as genome size and average recombination rate; and experimental factors such as map quality and assembly quality . In particular , we might expect to underestimate the strength of selection in species with low-quality assemblies or maps , and we might expect that on average , larger genomes and higher recombination rates would reduce the impact of selection . In order to account for these parameters that are not directly of interest , we use two approaches . First , we compare a model that includes both our parameters of interest and our parameters not directly of interest to a model that includes only the parameters not directly of interest , in order to test whether our proxies for Nc result in a significantly better fit . Second , we fit our proxies for Nc to the residuals of a linear model including only parameters not directly of interest , in order to determine how much variation proxies for Nc explain after accounting for all the variation that can be explained by genome size , average recombination rate , and quality parameters . We obtain assembly quality from NCBI , Phytozome , the original genome publication , or compute it directly from fasta files . C-values for plants come from http://data . kew . org/cvalues/ , and C-values for animals come from [69] . In all cases , most recent estimates , "prime" estimates , or flow cytometry estimates are preferred; where several seemingly equally good estimates are available , the average is used . In some rare cases , a related species is used instead of the sequenced species if the C-value for the sequenced species is not available . We focus on C-values instead of assembly size as using assembly size as a measure of genome size confounds genome size and assembly quality ( lower quality assemblies will be on average less complete and therefore smaller ) . Assembly parameters and sources are listed in S15 Table . We estimate average recombination rate as the overall map size divided by the size of the genome covered by the map . In order to determine which interactions among proxies for Nc ( size , range , and kingdom ) to include , we start with the full model including all interactions and remove all non-significant interactions . After doing so , our model is selection strength ~ log10 ( size ) + log10 ( range ) + kingdom + log10 ( size ) :kingdom ( 5 ) The data we analyze in this manuscript , and the scripts we used to produce our results , are available as follows . All genomes , polymorphism datasets , and GFF annotation files are publicly available from NCBI or other sources . Genome references and versions are listed in S1 Table , and URLs pointing to the location of genome sequence and GFF annotations are available in S2 Table . Sequence Read Archive ( SRA ) accessions for polymorphism datasets are listed in S10 Table , and references for polymorphism datasets , where available , are listed in S1 Table . Genetic maps for each species are available from the references listed in S1 Table , or as an R data file available at the GitHub page associated with this manuscript ( https://github . com/tsackton/linked-selection ) . All Perl scripts , R scripts , and C++ code associated with this manuscript are available from GitHub ( https://github . com/tsackton/linked-selection ) , and the function of each piece of code is documented both in comments in the code itself and in the Github README . Programs used for read mapping and genotyping , along with command line parameters , are described in the methods . The GitHub page also provides several intermediate data files , including range and size data for each species , neutral diversity and recombination rate for 100 kb , 500 kb , and 1 , 000 kb windows across each species , and the final dataset analyzed with the linear models described above . | A fundamental goal of population genetics is to understand why levels of genetic diversity vary among species and populations . Under the assumptions of the neutral model of molecular evolution , the amount of variation present in a population should be directly proportional to the size of the population . However , this prediction does not tally with real-life observations: levels of genetic diversity are found to be substantially more uniform , even among species with widely differing population sizes , than expected . Because natural selection—which removes genetically linked neutral variation—is more efficient in larger populations , selection on novel mutations offers a potential reconciliation of this paradox . In this work , we align and jointly analyze whole genome genetic variation data from a wide variety of species . Using this dataset and population genetic models of the impact of selection on neutral variation , we test the prediction that selection will disproportionally remove neutral variation in species with large population sizes . We show that genomic signature of natural selection is pervasive across most species , and that the amount of linked neutral variation removed by selection correlates with proxies for population size . We propose that pervasive natural selection constrains neutral diversity and provides an explanation for why neutral diversity does not scale as expected with population size . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2015 | Natural Selection Constrains Neutral Diversity across A Wide Range of Species |
The expression of long-term depression ( LTD ) in cerebellar Purkinje cells results from the internalisation of α-amino-3-hydroxy-5-methylisoxazole-4-propionic acid receptors ( AMPARs ) from the postsynaptic membrane . This process is regulated by a complex signalling pathway involving sustained protein kinase C ( PKC ) activation , inhibition of serine/threonine phosphatase , and an active protein tyrosine phosphatase , PTPMEG . In addition , two AMPAR-interacting proteins–glutamate receptor-interacting protein ( GRIP ) and protein interacting with C kinase 1 ( PICK1 ) –regulate the availability of AMPARs for trafficking between the postsynaptic membrane and the endosome . Here we present a new computational model of these overlapping signalling pathways . The model reveals how PTPMEG cooperates with PKC to drive LTD expression by facilitating the effect of PKC on the dissociation of AMPARs from GRIP and thus their availability for trafficking . Model simulations show that LTD expression is increased by serine/threonine phosphatase inhibition , and negatively regulated by Src-family tyrosine kinase activity , which restricts the dissociation of AMPARs from GRIP under basal conditions . We use the model to expose the dynamic balance between AMPAR internalisation and reinsertion , and the phosphorylation switch responsible for the perturbation of this balance and for the rapid plasticity initiation and regulation . Our model advances the understanding of PF-PC LTD regulation and induction , and provides a validated extensible platform for more detailed studies of this fundamental synaptic process .
The functional plasticity of neuronal synapses , including long-term potentiation ( LTP ) and long-term depression ( LTD ) , is essential for learning and the encoding of memories [1] . The focus of this study is LTD at the parallel fibre-Purkinje cell ( PF-PC ) synapse in the cerebellum , which is believed to play an important role in motor learning [2–4] . This form of LTD requires [5 , 6] the concurrent activation of a sufficiently large fraction of the around 175 , 000 excitatory en passant contacts made from cerebellar parallel fibres to the Purkinje cell dendritic tree [7] and of a climbing fibre comprising several thousand synaptic contacts [8] . PF-PC LTD is linked to the endocytic removal of α-amino-3-hydroxy-5-methylisoxazole-4-propionic acid receptors ( AMPARs ) from the Purkinje cell postsynaptic membrane [9–11] . The synaptic AMPAR population is dynamically controlled through lateral diffusion into and out of the synapse [8] , and receptor endocytosis and exocytosis between the cell surface and the endosome [12] . Endosomes store the internalised AMPARs before they are directed to either reinsertion into the membrane during plasticity [2–4] or degradation [13] . The AMPAR degradation [11 , 14] and de novo synthesis [15] provide additional regulation for the receptor population . PF-PC LTD is dependent on the increased internalisation of AMPARs relative to their reinsertion [16] . PF-PC LTD is induced by the activation of protein kinase C ( PKC ) [17] , elevated intracellular calcium [18] and the concurrent inhibition of serine/threonine phosphatase activity [19 , 20] . The mechanics of PF-PC LTD are partly controlled by two AMPAR-GluA2 subunit interacting proteins , glutamate receptor interacting protein ( GRIP ) and protein interacting with C kinase 1 ( PICK1 ) [5 , 6] , both of which bind at the same site via their C-terminal PDZ domains [21] . The three GRIP isoforms are functionally indistinguishable [22] , so we refer to them simply as GRIP . GRIP interacts with AMPARs , stabilising and clustering them both at the plasma membrane and at intracellular endosomal pools [23 , 24] . This interaction prevents AMPAR trafficking [21 , 23 , 25] , and AMPAR dissociation from GRIP is essential for the expression of PF-PC LTD [26] . AMPARs that lack the GRIP interaction are unable to stably incorporate into synapses [27] . PICK1 actively promotes AMPAR endocytosis in cerebellar Purkinje cells [6 , 26 , 28] and the PICK1-AMPAR interaction is indispensable for PF-PC LTD expression [6 , 10 , 28–30] . PICK1 also associates with the active form of PKCα [31] , which phosphorylates the S880 C-terminus residue of the AMPAR-GluA2 subunit [32 , 33] sustained by positive feedback mechanisms for at least 20 minutes during LTD induction [34 , 35] . GluA2-S880 phosphorylation , which is elevated after the induction of LTD in hippocampal slices [36] and is required for PF-PC LTD [37] , abolishes binding between GluA2 and GRIP . However , GluA2 binding to PICK1 is unaffected [38] . Dissociation of GRIP therefore allows PICK1 to bind at the same AMPAR-GluA2 site , promoting AMPAR internalisation . Disruption of the GluA2-GRIP interaction and AMPAR declustering are specifically associated with LTD induction [39] . PICK1 also interacts with GRIP and this enhances GluA2-S880 phosphorylation , possibly by directing PKCα to the GluA2 subunit [33] . The role of PICK1 in AMPAR reinsertion remains unclear , with several studies suggesting conflicting roles [21 , 28 , 40 , 41] . Phosphorylation of the tyrosine GluA2-Y876 by Src family kinases ( SFKs ) negatively interferes with the GluA2-S880 phosphorylation , suggesting a regulatory role of GluA2-Y876 in LTD induction [42] . GluA2-Y876 phosphorylation levels are determined by the balance between endogenous SFK and protein tyrosine phosphatase activities . The GluA2-Y876 site is predominantly phosphorylated during basal conditions [42] and actively dephosphorylated during mGluR1-mediated LTD induction [43] . The δ2-glutamate receptor ( GluD2 ) -associated tyrosine phosphatase , PTPMEG , actively dephosphorylates the GluA2-Y876 position in vitro [42] , and PTPMEG-null mice display impaired motor learning and LTD [44] . By dephosphorylating the GluA2-Y876 site and hence facilitating GluA2-S880 phosphorylation , PTPMEG gates the induction of LTD in the cerebellum [42] . To gain insight into the regulation of AMPAR mobility in cerebellar LTD , we constructed a bidirectional kinetic computational model of PF-PC LTD that emphasises AMPAR trafficking as a dynamic recycling loop , and the role of GRIP , PICK1 and the relevant kinases and phosphatases in maintaining this loop . This is the first model to explicitly account for the dynamic regulation of AMPAR mobility by the interaction of the GluA2-Y876 and GluA2-S880 phosphorylation sites , now known to be a key regulatory switch for PF-PC LTD induction . Our conceptually simple model sheds light on LTD signalling beyond the well-established data showing that PF-PC LTD is dependent on PKC activation , Ca2+ elevation and serine/threonine phosphatase inhibition [19] . We predict that PTPMEG cooperates with PKC to drive LTD expression by gating the effect of PKC on the dissociation of AMPARs from GRIP and thus their availability for binding to PICK1 and internalisation from the postsynaptic membrane . We also show that serine/threonine phosphatase inhibition increases the degree of LTD expression , in line with experimental data [45 , 46] , and that SFK is not required for the induction of LTD , but negatively regulates LTD expression , as demonstrated experimentally [47] . These results advance our understanding of PF-PC LTD regulation and induction , suggest new hypotheses for experimental validation and provide a platform for further computational studies .
We model AMPARs as embedded at the cell membrane or the endosome , with all interactions with protein partners occurring in the sub-membrane and the ‘sub-endosome’ regions , respectively . These regions constitute the two main compartments of the model , and the bulk cytosol merely acts as a source/sink for smaller molecules . The sub-membrane contains three sub-compartments–the postsynaptic density ( PSD ) , the extra-synaptic area and the endocytic zone , and AMPARs can diffuse laterally between these areas ( Fig 1 ) . The recruitment of AMPARs is a three-step process [48] comprising exocytosis at extra-synaptic areas , lateral diffusion to the PSD , and trapping by scaffold proteins ( GRIP ) . Only AMPARs within the endocytic zone can be internalised [49 , 50] . Trapping of AMPARs at the endocytic zone by dephosphorylated stargazin ( TARP-γ2 ) is essential for LTD expression [51–53] . In line with this data , LTD is well expressed in our model only when the diffusion rate out of the endocytic zone is kept very low ( <0 . 01s-1 ) . N-ethylmaleimide-sensitive factor ( NSF ) interacts with GluA2-containing AMPARs and has an essential role in the recruitment of AMPARs into the postsynaptic membrane , possibly by controlling SNARE-dependent exocytosis [54] or promoting lateral diffusion to the PSD [55 , 56] . We model the potentially manifold roles of NSF by requiring that AMPARs are bound to NSF in order to undergo exocytosis [57] . As NSF disrupts the AMPAR-PICK1 interaction [58] , and AMPARs bound to GRIP are not available for trafficking , only AMPARs bound to neither PICK1 nor GRIP can bind to NSF [54 , 57] . We model AMPAR trafficking exclusively as a recycling loop , and LTD as a perturbation of this dynamic trafficking equilibrium . Therefore , we do not consider de novo synthesis and degradation of AMPARs , whose inclusion is likely to occlude the effect of the phosphorylation switch on AMPAR mobility and LTD expression . Furthermore , degradation of internalised AMPARs does not have functional consequences for the regulation of LTD [59] , although the regulation of AMPAR recycling is essential for determining the degree of LTD expression [60] . Many published LTD models are unidirectional and measure LTD expression in terms of AMPAR internalization only , or even simply by the level of AMPAR phosphorylation [61] . This simplifies the modeling strategy but neglects the importance of the endocytosis-exocytosis balance in regulating the cell surface AMPAR population and the dynamic nature of AMPAR recycling . A sophisticated recent stochastic model of cerebellar LTD [35] does account for exocytosis of AMPARs , but disregards all other interactions within the intracellular compartment that are important in regulating AMPAR mobility and reinsertion . Our trafficking pathway is a bidirectional kinetic model ( Fig 2 ) that emphasises AMPAR trafficking as a dynamic recycling loop . As with other models of LTD , we measure LTD expression purely in terms of the reduction of the postsynaptic membrane AMPAR population [35 , 61] , although additional mechanisms , such as AMPAR desensitisation , may also play a minor role in the biological system [62] . Unique to our model , the dissociation from GRIP , and the mobilisation and availability of AMPARs for trafficking between compartments are regulated by the mutually exclusive phosphorylation of the GluA2-S880 and GluA2-Y876 sites ( Fig 3A ) [42] . GluA2-S880 is phosphorylated by PKC and dephosphorylated by PP2A , while GluA2-Y876 is phosphorylated by SFKs [42] and dephosphorylated by PTPMEG [42] . Phosphorylation of the GluA2-S880 site abolishes the interaction between the AMPAR and GRIP , allowing PICK1 to bind . PICK1 can also associate with GRIP directly to form a tripartite complex ( Fig 3B ) . The other interactions within the model are detailed in the Methods section . To observe the effect of PKC and PTPMEG on the endocytic rate alone , we initially selectively blocked exocytosis . Under basal conditions , when PKC is inactive , approximately 125 AMPARs populate the PSD [63] and around 40% of these are estimated to be internalised within 20 minutes [57] . When PKC is activated in the absence of active PTPMEG , the average rate of endocytosis is only slightly elevated relative to basal conditions ( 44% of AMPARs internalised over 20 minutes with activated PKC , versus 38% when PKC is inactive ) ( Fig 4A ) . However , activation of PKC in the presence of active PTPMEG increases the internalisation rate 2-fold above that generated by activated PKC alone , with 89% of AMPARs being internalised over 20 minutes . This suggests that the role of PTPMEG is to gate the effect of active PKC in promoting AMPAR dissociation from GRIP and subsequent internalisation . The result is in agreement with experimental data , which shows that elevated PKC alone does not increase the AMPAR internalisation rate in cerebellar Purkinje cells [64] . According to our LTD model , under basal conditions , the AMPARs trafficked between the cell surface and endosome are predominantly the unphosphorylated and GluA2-Y876-phosphorylated forms . During LTD induction , we expected a shift towards internalisation of the GluA2-S880-phosphorylated form of the receptor as PKC is activated . We reinstated exocytosis and measured the flow of the three different forms of AMPAR ( unphosphorylated , GluA2-Y876-phosphorylated and GluA2-S880-phosphorylated ) between the plasma membrane and endosomal compartments and vice versa in 3000-second simulations of the system under basal conditions , and during PKC-induced LTD ( Fig 4B ) . Under basal conditions , the cell surface AMPAR population remained stable and only the unphosphorylated form of AMPAR and the GluA2-Y876 phosphorylated form were internalised , each being trafficked at a rate of 0 . 03–0 . 04 receptors per second , equally in both directions . When PKC was activated in the presence of PTPMEG , the cell surface AMPAR population declined to 44% of its initial number over around 1000 seconds . This was followed by a steady state during which mainly the GluA2-S880-phosphorylated form of AMPAR was internalised , with 0 . 16 of these receptors being trafficked per second in both directions , in addition to a small number ( 0 . 04–0 . 06 per second for each ) of the unphosphorylated and GluA2-Y876 phosphorylated forms of the receptor ( Fig 4C ) . When PKC is inactive , the cell surface AMPAR population remains stable , both in the presence and absence of PTPMEG . To analyse the effect of PKC activation , we ran 3000-second model simulations comprising 1000 seconds under basal conditions , followed by a step function activation of PKC that was maintained for the remaining 2000 seconds . This represents the approximate period for which PKC activation is maintained by a positive feedback mechanism during LTD induction , in line with experimental data [35] . As late phase effects maintain LTD after the PKC activation window , we do not consider deactivation of PKC or the maintenance of LTD after this time . In the absence of PTPMEG , and in agreement with experimental results [42 , 44] , the activation of PKC does not result in a marked inward trafficking of plasma membrane AMPARs , with the cell surface population of AMPARs only falling to 92% of baseline when PKC is activated . Furthermore , there is no increase in the number of mobile AMPARs ( i . e . not bound to GRIP ) , with fewer than 6% of the AMPARs being mobile during the PKC activation period , as during basal conditions ( Fig 5A ) . When PTPMEG is present , the activation of PKC leads to an immediate increase in the average percentage of cell surface AMPARs that are mobile from ~6% to ~18% ( Fig 5B ) . This demonstrates cooperation between PKC and PTPMEG to mobilise the cell surface AMPARs for trafficking . Neither PKC activation nor PTPMEG alone is capable of eliciting LTD . Both enzymes are required concurrently , as suggested by experimental data demonstrating that LTD expression in cerebellar Purkinje cells requires PTPMEG activity [42] . The increase in mobile AMPARs during the PKC activation window triggers a decline in the cell surface AMPAR population towards a steady state as endocytosis dominates the trafficking dynamics ( Fig 5B ) . Experiments have shown that the population of GluA2-Y876-phosphorylated AMPARs declines during LTD induction [42] , with the GluA2-S880-phosphorylated form increasing concurrently [36] , as plasma membrane AMPARs are mobilised and internalised . Our simulations replicate and quantify this effect ( Fig 5C ) . Under basal conditions in our model , approximately 20% of membrane AMPARs are GluA2-Y876 phosphorylated , with none of the receptors phosphorylated at the GluA2-S880 site . However , immediately upon PKC activation , the population of GluA2-S880-phosphorylated AMPARs increased to 18% of the total PSD AMPAR population , and this was maintained throughout the PKC activation window . Comparable with experimental observations [42] , the population of GluA2-Y876 phosphorylated receptors declined from 20% to 9% upon PKC activation ( Fig 5C ) . It is well established that the inhibition of serine/threonine phosphatase activity accompanies LTD induction [19 , 45 , 65] . However , whether such inhibition is essential for LTD induction or merely augments is not understood . To study the effects of phosphatase inhibition on LTD induction , we performed simulations for PP2A concentrations ranging between 0–100% ( Fig 6 and Table 1 ) . Increasing phosphatase inhibition results in a corresponding increase in the degree of LTD achieved . Without PP2A inhibition , only a 39% reduction in cell surface AMPAR population is achieved after 20 minutes , rising to 77% reduction with 100% PP2A inhibition . This result is comparable to experimental results showing up to a 65% reduction in excitatory postsynaptic current amplitude in cerebellar Purkinje cells using PP2A inhibitors [45] , and suggests that tuning of phosphatase inhibition could regulate the degree of depression achieved during LTD . SFKs selectively phosphorylate the Y876 site of the AMPAR GluA2 subunit [42] . Under basal conditions , phosphorylation at this position limits GluA2-S880 phosphorylation . By allowing GRIP to bind , this stabilises the AMPARs at the cell surface or endosomal membrane . Active PTPMEG dephosphorylates GluA2-Y876 , enabling GluA2-S880 phosphorylation and hence the dissociation of the AMPAR from GRIP and its mobilisation for trafficking . We performed simulations under standard LTD induction conditions , in the absence of SFKs , and with increasing SFK concentrations up to 5-fold greater than the basal concentration . Removing SFKs from the system slightly enhanced LTD expression , with 38% of cell surface ( PSD ) AMPARs remaining after 20 minutes , compared to 44% for the wild-type conditions . Increasing concentrations of SFK caused a proportional decrease in the magnitude of the LTD response , which was directly related to the degree of GluA2-Y876 phosphorylation ( Fig 6B and Table 2 ) . This result is in agreement with experimental studies showing that SFK negatively regulates cerebellar LTD expression [47] , although it appears to contradict earlier studies showing that SFKs are essential for LTD expression [66 , 67] , with SFK inhibitors abolishing LTD . However , the broad-spectrum tyrosine kinase inhibitors used in these studies ( i . e . genistein and lavendustin A ) are likely to affect kinases other than SFKs [42] . If a more specific SFK inhibitor is used to reduce tyrosine ( GluA2-Y876 ) phosphorylation , LTD induction in cerebellar Purkinje cells is unaffected [42 , 47] , in agreement with our results . It should be noted that , in vivo , SFKs act on a broad range of substrates and , as such , their effect on AMPAR trafficking , both directly and indirectly , could be more complex than indicated by our model . However , the effect of SFK at the GluA2-Y876 phosphorylation site is sufficient to explain current experimental data . Knockout of PTPMEG or the PTPMEG-interacting GluD2 abrogates LTD [42] by preventing AMPAR mobilisation . Expression of the mutant subunit , GluA2-Y876F , which cannot be tyrosine phosphorylated , rescues LTD in GluD2-null Purkinje cells [42] . We replicated this result by blocking GluA2-Y876 phosphorylation . Under these conditions , even when PTPMEG was knocked out , LTD was fully expressed ( Fig 6C ) . This demonstrates the central role of GluA2-Y876 phosphorylation in the regulation of AMPAR mobility . The role of SFK activity thus appears to be in limiting AMPAR mobilisation under basal conditions , as well as being an active regulator of PF-PC LTD .
The AMPAR population at the Purkinje cell postsynaptic membrane is part of a continuous dynamic recycling loop . Even when the population is stable , under basal conditions , 90% of the internalised AMPARs are returned to the cell surface within 60 minutes [14] . It is this dynamism that ensures a rapid response to perturbation . Modelling both directions of AMPAR trafficking simultaneously is therefore essential for the accurate study of plasticity . Furthermore , a number of proteins and signalling pathways that regulate receptor internalisation may also affect reinsertion . Consequently , any LTD model that considers only the regulation of internalisation will necessarily be incomplete and may even produce misleading data . In a study of the effects of synaptic activity on AMPAR trafficking in cultured cortical neurons [14] , manipulating the rate of AMPAR internalisation–using tetrodotoxin and picrotoxin–had no effect on the size of the cell surface AMPAR population , as the reinsertion rate was similarly affected . It is thus clear that the regulatory mechanisms controlling AMPAR internalisation overlap with those controlling reinsertion . As such , AMPAR trafficking is best described as a unified recycling loop rather than two separate processes . The balance of kinase and phosphatase activity within cerebellar Purkinje cells is exquisitely poised to allow the AMPAR population to be stabilised at the cell surface and endosome , and yet rapidly mobilised for trafficking during LTD induction . Our simulations show that the GluA2-Y876 and GluA2-S880 phosphorylation sites together act as a ‘master switch’ both for the induction of PF-PC LTD and the regulation of its magnitude . Whilst PTPMEG acts as an overall facilitator of LTD induction , by gating the dissociation of AMPARs from their GRIP anchors , PP2A and SFK activity can tune the degree of depression achieved . This is an important insight that clarifies , and provides a straightforward molecular mechanism for , the role of kinase and phosphatase activity in LTD regulation . Experimental studies have established that PP2A inhibition enhances LTD expression [45] , and that SFK activity negatively regulates it [47] , in agreement with our simulations . Furthermore , Endo et al [68] produced mutant mice lacking the gene coding for G-substrate , a potent inhibitor of PP2A [69] . Surprisingly , the consequent elevated PP2A levels did not abolish LTD in cerebellar Purkinje cells . Our model explains this result , and demonstrates that PP2A inhibition regulates the magnitude of LTD achieved , but is not required for LTD induction ( Table 3 ) . Although the orphan glutamate receptor δ2 ( GluD2 ) is indispensable for PF-PC LTD expression [72] , its specific role remains unclear . However , by binding to and potentially activating PTPMEG , it may concentrate this phosphatase at the plasma membrane and thus facilitate the selective mobilisation of cell surface AMPARs . Whilst GluD2 is only expressed in cerebellar Purkinje cells , several brain regions express GluD1 [73] , which may function in a similar manner by binding and/or leading to PTPMEG activation , making this phosphatase a more global regulator of plasticity than currently known . Furthermore , PTPMEG has been shown to bind the NR2A subunit of NMDA receptors [74] , which could also support this function . The signalling pathways regulating synaptic plasticity are complex , both in terms of the number of signalling species involved and their spatiotemporal dynamics . This makes any bidirectional model of trafficking challenging to construct and implement , but essential for generating realistic data . Our model achieves this , is able to replicate a wide range of experimental observations of cerebellar parallel fibre-Purkinje cell LTD , sheds light on their underpinning mechanisms and provides a sound foundation for additional simulation experiments and for more detailed models of synaptic plasticity processes . Furthermore , our model is the first to explore the role of this type of mutually-exclusive phosphorylation switch , which is similar to switches found in other important systems , including receptors controlling insulin response [75] , and NMDA receptor function [76] .
The model was implemented in the well-established and validated open-source biochemical network simulator COPASI [77 , 78] , using kinetic parameters obtained from the literature ( see supplementary information S1 Table for details ) . We used deterministic simulation to efficiently and accurately establish the average system behaviour for a wide range of scenarios and parameter ranges [79] . These simulations were performed using the COPASI built-in LSODA ( Livermore Solver for Ordinary Differential Equations ) solver , with particle number to concentration conversions performed by COPASI . Model can be found in S1 Model . The model contains two compartments ( Fig 1 ) . The sub-membrane compartment comprises the volume of cytosol directly below the plasma membrane to a distance of 120nm [80] , and consists of three sub-compartments: postsynaptic density ( PSD ) , endocytic zone ( EZ ) and extra-synaptic area . The sub-endosome compartment is assumed to occupy the same volume as the sub-membrane . As AMPARs are entirely membrane-bound , they are concentrated in these regions and hence all of the key reactions occur here . The bulk cytosol , which is not explicitly modelled , merely acts as a source/sink for species that are distributed throughout the dendritic spine . Thus , when a species , such as GRIP or PICK1 , binds to an AMPAR , it is immediately replaced , by diffusion , by a spare from the bulk cytosol . This approach is supported by experimental and modelling data suggesting that AMPAR scaffolds are never saturated [12] . However , we also produced an alternative model in which GRIP and PICK1 numbers were finite . This model produced results qualitatively the same as those produced with the model used in our paper . The alternative model , together with representative results , is included in the supplementary information S2 Model and S1 Fig . The complete set of model reactions is summarised in Table 4 and is described below . Except where explicitly stated , these reactions occur in each compartment of the model , between species from the populations in that compartment . AMPARs exist freely or associated with GRIP or with PICK1 , forming an AMPAR-GRIP or AMPAR-PICK1 complex , respectively ( Table 4 , Reactions 1–6 ) . PICK1 may associate with the GRIP of an AMPAR-GRIP complex and thus a tripartite complex , AMPAR-GRIP-PICK1 , can form ( Reactions 7–12 ) . A dimeric GRIP-PICK1 complex is not considered , as preliminary experiments showed that it had no effect on the outcome of the simulations . The GRIP populations at the PSD and the endosome interact with AMPARs identically , anchoring the AMPAR to the PSD and the endosomal compartment , respectively [81] . AMPAR-GRIP interactions are not considered in the extra-synaptic area or the endocytic zone . PICK1 is a calcium sensor and the AMPAR-PICK1 binding rate increases 4-fold in the presence of a high calcium concentration [26] , as during PF-PC LTD induction . Endosomal AMPARs can also associate with NSF , but only when not associated with either GRIP or PICK1 ( Reactions 13 and 14 ) . All binding interactions are assumed to occur with mass action kinetics . AMPARs not bound to GRIP can diffuse laterally , in both directions , between the PSD and the extra-synaptic area ( Reactions 15–20 ) , and between the extra-synaptic area and the endocytic zone ( Reactions 21–26 ) . The rate constant for diffusion from one area to another is calculated as the ratio between the diffusion coefficient [12 , 48] and the area of the sub-compartment [35] . To undergo endocytosis ( Reactions 27 and 28 ) , a GRIP-bound plasma membrane AMPAR must detach from GRIP and bind to PICK1 . Furthermore , only AMPARs at the EZ can be internalised . AMPARs can only undergo exocytosis ( Reaction 29 ) when NSF is bound to the receptor , with AMPARs being reinserted into the extra-synaptic area . As we do not consider AMPAR-NSF interactions within the plasma membrane , AMPARs are assumed to detach from NSF when exocytosis occurs . We adopt a simple switch for activating and deactivating PKC ( Reactions 30 and 31 ) , in line with both experimental data [34] and computational simulations [35] , which show that positive feedback mechanisms maintain PKC activity for the duration of early LTD induction ( at least 20 minutes ) . PKC can exist freely in the cytoplasm or , when in its active form ( PKC* ) , combined in a reversible complex with PICK1 ( Reactions 32 and 33 ) . PKC* phosphorylates AMPAR at the GluA2-S880 site to generate AMPApS ( 880 ) ( Reactions 34–37 ) . The PICK1-PKC* complex can also phosphorylate the GluA2-S880 site . Once PICK-PKC* is bound to AMPAR , phosphorylation is assumed to occur at the turnover rate for PKC* ( Reactions 38 and 39 ) . The phosphorylation of GluA2-S880 reduces the affinity of the AMPAR for GRIP , as reflected by an increase in the AMPApS-GRIP unbinding rate ( Reaction 2 ) [35] . The AMPAR GluA2-S880 site is dephosphorylated by PP2A , which we assume constitutively active and inhibited ( 60% ) during LTD induction ( Reactions 40–45 ) . AMPAR is phosphorylated by SFKs at the GluA2-Y876 site to generate AMPApY ( 876 ) ( Reactions 46–51 ) . Dephosphorylation of GluA2-Y876 is performed by PTPMEG ( Reactions 52–57 ) . All phosphorylation and dephosphorylation reactions are assumed to occur with Michaelis-Menten kinetics . Experimentally , under basal conditions , the majority of AMPARs are unphosphorylated [42] . In line with experimental data [63] , the system was initially populated with 125 submembrane AMPARs and 125 sub-endosome AMPARs , all unphosphorylated . The kinetics of PTPMEG were calibrated such that the proportion of AMPARs phosphorylated at GluA2-Y876 was consistently approximately 25% , in line with experimental data [42] . However , simulations using alternative initial AMPAR populations–increasing the proportion of GluA2-Y876-phosphorylated AMPARs , for example–did not affect the results obtained , either qualitatively or quantitatively . Basal conditions were defined as corresponding to PKC inactive , PP2A uninhibited and AMPAR trafficking calibrated by setting the endocytosis rate such that approximately 40% of receptors were internalised over a 20-minute period when exocytosis was selectively blocked [57] . The exocytosis rate was set such that it balanced endocytosis under basal conditions . When simulating LTD induction , PKC was activated and PP2A was inhibited by 60% throughout the simulation . This inhibition was modelled by removing 60% of the PP2A from the model . For time course simulations , a step function was used to activate PKC ( Table 4 , Reactions 31 and 32 ) after allowing the simulation to run for 1000 seconds . As PTPMEG has no effect on LTD induction or expression in the absence of active PKC , PTPMEG was present and active throughout the 3000-second simulation . To simulate the knockout of specific species ( e . g . PTPMEG , Figs 4 and 5 ) , these species were removed from the model . We carried out standard sensitivity analysis to measure the impact of variations in the model parameters ( i . e . , the reaction rates from Table 4 ) on the simulation results . To this end , we established the sensitivity of the steady-state plasma membrane AMPAR population n during LTD induction to changes in each reaction rate ri from Table 4 . This involved calculating the scaled sensitivity coefficient of ri as the scaled partial derivative of the AMPAR population n by the reaction rate ri: SSC ( ri ) =δnδrinri The magnitude of the coefficient indicates the sensitivity of the AMPAR population n to changes in the reaction rate ri . The sign of the coefficient indicates whether n increases ( SSC ( ri ) > 0 ) or decreases ( SSC ( ri ) < 0 ) in response to an increase in the rate ri . Table 5 shows these coefficients for the system operating with the rates shown in the supplementary material ( S1 Table ) . Several model parameters have a small ( <0 . 1 ) scaled sensitivity coefficient , indicating that the model is robust to significant changes in these parameters . The model is sensitive to the remaining parameters: Experimental data from the literature was used to determine the values for these parameters that the model is sensitive to , as explained in the supplementary information S1 Table . | Changes in synaptic strength , which can include long-term potentiation and long-term depression , are important for learning and the encoding of memories across the brain . Long-term depression ( LTD ) , in particular , is thought to be essential for motor learning in the cerebellum , and disruption of this process , by disease or injury , can result in severe motor dysfunction . Cerebellar LTD is achieved by reducing the population of AMPA receptors at the Purkinje cell postsynaptic membrane . This population is maintained by a dynamic trafficking loop , in which AMPA receptors are continuously removed from the postsynaptic membrane by endocytosis and reinserted by exocytosis . Specific phosphorylation sites on the AMPA receptors regulate their interaction with proteins that either stabilise the receptors at the membrane or promote their trafficking . We develop a detailed bidirectional computational model of this trafficking loop and its regulation . The model shows how perturbing the trafficking balance towards AMPA receptor mobilisation and endocytosis can account for rapid induction of cerebellar LTD , and suggests mechanistic explanations for numerous features observed experimentally . This deepens our understanding of cerebellar LTD and provides a foundation for further experimental studies of this synaptic process . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"phosphorylation",
"vesicles",
"enzymes",
"cell",
"processes",
"enzymology",
"phosphatases",
"membrane",
"receptor",
"signaling",
"cellular",
"structures",
"and",
"organelles",
"endosomes",
"animal",
"cells",
"proteins",
"exocytosis",
"cell",
"membranes",
"endocytosis",
"biochemistry",
"signal",
"transduction",
"cell",
"biology",
"post-translational",
"modification",
"secretory",
"pathway",
"neurons",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"cell",
"signaling",
"purkinje",
"cells"
] | 2016 | A Computational Model for the AMPA Receptor Phosphorylation Master Switch Regulating Cerebellar Long-Term Depression |
The evolutionary timing and spread of the Mycobacterium tuberculosis complex ( MTBC ) , one of the most successful groups of bacterial pathogens , remains largely unknown . Here , using mycobacterial tandem repeat sequences as genetic markers , we show that the MTBC consists of two independent clades , one composed exclusively of M . tuberculosis lineages from humans and the other composed of both animal and human isolates . The latter also likely derived from a human pathogenic lineage , supporting the hypothesis of an original human host . Using Bayesian statistics and experimental data on the variability of the mycobacterial markers in infected patients , we estimated the age of the MTBC at 40 , 000 years , coinciding with the expansion of “modern” human populations out of Africa . Furthermore , coalescence analysis revealed a strong and recent demographic expansion in almost all M . tuberculosis lineages , which coincides with the human population explosion over the last two centuries . These findings thus unveil the dynamic dimension of the association between human host and pathogen populations .
The Mycobacterium tuberculosis complex ( MTBC ) is composed of closely related bacterial sub-species that have plagued human and animal populations for thousands of years . The most famous member of the MTBC is M . tuberculosis , the etiological agent of tuberculosis in humans that killed 1 . 7 million people in 2004 according to the World Health Organization [1] . A new threat is the worldwide emergence of multi-drug resistant ( MDR ) and extremely drug-resistant ( XDR ) strains . Recent data suggest that the propensity to gain drug resistance as well as the pathogen's transmissibility profile may be influenced by the genetic and evolutionary background of M . tuberculosis strains [2] . Thus , understanding the relationships and dynamics of the MTBC lineages will undoubtedly help to unravel the basis for the considerable success and spread of tuberculosis , in both humans and animals . The MTBC is essentially clonal with little evidence of horizontal gene exchange [3] , [4] , [5] , and probably derived from a pool of ancestral tubercle bacilli , collectively called “Mycobacterium prototuberculosis” [6] . However , despite the highly successful worldwide spread of the MTBC , the evolutionary timing of this spread remains largely unknown . This lack of knowledge is largely due to the limitations of the genetic markers used so far . All efforts to time MTBC evolution with single nucleotide polymorphisms ( SNPs ) have been based on a non-warranted hypothesis of universal bacterial mutation rates , itself extrapolated from a very hypothetical time of divergence between Escherichia coli and Salmonella enterica [7] . In this study , we used a completely new approach by employing genetic markers based on mycobacterial interspersed repetitive units ( MIRUs ) to determine the timing of divergence , population diversity and spread of the MTBC . MIRU loci comprise variable numbers of tandem repeat ( VNTR ) sequences , which allow them to be used as powerful genotyping markers [8] , [9] . In terms of genetic diversity and mutation rates , they resemble human microsatellites , which are widely used in human population genetics studies [10] . Similar to microsatellites , MIRUs behave as selectively neutral phylogenetic markers if sufficient numbers of loci are used to buffer against potential biases . Here we used experimental data on the variability and evolution of these markers in clinical isolates of infected patients , which allowed us to calculate the MIRU molecular clock and model their evolution in coalescence approaches . Based on this information and extensive analysis of a large collection of representative MTBC strains , we obtained new insights into the origin and demography of the MTBC and its dynamic association with the human host .
To infer the MTBC evolutionary history , we used a sample collection of 355 isolates , representative of well-identified primary branches of the MTBC world distribution ( Table S1 ) . A recently standardized combination of 24 MIRU loci ( Figure S1 ) , which does not comprise saturated loci [11] , was utilized . To illustrate the power of MIRUs to reconstruct geographical patterns of genetic differentiation and their level of resolution , a distance-based tree was constructed using individual genotypes and a neighbour-joining algorithm ( Figure 1A ) . The tree grouped all M . tuberculosis sensu stricto isolates ( all from human patients ) in a distinct lineage with the notable exception of the East African-Indian ( EAI ) population whose affiliation is unclear based on this approach . Another major lineage encompassed all MTBC strains from animals ( M . microti . M . bovis , M . caprae and M . pinipedii ) and the human isolates from West-Africa ( M . africanum West African 1 and 2 ) . From the resulting tree , it appears that the groupings of isolates within the primary MTBC branches based on SNPs , spoligotyping and large sequence polymorphisms ( LSPs ) [12] , [13] , [14] , [15] , [16] , [17] ( Figure S2 ) are highly congruent with those based on the MIRU typing , albeit the branch resolution was higher in the latter . In order to more robustly define the relationships between the lineages ( by reducing the number of individuals vs the number of markers ) , we then grouped individual isolates into the populations defined by the above groupings and built a tree based on MIRU allelic frequencies in these populations ( Figure 1B ) . The tree was rooted with samples of M . prototuberculosis ( including M . canettii ) , which was recently reported to represent the progenitor of the MTBC [6] . This approach clearly revealed the distinctiveness of the two major lineages with strong bootstrap support , called hereafter clades 1 and 2 . A further geographic sub-structuring within clade 1 became apparent , with distinct branches for the African ( Uganda , Cameroon and S ) , Asian ( Beijing and CAS ) , Latin American-Mediterranean and African-European populations ( X , Ghana and Haarlem ) . Clade 2 is composed of both animal and human pathogenic isolates . A basal position of EAI ( human tuberculosis ) in clade 2 has strong statistical support , indicating a human origin for this predominantly animal-associated MTBC lineage . However , low bootstrap values within clade 2 prevent us from drawing further inferences on the branching order . To confirm the groupings and the deep dichotomy obtained with the MIRUs , we used an independent approach , based on the ‘no-admixture’ model of the STRUCTURE program [18] . In this Bayesian approach , multilocus genotypic data are used to define a set of populations with distinct allele frequencies and assign individuals probabilistically to them , with or without prior knowledge of geographic sampling information . We applied STRUCTURE to the global data set ( including the outgroup ) and in ten independent runs , at K = 3 populations ( Fig . 1C ) STRUCTURE detected the same two deeply divergent clades 1 and 2 that were identified with the neighbour joining analysis ( see Figure 1B ) . Notably , this separation is independently supported by the fact that TbD1 ( M . tuberculosis deletion 1 ) is lacking in all clade 1 strains but present in all clade 2 strains , including those from EAI ( Figure 1B and S2 ) [12] . The robustness of these clades was further evidenced by STRUCTURE analysis , because each isolate derived all of its MIRU's from only one of the three ancestral sources of clade 1 , clade 2 or M . prototuberculosis ( see Protocol S1 ) . We further modelled the Bayesian assignments of the two main clades by sub-dividing them into additional clusters ( Figure S3A ) . The bacterial isolates were consistently split into the same major clusters as those defined by the distance-based approach ( see above ) . The highest likelihoods were obtained for K = 6 populations in each of the two main clades . Only three isolates ( 0 . 85% ) were assigned to unexpected clusters by the Bayesian approach ( Figure S3A ) , further illustrating the consistency of MIRU-VNTR cluster designations . To detect possible horizontal genetic transfer events , we used the STRUCTURE ‘linkage model’ as was done to detect ongoing genetic exchange in Helicobacter pylori [19] , [20] , Escherichia coli [21] and Moraxella catarrhalis [22] . Runs without prior knowledge of population source ( Figure S3B ) suggested that the vast majority of the MTBC strains are clonal , while some M . prototuberculosis strains might be hybrids with MTBC genotypes , in accordance with previous results [3] , [5] , [6] . To further assess the deep dichotomy , we calculated the allelic richness ( the number of alleles ) of the populations within the two main clades after correcting for sample size effects [23] ( Fig . 2 ) . High levels of genetic diversity are a surrogate indication of ancestral origins as illustrated in the highly divergent African human populations . The mean allelic richness per locus was close to five for both clades , and the difference was not significant ( Fig . 2C ) , arguing for a simultaneous split of the two clades . As expected , LAM and EAI , the most basal populations in clades 1 and 2 respectively , contained the highest number of alleles ( Fig . 2A , 2B ) . However , some uncertainty remains on a basal position for LAM because it conflicts with groupings based on internal deletions of the pks15/1 gene and on SNPs [13] . In order to estimate the time to the most recent common ancestor ( TMRCA ) in the MTBC , we made use of recent analytical tools [24] , [25] , which make these estimations possible . They rely on Bayesian statistics and apply a stepwise mutation model ( SMM ) for genetic markers . This model is a reasonable assumption for MIRU mutations , as initially shown for MIRU locus 4 in the BCG evolutionary framework [9] . To test the validity of this model for the total set of the MIRU loci used , we built a minimal spanning tree of all MTBC strains based on the degree of allele sharing . We then evaluated the proportion of strains that differed from their closest relative by one step ( single-locus variants- SLVs ) or by multiple steps , which would violate the SMM model . This simple method will certainly overestimate any violations of the SMM model because our sampling scheme is not exhaustive , resulting in some spurious missing links ( intermediate strains ) that falsely invalidate the SMM model . However , the data showed that at least 64% of the allelic changes fit the stepwise mutation model , a result that is close to the 75% and 81% observed in E . coli and yeast VNTRs , respectively [26] , [27] . To further evaluate the validity of the SMM model , eBURST analysis was performed on a much larger dataset comprising 1 , 733 MIRU-VNTR profiles from two population-based studies performed at regional and national levels ( see Materials and Methods ) . This analysis identified 142 groups and 1061 singletons . In order to determine whether tandem repeats evolve following a SMM model and to detect a potential bias towards increase or decrease in repeat numbers , we computed within each eBURST group all differences in number of repeats along the evolutionary path , starting from the putative founder of the group to its surrounding SLVs ( Figure S4 ) . For all but two of the 24 loci , the most frequent change was either −1 or +1 repeat unit , with the symmetric change generally being the next most frequent . The only minor exceptions were loci MIRU-VNTR 3007 and 2347 , which contain little information , because the only changes were one occurrence each of −2 and +1 repeat units , and four occurrences of −2 and two of +1 respectively . Both for the individual loci and for data cumulated over the 24 loci ( Figure S5 ) , the distribution of occurrences was unimodal and centered on 0 ( average of −0 . 07±0 . 23 , CI = 0 . 95 , for cumulated data ) . At least sixty-five percent of the allelic changes matched the stepwise mutation model . It is noteworthy that missing links falsely invalidating the SMM model probably occur even in this population-based dataset , because many patients from the population studied ( from the Brussels region and the entire Netherlands ) were foreign-born and have probably acquired their infection abroad . Therefore , tandem repeats in M . tuberculosis most frequently evolve by progressive gain or loss of single repeat units without significant general bias towards increase or decrease . To estimate MIRU mutation rates , we used data from large sets of serial or epidemiologically-linked isolates . The probability of showing a repeat change over periods of up to 7 years was estimated to be about 1% for five of the most variable loci [11] . This corresponds to a single-locus mutation rate of 1 . 4×10−3 per year . Consistently , 4 of these 5 loci composed the top 4 in the hierarchy of single-locus variation frequencies measured among the MIRU loci , both in a global MTBC isolate dataset [11] and in the above population-based dataset ( data not shown ) . This supports the use of these frequencies as a surrogate for estimating relative mutation rates of the different markers , and especially those of the less variable loci , for which repeat changes among serial or epidemiologically-linked isolates were not observed [11] . We therefore somewhat arbitrarily chose a lower mean mutation rate per year of 10−4 as a prior for the Bayesian inferences [25] over all loci , in order to accommodate the less variable loci which were associated with up to 38fold lower frequencies of single-locus variation . It is noteworthy that this initial value was well supported by posterior Bayesian analysis , as the calculated posterior mean for the mutation rate was 10−3 . 91 ( Figure 3 ) . By applying this mutation rate and a generation time of one day for the tuberculosis bacilli , we estimated a mean TMRCA of ≈40 , 000 years before present for the complex ( Table 1 ) . The TMRCAs for clades 1 and 2 were estimated as 21 , 000 and 33 , 000 years , respectively , and two of the oldest lineages , EAI and LAM coalesced at 13 , 700 and 7 , 000 years , respectively ( Table 1 ) . In a second step , we used the MSVAR software [25] that infers past demographic changes and calculates additional parameters , including TMRCA of monophyletic populations using slightly different algorithms . For this procedure , we focused on lineages for which at least 30 isolates were included in the study , in order to avoid small sample size artefacts . The use of this method confirmed the TMRCA of the EAI population at ≈7 , 000 years ( Figure 4B and Table 2 ) , albeit with very wide confidence intervals ( 150–190 , 000 years ) . Finally , genetic data can also unravel recent demographic change signatures in bacterial populations . By using Bayesian statistics , we tested whether a recent decline or expansion occurred in the MTBC population , and calculated ta , which reflects the time that has elapsed since the decline or expansion began . All MTBC populations from human sources that we considered displayed markedly consistent expansion rates and EAI is typical in that respect ( see Figure 3B ) . The detected growth rates ( on a log scale ) ranged from a modest 0 . 6 value , as seen in Africa , to 2 . 7 for Beijing , which is probably the most successful present day lineage . This latter value translates into a recent 500-fold population size increase . The mean modal value of log10 ta was 2 . 25 ( range 2 . 00–2 . 5 ) for the different populations , with the exception of the LAM lineage . This corresponds to a tuberculosis expansion that began 180 years ago ( see Table 2 ) .
Taken together , the findings presented in this study indicate that the MTBC is composed of two major lineages and has emerged approximately 40 , 000 years ago . This estimate is strikingly close to the proposed time of dispersal of founder modern human populations from the Horn of Africa [28] . However this dating must be considered with caution in the light of the large confidence intervals . Our results support the emergence of the MTBC clone from the M . prototuberculosis progenitor pool and its co-migration with modern humans out of Africa [6] . A similar trend was recently proposed for H . pylori and M . leprae [29] , [30] . We suggest that two main lineages arose later some 20 , 000 to 30 , 000 years ago from the common MTBC ancestor , one of which spread exclusively among humans , with subsequent waves of migration to Asia , Europe and continental Africa ( Figure 5 ) . This spreading scenario fits well with the current worldwide distribution of the main MTBC lineages , as reflected by the SpolDB4 database [12] , [13] , [14] , [15] , [16] , [17] and LSP analysis [14] , [17] . The second lineage ( clade 2 ) arose from a human EAI-like population some 30 , 000 years ago and is the probable source of animal tuberculosis [12] , [31] , a derivation that is strongly and convergently supported by both distance-based and probabilistic methods ( i . e . NJ and STRUCTURE ) . This conclusion is consistent with the finding that extant representatives of M . tuberculosis , which derived from the proposed progenitor of MTBC , are human pathogens [6] . Thus it is likely that humans infected their livestock and not the other way around . Clade 2 secondary branches include M . bovis and M . caprae , the infectious agents of tuberculosis in a wide variety of animals including cattle and goat , which were first domesticated in the Near East [32] , [33] . The transition from human to animal hosts may thus be linked to plant and animal domestication that took place in the Fertile Crescent some 13 , 000 years ago . This period corresponds to the estimated time of diversification of the oldest EAI and LAM populations ( Table 2 ) . In the Fertile Crescent , and during that era of human history , small nomadic hunter-gatherer groups were replaced by farming societies based on domesticated livestock and crops [34] . This paramount event in human history was probably not without consequence for an epidemic , infectious disease such as tuberculosis , where crowded farming populations may have promoted high infection rates , bacterial spread and transition to new niches and animal hosts [35] . Clade 2 also includes M . africanum strains that primarily infect humans . However , it has recently been speculated that M . africanum may not be primarily adapted to the human host but might have originated from an unknown animal reservoir [36] . All MTBC populations from human sources displayed markedly constant expansion rates , corresponding to an expansion that dates back to only about 180 years . Furthermore , the largest population size increase ( 500-fold ) was detected for Beijing , which is thought to be the most successful present day lineage . These results suggest that the expansion of the most recent form of human tuberculosis was coupled to Western urbanization and industrialization . This expansion was synchronous with the modern demographic explosion of Homo sapiens and modern intercontinental movements . Evidence for strong phylogeographical structuring of the pathogen population and preferential sympatric combinations of pathogen populations with particular ethnic groups has indicated a close association between M . tuberculosis and its human host [3] , [13] , [37] . Our results indicate recent parallel demographic changes between the pathogen and its host and reveal the tell-tale dynamic dimension of this association . The coalescence approach may also be useful in the future to monitor demographic changes in emerging MDR M . tuberculosis strains . Some of the conclusions presented here on the basis of MIRU data have also been reached previously , e . g . data from comparative genomics [12] after the completion of M . bovis genome [31] indicated that the MTBC did not arise as a zoonosis [38] . In contrast , the validity of efforts to date the origins of the common ancestor of MTBC by using SNP-based methods [39] , [40] , [41] , has remained questionable [42] . Furthermore , preliminary SNP-based phylogenetic reconstructions may have been affected by hitch-hiking , and ascertainment bias [43] , because those SNPs were associated with genes involved in drug-resistance [44] or were selected from a non-representative set of available genomes [14] , [17] , [45] . Such markers evolve too slowly for recent pathogens , as is also the case for LSPs and their use often results in uninformative phylogenies that consist of multifurcated unresolved trees [13] , [44] . Unlike previous studies , the novel analyses presented here rely on globally neutral markers with mutation rates that have been estimated from human M . tuberculosis infection cases , a descent-sampling scheme and multiple , convergent population genetic estimators . As they are based on intrinsically rare and stochastic VNTR changes in clonal populations , our mutation rate estimates do involve some special assumptions . The accuracy of the demographic and temporal estimates could be improved with long-term analyses , and we are aware that the use of a mean mutation rate for all loci is suboptimal , leading to an increase of the variance of parameters . However , our estimates were consistently corroborated by posterior Bayesian calculations in independent runs over different strain populations ( ranging from 10−4 . 19 for LAM to 10−3 . 82 for EAI ) , ruling out the risk of some local maxima . To gain further insights into the host-pathogen interactions , it would certainly be important to account for the biogeographic history and distribution of the different M . tuberculosis lineages , because recent adaptations to local host populations might play a major role [13] . Furthermore , it is known , that genetic diversity can influence the transmission dynamics of drug-resistant bacteria [2] , [46] , and , in terms of vaccination , it would be advisable to scrutinize independently the highly polymorphic clade 2 EAI strains that markedly differ in their genetic structure from the other human tuberculosis strains .
The 355 M . tuberculosis and M . prototuberculosis isolates were genotyped by multiplex PCR amplification as described previously [8] , [47] . The samples were subjected to electrophoresis using ABI 3100 and 3730 automated sequencers . Sizing of the PCR fragments and assignment of the VNTR alleles of the 24 loci was done using the GeneScan and customized Genotyper , as well as the GeneMapper software packages ( PE Applied Biosystems ) . The number of alleles ( allelic richness ) in each M . tuberculosis complex population was estimated and sample sizes were corrected by the rarefaction procedure using HP-RARE [23] . Comparison tests as well as P-values were estimated using the STATISTICA v . 6 . 1 package . Nei et al . 's DA distance [48] was used to construct both isolate and population trees using a neighbour-joining algorithm as implemented in the software Populations version1 . 2 . 28 . Support for the tree nodes was assessed by bootstrapping over loci ( 1 , 000 iterations ) . Using the no-admixture model [18] ( STRUCTURE version 2 ) , three to ten parallel Markov chains were run for all models of K with a burn-in of 100 , 000 iterations and a run length of 106 iterations following the burn-in . For each run , the ln likelihood of each model was calculated . The full data set was analysed for all models from K = 1 through to 3 without specifying prior information concerning the geographical sources or former designations . For K = 3 , a clear splitting solution was found in which the sampled populations clustered into two main tuberculosis groups plus the outgroup ( M . prototuberculosis ) ; a result fully consistent with the neighbour-joining population tree ( Figure 1B ) . For further analysis the data set was sub-divided into clades 1 and 2 , and these were subsequently tested for K = 1 through to 6 . Using the linkage model [49] of STRUCTURE version 2 , ten parallel Markov chains were run for each model with a burn-in of 100 , 000 iterations and a run length of 106 iterations following the burn-in . For each run , M . tuberculosis strains were specified as belonging to pre-determined source clusters . We estimated the ancestry in each source cluster and the proportion of each strain genome having ancestry in each cluster . To estimate the validity of SMM model , we built a minimal spanning tree of all MTBC strains based on the degree of allele sharing , by using BIONUMERICS ( Applied Maths , Belgium ) . We then evaluated the proportion of single-locus variants ( i . e . strains that differed from their closest relative ) that differed by one or by multiple repeat-changes . To further evaluate the validity of the SMM model and to detect a potential bias towards increase or decrease in repeat numbers , eBURST analysis was performed on a larger dataset from two population-based studies . The first one included 807 isolates from different TB cases notified in the Brussels-Capital Region ( Belgium ) from September 1st , 2002 to December 31st , 2005 [50] , while the second one is an ongoing study including 1907 isolates from different TB cases notified in the Netherlands over 2004 and 2005 ( Van Soolingen et al . , unpublished ) . In total , the dataset included 1 , 733 MIRU-VNTR profiles , with no missing data or incomplete repeats . On this dataset , the differences in the number of repeats were calculated for each pair of ancestor/descendant genotypes along the evolutionary path inferred by eBURST analysis [51] . The occurrence of each value of repeat difference was recorded for each group ( defined as groups of strains with at most one allelic mismatch with at least one other member of the group ) , and values were pooled over all eBURST groups . This analysis was performed using software Multilocus Analyzer ( S . Brisse , unpublished ) , which is an independent implementation ( coded in Python ) of the eBURST algorithm , to which the SMM test function was added . MIRU mutation rates were estimated by using data on VNTR changes among large sets of serial or epidemiologically-linked isolates [11] . Single-locus mutation rates of 5 most variable loci were estimated from corresponding frequencies of observed repeat changes . Repeat changes among serial or epidemiologically-linked isolates were not detected among the remaining , less variable loci . Therefore , the relative frequencies of single-locus variations among closely related isolates in a global MTBC isolate dataset [11] and in the population based dataset ( see above ) were then used as a surrogate for estimating mutation rates of less variable markers relatively to these most variable loci . In a first step , we used a Bayesian approach [25] that assumes a stepwise mutation model and estimates the posterior probability distributions of the genealogical and demographic parameters of a sample using Markov chain Monte Carlo simulations based on MIRU data . This method permits to extrapolate important biological parameters like the TMRCA of a given sample in years , the past and present effective population size and the latest demographic changes ( decline , constant population size or expansion ) . In order to assess the age of the main M . tuberculosis lineages , an alternative algorithm , YTime [24] was used to calculate the TMRCAs and their confidence intervals . For the MSVAR procedure [25] , [52] , we focused on lineages of which at least 30 isolates were available , in order to obtain a reliable coverage of the TMRCA and to avoid small sample size artefacts . The estimated parameters were scaled in terms of current population size , and two main demographic parameters were quantified: tf , which is a measure of time in generations , was defined as ta/N0 , where ta denotes the number of generations that have elapsed since the decline or expansion began , and r , which was defined as N0/N1 , where N0 is the current effective number of chromosomes , and N1 is the number of chromosomes at some previous point in time tf . For a declining population r<1 , for a stable population r = 1 and for expanding populations r>1 . The procedure also estimates θ , which is defined as N0μ , where μ is the mutation rate ( mutation locus−1 generation−1 ) . The analyses were performed assuming exponential demographic change . Three different chains were run for each analysis to confirm the convergence of the results . In the analyses , rectangular priors of the log parameter values have been used . The method was found to converge appropriately for both single and multilocus data sets and supported a model of population expansion for all MTBC populations . We present only the multilocus data in the present report . Expansion signatures were robust and were confirmed in runs where decline was assumed as a prior ( 10−2–10−3 ) . YTime [24]: YTime is a Matlab function which calculates the TMRCA for haplotype linked loci under the assumption of an S-SSM , which allows for unbiased +/−1 steps . YTime calculates confidence intervals using a simulation approach and is independent of the shape of the genealogy . We used all available loci ( N = 24 ) as an input . The strains were grouped according to their lineages ( obtained by phylogenetic analyses ) . The ancestral genotype for every subgroup was calculated as the mean of every single locus in the particular subgroup . The mutation rate was 10−4 per year per locus . For the growth rate parameters we assumed a mean effective population size of 108 for every sub-population and a growth of 103 ( the mean of the results is not affected by the growth rate , just the confidence intervals ) . | The causative agents of tuberculosis , grouped in the Mycobacterium tuberculosis complex , have infected one-third of the present human population and a wide range of other mammals . However , paradigmatic questions , such as why , where and when the disease began and expanded , have largely remained unanswered . In this study , we provide genetic evidence indicating that the most common ancestor of the bacterial complex emerged some 40 , 000 years ago from its progenitor in East Africa , the region from where modern human populations disseminated around the same period . This initial step was followed 10 , 000 to 20 , 000 years later by the radiation of two major lineages , one of which spread from human to animals . In more recent years ( approximately 180 years ago ) , coinciding with the human population explosion and the industrial revolution , the human-associated pathogen lineages have strongly expanded . These results thus reveal the strikingly parallel demographic evolution between humans and one of their primary pathogens . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"infectious",
"diseases/bacterial",
"infections",
"genetics",
"and",
"genomics/microbial",
"evolution",
"and",
"genomics",
"evolutionary",
"biology/microbial",
"evolution",
"and",
"genomics"
] | 2008 | Origin, Spread and Demography of the Mycobacterium tuberculosis Complex |
Antimicrobial peptides ( AMPs ) are an abundant and wide class of molecules produced by many tissues and cell types in a variety of mammals , plant and animal species . Linear alpha-helical antimicrobial peptides are among the most widespread membrane-disruptive AMPs in nature , representing a particularly successful structural arrangement in innate defense . Recently , AMPs have received increasing attention as potential therapeutic agents , owing to their broad activity spectrum and their reduced tendency to induce resistance . The introduction of non-natural amino acids will be a key requisite in order to contrast host resistance and increase compound's life . In this work , the possibility to design novel AMP sequences with non-natural amino acids was achieved through a flexible computational approach , based on chemophysical profiles of peptide sequences . Quantitative structure-activity relationship ( QSAR ) descriptors were employed to code each peptide and train two statistical models in order to account for structural and functional properties of alpha-helical amphipathic AMPs . These models were then used as fitness functions for a multi-objective evolutional algorithm , together with a set of constraints for the design of a series of candidate AMPs . Two ab-initio natural peptides were synthesized and experimentally validated for antimicrobial activity , together with a series of control peptides . Furthermore , a well-known Cecropin-Mellitin alpha helical antimicrobial hybrid ( CM18 ) was optimized by shortening its amino acid sequence while maintaining its activity and a peptide with non-natural amino acids was designed and tested , demonstrating the higher activity achievable with artificial residues .
Antimicrobial peptides ( AMPs ) are small evolutionally conserved molecules found among all classes of life , from multicellular organisms to bacterial cells [1] , [2] . In higher organisms , AMPs play a major role in innate immunity as a part of the first defence line against invading pathogens . In bacteria , AMPs provide a competitive advantage for the producer in certain ecological niches as weapons against other bacteria . Alpha-helical AMPs are among the most abundant and widespread membrane-disruptive sequences in nature and represent a particularly successful structural arrangement for innate defense , as it can easily afford peptide insertion into lipid bilayers [3] . In fact , the amphipathic structure facilitates electrostatic interactions between the peptide and the target cell membrane . Completion of the folding process involves hydrophobic interactions between the non-polar residues of the peptide and the hydrophobic core of the lipid bilayer [4] , [5] . AMP membrane perturbation activity can be explained by at least three major mechanisms , all leading to bacterial membrane's collapse and subsequent cell's death . Two of these models ( i . e . the ‘barrel-stave’ and the ‘toroidal-pore’ models ) rely on the peptide ability to form ordered transmembrane channels/pores , while the so called ‘carpet model’ implies that , at a critical threshold concentration , the peptides disrupt the bilayer in a detergent-like manner , eventually leading to the formation of micelles [6] . In recent years , AMPs are actively researched not only as direct antimicrobial agents , but also as potential endosomolytic moieties promoting the release of biomolecules into cells for delivery purposes [7]–[9] . On the other hand , the increasing prevalence of antibiotic resistance necessitates the development of new ways to combat bacterial infection . Although some AMPs are already in clinical and commercial use ( see Table S1 for a list of AMPs commercially available and in clinical trial ) , the future design of novel AMPs will need to minimize the toxicity against eukaryotic cells and enhance the resistance to proteolytic degradation , with a key opportunity being offered by the introduction of non-natural amino acids ( AA ) to contrast host resistance and increase compound's life . Thus far , several methods have been proposed for the rational design of AMP sequences with improved activity . In particular , computer-aided identification and design of AMPs played a crucial role in this area [10] . Most of these approaches are based on sequence alignment and calculation of amino acidic frequencies ( e . g . template-based approaches and some machine-learning methods ) [11]–[14] . However , several major disadvantages may limit the potential of these strategies . For instance , there is no understanding of the physicochemical requirements that play a crucial role in regulating the peptide activity . Also , peptide design is limited by the use of natural AAs only . In the effort to overcome these limitations , quantitative structure-activity relationship ( QSAR ) descriptors have been employed to correlate primary AA sequence with a given biological activity [15] , [16] . QSAR descriptors provide a reliable statistical model for prediction of the activities of new chemical entities . In particular , Hellberg et al . developed the so called ‘z-scale’ descriptors [17] , highly condensed variables derived from a principal component analysis ( PCA ) of several experimental or theoretical physicochemical properties for the 20 naturally occurring AAs . In detail , these z-scale descriptors correspond to the first three principal components explaining the variance in the set: z1 , z2 , and z3 represent the AA hydrophobicity , steric properties , and polarity , respectively . QSAR analysis of peptides using these descriptors has proven effective in predicting different physiological activities [18]–[20] . Z-scale descriptors were also successively expanded to include artificial AAs [21] . The new z-scales include 87 AAs and two extra variables ( z4 , z5 ) describing the electronic effects of the residues . All the mentioned descriptors combined with statistical analysis allow the design of novel AMP and the optimization of already existing ones in terms of desired characteristics such as improved activity , decreased toxicity , and easy synthesis . Due to the huge number of possible amino-acids combinations , stochastic optimization methods such as Genetic algorithms ( GA ) may be a preferred tool to perform directed random searches in large problem spaces , such as those encountered in drug design [22] , [23] . In particular , when simultaneous optimization of two or more characteristics is required , a class of GA called multi-objective evolutional algorithms ( MOEA ) can be used to provide an optimal solution [24][25] . In order to overcome current limitations and develop a flexible computational approach for AMP design , able to account for non-natural AAs , here we combine the representation of antimicrobial peptides in terms of physicochemical features with genetic algorithms . The novelty of this approach is in the unified treatment of both natural and non-natural AAs , allowing a systematic exploration of this enhanced combinatorial space . We target our approach to alpha-helical AMPs because of their advantages in terms of biological activity , mechanism of action , and ease of synthesis and manipulation [3] . Rather than training a single model on the existing alpha-helix AMPs we separately addressed the structure and the function characteristics , combining the two aspects in the design phase . Thereby , starting from two different sets of existing peptides , two statistical models were trained in order to account separately for structural and functional characteristics of alpha-helical AMPs . This approach has the advantage of considering broader and unbiased datasets , with respect to an alpha-helix only AMP dataset , enhancing the performance of the training phase . The first model represents antimicrobial physicochemical properties and was trained on a set of AMPs taken from the literature . The second model accounts for the all-helix conformation of the peptide and is based on a non-redundant set of all-alpha helix protein fragments . This in silico approach was used to design a set of five peptides with natural AAs ( GMG_01 , GMG_02 , GMG_03 , GMG_01_SCR , CM12 ) . GMG_01 and GMG_02 were designed ex-novo and predicted to be antimicrobic , while GMG_03 and GMG_01_SCR were designed as negative controls with no predicted bactericidal activity . CM12 was obtained by the reduction and optimization of the CM18 ( Cecropin ( 1–7 ) -Melittin ( 2–12 ) ) sequence , a well-characterized antimicrobial peptide . Finally , an AMP sequence containing non-natural AAs ( GMG_05Z ) was designed . Antimicrobial properties of all these peptides were experimentally validated in vitro by testing the minimum bactericidal concentration ( MBC ) against S . aureus and P . aeruginosa strains , representative of Gram-positive and Gram-negative bacteria respectively . In addition , Molecular Dynamic ( MD ) simulations were performed to predict and analyze the structural properties of the designed peptides , in particular to confirm the presence of helical motives .
Two different sets of peptides were prepared in order to represent functional and structural characteristics of alpha-helical AMPs . Dataset A , representing the functional requirements for AMP activity , was accurately compiled from the literature of existing and characterized antimicrobials . Dataset B accounts for the structural characteristics of alpha-helix AMPs and was assembled from a well-defined non-redundant set of proteins . Two types of descriptor encoding were utilized in order to present the training datasets to the learning algorithm ( see Table 1 ) : global descriptors and topological descriptors . Global descriptors are variables representing the whole molecule , while topological descriptors are variables representing the interaction of different residues along the amino acidic sequence . Charge and hydrophobicity related characteristics are among the most important properties for active peptides . [26] , [27] . Indeed , positively-charged peptides , rich in basic residues , particularly Lysines , can insert into the bacterial membrane more easily [5] , [28] . Hydrophobicity determines folding , binding to receptors , and interactions of proteins and peptides with biological membranes . Z-scale averages moments ( Equation 2 in Materials and Methods ) are used to account for hydrophobicity , as well as polarity and steric effects of each peptide . Topological description of the peptide sequence was accounted for by encoding QSAR descriptors into auto- and cross covariance ( ACC ) values . Classical ACC transformation was introduced by Wold et al [29] and results in two kinds of variables: auto covariance ( AC ) of the same descriptor and cross covariance ( CC ) between two different descriptors . Briefly , for a given protein sequence , ACC variables describe the average interactions between residues distributed a certain lag apart throughout the whole sequence . Besides describing the sequence order , ACC has the ability to transform each AA sequence of variable length into uniform equal-length vectors . This feature is very important in data mining methods , where a fixed-length vector describing each instance is required . However , averaging along the entire sequence may cause loss of information about strong and weak correlations . To cope with these limitations , the Maximum of auto- and cross-covariances ( MACC ) algorithm was introduced [30] , where positive and negative descriptor values are considered separately and only the maximum value of each lag is used . In this work we introduce an ACC descriptor accounting for both weak and strong correlations , the Minimum and Maximum of auto and cross-covariances ( mMACC ) descriptor ( Equation 1 ) . Equation 1 . Minimum and Maximum of auto and cross-covariance equations . Where Zki is the i-th descriptor of residue k in the sequence , d is the lag . As in the MACC algorithm , the maximum value of each interaction is taken into account . However , in the mMACC each z-scale descriptor is shifted by the absolute minimal value in order to have only positive interactions . This reduces the number of combinations , while maintaining both information of strong and weak interactions . In the following , the newly introduced ACC descriptor performances are compared with classical ACC and MACC descriptors . In the selection of the descriptors a tradeoff should be found between the performance of the encoding ( i . e . how well the statistical model based on a particular encoding is able to predict the peptide alpha-helix structure and/or antimicrobial activity ) and the requirement of minimizing the number of descriptors . Indeed , on equal terms of performance , a lower number of features is preferable , since the resulting model is less computationally expensive and the interpretation of resulting models is simpler . Figure 1 reports the performance of the three encodings ( ACC , MACC , mMACC ) as a function of the number of descriptors used . Prior to this analysis the descriptors were ordered by the mRMR ( minimum redundancy maximum relevance ) algorithm [31] . The performance is evaluated by the Mathews correlation coefficient ( MCC ) ( Equation 4 in the Materials and Methods ) , which assesses the prediction in terms of true and false positives and negatives . The maximum MCC value , corresponding to the optimal feature set , was compared for each encoding , as shown in Table 2 . The mMACC algorithm performed better both in the absolute MCC value , and in the ( smaller ) number of features . On the basis of these preliminary tests , the mMACC algorithm was chosen to encode topological descriptors for both Dataset A and Dataset B . mMACC plot of Dataset A showed a peak at 200 descriptors , whereas Dataset B reached its maximum of accuracy at 215 , and these subsets were selected to construct each training model . Results are summarized in Table 3 and the complete lists of features are reported in Tables S2 and S3 . Hereafter , the genetic algorithms for peptide sequence prediction will be based on these final optimal features . The distribution of the selected descriptors in terms of z-scales interactions and as a function of the lag between AAs is reported in the SI ( Figure S4 , Text S1 ) . RF has several properties that allow extracting relevant trends from data with complex variable relations . Proximity values are a measure of similarity between samples , calculated as the number of times the two samples end up in the same terminal node of the tree [32] . In this way , subclasses can be identified by finding peptides that have similar proximities to other AMPs . A matrix representing proximity values of each AMP in Dataset A was obtained from the final model . Cluster analysis resulted in five different clusters and the distribution of relevant properties was analyzed , as shown in Figure 2 . A dendrogram represents the subdivision into clusters ( Figure 2A ) , while a radial distribution of AA frequency is shown in Figure 2B , reflecting the average net charge at different pH ( Figure 2C ) . A heatmap showing the AA relative abundance of each cluster is represented in panel D . Cluster1 and Cluster2 present a high average net charge , with a different distribution of the charged residues . In particular , Lysine appears to be more frequent in Cluster 1 , as shown in the heatmap . Clusters 3 , 4 and 5 present a lower net charge , due to the higher abundance of negatively charged residues . This analysis stresses the role of overall charge and AA composition in classifying various AMP families . The algorithm can be asked to select sequences optimizing a particular property , in this case , presence or absence of antimicrobial activity and secondary structure . Two ab initio AMPs were chosen from two different trial sessions , hereafter named GMG_01 and GMG_02 . As a control , a session was performed aiming to the selection of non-antimicrobial , alpha-helical peptides ( GMG_03 ) . A second control was synthesized , GMG_01_SCR , a scrambled version of GMG_01 obtained by selection of the sequence with the lowest fitness upon permutation of the original sequence . This control was chosen in order to assess whether the algorithm was able to account for sequence order . After synthesis and purification of the above-mentioned peptides , MBC tests were performed in triplicate on S . aureus and P . aeruginosa ATCC strains . Peptides sequences , prediction scores and MBC values are summarized in Table 4 . GMG_01 and GMG_02 yielded an antimicrobial activity comparable to the most effective antimicrobial peptides described in literature . As expected , GMG_01_SCR yielded an antimicrobial activity approximately 16 times lower than the parental one , demonstrating that the AMP prediction model accounts for peptide's AA sequence . The residual antimicrobial activity was probably due to the reduced size of the peptide and the overall cationic nature of GMG_01_SCR . In fact , a short sequence with repeated AAs is not likely to present significant differences in its primary structure; consequently , its chemophysical profile results similar to the original sequence . Notably , GMG_01_SCR shows a peculiar sequence with all the charged residues concentrated on one terminus , demonstrating that the charge ‘spatial’ distribution is an important feature of functional alpha-helical AMPs . As expected , no antimicrobial activity was observed for the GMG_03 peptide , despite its alpha-helical secondary structure . This is likely related to the negative net charge of the peptide at physiological pH , which may not favour its adhesion to the bacterial cell surface . Optimization of an existing antimicrobial peptide ( CM18 ) was performed , in order to obtain an improved system at shorter length , thus with easier synthesis requirements . To this aim , a size constraint was added to preferentially select peptides with sequences shorter than 14 residues . Furthermore , a third objective was added to avoid an excessive difference from the original peptide , the Smith-Waterman normalized score ( Equation 5 in the Materials and Methods ) . This score measures the similarity between two amino acidic sequences , normalized by the sequence size and requires a measure of the similarity between two AAs . Instead of the commonly used BLOSUM or PAM , a score matrix obtained by the Euclidean distance between each amino-acid z-scale values was used ( Figure S2 ) . Interestingly , residues with similar physicochemical characteristics grouped together . This facilitates single-AA substitutions , particularly regarding non-natural residues . The sequence of CM18 was intentionally removed from the AMP dataset in order to avoid improper influence on the optimization process . A sequence of 12 AAs was selected ( CM12 ) , as reported in Table 4 . When tested for its MBC , CM12 retained full activity , notwithstanding a nearly 30% decrease in chain length . Finally , starting from the non-active ab-initio peptide - GMG_03 – the sequence was optimized for antimicrobial activity and all-alpha structure . In the optimization process , the amino-acid alphabet was extended to non-natural elements , including all the 87 AAs listed by the z-scale descriptors . The process was terminated after 300 generations , as described in SI ( Text S1 ) . From the candidate list , a sequence with two non-natural substitutions was chosen from a list of feasible solutions and synthesized . The selected peptide contains two norleucine ( Nle ) residues , one substituting the original leucine residue and the other one in the proximity of the C-terminus . MBC assays demonstrate an enhanced antimicrobial activity , significantly higher than the original one . It is worth noting that the overall net charge increased due to the elimination of two negatively charged and the insertion of two positively charged residues . This may facilitate the initial attachment of the peptide to the membrane . The new residue distribution confers a high amphipathicity to the resulting peptide sequence ( Figure 3 ) . Interestingly , Nle is an artificial AA frequently used in antimicrobial peptide design for research purpose [3] as well as in clinical studies [10] . Analysis of the AA z-scores heatmap ( Figure S2 ) revealed a clusterization of Nle with its natural precursor leucine , justifying the substitution choice . MD simulations were performed on a selected subset of peptides ( GMG_01 , GMG_03 , GMG_01_SCR and GMG_05Z ) to assess the accuracy of the proposed algorithm in terms of structural prediction . Different solvent conditions were simulated , either water or TFE/water mixture . The latter condition is known to stabilize secondary-structure elements and to partially account for the hydrophobic environment inside the lipid bilayer . The percentage of alpha-helix structure vs . other secondary-structure motives was monitored during 700 ns of molecular dynamics after suitable equilibration . The MD simulations fully support the structural predictions of the algorithm ( Figure 4 ) . In particular , both GMG_01 in TFE/water and GMG_03 in pure water assume rather stable helical conformations . GMG_01_SCR in TFE/water , by contrast , displays negligible alpha-helix propensity again confirming the algorithm prediction . GMG_01 was simulated both in pure water ( Figure S3 ) and in TFE/water mixture . Remarkably , the simulations predict a high percentage of helical structure only in the latter condition , as is the general behavior of linear alpha AMPs [3] . Finally , the TFE/water MD simulations of the NLE-containing peptide ( GMG_05Z ) show the formation of a very stable helical portion , but limited to residues 6 to 9 in the sequence , while the N-terminal portion results completely unstructured . Though MD simulations in the μs range should generally be sufficient for adequate exploration of the conformational landscape in the short peptides examined , it is not possible to rule out slower folding time for some sequences . In particular , in the GMG_05Z case , the simulated time was extended to 2 µs showing an unfolding of the helix followed by folding into an enlarged alpha-helix , also comprising residues 4 and 5 in the sequence . These results are shown in Figure S5 , also reporting the time series of secondary structure motives for GMG_01 , GMG_01_SCR and GMG_03 . In order to analyze the mechanism of action of GMG_05Z compared with the original peptide GMG_03 , it was of significant interest to determine their localization in bacteria following treatment . GMG_05Z and GMG_03 analogues with a C-terminal cysteine-atto633 insertion were synthesized and purified . Fluorescence and confocal microscopy were performed after treatment of an ATCC S . aureus strain ( ATCC33591 ) with both peptides separately . To determine the influence on the antimicrobial activity of the C-terminus cysteine-atto633 insertion , MBC test were repeated . No significant variations were detected for GMG_03 , while GMG_05Z MBC was twofold greater ( 0 . 25 µM ) , indicating that the addition of cysteine-atto633 had a minimal effect on the antimicrobial activity . MBC results are summarized in Table S4 of the supporting information . As a further control on the structural influence of the N-terminal cysteine residue , GMG_01 and GMG_03 MD simulations were repeated adding this AA to the sequence . The results ( Figure S3 ) confirm that the effect of this addition is very limited . Confocal images of S . aureus exposed to GMG_05Z revealed its ability to make contact with the membrane ( Figure 5A ) . As expected , the inactive peptide ( GMG_03 ) was instead unable to interact with bacteria , as shown in Figure 5B . Further studies are required in order to understand whether GMG_05Z acts by forming a transient pore or via metabolic mechanisms . In this work , a rapid and intuitive method for virtual screening of antimicrobial candidates was introduced . The method can be successfully applied to ab-initio prediction as well as peptide optimization with natural and non-natural AAs . Three different types of topological descriptors were applied for model construction of antimicrobial peptides and all-alpha peptides . The novel mMACC algorithm retained the best performance ( as assessed by its highest MCC values ) , thus lowering the number of needed descriptors . Furthermore , the identification of the optimal complexity of auto and cross covariance descriptors was achieved automatically by IFS , eliminating the tedious process of manual feature selection . The use of physicochemical descriptors allows the analysis and prediction of non-natural AAs insertions , extending the flexibility in peptide design . The ab-initio peptide prediction demonstrated the high degree of flexibility of the multi-objective evolutional algorithms ( MOEA ) approach , in which constraints and objectives can be added depending on the needs . The potential of this approach was demonstrated by transforming a non-antimicrobial peptide into a highly active AMP , using non-natural AAs . Finally , virtual screening was combined with MD simulations to gain insight into the structural properties of the predicted AMPs , and thus provide the molecular basis for understanding peptide-membrane interaction mechanisms . In conclusion , the combination of chemophysical descriptors and MOEA confers an elevated flexibility to antimicrobial peptide design , permitting to select highly active molecules .
Two different datasets , Dataset A and B , were constructed for model training and validation . Dataset A consists of antimicrobial peptides with a sequence length ranging from 11 to 40 residues extracted from YADAMP and CAMP databases [13] , [27] . After removal of peptides with disulfide bridges and non-standard residues sequences , 1884 peptides were left . The negative dataset was populated with non-secretory sequences randomly extracted from UniProt database , without ‘antimicrobic’ annotation and with a length ranging from 11 to 40 AAs . Dataset B represents all-alpha helical peptides . The CB513 dataset , a non-redundant set of 513 well-defined proteins [33] was used in a first step for extraction of all-alpha , all-beta and all-coil domains . Then a number of random sequences were extracted from the same database in order to account for mixed secondary structure states . The final dataset was then built using the simplest partition of the space into alpha and non-alpha peptides . For both datasets , a homology cutoff was imposed to exclude similar peptides in order to avoid redundant data that could influence the prediction performance . Peptides showing equal to or greater than 70% sequence identity to any other in the dataset were identified and removed by the CD-HIT program [34] . Final datasets composition is summarized in Table 5 . Global and topological descriptors were utilized in order to encode peptide sequences . Peptide charge at different pH conditions , isoelectric point and the number of positive and negative charges were used to describe charge-related characteristics . The z-scale moment ( μZi ) , an extension of Eisenberg's hydrophobic moment equation [35] , is introduced to represent z-scales distribution along peptide sequences . Equation 2 . Z-scale moment . In Equation 2 , δ is the angular frequency of the AA residues forming the structure ( 100° for alpha helix ) ; k is the number of the particular residue examined , L is the length of the sequence and Zik is the zi-scale value of the kth AA . In particular , μZ1 represents a measure of the hydrophobicity distribution along peptide sequence . Average sum of z-scale descriptors has been successfully used in QSAR analysis of bioactive peptides [36] , as it gives a general description of peptides physicochemical main features [37] . The aim of the study was to develop an alignment-independent method , therefore position specific score matrix ( PSSM ) as well as amino acidic and pseudo-amino acidic sequence descriptors were avoided . Both in the global and topological descriptors , Z-scale values were mean-centered and scaled prior to their use , as described by the following equation:Equation 3 . Z-scale descriptor normalization . Where Zi is the ith descriptor of z-scales variables , zi is the original z-scale value ( from [21] ) and N is the number of AAs in the z-scales descriptors table . The final list of descriptors is summarized in Table 1 . In this study , the Random Forest algorithm ( RF ) , implemented in the software suite WEKA [38] , was adopted as prediction engine . During the evaluation procedure , nine different variables combinations were tested for model building . In particular , the number of trees in the forest ( M ) and the number of random variables used for each tree ( T ) . Each model performance was measured with a 10-fold cross-validation analysis , where each dataset was divided into 10 parts - 9 parts for model learning ( training ) and the remaining part for validation ( testing ) . Four performance measures were used: true positive rate for sensitivity , false positive rate for selectivity , predictive accuracy and MCC , as defined below . Equation 4 . Performance evaluation equations Where TP , TN , FP and FN are the number of true positive , true negative , false positive and false negative , respectively , resulting from the model . MCC is an important index used to evaluate the performance of the predictor when the dataset is not balanced [39] . In order to obtain a non-redundant set of descriptors , the Maximum Relevance , Minimum Redundancy ( mRMR ) method [31] was employed to sort features in descending order of importance . Incremental Feature Selection ( IFS ) [40] was applied to the sorted descriptors list by incrementing consecutively the number of descriptor by 5 . Each descriptor set thus obtained was evaluated by tenfold cross-validation and the IFS curve was plotted to unveil the relation between the performance of the model and the feature subset . The optimal feature subset is defined as that showing the highest MCC value ( Figure 1 ) ; the selected model was used for peptides classification . The hierarchical list of the final descriptors for Dataset A and Dataset B is shown in Table S2 and Table S3 , respectively . For peptide optimization , a supplemental objective representing sequence similarity was added . Sequence similarity is defined by the Smith-Waterman score between the respective peptide sequences [41] . Since the Smith-Waterman score is dependent on input sequences length , the final score was normalized between 0 and 1 by dividing by the maximum score of the two self-alignments , as shown in Equation 5 [42] . Equation 5 . Smith-Waterman normalized score Here , SA , B is the similarity score between sequence A and B , SA , A and SB , B are the self-alignment score of sequence A and sequence B , respectively . In order to consider not only the identity between two amino acidic positions , a score matrix was defined by calculating the Euclidean distance between the five auto-scaled z-scale values of each AA pairs . The score was then normalized between 0 and 1 , where 1 is the identity . For visualization purpose , the resulting matrix was analyzed with R [43] and a heat map was produced , calculated as Log2 of the inverse AA distance normalized by AA median value ( Figure S2 ) . All peptides were prepared by solid-phase synthesis using Fmoc chemistry on an automatic peptide synthesizer and the crude peptides were purified by RP-HPLC , as previously described [8] . The cysteine residue added to the C-terminus of some peptides provided a sulfhydryl group for further ligation to the atto-633-maleimide fluorophore . The labelling of purified peptides was performed by incubating for 3 h with a 3-fold molar excess of atto-633-maleimide ( ATTO-TEC GmbH , Germany ) , 150 mM PBS buffer , TCEP , at pH 7 . 4 . Finally , atto-633-labeled peptides were purified by HPLC and then lyophilized overnight . The correct purified product was confirmed by electrospray mass spectroscopy with an API3200QTRAP a Hybrid Triple Quadrupole/Linear Ion Trap ( ABSciex , Foster City , California , USA ) . Peptides were stored at −80 C . Antibacterial activity of designed peptides was evaluated by a liquid microdilution assay in 10 mM sodium phosphate buffer ( SPB ) , pH 7 . 4 , as described previously [44] . Briefly , S . aureus ATCC33591 and P . aeruginosa ATCC27853 were grown in tryptone soy broth ( TSB; Oxoid ) . Exponentially growing bacteria were resuspended in SPB to obtain a density of 1∧106 colony forming units ( CFU ) /ml and exposed to different concentrations of peptide , ranging from 64 µM to 0 . 25 µM . After incubation of peptides for 1 . 5 h at 37°C , 0 . 2 ml of 10-fold serial dilutions of each samples were plated onto tryptone soy agar . As a control , bacteria were also incubated in the absence of peptides . After 24 h incubation at 37°C the number of CFU was assessed . Bactericidal activity was evaluated as minimal bactericidal concentration ( MBC ) defined as the lowest peptide concentration at which a reduction in the CFU/ml numbers of > = 3 logs was observed after 1 . 5 h of incubation in three independent experiments . Molecular dynamics simulations of selected peptide sequences ( GMG_01 , GMG_03 , GMG_01_SCR and GMG_05Z ) were performed with GROMACS 4 . 5 . 5 [45] . The force field used was Amber ff99SB-ILDN-NMR [46] . Random configurations of the peptides were solvated either with a box of TIP3P water molecules , or with a mixture of TFE ( trifluoroethanol ) and TIP3P water molecules ( 4 water molecules each TFE molecule , in order to obtain a ∼50% vol/vol solution ) . TFE force field parameters were taken from ref [47] . RESP HF/6-31G* charges of TFE and NLE ( Norleucine ) were taken from ref [47] and [48] respectively , while the other force field terms were taken from the already existing parameters . Either Cl− or Na+ ions were added to neutralize the system . The truncated octahedron solvation box dimension was chosen in order to keep a distance of at least 8 Å between the peptide and the box faces , and periodic boundary conditions were applied . For each examined peptide , simulations were performed under constant temperature ( 300 K ) and pressure ( 1 atm ) conditions , using the Nose-Hoover ensemble [49] for temperature coupling ( τ = 0 . 5 ps ) and the Parrinello-Rahman ensemble [50] for pressure coupling ( τ = 5 ps ) . The timestep was set at 2 fs , and the bonds involving hydrogen atoms were constrained using LINCS [51] . After an equilibration phase of 300–500 ns , the production runs lasted for 700 ns , and peptide snapshots were recorded each 10 ps . These 700 ns production runs were used for secondary structure analysis , performed using DSSP [52] . As previously described , S . aureus ATCC33591 strain was grown in tryptone soy broth and exponentially growing bacteria were resuspended in SPB to obtain a density of 1∧108 CFU/ml and exposed to 2 . 5 µM concentration of peptide labeled with ATTO633 ( Table S2 ) . After incubation for 15 min at 37°C , 5 µL of the solution were spotted onto a slice of 1% water agarose gel and placed on a glass bottom petri dish . Images were acquired using a Leica TCS SP5 SMD inverted confocal microscope ( Leica Microsystems AG ) interfaced with a HeNe laser for excitation at 633 nm and the sample was viewed with a 63×1 . 2 NA water immersion objective ( Leica Microsystems ) . The pinhole aperture was set to 0 . 5 Airy . All data collected were analyzed by ImageJ software version 1 . 44o . | In recent years , the increasing and rapid spread of pathogenic microorganisms resistant to conventional antibiotics especially in hospital settings spurred research for the identification of novel molecules endowed with antimicrobial activities and new mechanisms of action . Antimicrobial peptides ( AMPs ) received an increasing attention as potential therapeutic agents because of their wide spectrum of activity and low rate in inducing bacterial resistance . Currently , research is focused on the design and optimization of novel AMPs to improve their antimicrobial activity , minimize the cytotoxicity and reduce the proteolytic degradation , also in biological fluids . To this end , the introduction of non-natural amino acids will be a key requisite in order to contrast host resistance and increase compound's life . However , the amino acidic alphabet extension to non-natural elements makes a systematic approach to AMPs design unfeasible . A rational in-silico approach can drastically reduce the number of testing compounds and consequently the production costs and the time required for evaluation of activity and toxicity . In this article , AMP in-silico design with non-natural amino acids was performed and a series of candidates were tested in order to demonstrate the potentiality of this approach . | [
"Abstract",
"Introduction",
"Results/Discussion",
"Materials",
"and",
"Methods"
] | [] | 2013 | Antimicrobial Peptides Design by Evolutionary Multiobjective Optimization |
All cells respond to osmotic stress by implementing molecular signaling events to protect the organism . Failure to properly adapt can lead to pathologies such as hypertension and ischemia-reperfusion injury . Mitogen-activated protein kinases ( MAPKs ) are activated in response to osmotic stress , as well as by signals acting through G protein-coupled receptors ( GPCRs ) . For proper adaptation , the action of these kinases must be coordinated . To identify second messengers of stress adaptation , we conducted a mass spectrometry-based global metabolomics profiling analysis , quantifying nearly 300 metabolites in the yeast S . cerevisiae . We show that three branched-chain amino acid ( BCAA ) metabolites increase in response to osmotic stress and require the MAPK Hog1 . Ectopic addition of these BCAA derivatives promotes phosphorylation of the G protein α subunit and dampens G protein-dependent transcription , similar to that seen in response to osmotic stress . Conversely , genetic ablation of Hog1 activity or the BCAA-regulatory enzymes leads to diminished phosphorylation of Gα and increased transcription . Taken together , our results define a new class of candidate second messengers that mediate cross talk between osmotic stress and GPCR signaling pathways .
Cells routinely experience changing and often unfavorable conditions in their environment . The ability to adapt to environmental stress and re-establish homeostasis is essential not only to the survival of a cell , but also to the well-being of the organism . The response to such physical or chemical stresses is mediated by well-defined signaling networks . For example , changes in nutrient availability switch signaling between the opposing target of rapamycin ( TOR ) and AMP-activated protein kinase ( AMPK ) pathways [1 , 2] . Stressors such as UV irradiation , inflammatory cytokines , and osmotic shock promote signaling through activation of the p38 and c-Jun N-terminal Kinase ( JNK ) MAPK pathways [3 , 4] . While much is known about the mechanisms of stress-dependent signaling , less is known about coordination between the stress response and other cell signaling processes . In this study , we investigate cross-talk between osmotic stress and G protein-coupled receptor ( GPCR ) signaling pathways . Hyperosmotic stress causes water efflux and cell shrinkage in order to normalize the osmotic balance between the intracellular and extracellular space . Depending on the severity of the stress , cell shrinkage can lead to macromolecular crowding and alterations in cellular protein activity [5] , the production of reactive oxygen species ( ROS ) , DNA damage , cell cycle arrest , and apoptosis [6] . In addition to these negative effects , cells also initiate signaling events that promote adaptation . Most prominently , osmotic stress activates the MAPK p38 , which in turn phosphorylates myriad downstream targets that coordinate osmotic stress adaptations . Targets of p38 include the transcription factor NFAT5 , which promotes the expression of proteins associated with the synthesis and transport of osmolytes , antioxidants , and molecular chaperones [7 , 8] . Such changes ensure the survival of the cell , and they are likely to have important consequences for other signaling pathways via cross-talk mechanisms . As the largest receptor family in humans [9] , GPCRs are likely targets of cross-pathway regulation . These receptors respond to a wide variety of homeostatic cues , such as hormones and neurotransmitters , as well as to environmental signals such as odors and light . They signal primarily through intracellular heterotrimeric G proteins , comprised of Gα and Gβγ subunits . G proteins in turn activate downstream effectors leading to the production of second messenger molecules , such as calcium or cAMP , which bind to and activate intracellular protein kinases . Another mechanism of GPCR signaling entails the direct activation of protein kinases upstream of MAPKs [10 , 11] . The G protein α subunit is a molecular on/off switch for signaling processes . As such , it is likely to be a critical target for post-translational modifications that regulate GPCR signaling . In fact , several studies have shown that Gα proteins are phosphorylated , resulting in altered affinity for Gβγ subunits or guanine nucleotides [12–20] . In some cases , phosphorylation is the direct result of pathway activation , and thus constitutes a positive or negative feedback . In other cases , phosphorylation is triggered by parallel pathways , and thus constitutes a mechanism of signal coordination or cross-talk . Previously , we showed that Gα is phosphorylated in response to nutrient limitation [21] . Our focus here is Gα regulation by osmotic stress . Identifying how environmental stress can promote post-translational modification of Gα subunits is necessary to fully understand the mechanisms by which the pathways are coordinated and integrated . Studying the response to environmental stress is often challenging however , due to the expression of multiple protein isoforms and differences in expression among various tissues and cell types . Given these complexities , much can be learned from the analysis of orthologous signaling processes in simpler eukaryotes . The budding yeast Saccharomyces cerevisiae has a stress response pathway and a GPCR signaling pathway with component proteins that are evolutionarily conserved across eukaryotes . The High Osmolarity Glycerol , or HOG , pathway is comprised of a MAPK ( Hog1 ) , a MAPK kinase ( Pbs2 ) , and MAPK kinase kinases ( Ste11 , Ssk2/Ssk22 ) . Upon activation , Hog1 phosphorylates cytoplasmic and nuclear proteins that aid in the restoration of osmotic equilibrium through osmolyte synthesis and the induction of stress response genes [22–25] . Hog1 is the yeast ortholog of mammalian p38 [26 , 27] . Yeast use another , parallel MAPK pathway to initiate haploid cell fusion , or mating . This pathway is activated by pheromone binding to a GPCR to initiate exchange of GDP for GTP in the Gα subunit ( Gpa1 ) and subsequent dissociation of Gα from Gβγ . Gβγ activates a MAPKKK ( Ste11 , shared by the HOG pathway ) , a MAPKK ( Ste7 ) and a MAPK ( Fus3 , or Kss1 ) . Once activated , Fus3 promotes transcription of genes to initiate cell mating [28] . Fus3 is the yeast ortholog of mammalian ERK1 and ERK2 [29–32] . In the present study , we use yeast as a model system to investigate how crosstalk regulates G protein signaling during osmotic stress . We have shown previously that osmotic stress dampens the pheromone response pathway , and does so by Hog1-dependent and Hog1-independent mechanisms [33] . We have also shown that glucose limitation dampens the pheromone response pathway , and does so by reducing intracellular pH [34] . The increase in proton concentration is detected by the G protein directly , resulting in increased phosphorylation of Gpa1 and a dampened mating signal . Additionally , we have identified a family of three kinases ( Elm1 , Sak1 , and Tos3 ) and a PP1 phosphatase complex ( Reg1/Glc7 ) as the molecular machinery responsible for phosphorylating and dephosphorylating Gpa1 [21] . We show here that Gpa1 is likewise phosphorylated in response to osmotic stress , and that phosphorylation of Gpa1 requires the same protein kinases , but does not entail any changes in intracellular pH . In a search for alternative mediators of cross-pathway regulation , we conducted an unbiased metabolomics screen and found that 2-hydroxy branched chain amino acid metabolites are produced in a salt- and Hog1-dependent manner . Finally , we show that these metabolites are necessary and sufficient to promote Gpa1 phosphorylation and dampen downstream signaling . We propose that these metabolites represent a new class of second messengers of the stress-responsive HOG pathway .
To understand how cells adapt to environmental stresses , we sought to identify conditions that impact pheromone signaling through the phosphorylation of Gpa1 . We recently established that Gpa1 is phosphorylated by a family of three AMPK kinases ( Elm1 , Sak1 , and Tos3 ) , and dephosphorylated by the phosphatase complex Reg1/Glc7 [17 , 21] . These proteins were previously shown to phosphorylate and dephosphorylate the yeast AMPK , Snf1 [35–37] . Snf1 , is phosphorylated and activated in response to nutrient limitation , as well as heat shock , hyperosmotic shock , reactive oxygen species , ethanol , and changes in extracellular pH [38] . Accordingly , we asked whether the same environmental stressors would lead to phosphorylation of Gpa1 . We treated wild-type cells with the indicated stressor in a 2-hour time-course ( see Materials and methods ) , and analyzed cell lysates by western blot . As shown in Fig 1 , Gpa1 and Snf1 were phosphorylated in all stress conditions tested ( see also S1 Fig ) . However , among the stressors there were differences in the both the magnitude and duration of phosphorylation . In glucose-limiting conditions , approximately 90% of Gpa1 was phosphorylated by 2 minutes , with a gradual decline after 10 minutes ( Fig 1A ) . Heat shock ( at 42°C ) also promoted rapid phosphorylation but slow dephosphorylation . Osmotic stress promoted slow phosphorylation , but fast dephosphorylation . Heat and osmotic stress also promoted the phosphorylation of Snf1 , but the effects were comparatively weak and transient ( Fig 1B and 1C ) [38] . These data reveal that Gpa1 , like Snf1 , is phosphorylated in response to various stress signals . More broadly , the results indicate that physico-chemically distinct stimuli have a common ability to promote phosphorylation of two functionally distinct proteins , Snf1 and Gpa1 . It is well-established that the MAPK Hog1 is phosphorylated and activated in response to osmotic stress [22] . Hog1 is also activated by heat shock [39] , cold stress [40] , oxidative stress [41] , and hypoxia [42] . Given that many of these conditions also lead to phosphorylation of Gpa1 and Snf1 , we asked if Hog1 activation was required in either case . To this end , we replaced Hog1 with a mutant documented to lack catalytic activity , hog1K52R [43] , and then treated the cells with 0 . 5 M KCl . Whereas Snf1 phosphorylation was unperturbed , the phosphorylation of Gpa1 was almost completely abrogated in the hog1K52R strain ( compare Fig 1C , blue curve vs . red curve ) . It is unlikely that Hog1 phosphorylates Gpa1 directly , since the relevant site ( Ser200 ) does not adhere to the MAPK consensus sequence ( Ser/Thr-Pro ) [17] . Thus , Hog1 catalytic activity is required for the salt-induced phosphorylation of Gpa1 but not Snf1 . More broadly , these results implicate at least two distinct signaling pathways , and potentially two distinct second messengers , that mediate the response to osmotic stress . One potential second messenger is pH . Indeed , it is well established that glucose limitation leads to a substantial decrease in intracellular pH ( pHi ) [44] . We have shown recently that Gpa1 is a pH sensor , and that pH-dependent changes in conformation result in phosphorylation of the protein [34] . Since other stressors trigger phosphorylation of Gpa1 , we asked whether any of those conditions also cause a change in pHi . To that end we expressed the ratiometric fluorescent pH biosensor , pHluorin , in wild-type cells [34 , 45 , 46] . Consistent with earlier studies [34] , we observed a decrease in pHi from 7 . 0 to 6 . 4 upon glucose limitation ( Fig 1A , inset ) . In contrast , cells subjected to osmotic stress exhibited no change in pHi over the course of 60 minutes ( Fig 1C and S1 Fig ) . These data indicate that low glucose and osmotic stress each promote Gpa1 phosphorylation , but glucose alone affects pHi . We therefore postulated the existence of an additional second messenger of the osmotic stress response . The data presented above reveal that osmotic stress has no effect on pHi , yet is a potent inducer of Gpa1 phosphorylation . To identify potential second messengers of osmotic stress , we conducted a global , unbiased metabolomics analysis [47] . Based on results from the Gpa1 phosphorylation experiments above , we sought to identify metabolites that increased with osmotic stress and did so in a Hog1-dependent manner . To this end , we subjected wild-type and Hog1-deficient yeast cells to 0 . 5 M KCl for 20 minutes and then analyzed cell extracts by LC-MS/MS and GC-MS ( Fig 2A ) . This analysis identified 296 distinct entities representing each major class of biochemical molecules—amino acids , peptides , carbohydrates , lipids , nucleic acids , vitamins and cofactors , and xenobiotics ( S1 Table ) . Consistent with past findings , we found that the osmolytes trehalose [48] and glycerol [49] were induced substantially ( 32-fold and 2 . 5-fold , respectively ) ( Fig 2C ) . Using a comparable ( 2-fold ) cut off , we identified an additional 26 metabolites that increased in response to osmotic stress , and 13 that increased in the presence of Hog1 . Of these , only three required osmotic stress and Hog1 together ( Fig 2B and 2C , S2 Table ) : 2-hydroxyisovalerate ( HIV ) , 2-hydroxyisocaproate ( HIC ) , and 2-hydroxy-3-methylvalerate ( HMVA ) . All three compounds are 2-hydroxy carboxylic acid derivatives of the branched-chain amino acids ( BCAAs ) valine , leucine , and isoleucine , respectively ( Fig 2C and 2D ) . Thus , our analysis points to 2-hydroxy BCAA derivatives as candidate second messengers of osmotic stress . Our metabolomics study demonstrated that BCAA derivatives are produced in response to osmotic stress , and that their production requires Hog1 ( Fig 2D ) . In principle , deleting Hog1 could alter the production of additional second messengers that may not have been detected in our metabolomics screen . However , as BCAA derivatives were the most robustly increased metabolites that met our criteria for osmotic stress , we examined the consequences of disrupting BCAA catabolism through the so-called Ehrlich pathway in yeast [50] . The first step in the Ehrlich pathway is transamination to a 2-keto acid by the branched-chain amino acid transaminases , Bat1 and Bat2 . The second step is decarboxylation of the 2-keto acid to an aldehyde , which is subsequently converted to a fusel acid or fusel alcohol . The BCAA derivatives identified here retain the same carbon skeleton as the parent amino acids , suggesting the existence of an alternative metabolic route consisting of transamination followed by reduction to the 2-hydroxy acid ( Fig 3A ) . Products of the Ehrlich pathway are exported from the cell by the ABC transporter Pdr12 [50 , 51] . To test whether BCAA derivatives are required for phosphorylation of Gpa1 and/or Snf1 , we deleted the BAT1 and BAT2 genes individually ( Fig 3B ) . After osmotic stress , we observed a modest , but significant reduction in maximal phosphorylation of Gpa1 in the bat1Δ and bat2Δ mutants , as compared to wild-type cells ( Fig 3B and 3C ) . As expected , Snf1 phosphorylation was unaffected ( Fig 3B ) . Cells harboring deletion of both BAT genes are reported to be viable [52 , 53]; however in our hands , bat1Δbat2Δ double mutants arose at a lower-than-predicted frequency after tetrad dissection and likely harbored suppressor mutations . As an alternative approach , we attempted to use a tetracycline-repressible BAT1 in a bat2Δ background . However , the doxycycline used to repress BAT1 expression also promoted the phosphorylation of Gpa1 . Gpa1 phosphorylation was unaffected by loss of the transporter gene PDR12 ( Fig 3D ) , suggesting other routes of removal or of further metabolism . Together these results indicate that either Bat1 or Bat2 is necessary for cell viability . Both proteins , as well as their catalytic products , are necessary for a full response to osmotic stress . Our results indicate that Hog1 activity and BCAA catabolism are both needed for a full response to osmotic stress . In particular , we have shown that osmotic stress-dependent Gpa1 phosphorylation is reduced in mutants lacking Bat1 or Bat2 , and is eliminated in cells lacking Hog1 catalytic activity . Based on these findings , we hypothesized that Hog1 phosphorylates one or more components of the BCAA pathway . Indeed , Bat1 has five MAPK consensus sites ( S/TP ) , and Bat2 has three such sites . In support of our hypothesis , replacement of the MAPK consensus sites in Bat1 and Bat2 led to a significant reduction in Gpa1 phosphorylation ( Fig 3B and 3E ) . However , there were no changes in the electrophoretic ( phosphorylation-dependent ) mobility of Bat1 , Bat2 , Bat15A , or Bat23A , either in the absence or presence of salt stress . There was also no effect of osmotic stress on Bat23A in cells lacking Bat1 ( bat1Δ bat23A ) or Bat15A in the absence of Bat2 ( bat15A bat2Δ ) ( S2 Fig ) . Taken together , these results suggest that Hog1 does not target the transaminases , and instead plays an indirect role in promoting the production of BCAA derivatives . That role could be to induce phosphorylation , or regulate the transcription , of some other component of the metabolic pathway . One potential target is the reductase ( s ) ( as yet unidentified ) that converts the 2-keto acid to the 2-hydroxy acid . An intracellular second messenger should , by definition , be sufficient as well as necessary to evoke the response of the extracellular first messenger . Having demonstrated that BCAA derivatives are necessary for a full response to osmotic stress , we tested the ability of the BCAA derivatives to promote phosphorylation in the absence of salt stimulus . To better enable these compounds to traverse the cell membrane , we grew the cells at pH 5 , which is closer to the pKa of the metabolites . By favoring the protonated , uncharged species , the BCAA derivatives can more easily traverse the plasma membrane . Importantly , the lower external pH does not change the intracellular pH ( Fig 4B , inset ) [34] . Using this approach , we found that HIV , HIC , and HMVA promoted Gpa1 phosphorylation , but with varying efficacy . HIC showed the strongest effect while HIV had the weakest effect ( Fig 4A and 4B ) . Addition of HIC ( but not salt ) to Hog1-deficient cells promoted the phosphorylation of Gpa1 , consistent with the idea that BCAA derivative production is a consequence of Hog1 activation ( Fig 4A , 4C and 4D ) . Snf1 was likewise unaffected , consistent with the idea that it is regulated by a distinct second messenger ( Fig 4A ) . Taken together , these experiments indicate that BCAA derivatives are sufficient to promote the phosphorylation of Gpa1 and thus meet the criteria for second messengers of osmotic stress . Our results so far show that BCAA derivatives promote the phosphorylation of Gpa1 . We next attempted to delineate the mechanism by which BCAA derivatives act . We demonstrated previously that protons interact directly with the G protein α subunit , causing a conformational change that promotes its phosphorylation . Moreover , the pH dependent change is conserved in yeast and human Gα proteins [34] . We hypothesized that BCAA derivatives might likewise act by binding to the Gα subunit . To test this we collected 1H-15N 2D heteronuclear NMR spectra of Gα , both in the absence and presence of BCAA derivatives . These spectra allow for the detection of protons directly bonded to a 15N , including both backbone and side-chain NH resonances . As an NH resonance can be detected for every residue , with the exception of proline , the spectrum contains a “fingerprint” of the protein backbone and allows perturbations resulting from interactions to be detected on a per-residue basis . This approach is widely considered as a definitive method for detecting low- to intermediate-affinity binding of ligands to proteins [54] . Accordingly , we acquired the NMR spectra of 15N-enriched Gαi in its GDP-bound state , alone or in the presence of a 25-fold excess of individual BCAA derivatives . As shown in Fig 5 , there were no significant peak shifts when BCAA derivatives were present ( Fig 5A–5C ) . As a positive control , we acquired NMR spectra of Gαi-GDP at pH 6 and at pH 7 . As shown in Fig 5D , a substantial number of peaks were shifted at the lower pH , consistent with proton-dependent conformational changes in Gαi . These results indicate that BCAA derivatives likely act on another component of the G protein signaling pathway . Gpa1 is phosphorylated by the AMPK kinases Elm1 , Sak1 , and Tos3 . Whereas Elm1 phosphorylates Gpa1 in a cell-cycle-dependent manner [17] , Sak1 is responsible for phosphorylation during glucose limitation [21] . Our data presented above indicate that Gpa1 is likewise phosphorylated in response to osmotic stress . To determine which , if any , of the AMPK kinases mediates the response to osmotic stress , we compared Gpa1 phosphorylation in cells lacking each of the three kinases , alone or in combination . As shown in S3A Fig , deletion of ELM1 resulted in the greatest reduction of Gpa1 phosphorylation , while deletion of SAK1 had a comparatively small effect . We then performed the same experiment using BCAA metabolites in place of osmotic stress . As with salt stimulation , HIC promoted the phosphorylation of Gpa1 in cells , and phosphorylation was diminished in the elm1Δ mutant ( Fig 6 and S3B Fig ) . Taken together , these results indicate that both the primary messenger ( osmotic stress ) and the putative second messenger ( the BCAA derivatives ) act through Elm1 . More broadly , these results confirm a fundamental difference between glucose- and salt-dependent changes in the cell . While both conditions lead to Gpa1 phosphorylation , they lead to the production of two distinct second messengers ( protons and BCAA derivatives ) and to phosphorylation by two distinct protein kinases ( Sak1 and Elm1 ) . We have shown that osmotic stress leads to a diminished pheromone response [33] and phosphorylation of the Gα protein ( this work ) . Based on our model , the BCAA derivatives are responsible for many of the intracellular effects of osmotic stress signaling , including Gα phosphorylation . According to our proposed mechanism , the same metabolites should also dampen the response to pheromone . To test this hypothesis , we employed a transcriptional reporter assay using GFP under control of the FUS1 promoter , which is specific to the pheromone response pathway [55] . We then measured fluorescence in response to increasing concentrations of the α-factor mating pheromone , alone or in combination with KCl or the BCAA derivatives . Consistent with previous reports [33] , osmotic stress dampened the pheromone response by approximately 40% . Consistent with our present model , the addition of HIV , HIC , or HMVA also led to a diminished response of up to 40% ( Fig 7A ) . The capacity of each BCAA derivative to dampen transcription correlated directly with its ability to promote Gpa1 phosphorylation ( Fig 4B ) . Deletion of the Gpa1 kinases conferred an elevated signal at all but the highest concentrations of pheromone . At 10 μM pheromone the mutant strain was less sensitive to KCl and HIC ( a reduction of 27% and 26% ) compared to wild type ( 35 and 41% , respectively ) . At low and intermediate concentrations , the mutant strain was less responsive to salt and largely unresponsive to the BCAA derivatives ( Fig 7B ) . Thus , BCAA derivatives are produced in response to an osmotic stress stimulus and , by every measure used , approximates the biochemical effects of salt on Gpa1 . By these criteria the BCAA derivatives could function as second messengers of the osmotic stress response pathway and account for part of the osmotic stress response program .
Here , we present several novel features of the pheromone response pathway that we believe will be generally applicable to other MAPK signaling systems . First , we show that multiple environmental stressors lead to G protein phosphorylation . Phosphorylation of Gpa1 is accompanied by attenuated signaling through the effector MAPK , Fus3 [21 , 33 , 34] . Second , we show that many of these same stressors trigger the activation of another MAPK , Hog1 . When Hog1 is activated , Fus3 signaling is inactivated . Third , we present the results of a comprehensive screen for small molecule metabolite second messengers , and show that 2-hydroxy BCAA derivatives are generated in response to osmotic stress and Hog1 activation . We show further that these metabolites are sufficient to trigger Gpa1 phosphorylation and a dampening of the Fus3 pathway . Finally , we show that the protein kinase Elm1 is required for phosphorylation of Gpa1 in response to osmotic stress and by addition of the metabolites . These processes are clearly distinct from those reported previously for glucose stress , which leads to a decrease in cellular pH , direct binding of protons to the Gα subunit , and direct phosphorylation of Gα by Sak1 . While the target of the BCAA metabolites remains to be identified , we have largely excluded the kinase and Gα subunit substrate as candidates . Based on our findings , we propose that BCAA metabolites represent a newly described “second messenger” of stress-activated signaling . The concept of second messenger signaling stems from the work of Earl Sutherland in 1957 [56] when he discovered that the activity of liver phosphorylase is stimulated indirectly by hormones , requiring a “heat-stable factor” that was later identified as cAMP [57] . That work established a paradigm of cell signaling whereby a first messenger ( e . g . , hormone or neurotransmitter ) activates a receptor on the cell surface ( canonically a GPCR ) and activation of an intracellular effector protein that produces the second messenger molecule . This process serves to greatly amplify the intracellular response since activation of one receptor can lead to the production of multiple second messenger molecules . Since the discovery of cAMP , several other second messengers have been identified , including cGMP [58] , inositol trisphosphate [59 , 60] , diacylglycerol [61] , and calcium [62 , 63] . Each of these molecules was painstakingly identified through rudimentary biochemical methods . With advances in metabolomics technologies however , we now have the ability to examine a broad complement of small molecules in a single experiment . In yeast , BCAAs are catabolized through the Ehrlich pathway . The end products of this pathway are fusel alcohols or fusel acids [50] . Much like the catabolism of BCAAs by the Ehrlich pathway in yeast , BCAAs in mammals are metabolized to 2-keto acids by the branched-chain amino acid transaminases . The 2-keto acids primarily undergo oxidative decarboxylation by branched-chain keto acid dehydrogenase to yield substrates for further oxidation and generation of anaplerotic compounds for the TCA cycle [64] . However , the molecules characterized here appear to have undergone an alternative route , wherein 2-keto acids are reduced to form 2-hydroxy acids . Excess levels of 2-hydroxy acids are found in human patients with maple syrup urine disease , also known as branched-chain ketoaciduria . This is an autosomal recessive disorder caused by a deficiency in dehydrogenase activity . Without this enzyme , 2-keto acids accumulate and are shunted towards formation of 2-hydroxy acids [65 , 66] . Accumulation of 2-keto and 2-hydroxy acids often results in brain damage due to impaired neurotransmitter function caused by inhibition of glutamate uptake [67 , 68] , and neuronal energy metabolism dysfunction [69 , 70] . Although 2-hydroxy acids are produced , the accumulation of BCAAs and 2-keto acids seems to have the greater impact on the pathophysiology of maple syrup urine disease [71] . Previously we showed that osmotic stress dampens and delays the mating pheromone response in yeast [33] . Here we describe potential mechanisms of this cross-pathway regulation . While our analysis focused on yeast , several tissues routinely experience osmotic stress , and can develop disease if osmoregulation is impaired . For example , osmotic stress can promote dry eye disease [72] and diabetic retinopathy [73] . High osmolarity in the vasculature can lead to hypertension [74] and a hyperosmolar hyperglycemic state in diabetics [75] . Importantly , BCAA metabolism is also conserved in humans [76] . Reduced levels of the BCAAs are observed in heart failure , sepsis , trauma , and burn injury [77] . Moreover , a reduction in the expression of branched-chain amino acid transaminase and keto acid dehydrogenase , as well as an increase in the levels of 2-keto acids , have been identified as hallmarks of heart failure [78] . However the connection between osmotic stress signaling and BCAA metabolism is not clearly understood . Collectively , these examples highlight the need for a more complete understanding of the osmotic stress response and of BCAA metabolism . In summary , we identified 2-hydroxy BCAA derivatives as candidate second messengers of the osmotic stress pathway . As second messengers , these molecules are likely used to amplify the osmotic stress response and coordinate responses to hormones and neurotransmitters . A challenge for the future is to determine the mechanism by which Hog1 ( or p38 ) promotes BCAA derivative accumulation , their cellular target ( s ) in both yeast and humans , as well as their potential as lead molecules for pharmacological control of the stress response in a mammalian system .
All strains ( S3 Table ) were generated from the BY4741 wild type strain ( MATa his3Δ1 leu2Δ0 met15Δ0 ura3Δ0 ) [79] . Gene deletion strains were generated by homologous recombination of PCR-amplified drug resistance genes from the pFA6a-KanMX6 [80] or pFA6a-hphMX6 plasmids [81] , with flanking sequence homologous to the gene of interest [82] , or by the delitto perfetto method , leaving no selection marker [83] . Similarly , Flag-tagged strains were generated by homologous recombination of the PCR-amplified cassette from pFA6a-6xGly-3xFlag-HIS3MX6 [84] , with flanking sequence homologous to either side of the stop codon of the gene of interest . Bat15A Bat23A non-phosphorylatable mutants were generated using the delitto perfetto method . BAT1 was replaced with the counter selectable marker and reporter gene cassette and then with synthesized bat15A . The same steps were then used to replace BAT2 with bat23A . All cells were grown at 30°C unless otherwise noted . The pRS426-PFUS1-YeGFP3 plasmid was generated by subcloning the YeGFP3 gene [85] under control of the yeast FUS1 promoter from pDS30 ( from Daria Siekhaus , University of California , Berkeley ) [86] into pRS426 [87] , by digestion with BamHI and XhoI , and ligation of gel-purified products . pYEplac181-pHluorin ( 2μ , ampR , LEU2+ ) was the gift of Rajini Rao ( Johns Hopkins University ) ( S4 Table ) [34 , 45 , 46] . Cells were grown to saturation overnight in SCD medium , diluted to OD600 = 0 . 10 , grown to OD600 ~0 . 6–0 . 8 , then diluted again and grown to OD600 ~1 . 0 . One third volume of SCD containing 3x stress stimulus was added to the experimental cell cultures . Control samples were mixed with 1/3 volume of SCD alone . Aliquots were collected at the times indicated , mixed 19:1 with 6 . 1 N trichloroacetic acid ( TCA ) and placed on ice . Cell pellets were collected by centrifugation at 1962 x g for 2 minutes , and resuspended in 10 mM NaN3 . Cells were collected by centrifugation at 16 , 060 x g for 1 minute , the supernatant was removed , and cell pellets were stored at -80°C until use . Heat shock experiments were carried out by growing cells as indicated above , then transferring the cultures to a 42°C water bath incubator/shaker and adding 1/3 final volume of SCD medium pre-warmed to 42°C . Control samples were mixed with 1/3 volume of media at 30°C . For glucose limitation experiments , wild-type cells were grown as above to OD600 ~0 . 8 , collected by centrifugation at 1962 x g for 2 minutes , resuspended with one-quarter volume of glucose-free SC medium , centrifuged again and resuspended in the original volume of SCD medium containing either 2% or 0 . 05% glucose . Note that the centrifugation step leads to partial and transient Gpa1 phosphorylation ( Fig 1A , 2% Glucose curve ) . Cell pellets were thawed on ice , and resuspended in ice cold TCA buffer ( 10 mM Tris-HCl , pH 8 . 0 , 10% TCA , 25 mM ammonium acetate , 1 mM ethylenediaminetetraacetic acid ) . Cells were vortexed for 10 minutes , then collected by centrifugation at 16 , 060 x g for 10 minutes at 4°C . Pellets were reconstituted in resuspension buffer ( 100 mM Tris-HCl , pH 11 . 0 , 3% sodium dodecyl sulfate ( SDS ) ) , heated at 99°C for 10 minutes , cooled to room temperature for 10 minutes , and centrifuged at 16 , 060 x g for 1 minute . Lysates were transferred to new tubes and 5 μL was used in a Bio-Rad DC Protein Assay ( Bio-Rad #5000112 ) , carried out according to the manufacturer’s protocol , and compared against a bovine serum albumin standard curve . Lysates were normalized to 2 μg/μL with resuspension buffer and 6x SDS sample buffer ( 350 mM Tris-HCl , pH 6 . 8 , 30% ( v/v ) glycerol , 10% ( w/v ) SDS , 600 mM dithiothreitol , 0 . 012% ( w/v ) bromophenol blue ) , and used immediately or stored at -80°C . Cell lysates were heated at 99°C for 10 minutes , then 40 μg of protein was loaded onto 10% SDS-PAGE gels . Gels were then run in SDS electrophoresis buffer ( 25 mM Tris base , 20 mM glycine , 0 . 1% ( w/v ) SDS ) at room temperature for 20 minutes at 20 mA/gel after which , current was increased to 25 mA/gel for 110 minutes . Electrophoresed proteins were then transferred to nitrocellulose membranes at 100 V for 90 minutes at 4°C in transfer buffer ( 20% methanol , 25 mM Tris base , 200 mM glycine ) . Membranes were blocked in TBS-T ( 100 mM Tris Base , pH 7 . 5 , 150 mM NaCl , 0 . 1% Tween-20 ) containing 5% ( w/v ) milk and 10 mM NaN3 for 1 hour unless otherwise indicated . Western blots were probed with antibodies raised against Gpa1 ( in-house rabbit polyclonal antibody , 1:1 , 000 ratio ) [88] , phospho-Snf1 ( phospho-AMPKα ( Thr172 ) 40H9 Rabbit mAb , Cell Signaling Technology #2353 , 1:2 , 000 ratio ) , Snf1 ( poly histidine HIS-1 mouse mAb , Sigma-Aldrich #H1029 , 1:3 , 000 ratio ) , Hog1 ( Santa Cruz Biotechnology #sc-6815 , 1:500 ratio ) , phospho-Hog1 ( phospho-p38 MAPK ( Thr180/Tyr182 ) 28B10 Mouse mAb , Cell Signaling Technology #9216 , 1:500 ratio ) , and Glucose-6-phosphate dehydrogenase as a loading control ( G6PDH , Sigma # A9521 , 1:50 , 000 ratio ) . Blots were incubated with primary antibodies for 1 hour to overnight , washed 3 x 5 minutes with TBS-T , then incubated with horseradish peroxidase-conjugated secondary antibodies raised against rabbit ( Bio-Rad #1662408 ) , mouse ( Bio-Rad #1721011 ) , or goat ( Santa Cruz Biotechnology #sc-2768 ) at a 1:10 , 000 ratio in TBS-T containing 5% ( w/v ) milk , and washed 3 x 5 minutes with TBS-T . Blots were imaged on a Bio-Rad ChemiDoc MP imaging system after a 5 minute incubation with Clarity ECL Western Blotting Substrate ( Bio-Rad #1705061 ) . Wild type yeast were transformed with plasmid pYEplac181-pHluorin [34 , 45 , 46] and grown in SCD-Leu medium . For cells treated with BCAA derivatives ( 30 mM ) the medium was titrated to pH 5 . 0 with HCl . Experiments and pHi calculations were carried out as in [34] using the indicated stressor or metabolite at 3x stock concentration . Wild type and hog1Δ cells were grown to saturation overnight , diluted to OD600 = 0 . 10 grown to OD600 ~0 . 6 , diluted again to OD600 = 0 . 00075 , incubated overnight to OD600 ~0 . 9 . Cultures were then split in half and grown to OD600 ~1 . 0 and mixed 1:4 with SCD or SCD plus 2 . 5 M KCl . After 3 minutes the cultures were transferred to 250 mL conical bottles ( Corning #430776 ) and centrifuged for 3 minutes at 1819 . 3 x g in a Sorvall RC3C Plus centrifuge using an H6000A swinging bucket rotor . After aspirating the supernatant the cell pellets were snap-frozen in place with liquid nitrogen and stored at -80°C . The cells were exposed to KCl for a total of 20 minutes . Frozen pellets were submitted to Metabolon , Inc . for GC-MS and LC-MS/MS analysis of metabolites ( see S1 Methods ) . For NMR measurements , 15N-enriched Gαi-Δ31 produced as in [89] was exchanged into NMR buffer ( 20 mM sodium phosphate , pH 7 . 0 , 50 mM NaCl , 2 mM MgCl2 , 200 μM GDP , 5% D2O ) . Each NMR sample contained 50 μM Gαi-Δ31 and 1 . 25 mM ligand . NMR spectra were acquired at 25°C on a Bruker Avance 850 NMR spectrometer . Two-dimensional 1H–15N HSQC experiments were recorded with 1024 and 128 complex points in the direct and indirect dimensions , respectively , 44 scans per increment and a recovery delay of 1 . 0 seconds . Spectral widths used were 13586 . 957 Hz ( 1H ) and 3015 . 682 ( 15N ) Hz . Spectra were processed and analyzed using NMRPipe ( NIDDK , NIH ) and Sparky ( UCSF ) . Four colonies of the same strain transformed with plasmid pRS426-PFUS1-YeGFP3 and one colony of the untransformed background strain ( to use for background fluorescence subtraction ) were grown to OD600 ~1 . 0 . Samples were added in duplicate to black clear-bottomed 96-well plates containing 10x stocks of serially diluted α-factor mating pheromone ranging in concentration from 1x10-4 . 5 M to 1x10-10 M prepared in sterile water , and 5x stocks of stimulus solution prepared in growth medium . The OD600 for each well was measured for cell number normalization . After 3 hours , GFP fluorescence was measured at an excitation wavelength of 485 nm , and emission wavelength of 538 nm , using a cutoff of 530 nm , in a Molecular Devices Spectramax M5 plate reader . For data presentation , raw fluorescence values from each well were normalized to the number of cells in that well ( represented by the OD600 ) using the shorthand Taylor Series 11+x where x = OD600 . Normalized values of each technical duplicate were averaged , and normalized values from the background strain ( containing no fluorescence reporter ) were subtracted . Finally , each well was normalized as a percent to the average maximum fluorescence value in the α-factor treated positive control . Dose-response curves were fitted using a nonlinear Boltzmann function . All data are reported as mean ± the standard deviation . Statistical significance was determined by an unpaired two-sided student’s t-test . In all cases , a p-value ≤ 0 . 05 was considered to be statistically significant . | Just as organisms must adapt to a challenging environment , cells must respond to chemical or physical changes that occur within the organism . Regardless of the environmental cue , all cells use molecular signaling pathways to respond to those changes . Many are transmitted by G protein-coupled receptors ( GPCRs ) or the high osmolarity glycerol ( HOG ) pathway . While these pathways have been studied independently for decades , less is known about how they coordinate with each other to carry out the proper response , particularly when conflicting signals are present . One way coordination can be achieved is through “second messenger” molecules produced by one pathway to regulate another pathway . Here , we identify candidate second messengers of osmotic stress by global metabolite profiling analysis of the yeast S . cerevisiae . We find that three branched-chain amino acid ( BCAA ) metabolites increase in response to osmotic stress and require the stress response MAPK Hog1 . We show that these BCAA derivatives are necessary and sufficient to recapitulate the effects of osmotic stress on the GPCR pathway . Our results identify a new way that HOG and GPCR pathways communicate , and may in the future guide better treatment strategies for stress-related cell damage . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"phosphorylation",
"cellular",
"stress",
"responses",
"second",
"messenger",
"system",
"mechanisms",
"of",
"signal",
"transduction",
"cell",
"processes",
"g-protein",
"signaling",
"mapk",
"signaling",
"cascades",
"protein",
"kinase",
"signaling",
"cascade",
"stress",
"signaling",
"cascade",
"proteins",
"biochemistry",
"signal",
"transduction",
"cell",
"biology",
"post-translational",
"modification",
"biology",
"and",
"life",
"sciences",
"osmotic",
"shock",
"cell",
"signaling",
"signaling",
"cascades"
] | 2017 | Amino acid metabolites that regulate G protein signaling during osmotic stress |
Allostery is a fundamental process by which ligand binding to a protein alters its activity at a distinct site . There is growing evidence that allosteric cooperativity can be communicated by modulation of protein dynamics without conformational change . The mechanisms , however , for communicating dynamic fluctuations between sites are debated . We provide a foundational theory for how allostery can occur as a function of low-frequency dynamics without a change in structure . We have generated coarse-grained models that describe the protein backbone motions of the CRP/FNR family transcription factors , CAP of Escherichia coli and GlxR of Corynebacterium glutamicum . The latter we demonstrate as a new exemplar for allostery without conformation change . We observe that binding the first molecule of cAMP ligand is correlated with modulation of the global normal modes and negative cooperativity for binding the second cAMP ligand without a change in mean structure . The theory makes key experimental predictions that are tested through an analysis of variant proteins by structural biology and isothermal calorimetry . Quantifying allostery as a free energy landscape revealed a protein “design space” that identified the inter- and intramolecular regulatory parameters that frame CRP/FNR family allostery . Furthermore , through analyzing CAP variants from diverse species , we demonstrate an evolutionary selection pressure to conserve residues crucial for allosteric control . This finding provides a link between the position of CRP/FNR transcription factors within the allosteric free energy landscapes and evolutionary selection pressures . Our study therefore reveals significant features of the mechanistic basis for allostery . Changes in low-frequency dynamics correlate with allosteric effects on ligand binding without the requirement for a defined spatial pathway . In addition to evolving suitable three-dimensional structures , CRP/FNR family transcription factors have been selected to occupy a dynamic space that fine-tunes biological activity and thus establishes the means to engineer allosteric mechanisms driven by low-frequency dynamics .
Small regulatory molecules frequently bind proteins at regions remote from the active site . These allosteric events can switch proteins between inactive and active states [1] . Knowledge of the molecular basis of allostery derives from a wealth of theoretical and experimental studies and traditionally describes the process in terms of conformational change within the protein [2] , [3] . Combinations of X-ray crystallography and NMR have permitted analysis of the ligand binding sites , intermolecular interactions , and conformational fluctuations that underpin diverse allosteric systems [4] , [5] . There is also considerable evidence that allosteric cooperativity can be communicated between distant sites on proteins through modulation of their dynamic properties , even in cases where that are no structural changes between the ligand bound ( holo ) and unbound ( apo ) forms [6]–[12] . Since the original identification , by Cooper and Dryden [4] , of this alternative route of “allostery without conformational change , ” there has been considerable debate over the mechanisms by which dynamic fluctuations are communicated between allosterically coupled sites on proteins . One hypothesis for fluctuation-induced allostery is that binding modifies the structure of the thermally excited global normal modes and thence the coupling interaction between cooperative elements . This in turn affects the structural ensemble of the distant sites and so the free energy of binding [13]–[15] . Another view maintains that physically connected pathways of excited or repressed dynamics , coupled along their trajectories , connect allosteric sites [16]–[18] . Here we propose the hypothesis that the normal modes of protein structural motion , large-scale motions dispersed across the entire protein , are important carriers of the allosteric signal and act without requiring structural change . Previous studies of the normal modes have demonstrated that conformational transitions in proteins , including those that underpin allosteric regulation dependent on conformational change , are well described by one or a few low-frequency modes [19]–[25] . The normal modes , however , can also be used to describe the whole spectrum of internal fluctuations of a protein around a mean structure . The low-frequency global modes , in particular , can involve entire protein domains . Alteration of the normal modes might therefore be communicated to distant sites of a protein as a change in the degree of motion around a mean structure without overall conformational change . Global low-frequency fluctuation therefore represents an alternative theoretical approach to allosteric communication that does not depend upon conformational change . An important consequence of this alternative mechanism of allosteric communication is that it can be captured by coarse-grained representations and models , such as the elastic network model ( ENM ) . Here we develop this theory , and the validity of a coarse-grained model approach , through a computational and experimental study of the homodimeric CRP/FNR family transcription factors Catabolite Activator Protein ( CAP ) of Escherichia coli and GlxR of Corynebacterium glutamicum . CAP is a 210-amino-acid transcription factor that binds cAMP generated by adenylyl cyclase in response to the phosphorylated form of Enzyme IIAGlc ( phosphorylated in response to the phosphoenolpyruvate-carbohydrate phosphotransferase system ) [26] , [27] . cAMP-bound CAP regulates the transcription of over 100 genes crucial for carbon utilization through its binding to a specific promoter region and recruitment of RNA polymerase [28] . Previous studies of the ligand binding domain of CAP demonstrated negative cooperativity between cAMP binding sites in the absence of structural change within this domain [10] . The observed negative cooperativity in this isolated domain occurs through a conformational entropic penalty for binding the second molecule of cAMP , but there is no mechanistic description for how such a phenomenon can occur in the full-length protein . Seven of eight CAP mutants previously examined showed a direct correlation between ΔΔG and the adiabatic compressibility ( βs° ) where proteins with a higher βs° ( reflecting increased structural flexibility in solution ) demonstrated enhanced negative cooperativity [29] . While it is therefore reasonable to hypothesize a role for protein dynamics in allostery in CAP , there is no conceptual framework to describe how these changes in motion might arise , how they contribute to allostery , and how a resulting theory might translate to related molecules . CAP is therefore a suitable model system for a theoretical and experimental investigation of the contribution of the normal modes to allostery . Here we propose that changes to global low-frequency protein backbone fluctuations are carriers of an allosteric signal in CAP and present this in the context of a significant new quantitative theory for allosteric coupling . We produce coarse-grained models that describe global low-frequency protein backbone motions of CAP and show a strong correlation between negative cooperativity for cAMP and modulation of the delocalised normal modes on ligand binding without a requirement for a spatially distinct physical pathway or conformational change . We demonstrate experimentally that altered connectivity between backbone elements in CAP can give predictable alterations to cooperativity for cAMP binding through altered mode amplitudes . We further demonstrate a broader applicability for this theory using an additional CRP/FNR family transcription factor , GlxR of C . glutamicum . We unite our findings for CAP and GlxR to determine the extent to which key inter- and intramolecular parameters contribute to negative cooperativity in CRP/FNR family transcription factors . We further demonstrate that amino acids that contribute significantly to allosteric control are more likely to be conserved in variant proteins from diverse species . The theoretical and experimental work and associated data analysis provide both a significant advance in our understanding of the mechanisms that underpin the dynamic regulation of allostery and also a means for informed rational engineering of cooperativity in proteins .
To computationally address cases of allostery that arise from fluctuation-modification , without conformational change , requires a very different approach from those corresponding to the classic Monod-Wyman-Changeaux case of conformational switching . On the one hand , fully atomistic simulations are not capable of attaining , in most cases , the long dynamical time scales explored by the slow , global dynamic modes whose thermodynamics are essential for the effect . On the other hand , because these modes by their nature integrate many local interactions into their effective geometries and potentials , coarser-grained models of protein structure can possibly provide sufficiently accurate calculations of the relevant dynamics , while allowing the computation of dynamics to the necessary timescales . Models that represent protein structures by Cα-atom positions alone reproduce low-frequency modes well in comparison to experimental data [21] , [30] . We therefore used the co-ordinates from a high-resolution crystal structure determination of the full-length cAMP bound CAP homodimer to construct an ENM [31] for the apoprotein as well as single and double ligand bound holoprotein states ( Figure S1 ) . Free energies , ΔG , were calculated using the full harmonic solution , and the negatively cooperative binding of cAMP to wild-type full-length CAP confirmed by calculating a positive value for ΔΔG = ( ΔGholo2−ΔGholo1 ) − ( ΔGholo1−ΔGapo ) = 179 cal mol−1 consistent with experimentally obtained values ( Table S2 ) [32]–[35] . To confirm that the total motion within the ENM is not an artefact of coarse-graining , we also carried out molecular dynamics simulations [36] with full atomistic detail , including an explicit water model , and performed principle component analysis ( PCA ) on the generated trajectories [37] . B factors represent the convolution of static and dynamic disorder in the crystal . Dynamic disorder can be attributed to local motions of individual atoms , whereas static disorder represents different atomic positions in the individual protein molecules . The experimental B factors , albeit constrained by crystal packing , therefore represent a reasonable approximation of the local motions in solution [38] . ENMs and atomistic PCAs represent overall unconstrained dynamic motions and hence show much larger deviations in the termini and the flexible loop regions ( for example , residues 150–175 of Figure S2 ) . The crystallographic B factor data were qualitatively well represented at either scale of coarse-graining ( Figure S2a ) and the distribution of the normal mode frequencies agreed well between ENM and PCA ( Figure S2b ) . The total predicted motion within the ENM , at least at the level of B factors and low-frequency mode structure , is therefore similar to other methods of analysis and not an arbitrary feature of the model . Since the fluctuation-induced allosteric effect arises from the low-frequency structure of the protein dynamics , the ENM level of analysis applies to the experimental phenomena studied here . We hypothesized that if side-chain replacement on amino acids at sites distinct from the cAMP binding site of CAP do not cause conformational rearrangement , yet increase or decrease amino acid side chain hydrophobic or electrostatic forces in their local environments , the normal modes of protein motion would be altered without significant structural changes . If these changes to the normal modes have sufficiently global effects , they will in turn modify cooperativity between the cAMP binding sites through an entropic contribution to the binding free energy . Amino acid side chain replacement can therefore act as a sensitive probe of the contribution of side chain connectivity to cooperativity and the underlying mechanism for allostery within the elastic structure of the protein . The change in allosteric free energy ( ΔΔG ) as a function of altering the entire primary amino acid sequence ( one residue at a time ) can therefore be viewed as a quantitative map of the contribution of the normal modes to cooperativity . Such a quantitative map can be constructed either by simulation or experiment; in practice , it is convenient , as we demonstrate below , to use simulation of the entire allosteric map to guide mutagenesis for experimental study . We therefore performed a scanning computational mutagenesis of the entire CAP protein to investigate the influence of side chain connectivity on cooperativity via their influence on the normal modes . Changing the effective elastic potential between protein backbone carbon atoms in the neighbourhood of each residue of the ENM in turn and calculating effects on ΔΔG was used to determine the scanning computational mutagenesis map . The increase and decrease in elastic potential in the ENM was hypothesized to simulate the effects of local strengthening and weakening of side chain interactions in CAP . A color-coded map corresponding to altered cooperativity with changing local interaction strength is plotted graphically by amino acid residue ( Figure 1a ) and in real space ( Figure 1b ) . The global map for the ENM ( Figure 1a ) demonstrates large regions where cooperativity is susceptible to control by altering side chain connectivity . It is important to note that these control regions are not necessarily adjacent to the cAMP-binding site . For example , regions corresponding to amino acids 127–137 ( at the interface between the two monomers ) and 150–162 ( within the DNA binding domain , far from both the dimer interface and cAMP binding regions ) appear to exercise considerable control over cooperativity without contributing to a spatially distinct dynamic pathway and without direct interference with the cAMP binding site . To experimentally test the model and demonstrate rational engineering and control of allostery , we selected the residues of CAP highlighted in Figure 1b . We examined amino acids predicted to show altered ( V132 , H160 ) or neutral ( V140 ) responses to altered amino acid side chain interactions ( Table 1 ) . The removal ( V132A ) or addition ( V132L ) of a side chain methyl group of V132 was engineered to decrease and increase , respectively , the strength of hydrophobic interaction across the dimer interface . Computation predicted that these mutations would result in more negative and positive cooperativity in CAP , respectively ( Figure 2a ) and that the most important contacts contributing to this effect were with L62 and V132 of the opposing monomer ( Figure S3b ) . High-resolution X-ray crystal structures of CAP mutants V132A and V132L demonstrated that these variants possessed decreased and increased hydrophobic interactions across the dimer interface , respectively ( Figure 2b ) . Comparison of variant crystal structures with wild-type demonstrated that there was no statistically significant change in structure ( Figure S4 , Table S1 ) . Cooperativity for cAMP binding was studied by isothermal titration calorimetry ( ITC ) for wild-type , V132A , and V132L proteins to examine whether the experimentally observed changes in cooperativity matched computational predictions ( Figure 2c–e , Table 1 ) . The ITC data were well-described by a three-site model , with two major and one minor cAMP binding site ( Figure S5 ) [39] and allowed derivation of the thermodynamic parameters for all proteins ( Table S2 ) . The qualitative computational prediction for altered cAMP cooperativity was tested experimentally including a significant controlled inversion of the sign of the cAMP cooperativity ( V132L ) . The thermodynamic parameters for wild-type CAP demonstrated an overall favourable entropy change and unfavourable enthalpy change on binding the second molecule of cAMP consistent with a previous report [39] . A previous study of the truncated CAP ligand-binding domain demonstrated that binding of the second molecule of cAMP was entropically unfavoured [10] . The difference in thermodynamics between our experiments ( Table S2 ) and previous experiments using the ligand-binding domain alone [10] is therefore likely due to the contribution of motions of the DNA binding domain [40] . This interpretation is supported by previous analysis that has calculated the thermodynamic contribution of the DNA binding domains in the switch to the active conformation [41] . Previous calculations and experiments anticipate that , while the contribution of the normal modes to allostery is entropically controlled ( in terms of the net allosteric free energy ) , coupling of the low-frequency modes to side-chain motion generically gives rise to additional , but compensating , contributions to enthalpy and entropy and this is observed in our thermodynamic data ( Table S2 ) [9] . It is notable that , due to this self-cancelling of the contribution of local fast modes within the total free energy , the entropically driven ENM is able to predict qualitative changes to experimental cooperativity despite the local mode contribution of enthalpy to overall thermodynamics . The ENM calculations predicted a reduction in the negative cooperativity of CAP in response to a reduction in the strength of the local interactions of residue H160 ( Figure 3a ) . In particular , H160 was predicted to form interactions that contribute to allostery with D162 and Q165 ( Figure S3a ) . The mutation H160L was predicted to break these interactions while maintaining side chain bulk; this was confirmed by X-ray crystallography of the H160L CAP protein ( Figure 3b ) . No overall change in H160L protein structure was evident compared to wild-type ( Figure S4 , Table S1 ) . ITC experiments ( Figure 3c ) demonstrated that cooperativity for cAMP became less negative as predicted by computation ( Table 1 ) . This crucial experiment demonstrates that altering low-frequency motions at a site distant from both the ligand binding site as well as the dimer interface , and from any presumed physical pathway of structural change connecting these sites , can nonetheless give predictable effects on cooperativity . Altering local interactions associated with V140 was predicted by the ENM to have minimal effects on cooperativity ( Figure 4a ) despite significant local hydrophobic interactions; we therefore examined the effect of decreased and increased local hydrophobic interactions in V140A and V140L variants as a control experiment . The V140L mutant protein had no discernible effect on protein structure ( Figure S4 ) . As predicted by the ENM mutagenesis , measurement of cooperativity for cAMP in V140L by ITC ( Figure 4c ) showed no differences when compared to wild-type ( Table 1 ) . Interestingly , although V140A protein showed no global change in structure ( Figure S4 ) , there is , in this mutation , a significant local conformational change evident in the crystal structure where the mutated V140A residue formed a new hydrophobic contact with the rotated side chain of C179 that is not present in the wild-type or V140L proteins ( Figure 4b ) . When included in the model , simulated as kC179/k = 4 , this new contact revealed new interactions within the monomer ( Figure S3a ) that drove CAP towards positive cooperativity on simulation ( Table 1 ) . ITC experiments ( Figure 4d ) demonstrated that this CAP variant with the identified side chain rearrangement was positively cooperative , thus supporting the qualitative prediction of the model . A bar graph for the calculated and observed values for K2/K1 revealed the agreement in the direction of the change of cooperativity on simulation and experiment ( Figure S6a ) . A plot of the experimentally observed value for K2/K1 against that predicted from the ENM demonstrated a correlated relationship where observed increases to K2/K1 are associated with similar changes to K2/K1 by the ENM ( Figure S6b ) . The consistency in prediction by the ENM and the quantitative correlation between predicted and observed changes do not support the notion that the agreement between experiment and the ENM is due to a chance occurrence . The ENM can provide further insight into the mechanism by which allosteric control is associated with alterations to the normal modes . No global structural changes were induced in the ENM simulations or were evident from crystal structures of variant proteins; only the pattern of coupled low-frequency fluctuations was modified by the simulated side-chain mutations . This appearance of “control at a distance” in the CAP homodimer is explained , through contributions to binding entropy , if there are correlations in the low-frequency motions between cAMP binding sites and if ligand binding or side chain mutation modifies this correlation [42] . As all fluctuating systems dominated by locally harmonic interactions possess a structure of normal modes , with just such distant correlations , they suggest the mechanism for allostery in CAP . To examine whether the mutations studied here can have such distant effects , we calculated the change to local Cα flexibility in the case of tightening and loosening side chain interactions at V132 at the dimer interface ( Figure 5a ) . Modifications to simulated backbone flexibility are present throughout CAP with varying amplitude and furthermore follow opposite signs at kV132/k = 0 . 25 ( V132A ) and kV132/k = 4 ( V132L ) . For example , kV132/k = 4 shows significant tightening of the protein ( compare Figure 5a and Figure S3b ) . An examination of the effect of simulated mutations at V140 and H160 on nonlocal Cα flexibility reinforces this finding ( Figure S7 ) . The predominantly neutral mutation , V140L , simulated as kV140/k = 4 has little effect on protein backbone flexibility , except at sites where V140 has calculated interactions , consistent with the absence of any effect on allostery on both simulation and experiment . In the case of H160 ( kH160/k = 0 . 25; at a surface loop distant from both the cAMP binding site and dimer interface ) and V140A ( kC179/k = 4 , kV140/k = 0 . 25 ) , the simulated mutations create a uniform decrease in flexibility throughout the monomer except for the straightforward loosening/tightening at the site of the mutations . There is a general trend , therefore , for those simulated mutations that decrease negative cooperativity to be associated with decreased protein backbone motion nonlocally . A specific requirement of global low-frequency motion as an underpinning mechanism for allostery at a distance is a coupling between protein motion and the behaviour of the cAMP-binding site . We find that the loosening and tightening effects of simulated mutations is correlated with significant modulation of backbone flexibility in the region of the cAMP-binding site ( amino acids 71–74 , 83–85 , and 121 ) ( Figure 5b ) . The figure shows that , in general , changes in root-mean-square deviation ( rmsd ) at the ligand-binding site induced by mutation correlate ( in this case , kR/k = 0 . 25 ) with cooperativity . Mutations that increase motion at the ligand bind site are associated with an increase in the extent of negative cooperativity and vice versa . This is entirely consistent with the controlling entropic allosteric mechanism in these cases , providing that cAMP binding has the effect of increasing local rigidity . This interaction between the heightened local motions following the first cAMP-binding event creates an entropic contribution to negative cooperativity in ΔΔG [9] . Heightened fluctuation at the second binding site ( on binding the first molecule of cAMP ) is a general mechanism for negative cooperativity without conformational change [6] . Positive cooperativity without conformational change can be induced by reducing the fluctuation amplitude ( for example , the MetJ transcription factor of E . coli [9] ) . Studies using CAP have successfully demonstrated that changes to global low-frequency protein dynamics are associated with allostery . We investigated another protein to explore the more general applicability of the mechanism . GlxR of C . glutamicum is a cAMP binding homodimeric transcription factor of the CRP/FNR family that activates genes required for aerobic respiration , glycolysis , and ATP synthesis [43] , [44] . We solved the X-ray crystal structure of the GlxR apoprotein to produce an ENM for the non-cAMP bound state [45] . Coordinates from an available crystal structure determination of the full-length cAMP bound GlxR homodimer allowed us to construct an ENM for the single and double ligand bound holoprotein states . Examination of the structures for GlxR in the apo and holo forms revealed no significant difference in structure . GlxR therefore represents a new exemplar for allostery in the absence of conformation change . Free energies , calculated from ENMs for GlxR , predicted considerably greater negative cooperative binding of cAMP ( K2/K1 = 2 . 37; ΔΔG = 513 cal mol−1 ) than for CAP ( K2/K1 = 1 . 35; ΔΔG = 179 cal mol−1 ) . This prediction of enhanced negative cooperativity was confirmed on experiment with an observed value for K2/K1 of 19 . 47 ( Table 2 ) . A computational scanning mutagenesis map was produced for GlxR , as done previously for CAP , and altered cooperativity with changing local interaction strength is plotted graphically by amino acid residue ( Figure 6a ) and in real space ( Figure 6b ) . Both local tightening and loosening across the dimer interface , depending on the residue , was predicted to reduce negative cooperativity and therefore provides a robust experimental test of the model . We generated dimer interface loosening ( kL134/k = 0 . 25; L134V; Figure 7a ) and tightening ( kA131/k = 4; A131V; Figure 7b ) GlxR variants and compared simulated and experimental values for cooperativity in these proteins . Both L134V and A131V showed a clear reduction in negative cooperativity , as predicted , when compared to wild-type ( Table 2 ) by ITC ( Figure 7c–e ) , despite the fact that the mutants have opposing effects on hydrophobic interactions across the dimer interface . Allostery is therefore correlated with global low-frequency dynamics in an additional CRP/FNR family transcription factor . Our findings indicate general biophysical principles that describe the emergence of negative cooperativity in CRP/FNR family transcription factors through the allosteric modulation of normal modes . The property that allosteric effects are carried in general by the more globally distributed , and so typically longer wavelength , normal modes motivated the exploration of the underlying physics by coarse-graining the CAP and GlxR representations even further into rotational-translational block representations [46] . Two coarse-grained blocks per monomer ( one is the entire DNA-binding region , coupled only to the other block of its own monomer ) emerged naturally from the many residue–residue couplings internal to and between monomers at the molecular level . Figure 8a and 8b display the block structure and the corresponding “super-coarse-grained” model . A single representative internal mode within each dynamically tight block and the coupling strengths between the blocks ( including coupling across the dimer interface ) were investigated as “design parameters” for a general class of cooperative homodimer . Figure 8c ( CAP ) and 8d ( GlxR ) show allosteric cooperativity , calculated at this high level of coarse-graining , as a function of the integrated coupling strengths within the ligand binding domain ( k1 ) and between monomers ( k12 ) . Points below and above the z = 0 plane correspond to positive and negative cooperativity , respectively . The wild-type proteins for both CAP and GlxR are offset from the maxima of anti-cooperative ridges in the two-dimensional free energy landscapes that emerge . At this position , loosening coupling internal to monomers ( k1 ) moves the system into a basin of less negative cooperativity ( GlxR ) or positive cooperativity ( CAP ) , while loosening in the coupling region ( k12 ) moves the system for both CAP and GlxR to the top of the ridge ( red ) to increase negative cooperativity . Further analysis demonstrated consistency in the negative cooperativity arising through the normal modes in the ENM and in the super-coarse-grained model . For example , the simulated loosening ( kV132/k = 0 . 25; V132A ) and tightening ( kV132/k = 4; V132L ) mutations of the CAP ENM and the tightening ( kA131/k = 4; A131V ) mutation of GlxR alter cooperativity through generating effective changes in k12 at the super-coarse-grained level . The super-coarse-grained model therefore effectively reveals the critical intra- and intermolecular parameters that associate with cooperativity and how these parameters can be altered to move within the allosteric free energy landscape . If cooperativity confers a selective advantage on the organism , then the allosteric free energy landscape can also be viewed as evolutionary landscape . In this case , the position of a protein within the landscape depends upon selection pressures that impact upon k1 and k12 . This general hypothesis can be used to make an additional significant experimental prediction . If the similar position of CAP and GlxR within their respective free energy landscapes is the result of a selection pressure , then we predict that amino acids that contribute significantly to quantitative allosteric control ( Figure 1a and 6d ) will be more invariant in related proteins from different species . We therefore examined 163 CAP variants from diverse bacterial species and plotted the frequency of mutation of each amino acid residue against the contribution of that amino acid to allostery ( defined as absolute change ( Δ ) in K2/K1 for that amino acid in the canonical CAP ENM at kR/k = 0 . 25 ) . We found evidence that the rate at which an amino acid mutates is negatively related to ΔK2/K1 ( LRT , G2 = 33 . 7 , p<0 . 001; Figure 9 ) . The coefficient quantifying this decrease , β1 , was significantly different from zero [95% CI = ( −3 . 34 , −1 . 49 ) ] . Amino acids of CAP that contribute to allostery through regulation of low-frequency protein dynamics are therefore more likely to be conserved in CAP variants through their contribution to protein function . Note that a test for overdispersion was significant , even after allostery had been accounted for ( LRT , G1 = 1 , 663 . 9 , p<0 . 001 ) , suggesting that other variables also have an influence on mutation rates .
Here we demonstrate that negative allostery in CRP/FNR family transcription factors is correlated with modulation of the normal modes of protein motion on ligand binding in the absence of conformational change . The model makes key predictions that we test at select sites of the CAP and GlxR proteins , the latter identified as an important new exemplar for allostery in the absence of conformation change . The alterations in protein flexibility that are a signature for allostery in CRP/FNR family transcription factors are a consequence of the global nature of those normal modes responsible and mutations that predictably alter cooperativity do so by influencing protein backbone flexibility . Our theory describes how allostery can arise from changes to low-frequency dynamics in the absence of any mean structural change . The theory is particularly significant as it describes allostery as a natural consequence of the dynamic properties of a protein without a requirement for spatially localised dynamic pathways between allosteric sites . The allostery observed is unlikely to have microheterogeneity as an alternative explanation as all CAP proteins crystallised as a single superimposable structure . Any form of heterogeneity reduces the likelihood of forming ordered crystals [47] . Microheterogeneity is therefore not supported as a molecular cause for allostery in CAP . The possibility of a direct interaction between cAMP binding sites might also be considered as a mechanism to explain the allostery observed . The closest distance between the two cAMP molecules in the CAP dimer is 9 . 5 Å ( the distance between the N6 atoms of the adenine ring ) . Although it is impossible to conclusively eliminate small local changes that binding of the first molecule of cAMP has at the second site , no conformational changes have been reported in this region in previous NMR studies , making this explanation unlikely . The possibility of a direct interaction is made even more unlikely as , similar as to that described above , any invoked direct interaction between cAMP binding sites would have to consistently match not only the qualitative aspects of the computational predictions for the role of the global modes , but also their quantitative correlation with the observed experimental values . Analysis of the relationship between Cartesian distance and protein motions demonstrated strongly correlated motions between allosteric sites at distances of <10–20 Å [48] and the global normal modes are a suitable candidate to mediate such correlations in CRP/FNR family transcription factors . The range of available sites for side chain mutagenesis of CRP/FNR family transcription factors do not constitute as large a set of separate and independent control parameters as at first seems , but in a good approximation explore a lower dimensional space ( i . e . , reducing the very high dimensional parameter-space of the entire number of residues , just one slice of which is represented in Figures 1a and 6d , to the two-dimensional parameter spaces of Figure 8c–d ) . We hypothesize that this two-dimensional parameter space is , in turn , related to an evolutionary landscape for a protein . In the case of CAP and GlxR , our analysis reveals that evolutionary selection has resulted in the location of the proteins in a region close to maximizing negative cooperativity . The extent of negative cooperativity in CAP is generally small ( ΔΔG = 0 . 3 kcal mol−1 ) . However , the scale of biologically relevant cooperative effects is set by the thermal energy ( RT≈0 . 6 kcal mol−1 ) . The values of ΔΔG observed and manipulated experimentally are those that modulate the concentration range of cAMP to which the system is sensitive by an order of 1 . Engineering of cooperativity is therefore possible by manipulating ΔΔG , as described here , with the caveat that it is likely only possible over a thermodynamic range to which the protein is responsive . We find that there is a selection pressure against mutation of residues that contribute to allostery in CAP variants . A significant question that arises , therefore , is that of the selective advantage provided through negative cooperativity in CAP . In general , the advantages conferred by negative cooperativity in biological systems are not well resolved [49] . It is proposed that negative cooperativity reduces the sensitivity of a system and extends the concentration range over which a response can be observed [50] . In metabolism , recent modelling suggests that there is a significant overall advantage for metabolic pathway flux with components showing negative cooperativity [51] , [52] . In transcriptional regulation , negative cooperativity in the binding of D-camphor to the CamR repressor of Pseudomonas putida is proposed to enable coupling of high specificity for D-camphor with a physiological response to high concentrations of the metabolite [53] . Against this framework , it is reasonable to conjecture that negative cooperativity in CAP offers a selective advantage by increasing the concentration range over which a transcriptional response can be generated [54] . The decreased sensitivity of the response to cAMP in negative cooperativity might result in a selective advantage through resource conservation when compared to amplifying effect of a signal response in positive cooperativity [50] . The position within the effective parameter space can also allow CAP variants to further tune cooperativity in either direction without a potentially disastrous influence on protein structure and therefore function . Future experiments to experimentally validate the selective advantage provided by negative cooperativity will therefore be crucial and might typically combine high throughput sequencing of extensive mutational libraries of CAP , after selection in E . coli , with the simulated mutational map of this study [55] . The super-coarse-graining and finer-grained tools we have developed and tested in this work suggest a route to artificial protein design through modification of protein low-frequency fluctuations without compromise of structure . The mechanism also reflects an important balance between phenomena at different length scales within molecular biology . The role of the global normal modes in conveying allosteric signals requires a similarly coarse-grained picture of the protein to identify and discuss the mechanism . On the other hand , the exquisite specificity to local biochemistry is preserved in the mechanism; a set of single residues , themselves spatially distant from either binding site , exercise significant control on the size ( and sign ) of the underlying allosteric signal . The delicate interactions of effects at different length scales are missed without such a multiscale approach to the physics of protein dynamics . Changes to the normal modes are presented as an important new theory to describe how allostery can arise in the absence of structural change and provide an important theoretical context within which to frame global issues of allostery in proteins .
The open reading frame corresponding to the full-length CAP protein was cloned into the BamHI and HindIII sites of pQE30 and mutant variants constructed by site-directed mutagenesis . Wild-type and mutant recombinant protein was expressed from E . coli M182 ΔCAP F− Δ ( lacIPOZY ) X74 galE15 galK16 rpsL thi+ lambda− [pREP4] for 2 h at 37°C with 1 mM IPTG . Protein was purified using sequential nickel-chelated sepharose affinity and Superdex 75 16/60 size exclusion columns ( GE Healthcare ) . Protein concentration was calculated using the Beer-Lambert Law and a molar extinction coefficient of 20 , 065 M−1 cm−1 at 280 nm . Full-length GlxR protein was expressed and purified as previously described [56] . Protein was dialyzed against 100 mM KPO4 pH 7 . 8 , 200 mM KCl , 2 mM 1-thioglycerol at 4°C . Protein and buffer were degassed under vacuum and degassed buffer used to dilute cAMP ligand . cAMP concentration was calculated using the Beer-Lambert Law and a molar extinction coefficient of 14 , 650 M−1 cm−1 at 259 nm . Data were generated using an iTC200 ( MicroCal ) by typically 40 sequential 1 µL injections of 4–6 mM cAMP into 202 µL 130–400 µM protein . Data for the first injection was routinely discarded as this is affected by diffusion between the syringe and the protein solution during equilibration prior to data collection . Ligand binding for cAMP to CAP was described by a sequential three-site model ( two major and one minor binding site [39] ) . The presence of three cAMP binding sites in CAP was further confirmed in the crystal structures from this study ( Figure S5 ) . A sequential two-site model described ligand binding for cAMP to GlxR . The free ligand concentration , [L] , was calculated for each injection using the bisection method , which allowed calculation of the fraction of the protein in each bound state , Fi:Comparing the calculated heat content , Q , to the experimental value allowed calculation of the best fit of the binding constants , Ki , and the binding enthalpies , ΔHi , using the solver plug-in for Excel: ITC and ENM data for mutant proteins was compared to the wild-type by a comparison of means by one-way ANOVA . Normal distribution of the data was confirmed by the Shapiro-Wilk test . Homogeneity of variances was rejected for ITC data and confirmed for ENM data using the Levene test . ITC data were therefore examined using a Dunnett's T3 post hoc test for pairwise comparisons with unequal variances and ENM data examined using a two-sided Dunnett's post hoc test for pairwise comparisons with equal variances . CAP crystals were produced at pH 6 . 5 with 7–10% ( w/v ) polyethylene glycol 3350 and 15–20% ( v/v ) 2-methyl-2 , 4-pentanediol with 2 mM cAMP in 24-well hanging-drop vapour diffusion plates . Crystals were cryoprotected using mother liquor containing 30% ( v/v ) glycerol and flash cooled in liquid nitrogen [57] . Diffraction data for the wild-type protein were collected in-house using a Bruker MicroStar rotating anode and processed with SAINT [58] . All CAP mutant data were collected at the Diamond Light Source beams I-04 and I-24 and processed using Mosflm [59] and Scala [60] . CAP structures were solved using molecular replacement with Phaser [61] using CAP ( PDB 1I5Z ) . Model building and refinement were accomplished iteratively using COOT [62] and Refmac5 [63] in CCP4 [59] . CAP structures from crystals produced at pH 6 . 5 were indistinguishable from those previously produced at pH 7 . 5 [64] . Structural and refinement statistics are provided in Table S4 . Full details of GlxR crystallography and analysis of the structures will be reported elsewhere [45] . Members of the CAP family often crystallise with more than one protein chain in the asymmetric unit . In these cases the functional protein dimer is either generated by the crystallographic 2-fold axis on each of the protein chains or by noncrystallographic symmetry leading to a varying degree of asymmetry [65] , [66] . Significantly different conformations for each monomer have been observed in some homodimeric bacterial regulator proteins , most notably Mt-CRP [67] . The structures presented here contain one dimer ( wild-type CAP in space group P21 ) , two dimers ( wild-type in space group P1 ) , and three dimers ( V140A CAP in space group I2 ) ( see Table S4 ) . In all cases the dimers are symmetric with no significant differences between the two protein chains than for the functional dimer . ENM simulations were performed using our own code based on the regular implementation [31] , [68] . The spring constants were set to a constant value of 1 kcal mol−1 Å−2 with a cutoff radius of 8 Å , and only the Cα atoms in the protein were considered . The presence of cAMP effector at the binding site was treated by the addition of one node at the mass weighted average coordinate for each ligand . Varying the spring constant of any springs attached to a single residue of the protein was used to represent side chain mutations . The allosteric free energy was calculated by summing over modes 1 to n . n was determined by examining where values K2/K1 converged ( Figure S8 ) . The final results quoted used the converged value of K2/K1 . PDB files for constructing CAP ENMs were 1CGP , 1G6N , 1HW5 , 1I5Z , 1I6X , 1J59 , 1O3T , 1RUN , 1RUO , 1ZRC , 1ZRD , 1ZRF , 2GZW , 4HZF ( this work ) , and an additional in-house file isostructural to 2GZW . The PDB file for constructing the GlxR ENM was 3R6S . The CAP and GlxR proteins were modelled as two blocks for each monomer , one for the ligand binding domain and one for the DNA binding domain . We assigned one internal breathing mode to each subunit and allowed each subunit to move , producing seven degrees of freedom . For the apo-protein the internal subunit coupling strengths are characterized by k1 though k4 and the intersubunit couplings by k12 , k13 , and k24 ( Figure 4b ) . The effect of one ligand binding was included by modifying k1 by a factor β , k12 by α , and k12 by γ . The second ligand binding was therefore represented by further modifying k2 by β , k12 by a further factor of α , and k24 by γ . The allosteric free energy was determined from the determinant of the interaction matrix [69] . The couplings were defined from PCA analysis of 300 ns molecular dynamics simulations for the three states . In each case the protein was divided into the four zones by performing a rotational-translational-block approximation ( Figure 8a ) [46] , [70] . Examination of the couplings calculated for each of the three states allowed calculation of the apo values and the ligand binding factors . Varying the values of k1 , k2 , and k12 represents mutations in residues affecting the intra- and the interblock interactions . Wild-type values for CAP are: k1 = k2 = 13 . 70 , k12 = 27 . 08 , k3 = k4 = 3 . 98 , k13 = k24 = 5 . 19 kcal mol−1 Å−2 , α = 1 . 30 , β = 0 . 560 , and γ = 0 . 901 . Wild-type values for GlxR are: k1 = k2 = 12 . 85 , k12 = 24 . 67 , k3 = k4 = 3 . 98 , k13 = k24 = 4 . 21 kcal mol−1 Å−2 , α = 1 . 40 , β = 0 . 71 , and γ = 0 . 99 . Molecular dynamics ( MD ) simulations employed the harmonic force field equations used in the ff99SB and GAFF force fields within the AMBER simulation program [71] . The simulations employed the ff99SB force field for the CAP protein and the GAFF force field ( v . 1 . 4 ) for cAMP . ff99SB force field is used as the energetic interactions of side chains , which are reasonably represented by this force field [72] , and outperforms the ff03 force field [73] . MD calculations used a short-range cutoff of 10 Å , with the long-range portion of the Coulomb potential represented by an Ewald summation , and employed a time step of 2 fs . The bond lengths were constrained by the SHAKE algorithm . The initial starting structures were obtained directly from X-ray diffraction . These structures were then solvated in TIP3P water and energy minimized prior to simulation [74] . The system was heated to 300 K over a period of 20 ps and further equilibrated for 40 ns . Production runs at 300 K were carried out over 200 ns . PCA was performed by diagonalising the mass weighted covariance matrix of the atomistic simulations . The eigenvectors represent the shape of the atomistic motion and the corresponding eigenvalues the extent of the motion . To determine if ΔK2/K1 , hereon denoted x , is associated with the mutation rate of amino acids , we first estimated the relative amino acid mutation rate using the sequence data for CAP variants and we then statistically tested for an effect of x on this rate . Relative mutation rate was estimated by finding the minimum number of amino acid mutations needed to generate the observed variations in the sequence data , which we denote N . For each of the 165 proteins we found the protein having the smallest number of amino acid differences . The sum of these differences gave N . When summing differences , if more than one protein had the minimum difference , we included all the proteins having the minimum . We then determined the number of these mutations that were associated with each of the 210 amino acids , which we denote ni . Thus , ni estimates the relative mutation rate of amino acid i , and these estimates account for the evolutionary history of the proteins . If all amino acids had an equal mutation rate , then we would expect the ni to all be approximated by N/210 . We assumed that the true relative rate of mutation was related to x according to the logistic function: μ ( x ) = , where β0 , β1 , and β2 are constants . To account for overdispersion among the ni , which might be due to unmeasured covariates associated with the proteins , we assumed that the variation between the ni could be described by the beta-binomial distribution . Under these assumptions , the log-likelihood of the model described by the set of parameters θ = {β0 , β1 , β2 , φ} , is given by:where BB ( n|N , μ , φ ) is the beta-binomial distribution , which describes the probability of observing n successes from N trials when , on average , successes occur with probability μ and variation in this probability among replicates is described by the beta-distribution with variance μ ( 1−μ ) φ/ ( 1+φ ) [75] . Evidence that mutation rate was related to x was found by applying a likelihood ratio test ( LRT ) comparing the fit of the full model with the model that ignored x ( i . e . , when β1 = β2 = 0 ) . Let LL1 and LL0 be the maximum log-likelihood of the full model and the simpler model , respectively . Under the null hypothesis that x is not associated with mutation rate , the test statistic G = 2[LL1−LL0] is chi-square distributed with two degrees of freedom , as the more complex model has two additional free parameters: β1 and β2 . A LRT was also used to test for overdispersion by comparing the fit from the full model described above with the model that assumed variation had a binomial distribution ( φ is vanishingly small ) . This latter test , if significant , justifies the use of the beta-binomial distribution rather than the binomial . Confidence intervals for model parameters were estimated using the likelihood profile approach . The genome accession numbers analysed are: NP_232242 . 1 , NP_246094 . 1 , NP_249343 . 1 , NP_439118 . 1 , NP_458435 . 1 , NP_462369 . 1 , NP_671249 . 1 , NP_716257 . 1 , NP_760245 . 1 , NP_799172 . 1 , NP_873260 . 1 , NP_927748 . 1 , YP_052151 . 1 , YP_089126 . 1 , YP_128534 . 1 , YP_152459 . 1 , YP_205663 . 1 , YP_237645 . 1 , YP_262678 . 1 , YP_272974 . 1 , YP_455981 . 1 , YP_492074 . 1 , YP_526229 . 1 , YP_564189 . 1 , YP_588978 . 1 , YP_606222 . 1 , YP_690711 . 1 , YP_693743 . 1 , YP_718344 . 1 , YP_751967 . 1 , YP_855526 . 1 , YP_928876 . 1 , YP_941848 . 1 , YP_960806 . 1 , YP_001048976 . 1 , YP_001092716 . 1 , YP_001143048 . 1 , YP_001178491 . 1 , YP_001189422 . 1 , YP_001218107 . 1 , YP_001343325 . 1 , YP_001440391 . 1 , YP_001443362 . 1 , YP_001464812 . 1 , YP_001475605 . 1 , YP_001503357 . 1 , YP_001675803 . 1 , YP_001759053 . 1 , YP_001909102 . 1 , YP_002152521 . 1 , YP_002228709 . 1 , YP_002294894 . 1 , YP_002476451 . 1 , YP_002650381 . 1 , YP_002801694 . 1 , YP_002875051 . 1 , YP_002893931 . 1 , YP_002923696 . 1 , YP_002986005 . 1 , YP_003002662 . 1 , YP_003008634 . 1 , YP_003039145 . 1 , YP_003074496 . 1 , YP_003255073 . 1 , YP_003261368 . 1 , YP_003469961 . 1 , YP_003532766 . 1 , YP_003555253 . 1 , YP_003812150 . 1 , YP_003914673 . 1 , YP_004117516 . 1 , YP_004211044 . 1 , YP_004382110 . 1 , YP_004391469 . 1 , YP_004419866 . 1 , YP_004472683 . 1 , YP_004565203 . 1 , YP_004713013 . 1 , YP_004821770 . 1 , YP_005091541 . 1 , YP_005334361 . 1 , YP_005458526 . 1 , YP_005817463 . 1 , YP_006006755 . 1 , YP_006238931 . 1 , YP_006286710 . 1 , YP_006326252 . 1 , YP_006459298 . 1 , YP_006523113 . 1 , YP_006588319 . 1 , ZP_00134303 . 1 , ZP_00991497 . 1 , ZP_01161654 . 1 , ZP_01215522 . 1 , ZP_01815379 . 1 , ZP_01894180 . 1 , ZP_01898714 . 1 , ZP_02478644 . 1 , ZP_02958582 . 1 , ZP_03319669 . 1 , ZP_03611762 . 1 , ZP_03825776 . 1 , ZP_04636540 . 1 , ZP_04640765 . 1 , ZP_04752629 . 1 , ZP_04977551 . 1 , ZP_05043634 . 1 , ZP_05637197 . 1 , ZP_05774479 . 1 , ZP_05849758 . 1 , ZP_05879825 . 1 , ZP_05880998 . 1 , ZP_05919259 . 1 , ZP_05972068 . 1 , ZP_05990699 . 1 , ZP_06018230 . 1 , ZP_06051220 . 1 , ZP_06126446 . 1 , ZP_06542208 . 1 , ZP_06637662 . 1 , ZP_07161146 . 1 , ZP_07222409 . 1 , ZP_07266238 . 1 , ZP_07379670 . 1 , ZP_07395486 . 1 , ZP_07528968 . 1 , ZP_07744420 . 1 , ZP_07777878 . 1 , ZP_07888842 . 1 , ZP_08039455 . 1 , ZP_08068248 . 1 , ZP_08079426 . 1 , ZP_08100561 . 1 , ZP_08148040 . 1 , ZP_08310711 . 1 , ZP_08519301 . 1 , ZP_08725568 . 1 , ZP_08731411 . 1 , ZP_08745737 . 1 , ZP_08754750 . 1 , ZP_09013912 . 1 , ZP_09039716 . 1 , ZP_09185001 . 1 , ZP_09505069 . 1 , ZP_09557915 . 1 , ZP_09710329 . 1 , ZP_09778630 . 1 , ZP_09972449 . 1 , ZP_10075284 . 1 , ZP_10125383 . 1 , ZP_10128956 . 1 , ZP_10135899 . 1 , ZP_10142323 . 1 , ZP_10146384 . 1 , ZP_10426764 . 1 , ZP_10438900 . 1 , ZP_10622342 . 1 , ZP_10628430 . 1 , ZP_10630449 . 1 , ZP_10643899 . 1 , ZP_10655392 . 1 , ZP_10677933 . 1 , ZP_10700164 . 1 , and ZP_10763153 . 1 . | Allostery is a process by which a molecule binding to one site of a protein alters the activity of the protein at another site . Allostery is typically thought to occur through a change in protein structure , but there is now clear evidence that the dynamic properties of a protein can also regulate allostery without a change in overall conformation . Here we examine two members of a large family of bacterial transcription factors and provide a mechanism to describe the allosteric binding of their activating ligands . We demonstrate , in these systems , that allostery arises as a natural consequence of changes in global low-frequency protein fluctuations on ligand binding . We further demonstrate that the higher dimensional parameter space that describes all potential variant transcription factors can be reduced to a two-dimensional free energy landscape that determines the key molecular parameters that predominantly regulate allostery . We additionally show that the amino acids we determine as contributing sensitively to allosteric control tend to be conserved in diverse bacteria; thus we identify a link between residues that contribute to low-frequency fluctuations and evolutionary selection pressures . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2013 | Modulation of Global Low-Frequency Motions Underlies Allosteric Regulation: Demonstration in CRP/FNR Family Transcription Factors |
Progression of RNA polymerase II ( RNAPII ) transcription relies on the appropriately positioned activities of elongation factors . The resulting profile of factors and chromatin signatures along transcription units provides a “positional information system” for transcribing RNAPII . Here , we investigate a chromatin-based mechanism that suppresses intragenic initiation of RNAPII transcription . We demonstrate that RNAPII transcription across gene promoters represses their function in plants . This repression is characterized by reduced promoter-specific molecular signatures and increased molecular signatures associated with RNAPII elongation . The conserved FACT histone chaperone complex is required for this repression mechanism . Genome-wide Transcription Start Site ( TSS ) mapping reveals thousands of discrete intragenic TSS positions in fact mutants , including downstream promoters that initiate alternative transcript isoforms . We find that histone H3 lysine 4 mono-methylation ( H3K4me1 ) , an Arabidopsis RNAPII elongation signature , is enriched at FACT-repressed intragenic TSSs . Our analyses suggest that FACT is required to repress intragenic TSSs at positions that are in part characterized by elevated H3K4me1 levels . In sum , conserved and plant-specific chromatin features correlate with the co-transcriptional repression of intragenic TSSs . Our insights into TSS repression by RNAPII transcription promise to inform the regulation of alternative transcript isoforms and the characterization of gene regulation through the act of pervasive transcription across eukaryotic genomes .
Plasticity at the beginning and end of transcription units multiplies the RNA species that can be generated from genomes . Many RNA species result from RNA Polymerase II ( RNAPII ) activity at genes and abundant non-coding genomic regions [1 , 2] . Pervasive transcription results in overlapping transcripts , for example by initiating intragenic transcription leading to the production of alternative transcript isoforms [3] . Alternative Transcription Start Sites ( TSSs ) expand RNA isoform diversity , may result in functionally different RNA and proteins specific to disease , and allow for multiple transcriptional outputs from a single gene [4 , 5] . However , the mechanisms of alternative TSS activation , repression , and regulation are poorly understood in higher eukaryotes . Repression of a gene promoter by overlapping RNAPII transcription was originally described for two tandemly arranged human α-globin gene copies [6] . Read-through transcription from the upstream α-globin gene positions the downstream promoter in the middle of a transcription unit spanning both gene copies . Repression of a downstream promoter through the act of RNAPII transcription is referred to as Transcriptional Interference ( TI ) [7] . The core of this mechanism relies on the progression of RNAPII transcription through distinct stages [8] . Each stage is characterized by the co-transcriptional recruitment of factors involved in nascent RNA processing and chromatin modifications [9] . Dynamic phosphorylation of residues in the C-terminal YSPTSPS repeat region of the largest RNAPII subunit coordinates progression through the transcription cycle by recruiting stage-specific factors [10 , 11] . Metagene analyses of stage-specific transcription factors and chromatin signatures in diverse organisms strikingly visualize many common changes associated with RNAPII progression from the beginning to the end of active transcription units [12–17] . For example , histone 3 lysine 4 methylation ( H3K4me ) states decrease from tri- ( H3K4me3 ) to mono-methylation ( H3K4me1 ) from the beginning to the end of yeast genes [18] . Such signatures provide a “positional information system” ( POINS ) for RNAPII to coordinate molecular events required for each stage of transcription [8] . An important functional outcome of co-transcriptional chromatin changes involves the suppression of transcriptional initiation from within transcription units ( intragenic TSSs ) . Whereas TSSs in gene promoters are characterized by well-defined DNA cis-elements [19] , the activity of intragenic TSSs is connected to the co-transcriptional chromatin environment [20] . Histone 3 lysine 36 methylation ( H3K36me ) is characteristic of RNAPII elongation in many organisms [21–23] . H3K36 tri-methylation ( H3K36me3 ) prevents RNAPII transcription initiation from intragenic TSSs by mediating histone de-acetylation in yeast [24–26] . Chromatin-based repression of intragenic TSSs is also tightly linked to the activity of histone chaperones [27 , 28] . The FACT ( FAcilitates Chromatin Transcription ) complex , consisting of SSRP1 and SPT16 , contributes to this activity across taxa [29 , 30] . SPT16 was initially characterized as a SPT ( suppressor of Ty ) gene that is required for the suppression of gene promoters by read-through transcription initiating from adjacent upstream Ty or δ-element insertions [31 , 32] . RNAPII read-through transcription of upstream genes due to inefficient termination can elicit suppression of downstream gene promoters by TI [7 , 33 , 34] . Transcripts overlapping gene promoters may also arise from RNAPII transcription of long non-coding RNAs ( lncRNAs ) and suppress initiation by FACT-dependent TI [35–38] . In mammals , a combination of FACT , H3K36me3 , and gene-body DNA methylation suppress intragenic TSSs [39 , 40] . Co-transcriptional chromatin signatures are largely common across species , yet their roles in the regulation of intragenic TSSs often await experimental validation . Many factors characterizing POINS are active in plants [16 , 41] . The Arabidopsis FACT complex is physically associated with multiple RNAPII elongation factors , chromatin modifiers , and elongation specific RNAPII isoforms [42 , 43] . Reduced FACT activity results in developmental defects [44] that are linked to abnormal DNA methylation at heterochromatin [45] and imprinted loci [46] . However , the role of FACT in TSS selection in plants is unclear . Moreover , H3K36me3 localizes to promoter regions in Arabidopsis , whereas the di-methylated H3K36 variant ( H3K36me2 ) associates to RNAPII elongation zones [47] . These data indicate that mechanisms in addition to those previously described in budding yeast may have evolved to repress intragenic TSSs in plants . Genome-wide TSS mapping in Arabidopsis suggests that a choice between alternative TSSs exists for most transcripts [48] . Protein isoform diversity control in response to light through regulated TSS choice underpins the biological significance of this mechanism [49] . Moreover , TSS choice may also regulate gene expression at the level of translation by the inclusion of an upstream open reading frame ( uORF ) [50] . Despite the functional significance of alternative TSS choice , little is known about the molecular mechanisms regulating this phenomenon in plants . Here , we demonstrate the repressive effect of RNAPII elongation across gene promoters in Arabidopsis . We identify chromatin and RNAPII signatures associated with this form of gene regulation by “repressive transcription” . We uncover thousands of intragenic TSSs in fact mutants , revealing a role for FACT in preventing initiation of RNAPII transcription from within plant transcription units . Our analyses of chromatin signatures identify increased levels of the RNAPII elongation-associated H3K4me1 signal at intragenic sites that function as TSSs when FACT function is compromised . Thus , we resolve plant-specific molecular events repressing transcription initiation by the process of RNAPII elongation and highlight this mechanism for the first time in the context of a multicellular organism .
To investigate gene repression through the act of RNAPII transcription across promoter regions in higher organisms , we performed a literature screen of Arabidopsis T-DNA insertion mutants with loss-of-function phenotypes [51] . This specific type of T-DNA mutants must: 1 . ) be inserted upstream of gene promoter TSSs , 2 . ) show read-through transcription into downstream genes , and 3 . ) segregate as a recessive loss-of-function phenotype . Application of these criteria identified the quasimodo1-1 ( qua1-1 ) and red fluorescence in darkness 1–1 ( rfd1-1 ) mutants as candidate mutants for further analysis [52 , 53] . QUA1 encodes a glycosyltransferase required for the biosynthesis of cell-adhesion promoting pectins [52] . The qua1-1 T-DNA mutation is inserted 117 bp upstream of the annotated translational start site ( Fig 1A; S1A Fig ) . The cell-adhesion defect in qua1-1 results in dwarfed growth and ruthenium red staining of dark grown qua1-1 hypocotyls ( Fig 1B ) . We detect elevated QUA1 expression in qua1-1 compared to wild type by RT-qPCR ( Fig 1C ) . Northern blotting reveals an abundant T-DNA-QUA1 compound transcript in qua1-1 instead of the QUA1 mRNA ( Fig 1D ) [52] . The extended transcript detected in qua1-1 corresponds to a predicted transcript initiating within the T-DNA and extending into the downstream QUA1 gene ( S1B Fig ) . Next , we performed quantitative chromatin immunoprecipitation ( qChIP ) , which confirmed increased RNAPII levels across the QUA1 gene in qua1-1 , consistent with elevated levels of transcription initiating from within the T-DNA ( Fig 1E ) . RFD1 encodes RIBA1 , the first enzyme in the plant riboflavin biosynthesis pathway [53] . The T-DNA insertion is located 307 bp upstream of the RFD1 translational start site ( Fig 1F; S1C Fig ) . Under standard light conditions , most soil-grown homozygous rfd1-1 mutants die with white cotyledons ( Fig 1G ) [53] . However , we are now able to grow homozygous rfd1-1 mutants to seed under reduced light conditions , enabling comparative analysis of the RFD1 transcript pattern in wild type and homozygous rfd1-1 mutants . Although RT-qPCR analysis shows about 20-times higher RFD1 expression in rfd1-1 compared to wild type ( Fig 1H ) , northern blotting reveals an abundant T-DNA-RFD1 compound transcript with increased transcript size in rfd1-1 initiating from the upstream T-DNA insertion ( Fig 1I , S1D Fig ) [53] . Notably , the endogenous RFD1 mRNA isoform is not detected in rfd1-1 . Increased RNAPII levels across the RFD1 gene in rfd1-1 were also confirmed by qChIP ( Fig 1J ) . Together , these complementary analyses in qua1-1 and rfd1-1 are consistent with the hypothesis that initiation from the downstream gene promoter is repressed through the act of RNAPII transcription . To test if the genomic region between the rfd1-1 T-DNA insertion and the translational start site of RFD1 can function as a promoter ( designated as TIpRFD1 , Fig 1F; S1C Fig ) , we assayed transient marker gene expression in Nicotiana benthamiana and Arabidopsis thaliana leaves . We detected expression of β-glucuronidase ( GUS ) ( Fig 2A and 2B ) and enhanced Yellow Fluorescent Protein ( eYFP ) ( S2 Fig ) driven by TIpRFD1 in transient expression assays . To test if TIp can drive gene expression in relevant tissues and at sufficiently high levels , we performed a molecular complementation of the read-through mutants with genomic constructs driven by their respective short TIp . We detect RFD1-FLAG protein expression in independent transformant lines by western blotting ( Fig 2C ) . Importantly , RFD1 expression driven by TIpRFD1-RFD1-FLAG complements the rfd1-1 phenotype ( Fig 2D ) . Likewise , we detect QUA1-FLAG protein expression in independent TIpQUA1-QUA1-FLAG transformant lines by western blotting , and these lines complement the qua1-1 phenotype ( Fig 2E and 2F ) . Thus , TIp DNA regions provide necessary and sufficient promoter activity to drive functional RFD1 or QUA1 expression . Interfering RNAPII transcription across TIp is therefore a plausible mechanism to explain the repression of initiation despite transcriptional activity at these regions . Repressive RNAPII elongation across TIp in qua1-1 and rfd1-1 mutants may impact on molecular signatures associated with RNAPII elongation and initiation at TIp . To test this , we performed qChIP experiments to assay RNAPII initiation and elongation hallmarks . The elongating form of RNAPII ( RNAPII-Ser2P ) is enriched towards the 3’ end of the QUA1 gene and depleted from the QUA1 promoter in wild type Arabidopsis ( S3A and S3B Fig ) . H3K36me3 is enriched towards the 5’ end of genes in Arabidopsis , while H3K36me2 corresponds to the elongation phase and accumulates towards the 3’ end [47] . We find the same pattern along the QUA1 gene ( S3C and S3D Fig ) . Histone modifications of active promoters such as histone H3 acetylation ( H3ac ) and H3K4me3 are enriched towards the QUA1 promoter ( S3E and S3F Fig ) [47 , 54 , 55] . Thus , RNAPII initiation and elongation can be distinguished by our qChIP analyses . We profiled qua1-1 and rfd1-1 mutants by qChIP to determine the impact of upstream RNAPII transcription across TIpQUA1 and TIpRFD1 ( Fig 3A and 3B ) . Compared to their respective wild type ecotype , significantly higher levels of RNAPII-Ser2P were present at the position of promoter-proximal primer pairs in qua1-1 and rfd1-1 ( Fig 3C and 3D ) . These results support increased RNAPII elongation across the downstream promoter . Since bulk histone density remains largely unchanged across QUA1 and RFD1 in their respective mutants ( S3G–S3I Fig ) , we tested the presence of the Arabidopsis RNAPII elongation-specific chromatin signature H3K36me2 . The mutants displayed increased H3K36me2 levels at TIpQUA1 and TIpRFD1 ( Fig 3E and 3F ) . The increase of RNAPII elongation signatures at these promoters during repression indicates that these regions may now identify as zones of RNAPII elongation , rather than promoters . Consistent with this hypothesis , histone modifications associated with active promoters ( H3ac , H3K4me3 , H3K36me3 ) were significantly depleted at TIpQUA1 and TIpRFD1 in the mutants ( Fig 3G–3L ) . Collectively , these results demonstrate that upstream RNAPII transcription shifts the POINS to specify downstream promoters as intragenic regions . Our data suggest that promoter repression in these mutants could be driven by transcription-mediated chromatin state changes . Our analyses support that gene promoters can be repressed by interfering RNAPII elongation in Arabidopsis . We hypothesized that factors associated with RNAPII elongation , such as the FACT complex , may be required for repression . To test the role of FACT in promoter repression by read-through transcription in Arabidopsis , we combined the previously described knock-down alleles of spt16-1 and ssrp1-2 mutants with qua1-1 [44] . Ruthenium red staining comparing single and double mutants revealed patches of unstained hypocotyls in spt16-1 qua1-1 compared to qua1-1 ( Fig 4A ) . Importantly , spt16-1 qua1-1 alleviated the dwarf hypocotyl phenotype observed in qua1-1 ( Fig 4B ) . These results indicate tightened cell-adhesion and partial suppression of the qua1-1 phenotype . The rescue effect was even more pronounced in ssrp1-2 qua1-1 compared to spt16-1 qua1-1 ( Fig 4A and 4B ) . This can be explained by stronger knock-down of protein levels in ssrp1-2 compared to spt16-1 [44] . To test if FACT was required for read-through repression of RFD1 in rfd1-1 , we crossed spt16-1 with rfd1-1 . In our experimental conditions , about 20% of the progeny of heterozygous rfd1-1/RFD1 seeds segregate for the photo-bleaching phenotype ( S4A Fig ) . The progeny of seed segregating in addition for SPT16/spt16-1 reduced the photo-bleaching phenotype by about 25% , consistent with suppression of photo-bleaching in the rfd1-1 spt16-1 mutant ( S4A Fig ) . Collectively , these results support the conclusion that FACT is genetically required for interfering RNAPII elongation at the qua1-1 and rfd1-1 alleles . To test the roles of additional RNAPII elongation factors in repression , we assayed genetic interactions between qua1-1 and mutations in the Arabidopsis PAF-I ( Polymerase-Associated Factor I ) subunit VIP6 [56] and the Elongator subunit ELO3 [57] . To examine the role of H3K36me2 , we tested the interaction between qua1-1 and a mutation in the H3K36 methyltransferase SDG8/ASHH2 [58 , 59] . Interestingly , unlike spt16-1 , we find no evidence for suppression of qua1-1 in these mutants ( S4B Fig ) . Genetic linkage between QUA1 and SSRP1 precluded the inclusion of ssrp1-2 in this assay ( S4C Fig ) . All in all , these data argue for a key contribution of FACT during RNAPII elongation to trigger the qua1-1 phenotype . If phenotypic suppression of qua1-1 through fact mutants was mechanistically linked to gene repression through the act of upstream interfering transcription , we would predict transcriptional changes . To examine the pattern of QUA1 transcripts , we performed northern blotting in single and double mutants . While the transcript pattern in spt16-1 and ssrp1-2 is not clearly distinguishable from wild type controls , we observe new transcript patterns in spt16-1 qua1-1 and ssrp1-2 qua1-1 double mutants compared to qua1-1 ( Fig 4C ) . Importantly , variants of the high-molecular weight interfering transcripts remain detectable in fact qua1-1 double mutants , suggesting that upstream interfering transcription can still be initiated . The interfering transcript in fact qua1-1 double mutants appears to have a more broad size distribution than in qua1-1 , which is revealed most clearly by a reduced size of the main interfering transcript isoform in ssrp1-2 qua1-1 . While we find no evidence for the QUA1 mRNA in qua1-1 , we detect hybridization signal in fact qua1-1 double mutants overlapping the expected size of the QUA1 mRNA transcript . These data suggested one or more 5’-truncated transcripts initiating from cryptic TSSs in fact qua1-1 double mutants that could restore functional QUA1 expression . To resolve such transcripts , we performed 5’ Rapid Amplification of cDNA Ends ( 5’RACE ) in the ssrp1-2 qua1-1 double mutant compared to qua1-1 . Even though there appear to be differences in the main interfering transcript size in qua1-1 compared to ssrp1-2 qua1-1 ( Fig 4C ) , our 5’RACE identifies a common TSS ( TSS2 ) in these genotypes ( Fig 4D ) . Importantly , we identified a novel TSS ( TSS1 ) in ssrp1-2 qua1-1 ( Fig 4D ) . While TSS1 does not match the exact wild type QUA1 mRNA in ssrp1-2 qua1-1 , usage of TSS1 results in a short ( 182 nt ) 5’-extension of the QUA1 mRNA . It remains possible that the wild type QUA1 TSS may also be used in ssrp1-2 qua1-1 but was not captured by our 5’RACE experiments . Phenotypic suppression indicates that functional QUA1 mRNAs are produced from cryptic TSSs , such as TSS1 , that are accessible in fact mutants despite interfering transcription across the QUA1 promoter region . Overall , our results support the conclusion that the activity of the FACT complex as part of RNAPII elongation suppresses TSSs inside of transcription units . To test if FACT suppresses endogenous intragenic TSSs , we measured Arabidopsis TSSs by 5’-CAP-sequencing ( TSS-seq ) [60] . We obtained on average 47 million raw reads for two biological repeats of wild type , spt16-1 , and ssrp1-2 ( S1 Table ) . We identified 96232 TSS clusters and annotated them by genomic location . Many TSS clusters ( n = 30487 , or 31 . 7% ) mapped to annotated gene promoters ( Fig 5A; S5A Fig ) . The number of sequencing reads supporting TSS clusters showed a high degree of correlation between biological repeats ( S5B Fig ) . We examined the overlap of our TSS clusters with TSSs identified by CAGE ( Cap Analysis Gene Expression ) [48] . 76 . 7% of TSS clusters in annotated gene promoters overlap with at least one previously reported CAGE peak ( S5C Fig and S2 Table ) , indicating very good overlap across techniques and samples . Alternative mRNA isoforms of AT4G08390 are differentially targeted to mitochondria or chloroplast [61] . Our data resolve TSSs corresponding to these isoforms ( S5D Fig ) . Interestingly , our TSS-seq data reveals 17 . 4-fold more TSSs in exons ( n = 43414 , or 45 . 1% ) than in introns ( n = 2460 , or 2 . 5% ) ( Fig 5A and S3 Table ) . The Arabidopsis genome contains 2 . 6-fold more exonic bases ( 51 . 6 Mb ) than intronic bases ( 19 . 7 Mb ) , offering a partial explanation for the biased location of intragenic TSSs in exons . In conclusion , these data illustrate high reproducibility of our TSS-seq methodology , and its abilities to validate TSSs as well as to reveal novel TSSs . To test the role of FACT in regulating TSSs in Arabidopsis , we divided the TSSs into three groups ( S4 Table ) : i ) basal TSSs detected in both wild type and fact mutants ( n = 77738 , or 80 . 8% ) ; ii ) wild-type specific TSSs ( n = 1023 , or 1 . 06% ) ; and iii ) TSSs specifically detected in fact mutants ( i . e . fact-specific TSSs; n = 17471 , or 18 . 1% , S3 Table ) . The 17-fold increase of fact-specific TSSs over the wild-type specific TSSs suggests that the FACT complex largely represses TSSs . We frequently find fact-specific TSSs in intragenic locations ( Fig 5A ) . However , TSSs induced in fact mutants have a lower TSS-seq count compared to the basal TSS set indicating lower expression of transcripts derived by fact-specific TSSs ( Fig 5B and 5C ) . The large majority of fact-specific TSSs ( 9281 out of 11555 , or 80 . 3% ) were detected in both fact mutants ( Fig 5D ) . As much as 83 . 1% of fact-specific exonic TSSs do not overlap with a TSS identified by CAGE ( S5C Fig and S2 Table ) . The AT5G18500 gene illustrates the induction of an intronic TSS in fact mutants ( Fig 5E ) . The AT4G15260 UDP-glycosyltransferase gene reveals preferential usage of a downstream intragenic TSS in fact mutants that is normally regulated in response to light signaling ( Fig 5F ) [49] , suggesting that the promoter for the shorter transcript isoform is suppressed in a FACT-dependent manner from upstream RNAPII transcription . We next quantified TSS-seq peaks at canonical promoters for genes with or without fact-specific TSSs and compared their expression in wild type and ssrp1-2 . These analyses reveal that expression of the isoforms initiating at the canonical promoter TSSs for genes with fact-specific TSSs show no significant genome-wide decrease in ssrp1-2 ( Fig 5G ) . These data indicate that initiation from intragenic fact-specific TSSs does not necessarily result from reduced transcription initiating from upstream promoters , arguing against a promoter competition model . Overall , our TSS-seq data reveal thousands of intragenic regions that can function as TSSs depending on FACT activity . These results support a role of FACT as part of POINS in Arabidopsis , with a key function in suppressing intragenic TSSs . Common DNA sequences or chromatin signatures may predispose intragenic regions to function as fact-specific TSSs . We tested differential DNA-motif enrichment in exonic fact-specific TSSs compared to basal exonic TSSs . However , we detect no differentially enriched sequence motif or position bias within exons ( S5E Fig ) . To test if exonic TSSs may be characterized by promoter-like chromatin architecture , we re-analyzed available Arabidopsis ChIP-seq data of chromatin signatures in wild type [62–66] . We compared chromatin signatures centered on five sets of genomic locations: fact-specific exonic TSS positions , exonic control regions without TSSs in the same set of genes that have fact-specific exonic TSSs , basal exonic TSSs , exonic control regions without TSSs in the same set of genes that have basal exonic TSSs , and TSSs at gene promoters . Box plots capturing the median sequencing signal in 20 bp intervals around the positions are given to present data variability and associated statistical tests between the five genomic sets ( Fig 6 , S7 Fig ) . Metagene plots of the mean sequencing signal in a 400 bp interval centered on the positions are given with standard errors to visualize the dynamics of the chromatin signatures around the positions ( S6 and S7 Figs ) . Arabidopsis promoter-chromatin signatures clearly distinguish TSSs identified in gene bodies from TSSs at gene promoters ( Fig 6 ) , which is well-illustrated through the shape of accumulated signal in the metagene plots ( S6 Fig ) . Promoter TSSs show low nucleosome signal assayed by MNase-seq compared to intragenic TSSs and control regions ( Fig 6A , S6A Fig ) . fact-specific exonic TSSs show the highest MNase-seq signal compared to basal exonic TSSs and control regions . These data argue against promoter-like , low nucleosome density at fact-specific exonic TSSs in the repressed state . Moreover , the set of basal exonic TSSs is often enriched for promoter chromatin-signatures compared to fact-specific TSSs ( Fig 6B–6F , S6B–S6F Fig ) . These data argue against the idea that exonic regions we identify as fact-specific TSSs show the chromatin architecture of promoter TSSs in wild type . Of all ChIP-seq experiments assaying histone modifications that we analyzed , histone 3 lysine 4 mono-methylation ( H3K4me1 ) , associated with RNAPII elongation , represents the only post-translational histone modification that is enriched at fact-specific TSSs compared to control regions and basal exonic TSSs ( Fig 6G p = 1 . 8e-09 , S6G Fig , S7 Fig ) . The levels of H3K36me2 , an alternative Arabidopsis RNAPII elongation signature , between basal TSSs and fact-specific TSSs are indistinguishable ( Fig 6H ) . These data argue for a differential effect of Arabidopsis elongation-specific chromatin signatures , consistent with distinct contributions of the FACT complex among RNAPII elongation factors suggested by the genetics ( S4 Fig ) . To test if the detected increase of H3K4me1 at fact-specific TSSs could be explained by a bias in the particular ChIP-seq data [63] , we analyzed ChIP-seq data generated by an independent study that also assayed all three methylation states of H3K4 [67] . Consistently , the data for H3K4 di-and tri-methylation resulted in overall similar profiles ( Fig 6E and 6F , S6E and S6F Fig , S7A and S7B Fig ) . Importantly , the increase of H3K4 mono-methylation at exonic fact-specific TSSs could be confirmed ( S7C Fig p = 2 . 4e-08 ) . The combination of H3K4me1 and H3K27ac chromatin signatures characterizes enhancers in many systems . However , even though fact-specific TSSs appear enriched in H3K4me1 , these sites are reduced in H3K27ac compared to basal exonic TSSs ( Fig 6B , p = 2 . 4e-12 ) . Our analyses offer no evidence to support the idea that locations of fact-specific TSSs may represent intragenic enhancers . In summary , these analyses suggest that exonic fact-specific TSSs carry chromatin signatures of RNAPII elongation that are enriched for H3K4me1 . While the FACT complex directly interacts with residues in H3/H4 [68–70] , it interacts more strongly with H2A/H2B dimers and is considered a H2A/H2B chaperone in many organisms , including Arabidopsis [71 , 72] . To test if chromatin signatures based on H2A/H2B may participate in predisposing exonic sites as TSSs in fact mutants , we analyzed wild-type ChIP-seq data for H2A , ubiquitinylation at H2A lysine 121 ( H2AUb ) , H2B , mono-ubiquitinylation of H2B lysine 120 ( H2BUb ) and the H2A variant H2A . Z [73–76] . H2A . Z and H2Aub match the profiles of chromatin signatures of promoter TSSs , whereas we detect the strongest H2A signal in exons ( S7D–S7F Fig ) . However , H2A . Z levels at fact-specific TSSs are indistinguishable from those at basal exonic TSSs , arguing against a role of H2A . Z in specifying fact-specific TSSs . We note that basal exonic TSSs are enriched for H2AUb compared to fact-specific TSSs and control regions , consistent with elevated H3-based promoter TSSs chromatin signatures . The profile of H2B ChIP-seq data matches those of promoter TSSs-associated chromatin signatures , whereas H2BUb is enriched in exons , consistent with previously suggested roles in RNAPII elongation ( S7G and S7H Fig ) . Quantification of ChIP-seq signal identified no statistically significant changes between fact-specific exonic TSSs and basal exonic TSSs . Perhaps surprisingly , given the preferential activity of FACT as an H2A/H2B chaperone , our analyses found no evidence for H2A or H2B-based chromatin signatures distinguishing fact-specific TSSs that may mark these locations in concert with H3K4me1 . To test if fact-specific intragenic TSSs present in exons enriched for H3K4me1 may be a consequence of high RNAPII transcription , we assessed RNAPII occupancy using RNAPII ChIP-seq data [64] . To assay transcriptionally active populations of RNAPII we analyzed Global Run-On sequencing data ( GRO-seq ) [77] . We used GRO-seq data generated in nrpd1/nrpe1 double mutants to ensure the GRO-seq signal is specific to RNAPII , as previously described [77] . Interestingly , exonic regions identified as TSSs accumulate more RNAPII compared to exonic control regions in the same gene sets ( Fig 6I , S6I Fig ) , and this fraction of RNAPII is transcriptionally active ( Fig 6J , S6J Fig ) . Basal exonic TSSs correspond to more highly transcribed regions than fact-specific exonic TSSs ( Fig 6I and 6J ) , arguing against the idea that fact-specific TSSs represent regions with particularly high RNAPII activity . In conclusion , our chromatin-state analyses focused on exonic TSSs suppressed by FACT are consistent with a co-transcriptional mechanism that may be linked to the H3K4me1 chromatin signature . Our above analyses of available ChIP-seq datasets suggest that at least part of the specification mechanism that distinguishes exonic regions to function as TSSs in fact mutants from basal exonic TSSs may involve relatively high starting levels of H3K4me1 . As the chromatin signatures of the QUA1 and RFD1 promoter region in their respective qua1-1 and rfd1-1 mutants were not assayed by the wild type ChIP-seq data , we tested if these promoter regions also showed high H3K4me1 in qua1-1 and rfd1-1 read-through mutants . Indeed , we detected increased H3K4me1 in the mutants compared to their respective wild type controls ( Fig 7A–7D ) . These results are consistent with FACT-dependent repression of TSSs around the promoter regions of the RFD1 and QUA1 genes when these promoter regions acquire RNAPII elongation signatures such as H3K4me1 by read-through transcription in qua1-1 and rfd1-1 mutants . To test if the repression of gene promoter TSSs by RNAPII elongation shares molecular signatures of TSS repression within gene bodies , we performed targeted qChIP analyses at selected fact-specific intragenic TSSs comparing wild type and the ssrp1-2 mutant . To identify endogenous genes for qChIP analysis , we selected strongly induced intragenic fact-specific TSSs . We next performed RNA-seq in wild type and ssrp1-2 to refine our selection based on a visual increase of RNA-seq reads in exons downstream of fact-specific TSSs in the ssrp1-2 sample . The increased RNA-seq signal downstream of fact-specific TSSs implies that these TSSs generate bona fide alternative transcripts . We selected four genes with intragenic fact-specific TSSs ( AT5G18500 , AT4G15260 , AT3G56210 , and AT5G51200 ) and two control loci for basal exonic TSSs ( AT5G13630 and AT1G06680 ) ( Fig 7E and 7F and S8A–S8D Fig ) . We measured H3K4me1 and H3K4me3 levels by qChIP in wild type and ssrp1-2 at promoter TSSs and intragenic TSSs for the four genes with intragenic fact-specific TSSs and at the two basal exonic TSSs . We present these data normalized to H3 signal at these positions to account for potential changes in H3 levels . Importantly , triggering intragenic TSSs in ssrp1-2 corresponded to a significant decrease in H3K4me1 at the four fact-specific TSSs ( Fig 7G and 7H , S8E and S8F Fig ) , whereas we could detect no difference at the basal exonic TSSs ( S8G and S8H Fig ) . Conversely , H3K4me3 levels increase at all four fact-specific TSSs in ssrp1-2 mutants ( Fig 7I and 7J , S8I and S8J Fig ) , whereas we could not detect any change at the basal exonic TSSs ( S8K and S8L Fig ) . We note that the levels of these marks at the corresponding gene promoter TSSs are not significantly changed ( Fig 7G and 7J , S8 Fig ) , offering chromatin-based support that the overall expression of the gene isoforms starting at the promoter TSSs are largely unaffected by this mechanism . These findings are consistent with our genome-wide TSS-seq analyses of promoter TSS strength for genes with and without fact-specific TSSs ( Fig 5G ) . In conclusion , our qChIP analyses of H3K4 mono- and tri-methylation states suggest dynamic changes when FACT activity is compromised: fact-specific intragenic TSSs acquire H3K4me3 chromatin signatures of active promoters that are correlated with a reduction of H3K4me1 . Given the function of FACT as a histone chaperone , it seems plausible that a reduction in nucleosome density in fact mutants may facilitate the establishment of fact-specific TSSs . While we detected a trend of reduced bulk H3 levels at fact-specific TSSs in the ssrp1-2 mutant , these changes were statistically significant at only three of six intragenic loci tested ( S9 Fig ) . These data suggest that reduced nucleosome density in ssrp1-2 may aid the formation of fact-specific exonic TSSs , yet does not offer a satisfactory explanation for this phenomenon . We examined possible changes in the presence of other histone modifications by qChIP: two active promoter marks ( H3K36me3 and H3K27ac ) and an elongation mark ( H3K36me2 ) . We observed a general trend , although not always statistically significant , towards increased H3K36me3 and H3K27ac at fact-specific TSSs in ssrp1-2 , while H3K36me2 was generally reduced ( S10–S12 Figs ) . Importantly , we did not detect significant changes in the levels of any of the histone modifications tested at the control basal TSS positions in ssrp1-2 ( S8–S12 Figs ) . Collectively , our qChIP analyses suggest that FACT represses intragenic TSSs co-transcriptionally by regulating chromatin changes that favor a balance of relatively high intragenic H3K4me1 levels and low levels of chromatin signatures found at promoter TSSs , such as H3K4me3 . All in all , our data support that FACT is required for the repression of intragenic TSSs in plants . Read-through transcription blurs transcript boundaries that may re-define gene promoters as intragenic , which reconciles the genetic requirement of FACT for promoter TSS repression by read-through transcription . Repression of promoter TSSs coincides with a loss of initiation-specific RNAPII hallmarks and a gain of elongation-specific signatures . Similarly , the FACT complex represses initiation of transcription from several thousand intragenic fact-specific TSSs . We could not fully resolve what molecularly distinguishes intragenic sites that function as fact-specific TSSs from surrounding locations , but fact-specific intragenic TSSs show relatively high levels of H3K4me1 in the repressed state . We condensed our results characterizing the chromatin dynamics accompanying the transition from FACT-repressed intragenic TSSs to active TSSs in a cartoon summarizing our findings ( Fig 8 ) . In conclusion , we uncover a co-transcriptional chromatin-based mechanism shaping gene regulation and transcript isoform diversity by regulating TSS selection in plants .
TSSs shape RNA isoform expression , but little is known about the mechanisms regulating TSS choice within transcription units . RNAPII transcription across gene promoters has the potential to re-define gene promoters as “intragenic” and repress them by mechanisms inhibiting initiation from within transcription units . We leveraged Arabidopsis T-DNA read-through mutants to identify a role of the conserved FACT histone chaperone complex in the repression of intragenic TSSs in a multicellular organism . Consistently , we identify a large number of intragenic TSSs repressed by FACT , particularly from exonic regions enriched for the chromatin signature H3K4me1 . Three activities of the FACT complex that may explain a key role in repressing intragenic TSSs across species are: 1 . ) stimulation of RNAPII elongation , 2 . ) histone re-assembly in the wake of RNAPII transcription to avoid gaps in nucleosome density , and 3 . ) recycling of old histones to maintain chromatin-based signals of POINS . First , FACT stimulates RNAPII transcription of DNA templates packaged in nucleosome structures [30] . Structural analyses suggest that the FACT complex directly binds nucleosomes on several contacts of histone proteins , stabilizing otherwise energetically unfavorable nucleosome conformations that weaken nucleosome binding to DNA [78] . Stabilization of partly unfolded nucleosome intermediates facilitates RNAPII progression through nucleosome barriers . The ability to stabilize nucleosomes may distinguish FACT from other RNAPII elongation factors that did not score as hits in our assay , such as PAF-I or Elongator ( S4 Fig ) . Defective FACT may result in “transcription stress” through stalled or arrested RNAPII molecules in transcription units that may trigger proteolytic degradation of stalled RNAPII [79] . Associated chromatin changes may facilitate the initiation of RNAPII transcription that could help to explain elevated TSSs in fact mutants [80 , 81] . The relatively high H3K4me1 levels at fact-specific TSSs in wild type may indicate sites reliant on efficient RNAPII elongation . Consistently , we detect increased RNAPII ChIP-seq and GRO-seq signals at fact-specific TSSs compared to control regions in the same gene sets ( Fig 6I and 6J ) . Consequently , defective elongation may contribute to preferential RNAPII initiation from within transcription units at these sites . Second , FACT aids the re-assembly of nucleosomes from cellular histone pools in the wake of transcribing RNAPII to prevent gaps in nucleosome coverage [30 , 82] . Consistently , reduced nucleosome density within transcription units has been reported in human and yeast fact mutants [40 , 83] . Nucleosome Depleted Regions ( NDRs ) are associated with active promoter TSSs , and the establishment of intragenic NDRs may trigger the initiation of RNAPII transcription [28 , 40 , 84] . Relatively high MNase-seq signal at fact-specific TSSs compared to basal TSSs and control regions provides some evidence for nucleosomes blocking access to fact-specific TSSs ( Fig 6A ) . FACT activity may be needed for TSS repression as nucleosome positioning at fact-specific TSSs may be sensitive to FACT histone re-assembly activity . Our locus-specific H3-qChIP analyses provide some support for this idea , as we detect a trend towards reduced H3 levels in ssrp1-2 mutants at fact-specific TSSs . However , the reduction of H3 is statistically significant at only two out of four fact-specific TSSs ( S9 Fig ) . Intragenic NDRs resulting from reduced FACT histone re-assembly activity may contribute to the increase of transcriptional initiation from fact-specific TSSs . Third , the propensity of FACT to re-deposit histones back into their previous locations in the wake of RNAPII transcription represents an intuitive mechanism to maintain the co-transcriptional positional information provided by chromatin signatures [85] . The gradient of H3K4me at yeast genes from H3K4me3 at the beginning of genes towards H3K4me1 at the ends supports a role of differential methylation at H3K4 as a positional signal [18] . Old histones accumulate towards the H3K4me3-rich 5’ ends of yeast genes , so conceivably FACT may contribute to the co-transcriptional maintenance of this pattern [86] . Consistently , defective FACT disrupts POINS as is evidenced by the incorporation of the promoter-enriched histone variant H2A . Z within transcription units in yeast fact mutants [87] . Our analyses of H2A . Z ChIP-seq data found no evidence for high H2A . Z levels at fact-specific TSSs in wild type ( S7 Fig ) . However , promoter TSSs chromatin signatures , such as H2A . Z , may accumulate at these sites in fact mutants . We find support for this idea in our qChIP analyses focused on other known active promoter chromatin signatures , such H3K4me3 ( Fig 7 ) , and to a lesser extent also H3K36me3 and H3K27ac ( S11 and S12 Figs ) . Future studies will be required to dissect the contributions of defects in RNAPII elongation , nucleosome re-positioning , and POINS establishment in the up-regulation of intragenic TSSs observed in fact mutants . In yeast , histone deacetylases associate with elongation-specific H3K36me3 and elongating RNAPII to repress the activity of intragenic TSSs [24 , 25] . A reduction of histone acetylation in promoter regions in qua1-1 and rfd1-1 read-through mutants supports this observation in plants ( Fig 3 ) . Several histone deacetylases ( HDACs ) associate with RNAPII elongation complexes in Arabidopsis [42] . However , the plant HDAC complexes participating in the suppression of intragenic TSSs are yet to be identified . Our chromatin state analyses in qua1-1 and rfd1-1 support H3K36me2 as a chromatin signature of RNAPII elongation ( Figs 3 and 6 ) . Curiously , we find no evidence for a role of the Arabidopsis H3K36 methyltransferase SDG8/ASHH2 in gene repression through the act of RNAPII transcription ( S4 Fig ) . One of the 47 alternative SET-Domain Genes ( SDGs ) might contribute to the repression of intragenic TSSs [88] . Alternatively , since FACT-repressed intragenic TSSs are not specifically enriched for H3K36me2 compared to basal exonic TSSs and control regions in wild-type plants ( Fig 6H , S6 Fig ) , TSS repression by the act of RNAPII elongation in plants may be less dependent on H3K36 methylation-based signals . Instead , our screen for chromatin signatures characterizing intragenic regions poised to function as TSSs in fact mutants identifies H3K4me1 as the strongest candidate histone variant or post-translational histone modifications enriched at these sites . Signals of H3K4me1 at fact-specific TSSs show an inverse relationship with increasing H3K4me3 levels when RNAPII initiates transcription from fact-specific intragenic TSSs ( Fig 7 ) . Perhaps , FACT is involved in recycling old H3K4me1-containing nucleosomes , since we detect reduced H3K4me1 at these sites in fact mutants . Newly incorporated nucleosomes might be more poised to accumulate H3K4me3 at these positions in fact mutants when transcription initiation is triggered from these sites ( Fig 8 ) . Alternatively , FACT-linked H3K4me3 demethylase- and/or H3K4me1 methyltransferase activities would be consistent with our results . However , the exact molecular mechanism of chromatin-based repression of intragenic TSSs in plants remains an area for future experimental focus . The combination of H3K4me1 , H3K27ac , and low levels of bidirectional transcription are classically associated with enhancer regions [89] , however it is unclear if these features directly contribute to enhancer function [90] . Our analysis of H3K27ac ChIP-seq data showed reduced H3K27ac signals at fact-specific TSSs compared to basal TSSs ( Fig 6B ) . We therefore disfavor the hypothesis that fact-specific intragenic TSSs are decorated with the combinatorial chromatin signatures characterizing enhancers . Intragenic H3K4me1 in Arabidopsis correlates with RNAPII elongation and counteracts H3K9me2-mediated gene repression [63] . While initiation of transcription from fact-specific TSSs can result in poly-adenylated RNA ( Fig 7E and 7F ) , the overall expression is reduced compared to basal TSSs ( Fig 5B and 5C ) . Selective RNA degradation shown for cryptic transcripts may offer a partial explanation [2] . Overall , some regions identified as fact-specific intragenic TSSs might bear similarity to mammalian “primed enhancers” that are poised for activation when new gene expression programs are implemented [91] . FACT activity is targeted by cancer therapeutics [92] , yet the regulation of FACT activity in Arabidopsis is largely unexplored . We identified intragenic fact-specific TSSs using knock-down alleles , suggesting that relatively mild modulation of FACT activity elicits profound effects on intragenic initiation in Arabidopsis . Arabidopsis spt16-1 and ssrp1-2 mutants display similar phenotypic defects , indicating that regulation of intragenic TSSs may shape plant gene expression programs underlying environmental responses and development [44 , 46] . A prime example may turn out to be plant light signaling that relies on alternative TSS choices [49] , as we observed for the AT4G15260 gene . Furthermore , recent examples of gene regulation by the act of interfering lncRNA transcription in yeast and human emphasize a key role for FACT [7 , 37 , 38] . While such examples remain to be characterized in plants , we demonstrate that the underlying mechanism of repressive RNAPII transcription is operational . Our study illustrates striking similarities between the repression of promoter TSSs by interfering read-through transcription and the repression of intragenic TSSs . These similarities can be reconciled by the repressive effects of RNAPII elongation on TSSs . While the underlying mechanism bears some overlap with classical studies in budding yeast , there appear to be important differences at the level of RNAPII elongation-associated chromatin signatures , highlighting functional differences between species . Our study offers a platform to query the regulatory roles of intragenic TSSs in plants . We advance the molecular mechanism limiting intragenic TSSs by FACT . We map thousands of intragenic sites that initiate transcription when FACT function is compromised . Our data suggests that relatively high levels of H3K4me1 contribute to chromatin-based specification of these sites . Our insights into repressive RNAPII transcription promise to inform the characterization of gene regulation through the act of pervasive transcription throughout eukaryotic genomes .
All Arabidopsis thaliana lines used in this study are listed in the S5 Table . Arabidopsis thaliana and Nicotiana benthamiana plants were grown in greenhouses or climate chambers with a 16h light/8h dark cycle at 22°C for general growth and seed harvesting . For seedlings grown on plates , sterilized seeds were grown on 1/2 Murashige and Skoog ( MS ) medium containing 1% sucrose and supplemented with 1% Microagar . For analysis of the homozygous rfd1-1 phenotype , seeds were sown in 96 well trays stratified for 2–3 days at 4°C . Plants for F2 analysis were grown in high light conditions ( >100 μE ) . White seedlings were counted 10 days later . To propagate rfd1-1 homozygotes , heterozygous rfd1-1 seeds were sterilized and sown on MS plates with phosphinothricin selection , covered in foil , and stratified for 2 days at 4°C . Seeds were light induced for 6–8 h in a growth chamber with light strength of 80–100 μE . Plates were covered in foil for 3 days , the plates were unwrapped and grown in low light ( <50 μE ) for 3–4 weeks before transferring to soil . To isolate RNA , rfd1-1 homozygote seeds and corresponding wild type controls were sterilized and sown on MS plates as described above and grown in low light for two weeks . In order to collect enough material for ChIP , heterozygous rfd1-1 seeds were sterilized and sown on MS plates with phosphinothricin selection as described above and grown in low light for two weeks . Col-0 wild type controls were treated the same way , but without selection . Seeds were sown in 96-well plates containing 70 μl ddH2O . To synchronize germination , seeds were stratified at 4°C for 2–3 days . Germination of seeds was induced by light for 8–10 hours . The plates were wrapped in aluminium foil for 4 days . Etiolated seedlings were stained with 0 . 05% ruthenium red solution for 2 minutes . Seedlings were washed twice with ddH2O . Staining phenotype was recorded using a stereomicroscope . Marker gene constructs were generated using pGWB vectors [93] . TIpRFD1 was amplified from rfd1-1 genomic DNA using primers MLO414/422 . p35S was amplified from rfd1-1 genomic DNA using primers MLO538/MLO416 . TIpRFD1 and p35S were inserted into the pENTR-D-Topo vector through topo reaction to generate entry vectors SMC358 ( containing TIpRFD1 ) and SMC379 ( containing p35S ) . Entry vectors were used in a LR reaction with pGWB533 ( containing GUS ) and pGWB540 ( containing eYFP ) to generate expression vectors SMC371 ( TIpRFD1-GUS ) , SMC367 ( TIpRFD1-eYFP ) , SMC377 ( p35S-GUS ) and SMC373 ( p35S-eYFP ) . The expression vectors were transformed into Agrobacterium tumefaciens strain GV3850 by electroporation under 2 . 5kV , 400Ω resistance and 25uF capacitance . Agrobacteria harboring expression vectors were respectively co-infiltrated with the p19 suppressor of silencing into Nicotiana benthamiana and Arabidopsis thaliana efr mutant leaves [94] . GUS and eYFP signal was detected at 2 days after infiltration . Complementation constructs were generated using SMC330 , a version of pEG302 [95] enabling hygromycin selection following plant transformation . SMC330 was generated by replacing the Bialaphos resistance gene with the Hygromycin resistance gene of pCambia1300 . TIpQUA1:QUA1 and TIpRFD1:RFD1 were amplified from genomic wild type DNA using primers MLO727/728 and MLO414/442 , respectively . The resulting PCR products were introduced into pENTR-D-Topo by topo cloning to generate entry vectors ( SMC409 for TIpQUA1:QUA1 and SMC356 for TIpRFD1:RFD1 ) . The entry vectors were used in a LR reaction with SMC330 to generate expression vector SMC410 ( containing TIpQUA1:QUA1-FLAG construct ) and SMC380 ( containing TIpRFD1:RFD1-FLAG construct ) . The complementation constructs were then transformed into Agrobacterium tumefaciens strain GV3101 ( pMP90 ) by electroporation under 2 . 5kV , 400Ω resistance and 25μF capacitance . Agrobacterium-mediated transformation of Arabidopsis was performed as described in [96] . Homozygous qua1-1 and heterozygous rfd1-1 Arabidopsis were used for complementation . Seeds from transformed Arabidopsis were screened for T-DNA integration by hygromycin resistance . Multiple independent single-locus insertions were identified by segregation analysis and tested for complementation and protein expression . Phenotypic complementation was tested using progeny of lines homozygous for qua1-1 or rfd-1 , and hemizygous for the complementation constructs ( Fig 2D and 2F ) . The GUS staining assay was performed as previously described [97] . X-Gluc ( 5-bromo-4-chloro-3-indolyl glucuronide ) substrate was vacuum infiltrated into A . thaliana and N . benthamiana leaves . After staining , leaves were rinsed in 70% ethanol at room temperature until the chlorophyll was washed off . eYFP fluorescence was quantified using a Biorad imager Gel Doc . Equal amounts of plant material were harvested from plant tissue . Proteins were extracted in 2 . 5x extraction buffer ( 150 mM Tris-HCl pH 6 . 8; 5% SDS; 25% Glycerol; 0 . 025% Bromophenol blue; 0 . 1 mM DTT ) . Proteins were separated by SDS-PAGE using precast 4–15% Criterion TGX stain-free protein gels ( Biorad ) and transferred to PVDF membrane using a semi-dry Trans-blot Turbo transfer system ( Biorad ) . Membranes were blocked ( 5% non-fat dried milk in PBS ) for 1 hour at room temperature . Anti-FLAG ( Sigma F3165 ) was added overnight at 4°C with rotation . Membranes were washed with PBS before incubation with the anti-mouse HRP-conjugated secondary antibody ( Dako P0161 ) for 1 hour at room temperature . Membranes were washed in PBST . Chemiluminescent signals were detected using Super-Signal West Pico Chemiluminescent ( Thermo Fisher Scientific ) according to manufacturer’s instructions . qChIP experiments were performed essentially as described in [98] , with minor modifications . For immunoprecipitations , Protein A magnetic beads ( GenScript ) and 2 μg of an antibody ( Anti-Histone H3 , ab1791; Anti-RNA polymerase II CTD YSPTSPS phosphor S2 , ab5095; Anti-RNA polymerase II subunit B , AS11 1804; Anti-Histone H3 mono methyl K4 , ab8895; Anti-Histone H3 tri methyl K4 , ab8580; Anti-Histone H3 tri methyl K36 , ab9050; Anti-Histone H3 di methyl K36 , ab9049; Anti-Histone H3 pan-acetyl , ab47915; Anti-Histone H3 lysine 27 acetylation , ab4729 ) were added to solubilized chromatin . Quantitative analysis was performed on captured DNA by qPCR ( Biorad ) . See S5 Table for oligonucleotide sequences . ChIP enrichments were calculated as the ratio of product of interest from IP sample to the corresponding input sample ( %IP ) . For qua1-1 and corresponding wild type ( ecotype WS ) , %IP results were further normalized to %IP for an internal reference gene ( ACT2 ) to account for different fixation conditions stemming from the qua1-1 cell wall defect . Error bars represent standard error of the mean resulting from at least three independent replicates . RNA was isolated from 14 day old seedlings using Plant RNeasy Mini-Kits as per manufacturer’s instructions ( Qiagen ) . For RT–qPCR experiments , first strand complementary DNA synthesis was performed on Turbo DNase-treated ( Ambion ) RNA using oligo-dT primers and Superscript III ( Invitrogen ) as per manufacturer’s instructions . Negative controls lacking the reverse transcriptase enzyme ( -RT ) were performed alongside all RT–qPCR experiments . Quantitative analysis was performed by qPCR ( Biorad ) . Data was normalized to an internal reference gene ( ACT2 ) . Levels in mutants represent relative expression compared to corresponding wild type . Northern analyses were performed as previously described with minor modifications [99] . Briefly , 5 micrograms of total RNA were separated by electrophoresis on agarose-formaldehyde-MOPS gels and transferred to a nylon transfer membrane by capillary blotting in 10x SSC overnight . RNA was crosslinked to the nylon membrane by UV irradiation . Membranes were probed with single stranded cDNA probes generated by incorporation of radioactive α-32P-dTTP . A Typhoon phosphoimager ( GE Healthcare Life Sciences ) was used for analysis . The general transcriptome sequence library ( poly ( A ) -enriched ) for RNA-seq of 2-week old ssrp1-2 and wild type Arabidopsis seedlings were constructed using Illumina TruSeq Sample Prep Kit v2 following the manufacturer's protocol . Sequence library were measured on Agilent 2100 Bioanalyzer . The sequencing was performed on a HiSeq 4000 ( Illumina ) platform for paired-end 100 ( PE100 ) run . 5’RACE experiments were performed using the SMARTer RACE 5'/3' Kit ( Takara , Japan ) according to manufacturer’s instructions . See S5 Table for oligonucleotide sequences . TSSs were mapped genome-wide in Arabidopsis using 5’-CAP-sequencing [60] , with some minor changes as previously described [100] . Briefly , 5 micrograms of DNase-treated total RNA were treated with CIP ( NEB ) to remove all non-capped species in the sample . Next , 5’ caps were removed using Cap-Clip ( CellScript ) to permit ligation of single-stranded rP5_RND adapter to 5’-ends of previously capped species with T4 RNA ligase 1 ( NEB ) . Poly ( A ) -enriched ligated RNAs were captured with oligo ( dT ) Dynabeads ( Thermo Fisher Scientific ) according to manufacturer’s instructions and fragmented in fragmentation buffer ( 50 mM Tris acetate pH 8 . 1 , 100 mM KOAc , 30 mM MgOA ) for 5 mins at 80°C . First-strand cDNA was generated using SuperScript III ( Invitrogen ) and random primers following manufacturer’s instructions . Second-strand cDNA was generated using Phusion high-fidelity polymerase ( NEB ) and the BioNotI-P5-PET oligo as per manufacturer’s instructions . Biotinylated PCR products were captured by streptavidin-coupled Dynabeads ( Thermo Fisher Scientific ) , end repaired with End Repair Enzyme mix ( NEB ) , A-tailed with Klenow fragment exo- ( NEB ) , and ligated to barcoded Illumina compatible adapter using T4 DNA ligase ( NEB ) . Libraries were amplified by PCR , size selected using AMPure XP beads ( Beckman Coulter ) , pooled following quantification by Bioanalyzer ( Agilent ) , and sequenced in single end mode on the following flowcell: NextSeq 500/550 High Output Kit v2 ( 75 cycles ) ( Illumina ) . All custom code used in this study is available from https://github . com/Maxim-Ivanov/Nielsen_et_al_2018 . Quality of raw TSS-seq data was consistently high as reported by the FastQC software ( https://www . bioinformatics . babraham . ac . uk/projects/fastqc/ ) . In brief , the TSS-seq data analysis pipeline was as follows: FASTQ files were subjected to quality and adapter trimming at 3' ends using Trim Galore v0 . 4 . 3 ( —adapter "ATCTCGTATGCCG" ) ( https://github . com/FelixKrueger/TrimGalore ) . UMI barcodes ( 8 nt ) were trimmed from 5' ends and appended to FASTQ headers using UMI-Tools extract [101] . The adapter- and UMI-trimmed reads were aligned to TAIR10 genome assembly using STAR v2 . 5 . 2b ( —outSAMmultNmax 1—alignEndsType Extend5pOfRead1 ) [102] . The output SAM files were sorted and converted to BAM using Samtools v1 . 3 . 1 [103] . Reads aligned to rRNA , tRNA , snRNA or snoRNA loci were filtered out using BEDTools v2 . 17 . 0 [104] . The resultant BAM files were filtered for reads with MAPQ≥10 using Samtools . Finally , BAM files were deduplicated using UMI-Tools dedup . The "clean" BAM files were converted to stranded Bedgraph files using BEDTools genomecov ( -bg -5 -strand + for forward strand , -bg -5 strand—for reverse strand ) . Bedgraph files were compressed to BigWig format using kentUtils bedGraphToBigWig ( https://github . com/ENCODE-DCC/kentUtils ) . For more details on the TSS-seq read alignment pipeline , see the 01-Alignment_of_5Cap-Seq_data . sh file in the code repository . At the next step , TSSs were called from BigWig files using the CAGEfightR package v1 . 0 . 0 [105] which is available from Bioconductor ( https://bioconductor . org/packages/release/bioc/html/CAGEfightR . html ) and also from author's repository on Github ( https://github . com/MalteThodberg/CAGEfightR ) . Only genomic positions supported by at least two 5' tags in at least two libraries from the same genotype were considered as TSS candidates . Adjacent TSSs separated by not more than 20 bp were merged together into TSS clusters . The TSS clusters were annotated by intersection with various genomic features which were extracted from the TxDb . Athaliana . BioMart . plantsmart28 package . The package contains annotations from ENSEMBL Plant version 28 which combines TAIR10 and Araport11 . In particular , proximal upstream regions were defined as [ ( gene start ) -500bp , ( gene start ) -100bp] and promoters as [ ( gene start ) -100bp , ( gene start ) +100bp] . Called TSSs were annotated by genomic location as either genic ( "promoter" , "proximal" , "fiveUTR" , "threeUTR" ) , intragenic ( "exon" , "intron" , "antisense" ) , or intergenic . In case of conflicting annotations , a single annotation was chosen according to the following hierarchy: intergenic < antisense < intron < exon < threeUTR < fiveUTR < proximal < promoter . The full TSS calling pipeline was detailed in the 02-Calling_TSS_with_CAGEfightR . R script . The statistical analysis of genomic distribution of the called TSS was described in the 03-Exploratory_analysis_of_exonic_TSS . R file in the code repository . The differential motif enrichment analysis was done using the DREME software [106] . To investigate the possible correlations between fact-specific exonic TSS and various histone modifications in Arabidopsis , we re-analyzed the available histone H2A , H2B , H3 and RNAPII ChIP-seq datasets [62–65] [73–76] , as well as an MNase-seq dataset [66] . All accession numbers are available from the S6 Table . Two of these ChIP-seq datasets are paired-end [66] [75] and the rest are single-end . The pipelines for remapping paired-end and single-end ChIP-seq data were detailed in 04-Remapping_Paired-End_ChIP-Seq_and_MNase-Seq . sh and 05-Remapping_Single-End_ChIP-Seq . sh files , respectively . In brief , the alignment to the TAIR10 genome was done using STAR v2 . 5 . 2b . The BAM files were sorted and filtered for MAPQ≥10 . To convert BAMs into Bedgraph files which correctly represent the source of ChIP-seq or MNase-seq signal , one has to infer the coordinates of original inserts . Otherwise , if read length was smaller than the average insert size , then the sequencing depth is expected to peak around the true source of ChIP-seq signal instead of coinciding with it . This operation is trivial for paired-end data , because the insert size for each pair of reads is directly available from the TLEN field of BAM files ( see 04-Remapping_Paired-End_ChIP-Seq_and_MNase-Seq . sh ) . However , for single-end data the average insert size first has to be guessed from the data itself , and then each read has to be resized from its 3' end to half of the insert size . Therefore , we used single-end ChIP-seq BAM files as input for MACS2 software ( -g 1 . 35e+08 -m 3 , 50—half-ext—bdg ) [107] and continued with the output Bedgraph files ( see 05-Remapping_Single-End_ChIP-Seq . sh ) . Two of the single-end ChIP-seq datasets mentioned above were treated in a slightly different way: i ) The raw data in Solexa and SCARF formats [67] were converted to FASTQ as detailed in the 06-Convert_Solexa_and_SCARF . sh file; ii ) The color space data from ABI Solid platform [62] were aligned with Bowtie v1 . 2 . 2 ( -C—best ) . Otherwise these special datasets were processed as described in 05-Remapping_Single-End_ChIP-Seq_and_MNase-Seq . sh . In addition , to investigate the expression level of genes containing exonic TSS of interest , we converted the original tracks from an Arabidopsis GRO-seq study to Bedgraph format [77] ( see the 07-Convert_GRO-Seq_data . sh ) . Finally , all the ChIP-seq , GRO-seq and MNase-seq Bedgraph files were used as input for the custom boxplot and metagene plotting pipelines ( see 08-Boxplot_and_metagene_pipeline . R in the code repository ) . The control intervals shown on boxplots and metagenes were produced by choosing random positions in exons of two gene sets: i ) Genes with basal TSS ( 9221 genes ) ; ii ) Genes with fact-specific TSS ( 5604 genes ) . For GRO-seq plots , we removed control positions located less than 200 bp from gene ends , because plant GRO-seq is known to produce exaggerated signal at pA sites [77] . For RNA-seq data processing , standard Illumina adapters were trimmed from both R1 and R2 by Trim Galore v0 . 4 . 3 ( —paired—Illumina ) . Then the read pairs were aligned to TAIR10 using the STAR aligner v2 . 5 . 2b in the local mode ( —outSAMmultNmax 1—alignEndsType Local ) . The output SAM files were sorted , filtered for MAPQ≥10 and converted to BAM format using SAMtools v1 . 3 . 1 . Finally , Bedgraph files for visualization in the IGV browser were generated from BAM files using BEDtools v2 . 17 . 0 ( -bg -split ) . | Genes represent DNA elements that are transcribed into mRNA . However , the position where transcription actually starts can be dynamically regulated to expand the diversity of RNA isoforms produced from a single gene . Functionally , alternative Transcription Start Sites ( TSSs ) may generate protein isoforms with differing N-terminal regions and distinct cellular functions . In plants , light signaling regulates protein isoforms largely through regulated TSS selection , emphasizing the biological significance of this mechanism . Despite the importance of alternative TSS selection , little is known about the underlying molecular mechanisms . Here , we characterize for the first time how transcription initiation from an upstream promoter represses alternative downstream promoter activity in plants . This repression mechanism is associated with chromatin changes that are required to maintain precise gene expression control . Specific chromatin signatures are established during transcription via dynamic interactions between the transcription machinery and associated factors . The conserved histone chaperone complex FACT is one such factor involved in regulating the chromatin environment along genes during transcription . We find that mutant plants with reduced FACT activity specifically initiate transcription from thousands of intragenic positions , thus expanding RNA isoform diversity . Overall , our study reveals conserved and plant-specific chromatin features associated with the co-transcriptional repression of downstream intragenic TSSs . These findings promise to help inform the molecular mechanism underlying environmentally-triggered TSS regulation in plants . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"biotechnology",
"engineering",
"and",
"technology",
"brassica",
"dna-binding",
"proteins",
"dna",
"transcription",
"plant",
"science",
"model",
"organisms",
"experimental",
"organism",
"systems",
"plant",
"genomics",
"epigenetics",
"plants",
"chromatin",
"arabidopsis",
"thaliana",
"research",
"and",
"analysis",
"methods",
"bioengineering",
"animal",
"studies",
"chromosome",
"biology",
"proteins",
"gene",
"expression",
"plant",
"genetics",
"histones",
"nucleosomes",
"biochemistry",
"eukaryota",
"plant",
"and",
"algal",
"models",
"cell",
"biology",
"phenotypes",
"genetics",
"biology",
"and",
"life",
"sciences",
"genomics",
"plant",
"biotechnology",
"organisms"
] | 2019 | Transcription-driven chromatin repression of Intragenic transcription start sites |
The knowledge of the tertiary structure of RNA loops is important for understanding their functions . In this work we develop an efficient approach named RNApps , specifically designed for predicting the tertiary structure of RNA loops , including hairpin loops , internal loops , and multi-way junction loops . It includes a probabilistic coarse-grained RNA model , an all-atom statistical energy function , a sequential Monte Carlo growth algorithm , and a simulated annealing procedure . The approach is tested with a dataset including nine RNA loops , a 23S ribosomal RNA , and a large dataset containing 876 RNAs . The performance is evaluated and compared with a homology modeling based predictor and an ab initio predictor . It is found that RNApps has comparable performance with the former one and outdoes the latter in terms of structure predictions . The approach holds great promise for accurate and efficient RNA tertiary structure prediction .
RNAs are a type of macromolecule of crucial and versatile biological importance . In addition to their long-discovered functions of carrying genetic information and acting as a part of translation machinery , recently they were found to be able to participate in the regulation of gene expressions and protein synthesis , and to act as scaffolds for higher-order complexes and transmit signals between cells , etc [1–4] . To fully understand the function of RNAs , knowledge of their three-dimensional ( 3D ) structure is often required . Although the most reliable sources of RNA structural information are experimental measurements from X-ray crystallography , nuclear magnetic resonance ( NMR ) spectroscopy , and cryoelectron microscopy , such experiments are costly or technically challenging due to the physical chemical nature of RNAs . As a result , computational prediction of RNA structures provides a valuable alternative source of gaining structural information . Many programs have been developed to assist RNA 3D structure prediction , including YAMMP [5] , NAB [6] , ERNA-3D [7] , MANIP [8] , S2S [9] , FARNA [10] , MC-Fold/MC-Sym [11] , RNA2D3D [12] , iFoldRNA [13 , 14] , NAST [15] , Assemble [16] , HiRE-RNA [17] , FARFAR [18] , RNABuilder [19] , ModeRNA [20] , OxRNA [21] , 3dRNA [22] , a coarse-grained model that includes salt effect [23] , our pk3D [24] , etc . However , this is not a complete list due to the rapid development in the field . The prediction of the tertiary structures of RNA loops , as part of the effort of structure prediction , merits specific attention . First , this is because RNA functions often reside in loop regions and about 46% of nucleotides in an RNA chain remain unpaired , according to Dima and her colleagues’ research [25] . Some people mistake loop regions as unstructured . Some people mistake loop regions as structured . There are stacking interactions between neighboring bases , and other non-canonical contacts . These enthalpies sometimes dominate the entropic components so that order dominates . Other times entropy dominates . Second , the prediction accuracy of the loop regions is usually lower than that of the helical regions , mostly due to the high flexibility of loops and the difficulty in calculating the energy of non-canonical base pairs and base triples frequently observed in loops . Previously , several methods have been developed for loop structure predictions . For example , Das and co-workers developed a deterministic stepwise assembly ( SWA ) method , and with the Rosetta statistical potential this brute-force method either reaches atomic accuracy or exposes flaws in the energy function for a testing set of 15 RNA loops [26] . Liu and Chen designed a set of dinucleotide-based statistical potentials for RNA loops and junctions and combined them with their Vfold model , then they made predictions for the coarse-grained ( CG ) 3D structures of both RNA loops and junctions [27] . One unique advantage of their approach is its ability to go beyond the native structures by accounting for the full free energy landscape , including all the non-native folds . Frellsen and colleagues developed a program BARNACLE ( BAyesian network model of RNA using Circular distributions and maximum Likelihood Estimation ) to remove some important limitations associated with the discrete nature of fragment assembly methods [28] . BARNACLE is a probabilistic RNA model based on dynamic Bayesian network and is able to efficiently sample conformations in a natural , continuous 3D space . It is shown that it captures several key features of RNA structure , and readily generates native-like conformations for 9 out of 10 test structures , solely using coarse-grained base-pairing information . In this work we develop an approach that is specifically designed for predicting the tertiary structures of RNA loops or missing fragments which can be part in helical regions and part in loop regions . The loops can be hairpin loops , internal loops , or multi-way junction loops . The approach includes a probabilistic CG RNA model , a sequential Monte Carlo growth method , a simulated annealing strategy , and a statistical potential used for scoring the generated structural candidates . The probabilistic model is inspired by BARNACLE . However , the original BARNACLE model considers all the seven torsional angles in the RNA chain , as is not always necessary since some dihedral angles are relatively rigid . The advantage of the new model developed here is multifold . First , the probabilistic nature of the model allows a continuous sampling in the conformational space and therefore is able to cover all the relevant conformations , as is difficult for fragment assembly methods or our previous one based on a discrete-state model [24 , 29] . Second , by adjusting the probabilistic functions in the model , we are able to deliberately enhance the sampling in a specific phase space . Third , the coarse-grained model further increases the efficiency , making it possible to deal with large RNAs . We name this new approach as RNApps , short for RNA structure Prediction with a Probabilistic Sampling strategy . We test this approach with a benchmark set including nine loops , a 23S ribosomal RNA , and a non-redundant RNA 3D structure dataset containing 876 RNAs [30] . We also compare the performance of our approach with other predictors .
The atomistic structure of an RNA is defined by seven torsional angles , as illustrated in Fig 1A . However , the usage of such an all-atom model is expensive for structure modeling , particularly when high-throughput structure prediction is needed . In our approach , an RNA molecule is described by a CG model that includes a virtual bond model for the backbone and a reduced model for the nucleobases . The virtual bond model can be traced back to Olson [31] , where the RNA backbone includes only two types of atoms , i . e . , P and C4’ . The model has been used in much work [32–35] and is also used here to describe the RNA backbone . For the representation of the RNA bases , three beads are used , including the atoms N1 and C2 in the purine ( or pyrimidine ) ring , and a virtual bead Bc ( short for base center ) at the geometric center of the six-membered ring containing atoms N1 , C2 , N3 , C4 , C5 , and C6 . In total there are five beads for each nucleotide in the CG model . In the current model all the bond lengths and bond angles are fixed . Five torsional angles are defined based on the CG model , including θ , η , μ , ϕ , and ω , as illustrated in Fig 1B . Note that in the implementation there are actually two sets of such angles used for the purpose of growing RNA chains . The first set is defined in a “forward” way , i . e . , from the 5’-end of the RNA to its 3’-end; and the other is defined in a “reverse” way , i . e . , from the 3’-end to the 5’-end . These two sets are denoted by the subscript “+” and “−” , respectively . More specifically , from the 5’-end to the 3’-end , θ+ is calculated as the torsional angle formed by Pi-C4’i-Pi+1-C4’i+1 , η+ by C4’i-Pi+1-C4’i+1-Pi+2 , μ+ by C4’i-Pi+1-C4’i+1-Bci+1 , ϕ+ by Pi+1-C4’i+1-Bci+1-N1i+1 , and ω+ by C4’i+1-Bci+1-N1i+1-C2i+1; from the 3’-end to the 5’-end , θ- is calculated as the torsional angle formed by C4’i+1-Pi+1-C4’i-Pi , η- by Pi+2-C4’i+1-Pi+1-C4’i , μ- by C4’i+1-Pi+1-C4’i-Bci , ϕ- by Pi+1-C4’i-Bci-N1i , and ω- by C4’i-Bci-N1i-C2i . Once all the torsional angles along the RNA chain are known , the tertiary structure of the RNA molecule can be built in a “growing” way by calculating the coordinates of the to-be-grown nucleotide based on the values of the torsional angles and the previously grown nucleotides . The growing process is realized in both the “forward” way and the “reverse” way , to take advantage of the position constraints as much as possible . In the five torsional angles defined above , only three are independent , i . e . , θ , η , and ϕ . This is because μ is linearly dependent on η ( if grown in a forward direction ) or θ ( if grown reversely ) , and ω is approximately linearly dependent on ϕ , as supported by the statistics given in Fig 2 and S1 Table . The statistics are based on the RNA09 database compiled by Murray and co-workers by applying quality-filtering techniques ( using resolution , crystallographic B factor , and all-atom steric clashes ) to the backbone torsional angle distributions from a large RNA database [36] . Based on the above results the task of growing an RNA structure is reduced to determining the three torsional angles θ , η , and ϕ , and it is done in an iterative and probabilistic way , as illustrated in Fig 3 . For example , if we are growing a loop from the 5’-end to the 3’-end , for a given θ + i , the value of η + i can be obtained probabilistically from the conditional probability p ( η + | θ + = θ + i ) . Similarly , the value of ϕ + i can be calculated from the just obtained η + i and the conditional probability p ( ϕ + | η + = η + i ) . The value of θ + i + 1 of the next nucleotide can be obtained from η + i and p ( θ + | η + = η + i ) , so on and so forth . The loop may also be grown in a reverse direction , i . e . , from the 3’-end to the 5’-end , and the procedure is similar but with a different set of conditional probabilities defined in a reverse way . The conditional probabilities are calculated from the loop structures in the RNA09 database . Before the calculation , we first strip the helices in the structures . If two consecutive base pairs are formed , they are determined to be a helix . Three types of base pairs are considered , including A-U , G-C , and G-U . Take p ( η+|θ+ ) as an example . We first calculate all the ( θ+ , η+ ) pairs in the loops in the database and plot the data points in a two-dimensional surface , as shown in Fig 4A . We then equally divide both torsional angles into K bins and calculate the conditional probability numerically as follows p ( η + ( j ) | θ + ( i ) ) = n i j N i ( i , j = 1 , 2 , … , K ) , ( 1 ) where θ+ ( i ) and η+ ( j ) are the center of the i-th and j-th bins , respectively , Ni is the number of data points with their θ+ values falling into the i-th bin , and nij is the number of data points with their θ+ values falling into the i-th bin and η+ falling into the j-th bin simultaneously . K is set to 72 , and thus the bin width is Δ = 360°/K = 5° . The other conditional probabilities are similarly calculated . The conditional probabilities between η+ and θ+ and between ϕ+ and η+ are shown in Fig 4B and 4C , respectively . The latter are classified into four types according to the nucleotide to which ϕ+ belongs , while the former are not . This is because one ( θ , η ) pair involves two nucleotides and thus there are 16 types of dinucleotides; to classify the data into 16 types will lower the statistical quality since the number of data is limited . Based on the probabilistically generated torsional angles , we build the CG structure of RNA loops/fragments by adding nucleotides one by one on the previously grown nucleotides with a sequential Monte Carlo ( SMC ) method [37] , which is employed to prune the growing tree and bias the conformations to those having low energies and satisfying specific constraints . The constraints may include the excluded volume effect , the requirement of chain connectivity , experimental information , etc . The specific algorithm is similar to that used in our previous works [24 , 29 , 32 , 38–41] , but is improved here by growing the chain in two directions alternatively . Specifically , for a loop sequence X1 X2⋯Xn whose structure is to be predicted , we grow it alternatively in two directions in 3D space from two anchor nucleotides to which this loop is attached . The algorithm also works if there is only one anchor , which happens when the loop is at one end of the RNA . To be more specific , for the i-th nucleotide to be grown , we randomly determine D = 10 possible positions of attaching it to the ( i − 1 ) -th nucleotide ( if grown in a forward way ) by probabilistically generating the corresponding θ , η , and ϕ values based on the algorithm described in Fig 3 and the conditional probabilities exemplified in Fig 4 . In order to increase the success rate of connecting the growing loop to the anchor nucleotide , we exclude the newly grown nucleotide whose P atom is farther from the P atom in the target anchor than a threshold , which is related to the number of nucleotides between these two nucleotides . This early pruning of the unlikely partial chain will also save a lot of computational time . After appending the D random configurations , we obtain a total of N = L × D partial chains , where L is the number of chains before the growth of the i-th nucleotide . If N is greater than a threshold M , we do a resampling procedure by randomly choosing M partial chains out of N with the probability of choosing the m-th one proportional to exp ( −Em/RT1 ) , otherwise we keep all the N chains . Here Em is the statistical energy of the m-th partial chain that will be described later . The temperature T1 is used to control the relative probabilities between the candidates and RT1 is set to 300 RASP energy unit . The above procedure works for a single chain . For internal loops or multi-way junction loops with more than one chain , the procedure is similar . Take three-way junction loop as an example . We first build M conformations for the first chain using the SMC method described above . Next for each in M conformations , we grow D conformations for the first nucleotide in the second chain and perform resampling to select M conformations out of N = M × D ones . We repeat these steps for the rest nucleotides in the second chain and then for the third chain . Finally we get M chains after the whole three-way junction loop is built . The parameters M and D are chosen as a compromise between the accuracy and efficiency of the SMC algorithm . A large value increases the accuracy while deteriorates the efficiency , whereas a small value does the opposite . We use M = 80 and D = 20 for reasonable run time . These values are the default in this work unless otherwise stated . The growth procedure based on sequential Monte Carlo gives M structural candidates with low energies . These structures are further optimized with simulated annealing ( SA ) [42 , 43] . During the SA procedure , we adopt the FRESS method ( fragment regrowth via energy-guided sequential sampling ) developed by Zhang et al . [44] to update the structures . In detail , for each among M structural candidates , In both the SMC and SA procedures , we adopt the Ribonucleic Acids Statistical Potential ( RASP ) developed by Capriotti et al . [46] . In brief , RASP is an all-atom knowledge-based or statistical potential derived from a non-redundant set of 85 RNA structures . The statistical energy is calculated as a function of atom types , distance and sequence separation . It also includes a representation for local and non-local interactions in RNA structures . The threshold of sequence separation used to differentiate these two interactions is optimized with information theory . The base pairing and base stacking interactions are implicitly incorporated in the potential . More details of the RASP potential can be found in the reference [46] . The reason for choosing RASP over the other all-atom energy functions such as AMBER force field is multifold . First , it is a statistical potential and therefore more consistent with the statistical model developed here . Second , according to the authors RASP performs better than ROSETTA FARFAR force field in the selection of accurate models . Third , it is easier to be incorporated into our C++ code . Since RASP is an all-atom statistical potential , the corresponding all-atom structures of the CG conformations generated by our approach need to be reconstructed . The procedure is as follows . Based on the atoms of the nucleobases , namely N1 , C2 , and Bc , the coordinates of the heavy atoms in the nucleobases are calculated with the pre-built templates taken from the RNA09 database . There are four such templates , built for A , U , G , and C , respectively . The coordinates of atoms in the backbone are calculated based on the coordinates of the atoms P , C4’ , and N1 ( in the bases of U and C ) or N9 ( in the bases of A and G ) with a pre-built template . After the reconstruction we check if there are steric clashes between atoms . If any clash is found , the structure is discarded .
Here we test our approach and compare it with two other RNA tertiary structure predictors , i . e . , RLooM [47] and iFoldRNA [13 , 14] , and also with RNAnbds [29] , which was developed previously in our lab . RLooM is based on homology modeling and utilizes template structures extracted from a PDB database . iFoldRNA represents an ab initio way of prediction based on physical interactions and discrete molecular dynamics ( MD ) simulation . These two are selected to represent two very different strategies of structure prediction in the field . The testing set that appears in the original RLooM paper [47] is used for the comparison . This set contains 9 RNA loops with their length ranging from 6 to 17-nt . For RLooM , the RMSD values of the predicted loops with respect to the experimental ones were directly taken from the RLooM paper [47] . For iFoldRNA , they were obtained by feeding to the iFoldRNA-v2 server the whole sequences plus the corresponding base pairing information as constraints [14] . For RNApps , the input includes the whole sequence and the structure other than the loops . The output is the center of the largest cluster calculated from the trajectory of the SA simulation , as described in the Methods section . For a fair comparison , the RMSDs of the reduced backbones ( P , O5’ , C5’ , C4’ , C3’ , and O3’ atoms ) given optimal superposition with the experimental ones are calculated , following the same methodology as in RLooM and iFoldRNA . For the comparison with our previous RNAnbds , the RMSDs of all heavy atoms are used . The results are summarized in Table 1 . In the first part of the table , RNApps is compared with RLooM and iFoldRNA . It can be seen that for this specific testing set , RNApps performs comparably with RLooM or slightly worse in a few cases . This is normal since the homology modeling based predictors usually perform better than those based on energy rules . However , such predictors are sensitive to the testing set . For example , we tested RLooM with 28 fragments of length of 8-nt taken from the RNA pseudoknot 1E95 and found that RLooM failed for all of them . The reason is that there are no homological partners of this RNA in the RLooM database . In contrast , our method guarantees to give a reasonable prediction ( data shown in S1B Fig ) . The comparison with iFoldRNA shows that RNApps gives smaller RMSDs in seven out of nine RNAs , and slightly larger RMSDs in two cases . Therefore , for this testing set taken from the third party ( the RLooM paper ) , our method shows a better performance than iFoldRNA . However , it is also worth mentioning that iFoldRNA is not only a structure predictor , but also can be used for studying folding dynamics as a result of the employed molecular dynamics simulation . We then compare RNApps with our previous RNAnbds . It can be seen that the new approach has a considerably better performance , giving smaller RMSDs for seven out of nine loops and two larger RMSDs . The improvement is mostly attributed to the feature that RNApps is able to sample the structural space in a continuous way , while RNAnbds is based on a discrete-state RNA model and thus unable to reach some structural regions . The different energy functions employed may also contribute . In Table 1 we also give the RMSDs of heavy atoms calculated after optimal superposition of the anchors—the predicted and the experimental loops themselves are not superimposed . These RMSDs characterize how well the relative positions of the loops with respect to their environment are reproduced . It can be seen that the RMSDs are only slightly larger than those calculated after optimal superposition of the loops , indicating that this performance is good . In Fig 5 we superimpose the predicted structures with the experimental ones for the nine RNAs to make a visual comparison . Since the approach actually predicts a structural ensemble , for each sequence we also show the largest cluster to which the predicted structure belongs . It can be seen that for seven out of nine loops , the predicted structures are similar to their corresponding experimental ones . As for the structural feature of the cluster , the trends of the backbones correlate well with each other , while the bases are somehow dynamic if they do not form base pairs . The loops in 1Q9A and 2CKY ( Fig 5F and 5H ) are exceptional . In 1Q9A the concerned loop is known as the bulged-G motif , which is a ubiquitous loop and provides specific recognition sites for proteins and RNAs [48] . More specifically , it is an essential part of the binding site for elongation factors in Escherichia coli 23S ribosomal RNA . In this motif , one strand forms a kink so that the bulged-G can form a base triple by non-canonical interactions , permitting interstrand , but disrupting intrastrand stacking . The kink in one strand , together with the S-turn in the opposite strand , appears to make the motif amphipathic , with one surface more polar and ready for interacting with positive patches on proteins or a divalent metal ion , and the other surface less polar and suitable for making RNA-RNA contacts [48] . The above features indicate that this motif is highly optimized for binding its partners , which should be considered when making structure predictions . The RNA 2CKY , which is a riboswitch specifically recognizing the thiamine pyrophosphate , has a similar situation [49] . The loop shown in Fig 5H is in a highly distorted region that binds its ligand . These results suggest that , the consideration of the binding partners , if they have , is necessary for a correct prediction of the tertiary structures of RNAs . Our approach can also be used for reconstructing a missing RNA fragment , in addition to a single loop . The fragment can be a mixing of helical and loop regions . Here we show its performance with a 23S ribosomal RNA ( PDB ID 1FFK ) [50] consisting of about 3 , 000 nucleotides . In total we make 105 tests , and in each test we delete one fragment of length N and then reconstruct its tertiary structure , pretending that the fragment is missing in the RNA structure . The fragments are chosen every S nucleotides along the sequence , where S is set to 25 so that the fragments are uniformly distributed in the whole RNA . The length N can be 5 , 8 , and 10 , and the percentage of nucleotides in the helical region in the 105 fragments is 36% , 39% , and 39% , respectively . According to Fig 6 , the percentage of RMSDs smaller than 4 Å is 76 . 2% , 71 . 4% , and 56 . 2% for the fragments of length 5 , 8 , and 10 , respectively . The average RMSDs in the three cases are 2 . 55 Å , 3 . 23 Å , and 3 . 97 Å , respectively . We further test our approach with a much larger dataset—RNA 3D Hub [30] . It contains 876 non-redundant RNAs ( release ID 1 . 89 ) , including all types of base-pairing , base-stacking , and base-backbone interactions . Out of this dataset we select 1768 hairpin loops , 1649 internal loops , 230 three-way junction loops and 113 four-way junction loops . We then delete these loops from the molecules and reconstruct them following the procedure described in the Methods section . For each loop , the structure at the center of the largest cluster is taken as the predicted structure . In Fig 7 we show the RMSDs of the predicted structures with respect to the experimental ones as a function of loop length . Overall , for hairpin loops , internal loops , and three-way junction loops , the RMSDs increase gradually as the loop length increases . The average RMSDs are around 4 Å for hairpin loops of length smaller than 10-nt and increase from 5 to 9 Å for loops of length from 10- to 20-nt . For even longer loops , the average RMSDs are around 11 Å . For internal and three-way junction loops , the performance is better than that for hairpin loops , as may be attributed to the increased number of anchor nucleotides in these cases . For four-way junction loops , the average RMSDs undergo large fluctuations , due to the relatively smaller number of such loops in the dataset . The average RMSDs generally range from 3 Å to 6 Å for loops of length smaller than 20-nt . In Fig 8 we show two examples for each one of four cases for visual comparison—one example has a medium loop length ( 9-nt ) and one has a long loop length ( 17-nt ) . The loops with their RMSDs close to the average ( marked by the red dots near the red lines in Fig 7 ) are selected , so that they are most representative . For the eight cases shown here , six loops are predicted with a reasonable accuracy , with RMSDs around 4 Å . Visual inspection of the figures shows that , the backbones of the predicted structures match those of the experimental ones , while the bases are somehow dynamic ( e . g . , Fig 8A and 8C ) . Two loops are less well predicted . The loop shown in Fig 8B is an essential part of the binding site for elongation factors in rat 28S rRNA . The RMSD between the predicted structure and the experimental one is 6 . 46 Å . This relatively large value is due to the failure of reconstructing the bulge-G motif , similar to the case discussed in Fig 5F . The loop of length 17-nt shown in Fig 8H belongs to a four-way junction , locating at the surface of a large ribosomal RNA . The loop is formed by four sub-loops of length 2- , 3- , 3- , and 9-nt , respectively . The RMSD between the predicted structure and the experimental one is 5 . 17 Å . In the experimental structure , the longest sub-loop forms little interaction with the other three ones and extends out into the solvent . The prediction well reproduced a conformation with this kind of characteristics . The largest deviation between prediction and experiment is in the 2-nt sub-loop G612-A613 and 3-nt sub-loop G647-A648-G649 , where A613 and A648 are both in a sharp turn and form a stacking interaction with A668 and A638 , respectively .
In this work we develop an approach named RNApps specifically designed for loop structure prediction . The approach includes a probabilistic coarse-grained RNA model , a sequential Monte Carlo growth algorithm , a simulated annealing procedure and an all-atom statistical energy function . We tested the approach with a set of nine RNA loops , a 23S ribosomal RNA , and a large RNA dataset containing 876 RNAs ( RNA 3D Hub , release ID 1 . 89 ) . For the testing set including nine RNA loops , six loops can be predicted with good accuracy ( RMSD < 2 . 5 Å ) , one loop has an RMSD of 3 . 01 Å and two have RMSDs around 6-7 Å . We compared the results with a homology modeling based predictor RLooM and an ab initio predictor iFoldRNA . It was found that RNApps performs comparably with RLooM while considerably better than iFoldRNA . However , we also noted that RLooM cannot guarantee a return of valid structures for some targets , due to the lack of their homology information in the database . In contrast , RNApps and iFoldRNA guarantee a result , and iFoldRNA can also be used for studying the folding dynamics . The tests with a ribosomal RNA showed that the average RMSD is 2 . 55 Å , 3 . 23 Å , and 3 . 97 Å for the rebuilt fragments of length 5- , 8- , and 10-nt , respectively . The tests with RNA 3D Hub showed that the average RMSDs for hairpin loops are around 4 Å and increase slightly with the loop length . The performance for internal loops , three-way and four-way junction loops is even better than that for hairpin loops , mostly due to the increased number of anchor nucleotides . Further analysis showed that although most RNA loops are predicted with good accuracy , some ones with non-canonical base pairs , base triples , or rare torsional angles are reproduced with a lower quality . We note that our approach in it current form does not perform considerably better than the other predictors and even slightly worse than the homology modeling based one . However , our approach has many unique and promising features . First , it is not only designed for predicting structures of hairpin loops , but also for internal loops , three-way and four-way junction loops and even complex cases . One motivation is that many predictors have been designed to predict the relative position and orientation of the helices of an RNA molecule , however , less attention was paid to the construction of the loops connecting the helices . This makes our work necessary . Second , the efficiency of our approach is high . With the parameters used in this work , a prediction of loop with medium length takes several minutes without SA optimization or several tens of minutes with SA optimization . For example , the computational time for the longest loop ( 26-nt ) in three- and four-way junctions is approximately one hour . This efficiency is much higher than that of most MD based algorithms and the approach is free from the problems associated with homology modeling based methods . The efficiency will be further improved by optimizing the SMC and SA algorithms . Third , the probabilistic and continuous nature of the approach guarantees the sampling of all the relevant phase space in principle , and allows a dynamic adjustment between accuracy and efficiency , which can be determined by users based on their own computational capacity . Fourth , the SMC framework of the approach makes the incorporation of constraints very easy . The constraints may be the experimental information of atomic distance , base pairs or base stacking , or information from users’ experience . With the introducing of such information , the sampling space can be greatly reduced and both accuracy and efficiency will be significantly improved . The way of incorporation of constraints into the SMC framework can be found in previous work [41] . We believe our approach is useful for predicting the tertiary structure of RNA loops . We also noticed that there are two recent works in protein loop predictions that are similar to ours . In the first one , Tang et al . developed an approach named DiSGro based on sequential Monte Carlo method [51] , which is the same as ours . With this approach , they are able to efficiently generate high quality protein loop conformations . The average minimum global backbone RMSD for 1 , 000 conformations of 12-residue loops is 1 . 53 Å , with a lowest energy RMSD of 2 . 99 Å , and an average ensemble RMSD of 5 . 23 Å . In the second work , the same authors upgraded their approach for the prediction of conformations of multiple interacting loops in proteins [52] . For the most challenging target proteins with four loops , the average RMSD of the lowest energy conformations is 2 . 3 Å . One novel feature of the approach is the simultaneous construction of multiple loops , making it less likely to over-sample conformations in certain local energy minima . This idea is naturally compatible with the framework of our approach and can be easily incorporated into it . We believe their approaches and ours can borrow ideas from each other and then improve the performance of both . | RNA is an important and versatile macromolecule participating in a variety of biological processes . In addition to experimental approaches , computational prediction of 3D structure of RNAs and loops is an alternative and important source of gaining structure information and insights into their functions . The prediction of RNA loop structures is of particular interest since RNA functions often reside in the loop regions and about 46% of nucleotides in an RNA chain remain unpaired . For this purpose , we develop an approach RNApps based on a probabilistic coarse-grained RNA model . The probabilistic nature of the model , together with a sequential Monte Carlo ( SMC ) growth algorithm , allows a natural and continuous sampling of structures in 3D space , making the approach unique . The coarse-graining nature of the model further increases the efficiency . Here we test this new approach with various types of loops , including hairpin loops , internal loops , and multi-way junction loops , and make comparisons with other structure predictors . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"particle",
"physics",
"annealing",
"(genetics)",
"rna",
"structure",
"prediction",
"atoms",
"mathematics",
"statistics",
"(mathematics)",
"protein",
"structure",
"prediction",
"protein",
"structure",
"composite",
"particles",
"cellular",
"structures",
"and",
"organelles",
"rna",
"annealing",
"research",
"and",
"analysis",
"methods",
"rna",
"structure",
"proteins",
"mathematical",
"and",
"statistical",
"techniques",
"monte",
"carlo",
"method",
"biophysics",
"molecular",
"biology",
"ribosomes",
"physics",
"biochemistry",
"rna",
"rna",
"folding",
"ribosomal",
"rna",
"cell",
"biology",
"nucleic",
"acids",
"nucleic",
"acid",
"thermodynamics",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"non-coding",
"rna",
"statistical",
"methods",
"macromolecular",
"structure",
"analysis"
] | 2016 | Structure Prediction of RNA Loops with a Probabilistic Approach |
Due to toxicity and compliance issues and the emergence of resistance to current medications new drugs for the treatment of Human African Trypanosomiasis are needed . A potential approach to developing novel anti-trypanosomal drugs is by inhibition of the 6-oxopurine salvage pathways which synthesise the nucleoside monophosphates required for DNA/RNA production . This is in view of the fact that trypanosomes lack the machinery for de novo synthesis of the purine ring . To provide validation for this approach as a drug target , we have RNAi silenced the three 6-oxopurine phosphoribosyltransferase ( PRTase ) isoforms in the infectious stage of Trypanosoma brucei demonstrating that the combined activity of these enzymes is critical for the parasites’ viability . Furthermore , we have determined crystal structures of two of these isoforms in complex with several acyclic nucleoside phosphonates ( ANPs ) , a class of compound previously shown to inhibit 6-oxopurine PRTases from several species including Plasmodium falciparum . The most potent of these compounds have Ki values as low as 60 nM , and IC50 values in cell based assays as low as 4 μM . This data provides a solid platform for further investigations into the use of this pathway as a target for anti-trypanosomal drug discovery .
Trypanosoma brucei is the etiological agent of Human African Trypanosomiasis ( HAT ) also known as “sleeping sickness” . HAT is a neglected disease that mainly affects Sub-Saharan countries , with ~70 million people at risk of infection [1–3] . The metacyclic trypomastigote form of T . brucei is primarily transmitted to humans by the bite of an infected tsetse fly . Once inside the mammalian host the parasite invades the bloodstream and lymph system . At this stage , the human host is mainly asymptomatic , a period that can last for months and up to years . However , when T . brucei crosses the blood-brain barrier , a degenerative neurological breakdown occurs characterized by continuous sleep-wake patterns . In the last stage , the human host falls into a coma and at this point the disease is fatal . A handful of drugs ( pentamidine , eflornithine , nifurtimox , melarsoprol and suramin ) is available to treat HAT at the different stages of the disease ( e . g . haemo-lymphatic and brain infections ) . However , they are far from perfect drugs due to their low selectivity , high cost of production , high levels of toxicity , adverse side-effects and can have less than ideal routes of administration [4] . The increasing occurrence of resistance to these drugs is also of growing concern [5 , 6] . Therefore , new and more effective drugs that can be co-administered or replace the current treatments for this disease are urgently needed . The complete sequencing of the T . brucei genome has identified some differences in metabolism between the parasite and the human host , which could lead to the discovery of new drug treatments [7 , 8] . One significant difference between the human host and this parasite is in the respective enzymes they have available for the synthesis of the nucleoside monophosphates required for the production of their DNA and RNA . In T . brucei there is a complete reliance on the purine salvage pathways , obtaining the purine bases from the host , whereas in humans both the de novo pathway and the salvage pathways are present [9–12] . The trypanosome purine salvage pathway is comprised of several salvage enzymes ( i . e . nucleoside hydrolases , 6-oxopurine PRTases , adenine PRTase , adenosine kinase ) and interconversion enzymes ( i . e . AMP deaminase , adenylosuccinate lyase ( ADSL ) , adenylosuccinate synthetase ( ADSS ) , guanine deaminase , GMP synthase ( GMPS ) , GMP reductase and inosine-5´-monophopshate dehydrogenase ) ( Fig 1 ) . Importantly , there are constitutive differences between humans and T . brucei within the salvage pathways themselves . For example , T . brucei has three 6-oxopurine PRTase isoforms whereas there is only one 6-oxopurine PRTase in humans . These enzymes catalyze the transfer of the ribose 5'-phosphate moiety from 5-phospho-α-D-ribosyl-1-pyrophosphate ( PRib-PP ) to their respective purine bases ( hypoxanthine , guanine or xanthine ) ( Fig 2A ) . Thus , due to the central importance of these proteins in RNA/DNA synthesis , inhibition of these enzymes is suggested to be a viable approach to antiprotozoal drug therapy [13–16] . Acyclic nucleoside phosphonates ( ANPs ) are a family of antiviral compounds that have been shown to also inhibit plasmodial and mycobacterial 6-oxopurine PRTases [18–20] . The basic structure of these compounds consists of a nucleobase connected to a phosphonate group by a variety of chemical linkers . In some ANPs , this linker consists solely of carbon atoms while others have oxygen or nitrogen atom ( s ) to replace the carbon atoms [16 , 21–23] ( Fig 2B ) . More elaborate ANPs have a functional attachment at the position one , two or three atoms along the linker ( Fig 2C and 2D ) . A crucial property of the ANPs is the presence of a stable CH2-P bond in the phosphonate moiety [19 , 24] ( Fig 2B , 2C and 2D ) . This means such compounds are not susceptible to hydrolysis within the cell of the host or pathogen [19 , 21 , 22 , 25–28] . However , the ANPs possess negative charges on the phosphonic oxygen atoms , restricting cell permeability . In the case of the bisphosphonates ( Fig 2C and 2D ) , transport across the cell membranes is even more problematic [29] . To overcome this issue , the cell permeability was enhanced by attaching lipophilic or hydrophobic groups to the phosphorus atom either by a phosphoramidate or ester bond . Once inside the cell , these groups are hydrolyzed to produce the active parent compound [29–31] . Here , we show that two of the three T . brucei isoforms of the 6-oxopurine PRTases are identified as HGPRTs as a reflection of their substrate preference for hypoxanthine and guanine and the third isoform is identified as HGXPRT reflecting that it uses all three naturally occurring bases . We have performed double RNAi silencing of the HGPRTs and HGXPRT to show that the 6-oxopurine PRTases are crucial for the survival of T . brucei in vitro . We have also determined the crystal structures of two of the 6-oxopurine PRTase isoforms in complex with several ANPs and showed that prodrugs of these inhibitors possess antitrypanosomal activity in cell based assays .
The T . b . brucei Lister 427 procyclic form ( PF ) was cultivated in SDM-79 medium containing 10% FBS at 27°C [32] . The bloodstream form ( BF ) T . b . brucei Lister 427 and single marker ( SM ) strains were cultivated in HMI-9 medium and 10% FBS at 37°C in a humidified atmosphere at 5% CO2 [32] . The HMI-9 medium contains hypoxanthine at 1 mM concentration , while other purines are present in negligible amounts [33] . The SM cell line constitutively expresses the ectopic T7 RNA polymerase and tetracycline repressor and was used for the tetracycline inducible expression of dsRNA and the v5-tagged proteins . The 431 bp and 481 bp fragments of hgprt-I ( Tb927 . 10 . 1400 ) and hgxprt ( Tb927 . 10 . 1390 ) open reading frames , respectively , were PCR amplified from T . brucei BF427 genomic DNA with the oligonucleotides listed in S1 Supplementary File and cloned into the p2T7-177 plasmid [34] utilizing BamHI , HindIII or XhoI restriction sites inherent in the primers . For the inducible expression of v5-tagged HGPRT-I , HGPRT-II and HGXPRT , the coding sequences of the respective genes were PCR amplified using oligonucleotides listed in S1 Supplementary File . The obtained PCR fragments were cloned into the pT7_N-term-v5 vector derived from pT7 vector [35] using BamHI and XbaI restriction enzymes in order to maintain the open reading frames of the N-terminal 3 x v5 tag and the respective genes . The p2T7-177 and pT7_N-term-v5 plasmids containing fragments or open reading frames , respectively , of hgprt-I , hgprt-II and hgxprt genes were linearized with NotI enzyme and transfected into the BF SM cell line as described previously . The induction of dsRNA and v5-tagged protein expression was triggered by the addition of 1 μg/ml of tetracycline into the medium . Cell densities were measured using the Z2 Cell Counter ( Beckman Coulter Inc . ) . Throughout the analyses , BF cells were maintained in the exponential mid-log growth phase ( between 1x105 to 1x106 cells/ml ) . For immunofluorescence 1 × 108 of BF trypanosomes were washed with PBS supplemented with 10 g/l glucose , and spread onto slides coated with polylysine ( 100 μg/ml; Sigma ) . The cells were fixed with 3 . 7% formaldehyde in phosphate-buffered saline ( PBS ) , washed with PBS , and permeabilized with 0 . 1% triton X-100 in PBS . The cells were then washed with 1xPBS-T ( 0 . 05% Tween ) . After blocking with 5 . 5% FBS , the respective slides were incubated for 1 hr in PBS-T plus 3% BSA with the following primary antibodies: monoclonal anti-v5 ( 1:200 , Invitrogen ) , anti-enolase ( 1:400 , [36] ) and anti-hexokinase ( 1:400 , a gift from Paul Michels ) . After washes , the slides were incubated in the dark for 1 hr with the following secondary antibodies: goat anti-rabbit IgG ( H + L ) Texas Red conjugate ( 1:400 , Life Technologies ) or goat anti-mouse IgG ( H + L ) FITC conjugate ( 1:400 , Sigma ) . The slides were washed and mounted in Vectashield containing 1 . 5 μg/ml DAPI ( Life Technologies ) . Protein samples were separated on SDS-PAGE , blotted onto a PVDF membrane ( PALL ) and probed with the appropriate monoclonal ( mAb ) or polyclonal ( pAb ) antibodies . This was followed by incubation with a secondary HRP-conjugated anti-rabbit or anti-mouse antibody ( 1:2000 , BioRad ) . Proteins were visualized using the Clarity Western ECl Substrate ( Biorad ) on a ChemiDoc instrument ( BioRad ) . ImageLab software version 4 . 1 ( Bio-Rad ) was used for image acquisition and densitometric analysis of the blots . The PageRuler prestained protein standard ( ThermoFisher Scientific ) was used to determine the size of detected bands . Primary antibodies used in this study were: mAb anti-v5 epitope tag ( 1:2000 , ThermoFisher Scientific ) , mAb anti-mtHsp70 ( 1:2000 , [37] ) , pAb anti-enolase ( 1:1000 ) , pAb anti-hexokinase ( 1:1000 ) , pAb anti- HGPRT-I ( 1:100 , this work ) and pAb anti-HGXPRT ( 1:1000 , this work ) . Whole cell lysates ( WCLs ) were prepared from T . brucei PF427 and BF427 strains . SoTe/digitonin fractionation was performed to obtain cytosolic and organellar fractions . Briefly , 1x108 cells were harvested by centrifugation , washed in PBS-G , resuspended in 500 μl SoTE ( 0 . 6 M Sorbitol , 2 mM EDTA , 20 mM Tris-HCl pH 7 . 5 ) and lysed with 500 μl SoTE containing 0 . 03% digitonin . The cells were then incubated on ice for five minutes followed by centrifugation ( 4°C , 4500 g , 3 min ) . WCL , supernatant representing the cytosolic fraction ( CYTO ) and pellet representing the organellar membrane fraction ( ORG ) were fractionated by SDS-PAGE and analysed by immunoblotting . An Alamar Blue Assay was performed in a 96-well plate format to screen for sensitivity of T . brucei BF427 strain to ANPs . Parasites at 5×103 cells/ml were incubated with different drug concentrations ( two-fold serial dilutions ) at 200 μl final volume of HMI-9 with 10% FBS / well for 48 hours at 37°C and 5% CO2 . Afterward , Alamar Blue reagent ( 20 μl ) was added to each well and cells were incubated for an additional 12 hours under the same conditions . The fluorescence signal was quantified using a Tecan Infinite M200 plate reader ( Excitation 560 nm–Emission 590 nm ) . The EC50 values were calculated using non-linear regression in GraphPad Prism5 . 5 × 108 cells were washed in PBS and lysed in immunoprecipitation ( IP150 ) buffer ( Tris 10 mM pH 7 . 2 , 10 mM MgCl2 , KCl 150 mM , 1% Triton , and protease inhibitor cocktail ( Roche Applied Science ) for 1 hour on ice . The v5-tagged HGPRT-I , -II and HGXPRT complexes were purified using magnetic beads ( Dynabeads M-280 , sheep anti-mouse IgG , ThermoFisher Scientific ) charged with monoclonal anti-v5 antibody . The lysate was incubated with the charged beads for 2 h ( 4°C ) . The beads were then washed three times in IP150 buffer before the bound protein complexes were released by the addition of SDS-PAGE loading buffer and boiling ( 97°C , 10 min ) . Purified HGPRT-I , HGPRT-II and HGXPRT were dialyzed to 20 mM HEPES , 1 mM EDTA , at pH 8 . 0 and diluted to 0 . 1 mg/ml . A non-crosslinked sample was set aside and dimethylsuberimidate ( DMS ) was added from fresh 20 mg/ml stock solution in water to give a final concentration of 1 mg/ml . Crosslinking was stopped at different time points by mixing samples with SDS-PAGE loading buffer . The crosslinked proteins were resolved by SDS-PAGE on 15% gels and stained by Coomassie Brilliant Blue dye . The inhibitors and their prodrugs described here were synthesized as described previously [21 , 23 , 28 , 29] . The T . brucei hgprt-I , hgprt-II and hgxprt genes were PCR amplified using oligonucleotides listed in S1 Supplementary File , digested with BamHI and XhoI enzymes and ligated into the pSKB3 expression vector . The pSKB3 vector carries a 6x His-tag and an AcTEV cleavage site upstream of BamHI restriction site . A BamHI restriction site was introduced at the 5' end of the respective genes instead of the initiation methionine codon to maintain the 6x His-tag and AcTEV recognition motif in frame with the genes . The verified plasmids were transformed into the One Shot BL21 ( DE3 ) Chemically Competent E . coli cells ( Life Technologies ) . The bacterial cells were grown in LB medium until the OD600 reached ~ 0 . 4 . The expression of the recombinant proteins was induced by the addition of 1 mM isopropyl β-D-1-thiogalactopyranoside ( IPTG ) . After three hours of induction the cells were harvested , washed in PBS and resuspended in the STE buffer ( 50 mM Tris-HCl pH 8 . 0 , 150 mM NaCl , 1 mM EDTA ) supplemented with the cOmplete EDTA-free protease inhibitor cocktail ( Roche , Switzerland ) . Cells were lysed by 1% Triton X-100 and lysozyme ( 1 mg/ml ) for 30 min at 4°C , followed by DNase I treatment ( 5 μg/ml , 30 min rotation at 4°C ) in the presence of MgCl2 ( 30 mM ) . The lysates were then centrifuged and filtered before loading onto the Ni-NTA column ( 1ml Histrap HP , GE Healthcare ) . The 6xhis-tagged recombinant proteins were purified under native conditions using the ÄKTA prime plus instrument accordingly to the product manual . Elution fractions were analyzed by SDS-PAGE , and those containing HGPRT-I , -II or HGXPRT were pooled and dialyzed against 50 mM Tris-HCl , pH 8 . 0 , 150 mM NaCl , 10% glycerol . Protein concentrations were determined using Bradford protein assay ( BioRad ) with BSA as standards . The HGPRT-I and HGXPRT recombinant proteins were sent for antibody production to Davids Biotechnology ( Regensburg , Germany ) . Recombinant HGPRT-I , HGPRT-II and HGXPRT were subsequently used in all in vitro assays . The activity of HGPRT-I , HGPRT-II and HGXPRT was measured by a continuous spectrophotometric assay measuring the conversion of hypoxanthine , guanine and xanthine to IMP , GMP and XMP , at 245 nm , 257 . 5 nm and 255 nm , respectively [38] . The concentration of enzyme in the assay ( varied from 75–200 μg/ml for the two enzymes ) and the concentration of PRib-PP was kept at 1 mM . The reaction was carried out using a 1 . 0 cm path length with a UV Visible spectrometer 1601 ( Shimadzu ) in the reaction buffer ( 0 . 1 M Tris , pH 8 . 4; 0 . 11 M MgCl2; at 25°C ) . The initial velocity ( V0 [μM/min] ) was calculated using Beer-Lambert law for different concentration of the purine substrate molecules: A = Δε*L*c where A is measured absorbance , L is length of cuvette , Δε is constant ( 5817 M-1*cm-1 for guanine , 2283 M-1*cm-1 for hypoxanthine and 4685 M-1*cm-1 for xanthine [39] and c is concentration of the product formed in one minute ( V0 ) . The Km , Vmax and kcat values were calculated using GraphPad Prism5 according to Michaelis-Menten kinetics . The Ki values for HGPRT-I , HGPRT-II and HGXPRT were determined in buffer containing 0 . 1 M Tris , pH 8 . 4; 0 . 11 M MgCl2; at 25°C . The concentrations of inhibitors in the assay ranged from 1 nM to 100 μM , depending on the steady-state kinetics of the respective enzyme at that concentration of inhibitor . The reaction was preceded by 1 minute incubation of the respective enzyme with the given inhibitor . The Ki values were calculated using non-linear regression in GraphPad Prism5 . Expression and purification of HGPRT-I was performed following a procedure described previously [16] . Expression and purification of HGXPRT was carried out as above but with some minor modifications . Induction of expression was at 37°C for 3 hours upon the addition of 0 . 5 M IPTG . Cells containing the expressed protein were centrifuged at 4°C at 4000 x g for 15 minutes . The cell pellet was washed with 1xPBS and centrifuged a second time at 4°C at 4000 x g for 10 minutes . It was then stored at -80°C until purification . The cell pellet was then re-suspended in 50 mM Tris-HCl , 150 mM NaCl , 1 mM EDTA , 1% Triton X-100 , pH 8 in the ratio of 39 mL of buffer to 1 g of pellet . Cells were lysed by the addition of lysozyme ( 1 mg per ml of cells ) in the presence of protease inhibitor cocktail ( Roche ) ( 1 tablet per 50 mL of cells ) and incubated on ice for 30 min . The cells were then sonicated for four cycles for 30 seconds at 20% power with 1 minute between cycles . 1 M MgCl2 and DNase at 50 μg/ml of cells were then added prior to incubation at 4°C for 10 minutes . The soluble fraction was separated by centrifugation at 4°C at 10000 x g . The recombinant enzyme was purified using a Ni2+ chromatography column ( Profinity IMAC resin ) at 4°C . The elution buffer was 50 mM Tris-HCl , 500 mM NaCl , 30 mM imidazole , 10% glycerol , pH 8 . The eluent was then dialyzed into 50 mM Tris-HCl , 12 mM MgCl2 , 10% glycerol , pH 8 at 4°C overnight . For crystallographic studies , the enzyme was concentrated to 10 mg/mL using an Amicon concentrator ( model 12–76 psi ) and an Amicon Ultra-15 centrifugal device . All samples were subsequently stored at −80°C . Sample purity was assessed by ( 12% ) SDS-PAGE . A Direct Detect spectrometer used to determine protein concentration . This method measures the absorbance due to the presence of amide bonds in the polypeptide at A220 . Initial screening of crystallization conditions was performed by the hanging drop vapor-diffusion method using PACT premier , INDEX HT ( Hampton research ) , Proplex , PEG ion crystallization screens providing the lead crystals for optimization . Crystals for diffraction studies were obtained by incubating HGPRT-I or HGXPRT with the different ANPs on ice for 15 min . 1 μL of the HGPRT-I . ANP complex or HGXPRT . ANP complex was then mixed with 1 μL of well solution . For HGPRT-I the enzyme concentration was 28 mg/ml and the inhibitor concentration was 1 . 3 mM . The reservoir solution was 25% PEG 3350 , 0 . 2 M lithium sulfate and 0 . 1 M Bis-Tris with the pH between 5 and 7 . For HGXPRT the enzyme concentration was 5 mg/ml and the inhibitor concentration was 1 . 3 mM . The reservoir solution was 25% PEG 3350 , 0 . 2 M lithium sulfate and 0 . 1 M Bis-Tris with the pH 5 . Prior to cryocooling , all crystals were transferred to a cryoprotectant solution containing the reservoir , 20% glycerol and 2 mM of each ligand . Crystals were subsequently transported to the Australian Synchrotron and robotically placed in the cryostream ( 100 K ) on beamline MX2 for HGPRT-I . ANPs complexes and MX1 for HGXPRT . ANPs complexes . X-ray data were collected remotely by BLU-ICE [40] . Data were , integrated , scaled and merged using XDS [41] . The structure of T . brucei HGPRT-I in complex with single chain ANP ( PDB code 5JSQ ) was used as starting model for molecular replacement for HGPRT . The structure of Trypanosoma cruzi HGPRT in complex with PRib-PP and HPP ( PDB code 1TC2 ) was used as starting model for molecular replacement for HGXPRT , which was then phased using the program PHASER [42] . Subsequent refinement and model building was performed with PHENIX 1 . 7 . 3 [43] and COOT 0 . 8 . 2 [44] . The coordinates and structure factors for all six complexes have been deposited in the protein data bank . The access codes for HGPRT . 6 , HGPRT . 7 , HGPRT . 1 HGPRT . 2 , HGXPRT . 7 , HGXPRT . 2 are 6APS , 6APT , 6APU , 6APV , 6AQO and 6AR9 , respectively .
T . brucei genome [7] is annotated to contain three putative HGPRT genes: Tb927 . 10 . 1400 ( HGPRT-I ) , Tb927 . 10 . 1470 ( HGPRT-II ) and Tb927 . 10 . 1390 ( HGPRT-III ) . The genes for hgprt-I and hgprt-II are almost identical sharing 98% identity at the nucleotide level . At the amino acid level the HGPRT-I and HGPRT-II proteins differ at amino acid positions 49 , 54 , 206 , 207 and 210 . The last three amino acids are located at the C-terminus . Hence , HGPRT-I ends with the GEAKR sequence while HGPRT-II ends with VKAKL . The tripeptide , AKL , follows the highly conserved consensus sequence S/A-K/R-L/M for peroxisomal targeting and indeed this protein is predicted to be localized in peroxisomes by the PTS1 Predictor programme . The gene for hgprt-III differs significantly from the hgprt-I and -II isoforms sharing only 56% identity at the nucleotide level and 35% identity at the amino acid level . The product of this gene uses all three naturally occurring 6-oxopurines ( i . e . guanine , hypoxanthine and xanthine , see below ) . Therefore , we identify this isoform as HGXPRT . To explore the specificity of HGPRT-I , HGPRT-II and HGXPRT steady-state kinetic constants were determined for the three naturally occurring bases ( Table 1 ) . HGXPRT catalyzes the synthesis of GMP , IMP and XMP with kcat values of 1 . 36 , 2 . 4 and 3 . 66 s-1 , respectively . These rates are comparable with those for Leishmania donovani HGPRT and XPRT [45] and P . falciparum HGXPRT [38] . HGPRT-I and HGPRT-II catalyze the conversion of guanine and hypoxanthine at much slower rate ( Table 1 ) and show no activity for xanthine at the concentrations tested confirming the annotation of these enzymes as HGPRTs . Considering that HGPRT-I and HGPRT-II are almost identical at the amino acid level , the Km and kcat values differ significantly for the tested purine nucleobases ( Table 1 ) . Apart from the three residues in the peroxisome signalling sequence , these two enzymes also differ at amino acid position 49 and 54 , the latter one being in the middle of the PPi binding pocket . In summary , this data shows that all three T . brucei 6-oxopurine PRTases are able to bind hypoxanthine and guanine bases and that HGXPRT is a promiscuous enzyme possessing affinity for all three naturally occurring substrates with no strong preference for either one . The surprising promiscuity of HGXPRT in vitro prompted us to investigate the properties of the T . brucei 6-oxopurine PRTases in vivo . To decipher the subcellular localization of HGPRT-I , -II and HGXPRT , the respective genes were N-terminally tagged with a v5 tag and their expression was triggered by adding tetracycline into the medium . Differential cell lysis was performed using digitonin , which at low concentrations , only dissolves the plasma membrane . The subsequent Western analyses of the cytosolic ( CYT ) and organellar ( ORG ) fractions established the purity of the extracted fractions . The cytosolic enolase , the mitochondrial hsp70 and glycosomal hexokinase were confined within their respective subcellular fractions and served as suitable controls . Notably , the v5-tagged HGPRT-II and HGXPRT were detected in the organellar fraction while HGPRT-I was present in the cytosolic fraction ( Fig 3A ) . To determine if HGPRT-II and HGXPRT are localized to the mitochondrion or in the glycosome , the cells expressing the v5-tagged proteins were fixed and the respective enzymes were detected by immunostaining using an anti-v5 antibody ( Fig 3B ) . Direct immunofluorescence showed that HGPRT-II and HGXPRT co-localize with hexokinase , indicating that these isoforms are in the glycosome . This is in agreement with the presence of the PTS1 tripeptide motif at the C-terminal end of these two proteins . Localization of HGPRT-I was confirmed to be cytosolic as the v5 signal co-localized with the cytosolic marker , enolase ( Fig 3B ) . Since the purine availability differs between the culture media used to grow PF and BF T . brucei cells , we examined the expression levels of the three different isoforms of 6-oxopurine PRTases in the two life stages . Whole cells lysates from 107 , 5x106 and 106 cells were fractionated on SDS-PAGE . The steady-state abundance of HGPRT-I , -II and HGXPRT was detected using Western blot analysis with polyclonal antiserums against HGXPRT and HGPRT-I . The antiserum raised against HGPRT-I also cross-reacts with HGPRT-II as the amino acid sequences of these two isoforms are virtually identical . While expression of HGXPRT appears to be unaffected in the two life cycle stages , the steady state abundance of HGPRT-I and–II was significantly decreased ( approximately by 40–50% ) in BF cells ( Fig 4A ) . To distinguish between HGPRT-I and HGPRT-II isoforms , the 1x108 PF and BF cells were lysed using a low concentration of digitonin allowing the cytosolic and organellar fractions to be examined using the anti-HGPRT-I serum ( Fig 4B ) . The cytosolic signal for HGPRT-I remained similar between the PF and BF samples , but the signal for the glycosomal HGPRT-II was not detectable in the BF sample suggesting that expression of this protein is down-regulated in the infectious stage of the parasite . In order to investigate if the activity of HGPRT-I and HGXPRT is essential for in vitro growth of BF cells we generated three RNAi cell lines: single knock-down ( SKD ) of HGPRT-I , SKD of HGXPRT and double knock-down ( DKD ) of both HGPRT-I and HGXPRT . It should be noted that RNAi against HGPRT-I also silences the residual expression of HGPRT-II in BF cells as the DNA fragment used for dsRNA expression is identical for these two isoforms . However since the HGPRT-II expression is undetectable in BF cells ( Fig 3B ) we kept the simplified labelling SKD HGPRT-I and DKD HGPRT-I/HGXPRT . No significant growth phenotype was observed for SKD HGPRT-I and SKD HGXPRT cell lines in HMI-9 medium containing 1 mM hypoxanthine ( Fig 5A and 5B ) . The efficiency and specificity of RNAi was verified by Western blot using anti-HGPRT-I and anti-HGXPRT serums . Upon RNAi HGPRT-I silencing virtually none of this protein remained after two days , while the levels of HGXPRT remained largely unaffected ( Fig 5A , bottom panel ) . In the flipped experiment , where HGXPRT was silenced by RNAi induction no signal for this enzyme was detected by the second day , while the levels of HGPRT-I remained unchanged ( Fig 5B , bottom panel ) . These results show that HGPRT-I and HGXPRT possess interchangeable roles in hypoxanthine salvage and they are in full agreement with the enzyme kinetics that show both HGPRT-I and HGXPRT are able to efficiently convert hypoxanthine to IMP . When both , HGPRT-I and HGXPRT were silenced simultaneously , a significant growth phenotype appeared 72 hours after the tetracycline addition ( Fig 5C ) . Western blot analysis revealed elimination of HGPRT-I signal and a significant decrease in HGXPRT expression ( Fig 5C , bottom panel ) . Since the 10% FBS present in regular HMI-9 medium most likely contains diminutive amounts of additional purines ( i . e . adenine and adenosine ) [33] which could have supported the observed reduced growth of the DKD HGPRT-I/HGXPRT cell line , we tested all generated RNAi cell lines in media containing dialyzed FBS serum and only one defined source of the purine bases , i . e . hypoxanthine ( 50 μM , HMI-9hypoxanthine ) or xanthine ( 50 μM , HMI-9xanthine ) . While the noninduced cells grew normally with either of the added purines , the SKD HGPRT-I and SKD HGXPRT exhibited mild growth phenotypes in HMI-9hypoxanthine confirming that either enzyme can utilize hypoxanthine ( Fig 6A , left and middle panels ) . The slightly more profound growth phenotype in SKD HGPRT-I cells can be explained by the higher Km value of HGXPRT for hypoxanthine compared to the Km of HGPRT-I ( 13 . 7 vs 30 . 6 μM , Table 1 ) . Western analysis of whole cell lysates of noninduced and RNAi induced cells confirmed effective and specific silencing of HGPRT-I and HGPXRT in the respective SKD cell lines ( Fig 6A , left and middle panels ) . Densitometric analysis of protein bands did not reveal any significant changes in the steady state abundance of HGPRT-I and HGXPRT in the SKD HGXPRT and SKD HGPRT-I cell lines , respectively . Importantly , the DKD HGPRT-I/HGXPRT manifested significant growth phenotype in medium containing 50 μM hypoxanthine . Using the HGPRT-I and HGXPRT antiserums , a significant reduction in the expression of the targeted proteins was detected ( Fig 6A , right panels ) . Our data confirms that there are no other enzymes that are able to convert hypoxanthine to IMP , a nucleoside monophosphate that can be further converted to XMP and other nucleoside monophosphates ( Fig 1 ) . Since xanthine becomes the dominant purine base to be salvaged when the parasite reaches the spinal fluids [46] we also tested the generated RNAi cell lines in the medium containing only this molecule ( HMI-9xanthine ) . Surprisingly , xanthine is the least preferred purine for in vitro grown BF cells and the cells had to be cultured in this medium for two weeks to obtain regular growth . The noninduced and RNAi induced SKD HGPRT-I grew without any significant growth delays while the RNAi induced HGXPRT cells displayed a significant growth phenotype at day three upon the silencing ( Fig 6B , left and middle panels ) . This result is in with agreement with the in vitro data showing that HGPRT-I has no affinity for xanthine . Thus , the cells expressing only HGPRT-I are not able to survive in medium exclusively containing this purine . Western blot analysis of the SKD HGXPRT cell line shows that while there was a significant down-regulation of HGXPRT expression two days after the tetracycline addition , at day seven we detected its noticeable re-expression . This observation suggests that some RNAi-induced cells bypassed the tetracycline repression , a phenomenon that has been commonly observed in T . brucei regulated expression systems [47] . The DKD HGPRT-I/HGXPRT cells grew poorly in the HMI-9xanthine displaying a growth phenotype at day 2 after the RNAi silencing . The effective down-regulation of both enzymes was confirmed by Western blot ( Fig 6B , right panels ) . These results highlight the importance of HGXPRT for the infectious stage of the parasites especially in the late phase of infection when the parasite crosses the blood brain barrier and xanthine becomes the main purine source . The 6-oxopurine PRTases function as either dimers or tetramers in various protists , e . g . Leishmania , Plasmodium and Toxoplasma [48–50] . To establish the oligomeric properties of T . brucei HGPRT-I , -II and HGXPRT we used the purified recombinant proteins and performed a cross-linking experiment using dimethyl suberimidate ( DMS ) . The cross-linked proteins were visualized by Coomassie staining on SDS-PAGE . Fig 7A shows that all three individual proteins are able to form homodimers . This result has subsequently been confirmed by analysis of the crystal structures of the two enzymes ( see below ) , both of which reveal dimeric associations . When HGPRT-I or HGPRT-II were mixed together with HGXPRT no heterodimerization was observed as deduced from the positions of individual proteins on SDS PAGE gels ( Fig 7B ) . These results were confirmed in vivo using BF T . brucei cells overexpressing the v5-tagged versions of individual enzymes . Fig 8 shows that immunoprecipitation of v5-tagged HGPRT-I recovered the tagged and endogenous cytosolic HGPRT-I ( Fig 8 , left panel ) , while the immunoprecipitation of the v5-tagged glycosomal HGPRT-II pulled down only a negligible amount of the cytosolic HGPRT-I ( Fig 8 , middle panel ) . Therefore , it is possible that heterodimerization may happen between the HGPRT-I and HGPRT-II isoforms , but given the fact that expression of HGPRT-II is significantly reduced in the BF cells , and the HGPRT-I and HGPRT-II enzymes are confined to two different cellular compartments ( i . e . cytosol and glycosome ) it seems unlikely that these two proteins would form heterodimers in vivo . Immunoprecipitation of v5-tagged HGXPRT recovered the tagged and endogenous HGXPRT , but not the HGPRT-I ( Fig 8 , right panel ) . We can therefore conclude that heterodimerization between HGPRT-I and HGPXRT does not occur either in vivo or in vitro . Over 100 acyclic nucleoside phosphonates were screened for their cytotoxic activity in BF T . brucei cells . Considering the polar character of the phosphonate group , the ANPs ( examples 1–7 ) were tested mostly in the form of their cleavable prodrugs ( examples 1p – 7p ) to facilitate the transport across cell membranes ( Fig 9 , Fig 10 ) [30] . The most potent prodrug is compound 4p with an EC50 of 2 . 6 μM and with a selectivity ratio of >112 , when compared with cytotoxic concentrations ( CC50 ) in human cell lines [21 , 29 , 30] . The ANPs 1–7 corresponding to phosphoramidate prodrugs 1p-7p with the lowest EC50 values were further tested for their inhibition of HGPRT-I , HGPRT-II and HGXPRT . Both guanine- and hypoxanthine-based ANPs inhibited all three enzymes with Ki values ranging from 0 . 06 μM to 19 . 88 μM . The lowest Ki value ( 0 . 06 μM ) was measured for compound 3 with HGPRT-II . There is a lack in correlation between the Ki values for the purified enzymes and the EC50 values for the prodrug ( Fig 10 ) . This could be partially explained by differences in the pharmacokinetics of the prodrug , i . e . the ability to enter the cells and to be cleaved . These attributes can significantly influence the correlation between the enzyme inhibition and prodrug activity . Nevertheless , the data illustrate that regardless of the purine base , individual ANPs can inhibit all three isoforms . This result may explain the observed cytotoxic effect of the tested ANPs in the culture as these compounds inhibit all three 6-oxopurine PRTases . To explain the mode of binding of the ANPs , crystal structures of HGPRT-I in complex with compounds 1 , 2 , 6 and 7 , and HGXPRT in complex with compounds 2 and 7 were determined . Given the conservation of active site residues in HGPRT-I and HGPRT-II , the crystal structure of HGPRT-II was not determined . The structure of HGPRT-I was previously solved in complex with five single chain ANPs and with nucleotide products GMP and IMP [16] . However , the structure of HGXPRT has not previously been determined . We defer a detailed analysis of the overall structure of this enzyme to a subsequent publication . Here , we describe the active site of this enzyme and its interactions with the ANPs . The structures were determined at resolutions that range from 1 . 76–1 . 99 Å for HGPRT-I and 2 . 26–2 . 64 Å for HGXPRT ( Table 2 ) . In both enzymes , the active site is situated at the interface of two domains referred to as the hood and core [16] and is primarily comprised of three pockets identified as the purine base binding pocket , the 5′-phosphate binding pocket and the pyrophosphate ( PPi ) binding pocket . Fig 11 shows the Fo-Fc electron density and binding of 6 and 1 to HGPRT-I ( Fig 11A and 11B ) and the binding of 7 and 2 to both HGPRT-I and HGXPRT ( Fig 11C , 11D , 11E and 11F ) . The guanine bases in 1 , 2 , 6 and 7 , bind in a similar fashion to HGPRT-I and HGXPRT , being held in place by π-π stacking interactions with F166/Y201 ( HGPRT-I/HGXPRT ) ( Fig 12 ) and forming at least four and up to six hydrogen bonds with the amino acid residues and a neighbouring water molecule . These hydrogen bonds are formed between ( i ) the NZ of K145/K180 and the 6-oxo group , ( ii ) the NZ of K145/K180 and N7 of the purine base in four of the complexes , ( iii ) the amide nitrogen of V167/V202 and N1 of the purine base ( iv ) the main chain oxygen of D173/D208 and the 2-amino group of the purine base , ( v ) the side chain of D208 to N2 and ( vi ) a water molecule and N3 of the purine base , in two of the complexes ( Fig 12 ) . Thus , there are few differences in how this base binds to both enzymes . This is a reflection of the fact that the Km values for guanine are similar for these two enzymes ( Table 1; 16 vs 32 μM ) . In all six complexes , a phosphonate moiety is located in the 5´-phosphate pocket ( Fig 12 ) . In HGPRT-I and HGXPRT , this pocket is formed by D117 to T121 and D151 to T155 , respectively . In both complexes the backbone amide nitrogen atoms in this pocket hydrogen bond with the phosphoryl oxygen atoms of the ANP . Additional hydrogen bonds to the phosphoryl oxygen atoms also form with the side-chains of T118 and T121 in HGPRT-I and their counterpart residues , T152 and T155 , in HGXPRT . In addition , in HGPRT-I there are two conserved water molecules that also hydrogen bond to the phosphoryl oxygen atoms . However , in the HGXPRT complexes no such water molecules are observed . We note here though , that the resolution of these structures is slightly lower than that for HGPRT-I . Therefore , the location of all the nearby waters may not be fully apparent in HGXPRT . For compounds 2 and 7 , we have determined structures of complexes with both HGPRT-I and HGXPRT , thus allowing a direct comparison of the binding modes of these two ANPs . As indicated above , a phosphonate group is observed in the 5´-phosphate pocket in all four structures ( Fig 12B , 12C and 12D ) . When compound 7 binds , the phosphonate attached to the linker binds into the 5´-phosphate pocket in both complexes . However , compound 2 binds differently to the two enzymes . In HGPRT-I the phosphonate group attached to the linker binds into the 5´-phosphate pocket and the phosphonate in the attachment binds into the PPi pocket ( Fig 12D ) , whereas in HGXPRT the situation is reversed , with the phosphonate groups occupying alternate positions ( Fig 12F ) . This is the location where most of the differences are observed when the binding modes of the six ANPs are compared ( Fig 12 ) . In the HGPRT-I . 6 complex the PPi binding site is occupied by the phosphonate connected to the attachment . This phosphonate interacts with the side chain atoms of D114 , D173 , R179 and two water molecules ( Fig 12A ) . In this complex the side chain of L53 points inwards towards the active site and the peptide bond between L53-K54 is in the trans conformation . This allows L53 to form hydrophobic interactions with three carbon atoms from the linker and attachments ( Fig 12A ) . This trans peptide bond is also observed when HGPRT-I is in complex with ANPs containing a single phosphonate moiety with five or six carbons between the 6-oxopurine base and the phosphonate moiety [16] . However , in other crystal structures of HGPRT-I ( e . g . in complex with GMP and IMP ) the peptide bond between L53 and K54 can exist in the cis conformation , resulting in the side-chain of L53 being rotated out of the active site , allowing space for PRib-PP to bind [16] . It would appear that in HGPRT-I the binding of PRib-PP is the trigger for this conformational change , as has also been suggested to occur in other 6-oxopurine PRTases , including the human and T . cruzi enzymes [48 , 51] . In the HGPRT-I . 1 complex , the PPi binding site is occupied by a Mg2+ ion which ligands with the side chain of D173 , three water molecules , and with the cyano moiety of compound 1 ( Fig 12B ) . A sulphate from the crystallization buffer is bound in the PPi binding site . This sulphate ion interacts with side chain of R179 , and the amide nitrogen atoms of K54 and G55 . In this complex the peptide bond between L53 and K54 is in the cis conformation ( Fig 12B ) . The phosphonate group of the attachment of compound 7 makes similar hydrogen bonding interactions with both enzymes ( i . e . to the side chain atoms of D173/D208 , R179/R214 , and the amide nitrogen between L53/L86-K54/K87 and between G55/G88-K54/K87 ( Fig 12C and 12E ) . However , in the HGPRT-I . 7 complex the peptide bond between L53-K54 is in the trans conformation and the side chain of L53 is inside the active site ( Fig 12C ) . This allows L53 to form hydrophobic interactions with three carbon atoms in the linker and the attachment . In HGXPRT the peptide bond between L86-K87 is in the cis conformation , with the side chain of the L53 rotated out of the active site ( Fig 12E ) . In the HGPRT-I . 2 complex the phosphonate attach to the linker occupies the 5´-phosphate pocket and the phosphonate connected to the attachment occupies the PPi pocket ( Fig 12D ) . However , in the HGXPRT . 2 complex the situation is reversed and the phosphonate attached to the linker occupies the PPi pocket and the phosphonate of the attachment occupies the 5´-phosphate pocket ( Fig 12F ) . Although the phosphonate groups are swapped in the two structures the hydrogen bonding networks in the 5´-phosphate pocket are virtually identical in the two complexes . Within the purine binding pocket , there are more hydrogen bonds in the HGXPRT complex compared to the HGPRT-I complex . Of additional note is the fact that HGXPRT possesses a tyrosine at position 201 , which forms a hydrogen bond with E208 . In HGPRT-I this residue is phenylalanine and therefore no such bond can form . In the PPi pocket , the phosphonate groups are oriented at slightly different angles in the two structures , this is due to the fact that the linkers/attachments are different and the presence of Mg2+ in the HGPRT-I complex . Nonetheless , both binding modes result in the L53-K54/L86-K87 peptide bond adopting a cis conformation . In this location , there are fewer interactions between the compound 2 and HGXPRT compared to HGPRT-I and this difference is the likely reason for this compound having a higher Ki value for HGXPRT compared to HGPRT-I ( 6 . 54 μM vs 0 . 11 μM ) . On the other hand , the binding mode of compound 6 in the two enzymes is very similar ( Figs 12C and 11E ) which is reflected in their almost identical Ki values ( 0 . 22 μM and 0 . 40 μM ) . The crystal structures of human HGPRT in complex with compounds 1 , 2 and 6 have previously been determined ( PDB codes 4IJQ , 4RAC and 4RAN ) . When the binding mode of compound 6 is compared with its binding mode to HGPRT-I there is a general displacement of the purine base , the linker and the attached phosphonate , as well as a different binding mode of the phosphonate joined to the attachment ( Fig 13A ) . In addition , in HGPRT-I the L53-K54 peptide bond is in trans conformation and the side chain of L53 points into the active site , whereas in the human enzyme when compound 6 binds the equivalent peptide bond is in the cis conformation . Further , in the crystal structure of the human enzyme a sulphate ion fills the PPi binding pocket , whereas there is no sulphate in the HGPRT-I and HGXPRT structures ( Fig 13A ) . This difference appears to be a contributing factor in placing compound 6 in the two structures . By contrast , the binding of compounds 1 and 2 is very similar in HGPRT-I and human HGPRT ( Fig 13B and 13C ) . However , the cyano group of the compound 1 does adopt a different orientation in the two enzymes . In HGPRT-I it is locked in place by the nearby magnesium ion , whereas in the human enzyme there is no such interaction . Since the crystal structure of human HGPRT in complex with compound 2 is known this complex can be compared with HGXPRT . The main difference between the mode of binding of this compound in the two enzymes is that in HGXPRT the phosphonate group that belongs to the linker fills the PPi pocket and the phosphonate group that belongs to the attachment is in the 5´-phosphate pocket , whereas in human HGPRT , the positions of the phosphonate groups of compound 2 are inverted ( Fig 13D ) . Thus , there are several differences in the structures of human HGPRT and the trypanosomal enzymes that could be exploited for selective drug design .
The purine salvage pathway has been suggested as a possible drug target for a number of different parasites including Plasmodium , Leishmania and Trypanosoma species [52 , 53] . This is based on the fact that the de novo pathway for the synthesis of the nucleoside monophosphates is not present in any of these organisms . However , because of the redundancy in this pathway it was thought that none of the individual enzymes would be essential for the parasites . Here , we showed that double RNAi silencing of T . brucei HGPRT/HGXPRT caused significant growth phenotype in vitro . Moreover , selected prodrugs of potent ANPs had the ability to block the growth of BF parasites grown in culture . Interestingly , related prodrugs of aza-acyclic nucleoside inhibitors exerted cytotoxic activity against P . falciparum parasites also via inhibition of HGXPRT [29 , 30] . Although Plasmodium parasites utilize adenosine and inosine in addition to hypoxanthine , the first two are quickly converted to hypoxanthine by sequential actions of adenosine deaminase and purine nucleoside phosphorylase . Thus the enzymatic action of HGXPRT is crucial to produce IMP as the only precursor to all required nucleoside phosphate products . The indispensability of the plasmodial HGXPRT was further corroborated by metabolic studies [54] . Interestingly , while the Leishmania purine salvage pathway exhibits higher level of complementarity and redundancy compared to Trypanosoma and Plasmodium [52] , the simultaneous knock-out of L . donovani HGPRT and XPRT enzymes was also lethal for the parasite grown in vitro [55] . Furthermore , uptake and salvage of adenine and adenosine was not sufficient to sustain the purine nucleoside pool suggesting that the interconversion enzymes ( i . e . ADSL , ADSS , or GMP synthase ) , could be actually indispensable for this parasite as well . Thus , regardless of initial views , the purine salvage enzymes are possible targets in these parasites . In the mammalian host T . brucei lives extracellularly and exploits environments rich mainly in hypoxanthine , xanthine and inosine . Other purine bases and nucleosides ( e . g . adenine and adenosine ) are present in negligible concentrations [46] . Therefore , the parasite should be susceptible to inhibition of not only the salvage enzymes , like HGPRT and HGXPRT ( shown here ) , but also to the key interconversion enzymes , such as GMPS . Indeed , GMPS knock-out cell line was not able to produce infection in mice and specific GMPS inhibitors arrested the parasite´s growth in vitro [53] . More importantly , RNAi silencing of ADSS and ADSL , enzymes responsible for interconversion of IMP to AMP , caused attenuated virulence of T . brucei in mice suggesting that in mouse plasma there are limited concentrations of adenine and adenosine [56] . As a result , it is plausible to expect that HGPRT/HGXPRT would be also essential in the animal model . Moreover , the obvious promiscuity of T . brucei HGXPRT offers an elegant solution of one drug to target both HGPRT and HGXPRT enzymes and thus pharmacological inhibition of both enzymes can be proposed . Crystal structures of the two 6-oxopurine PRTs in complex with the ANPs ( Table 2 ) provide the specific details of their binding modes . As it has been also observed in studies of P . falciparum HGXPRT [25] , five atoms appears to be the optimal length of the linker to form interactions between a phosphonate end and residues in the 5´-phosphate pocket . The number of atoms required to be attached for the linker to reach into the PPi pocket is of the order of 5 to 6 depending on the location of the attachment , the presence or absence of magnesium and whether or not the PPi binding loop has a cis peptide bond present . The kinetic data show that compounds that possess guanine as the base are effective at inhibiting all three isoforms of the 6-oxopurine PRTs , and this appears to translate into compounds that are effective in killing T . brucei parasites in culture with the prodrug versions of these compounds . Compounds 3 and 6 are the most broadly effective inhibitors tested so far with Ki values lower than 1 μM for all three enzymes . Another promising inhibitor is ANP 4 with a very favourable selectivity index . Compound 1 is also of interest for further development since its Ki value is quite high compared to the antitrypanosomal activity of its prodrug ( relative to the other compounds ) . This is likely due to the fact that there is only one phosphonate group that is required to be masked by the prodrug formulation . The structure of HGPRT-I has a highly unusual PPi binding site in that , that in the trans conformation leucine points inward toward the 5’-phosphate pocket . This is a feature that could be utilized to develop inhibitors that are highly specific for this enzyme . Moreover , the PPi binding pocket differs at amino acid position 54 between HGPRT-I and HGPRT-II isoforms . HGPRT-I contains a lysine at this position whereas HGPRT-II contains arginine . Based on previous crystal structures of the 6-oxopurine PRTases , movement of this side chain is critical for opening up the active site to allow access to PRib-PP . It is possible that an arginine in this position could form more intramolecular interactions retarding the ability for PRib-PP to rapidly bind to the enzyme and or for the enzyme to rapidly release the product [57] . In this context , a sequence comparison across the 6-oxopurine PRTases shows that lysine is the preferred residue at this location [21 , 28 , 49 , 51 , 57 , 58] . HGXPRT is markedly different from other 6-oxopurine PRTases in that it possesses the ability to use xanthine as a substrate . This is a difference compared to the human enzyme that could also be exploited for the development of specific inhibitors . Importantly , the need to develop highly selective inhibitors of the parasite enzyme does not apply here . Humans also possess the de novo pathway for synthesis of their DNA/RNA and thus they are not fully reliant on the salvage pathway . This is demonstrated by individuals in the community where mutations occur to the human HGPRT resulting in a loss of 90% of activity of this enzyme , yet individuals with this affliction experience few symptoms , all of which are readily treatable by application of allopurinol [59] . In conclusion , we have demonstrated that T . brucei parasites require activity of HGPRT and HGXPRT enzymes for the normal cell growth in vitro . ANPs represent a promising class of potent and selective compounds as they inhibit the enzymes with Ki values in nanomolar range and exert cytotoxic effects on T . brucei cells grown in vitro with EC50 values in the single digit μM range . Our results provides a foundation for further investigations of these compounds in vivo and suggests that HGPRT-I/HGPRT-II/HGXPRT could be possible targets for future drug discovery efforts directed at controlling HAT . | Human African Trypanosomiasis ( HAT ) is a life-threatening infectious disease caused by the protozoan parasite , Trypanosoma brucei . Current treatments suffer from low efficacy , toxicity issues and complex medication regimens . Moreover , an alarming number of these parasites are demonstrating resistance to current drugs . For these reasons , there is a renewed effort to develop new classes of modern therapeutics based upon the unique T . brucei cellular processes . One potential new drug target is 6-oxopurine phosphoribosyltransferase ( PRTase ) , an enzyme central to the purine salvage pathway and whose activity is critical for the production of the nucleotides ( GMP and IMP ) required for DNA/RNA synthesis within this protozoan parasite . We demonstrated that T . brucei encodes two isoforms of hypoxanthine-guanine PRTases ( HGPRT ) and one hypoxanthine-guanine-xanthine PRTase ( HGXPRT ) . The concurrent activity of these enzymes is required for the normal cell growth in vitro . Moreover , acyclic nucleoside phosphonates represent a promising class of potent and selective compounds as they inhibit the enzymes with Ki values in nanomolar range and exert cytotoxic effects on T . brucei cells grown in vitro with EC50 values in the single digit micromolar range . Our results provide a new foundation for further investigations of these compounds in vivo and suggest that 6-oxopurine salvage pathway represents a possible target for future drug discovery efforts directed at eliminating HAT . | [
"Abstract",
"Introduction",
"Material",
"and",
"methods",
"Results",
"Discussion"
] | [
"rna",
"interference",
"crystal",
"structure",
"chemical",
"compounds",
"condensed",
"matter",
"physics",
"enzymology",
"phosphonic",
"acids",
"nucleotides",
"organic",
"compounds",
"parasitic",
"protozoans",
"purines",
"trypanosoma",
"brucei",
"protozoans",
"epigenetics",
"crystallography",
"enzyme",
"inhibitors",
"organophosphorus",
"compounds",
"guanine",
"solid",
"state",
"physics",
"genetic",
"interference",
"gene",
"expression",
"chemistry",
"physics",
"biochemistry",
"rna",
"trypanosoma",
"eukaryota",
"organic",
"chemistry",
"nucleic",
"acids",
"genetics",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"trypanosoma",
"brucei",
"gambiense",
"organisms"
] | 2018 | Evaluation of the Trypanosoma brucei 6-oxopurine salvage pathway as a potential target for drug discovery |
The mosquito is the obligate vector for malaria transmission . To complete its development within the mosquito , the malaria parasite Plasmodium must overcome the protective action of the mosquito innate immune system . Here we report on the involvement of the Anopheles gambiae orthologue of a conserved component of the vertebrate immune system , LPS-induced TNFα transcription factor ( LITAF ) , and its role in mosquito anti-Plasmodium immunity . An . gambiae LITAF-like 3 ( LL3 ) expression is up-regulated in response to midgut invasion by both rodent and human malaria parasites . Silencing of LL3 expression greatly increases parasite survival , indicating that LL3 is part of an anti-Plasmodium defense mechanism . Electrophoretic mobility shift assays identified specific LL3 DNA-binding motifs within the promoter of SRPN6 , a gene that also mediates mosquito defense against Plasmodium . Further experiments indicated that these motifs play a direct role in LL3 regulation of SRPN6 expression . We conclude that LL3 is a transcription factor capable of modulating SRPN6 expression as part of the mosquito anti-Plasmodium immune response .
In virtually all species , the innate immune system is responsible for the primary response against pathogens . Unlike adaptive immunity , the innate immune response does not confer long-lasting protection but instead , relies on the recognition of pathogen–associated molecular patterns ( PAMPS ) . Following recognition , cell-mediated responses eliminate the pathogen . In vertebrates , these responses involve inflammation and the recruitment of specialized cells to the site of infection via the production of effector molecules such as cytokines . As an important mediator of immune regulation , the cytokine tumor necrosis factor-alpha ( TNF-α ) has a variety of functions including apoptotic cell death , inflammation , immune signaling via NF-κB , and cellular proliferation/differentiation [1] . With such pleiotropic functions , it is critical that the expression of TNF-α be tightly regulated . Several components have been identified that are involved in the regulation of TNF-α , including the LPS-induced TNF-α factor ( LITAF ) [2] , [3] . Identified as a transcription factor , LITAF binds to the TNF-α promoter in response to bacterial LPS stimulation to influence the expression of TNF-α [3] , as well as additional LPS-induced cytokines [4] . The components of the mosquito innate immune system are of important biological relevance but are incompletely characterized . Much of our knowledge of the mosquito innate immune system is based on homologous innate immune pathways first described in Drosophila . Although evolutionarily distant from the well-characterized vertebrate TLR innate immune pathways , analogous mosquito Toll and IMD pathways drive the nuclear translocation of NF-κB-like transcription factors to provide defense against invading pathogens via expression of anti-microbial peptides . The activation of the Toll , IMD , and JAK-STAT pathways [5]–[7] , have been shown to limit the success of the malaria parasite Plasmodium . Several effector genes have been identified that influence Plasmodium development in the mosquito [8] , yet many questions remain as to how the mosquito immune response recognizes and destroys invading pathogens . Here we report on the first identification of a LITAF-like gene in insects and investigate its role in the mosquito immune response to Plasmodium . Similar to vertebrate LITAF , LL3 seems to act as a transcription factor involved in the regulation of the mosquito immune response , as evidenced by its direct effects on the expression of SRPN6 , a known anti-Plasmodium effector gene [9] , [10] . These findings demonstrate for the first time the role of a LITAF-like gene in insects and suggest that LL3 is an integral component of the mosquito immune response to limit Plasmodium infection .
An expressed sequence tag ( EST ) corresponding to the annotated gene AGAP009053 in An . gambiae was originally identified using a subtractive hybridization cDNA library enriched for mosquito genes following Plasmodium infection [11] . The predicted protein product of 82 amino acids , shares sequence similarity to LITAF , a transcription factor involved in the activation of TNF-α and other cytokines in vertebrate organisms [2] , [4] . Furthermore , BLAST analysis revealed six highly conserved LITAF-like sequences in the An . gambiae genome . Phylogenetic analysis of genes encoding LITAF domain-containing proteins across taxa revealed an expansion within dipteran insects , likely due to an ancient gene duplication event , in contrast to mammals and other invertebrates that contain a single LITAF gene ( Figure S1A ) . Each of the six identified LITAF-like proteins in An . gambiae have direct orthologues in other mosquito species . RT-PCR was used to examine the expression pattern of the An . gambiae LITAF-like genes in response to Plasmodium infection ( Figure S1B ) . Four of the genes , all located within an approximate 50 kb stretch on chromosome 3R , produced specific PCR products and were named LITAF-like 1–4 . Expression of AGAP009053 , or LITAF-like 3 ( LL3 ) , was strongly induced in the midgut of P . berghei-infected mosquitoes suggesting that LL3 is involved in the immune response against P . berghei parasites . To more closely characterize the role of LL3 in the mosquito response to Plasmodium infection , qRT-PCR was used to quantify the changes of gene expression in response to a Plasmodium-infected blood meal . At the onset of P . berghei ookinete midgut invasion ( ∼18 hours ) , LL3 expression is significantly increased and remains high ( Figure 1A ) . Parasites unable to sexually differentiate ( ANKA 2 . 33 ) or MAOP mutant parasites that produce ookinetes that attach but are unable to invade the midgut [12] fail to induce a response , suggesting that the expression of LL3 is triggered by the physical invasion of the mosquito midgut . Related experiments with P . falciparum show that LL3 is also induced within a similar time frame ( Figure 1B ) . For both species , a similar pattern of SRPN6 ( AGAP009212 ) expression was detected , raising the possibility that a common mechanism may regulate the expression of both genes . LL3 induction following P . berghei infection was evaluated by immunofluorescence assays using a peptide-derived LL3 antibody . LL3 signal above background was detected only in midgut cells in close proximity to ookinetes ( Figure 2A ) , suggesting that LL3 expression is induced by parasite invasion . This response was further confirmed by immunofluorescence assays following the dsRNA-mediated silencing of GFP ( control ) or LL3 to confirm the specificity of the LL3 signal ( Figure 2B ) . These results are consistent with the weak fluorescence obtained when mosquitoes were fed with the invasion-deficient MAOP mutant parasites suggesting that LL3 protein expression correlates directly with LL3 transcript abundance ( Figure S2 ) . Moreover , the overall response of LL3 to parasite invasion resembles that previously described for SRPN6 [9] . In cells that strongly express LL3 , fluorescence is detected in both the cytoplasm and nucleus ( Figure 2A and S2 ) . Although expression is primarily localized to the cytoplasm , a small proportion of the signal is detected in the nucleus that may be sufficient for transcriptional activation . While the mechanisms mediating LL3 nuclear translocation remain undefined , this may be regulated by post-translational modifications similar to LITAF in mammals [4] , and as is the case for REL1 , REL2 and STAT1 in An . gambiae [13] . No signal was obtained following the incubation with pre-immune sera ( Figure S2 ) . Previous experiments have shown in mammals that LITAF translocation to the nucleus in response to LPS treatment is phosphorylation dependent [4] . As a result , we wanted to determine the role of phosphorylation on the translocation of LL3 to the nucleus . However , due to the variability in the kinetics of Plasmodium midgut invasion and the often transient nature of transcription factor activation , we employed an alternative approach through pervanadate treatment as previously done to examine STAT translocation [14] . As a mixture of sodium orthovanadate and hydrogen peroxide , pervanadate induces oxidative stress mimicking the environment of midgut cells following Plasmodium invasion and acts as a potent phosphatase inhibitor . To investigate the LL3 response to pervanadate , immunofluorescence assays were performed to determine LL3 activation and nuclear translocation ( Figure 2C ) . In control mosquitoes ( −PV ) , only a weak fluorescence signal was detected with LL3 immune sera . Upon pervanadate treatment ( +PV ) , LL3 was strongly expressed in all cells and appears to be localized in both the nucleus and the cytoplasm ( Figure 2C ) . This suggests that LL3 expression is quickly induced in response to pervanadate treatment likely due to cell stress caused by increased reactive oxygen , to its strong phosphatase inhibitor activity , or a combination of the two . While this does not directly link nuclear translocation to LL3 phosphorylation , it does provide preliminary results to further explore the basis of LL3 activation . We used RNAi-mediated gene silencing to determine whether LL3 plays a role in the mosquito response to Plasmodium infection . Silencing of LL3 led to a substantial increase in P . berghei oocyst numbers and infection prevalence when compared to dsGFP controls ( Figure 3A ) . Infection by P . falciparum parasites was similarly affected as the LL3 knockdown mosquitoes displayed double the oocyst load when compared to controls ( Figure 3B ) . A small , but non-significant increase in the P . falciparum infection prevalence was detected , despite the high intensity of the controls . Previous reports have suggested that the immune responses of the mosquito to P . berghei and P . falciparum are quite divergent [15] . Several molecules that have been implicated in anti-Plasmodium defenses only function against a specific parasite species [8] . However , the dsRNA-mediated silencing of LL3 results in a significant increase in the number of developing oocysts for both rodent and human parasite species , suggesting that LL3 is a universal component of the mosquito anti-Plasmodium response . The presumed role of LL3 as a transcription factor suggests that it regulates the immune response at the transcriptional level . LL3 knockdown efficiency was verified by qRT-PCR and resulted in an approximate 80% reduction in mRNA abundance ( Figure 3C ) . To determine the specificity of LL3 silencing , the expression of the other LITAF-like genes was monitored by RT-PCR ( Figure S3 ) . Only LL4 displayed a slight decrease in expression , but further experiments are needed to determine if this is a downstream target of LL3 activation . Significantly , LL3 knockdown also resulted in a considerable decrease in SRPN6 expression ( Figure 3C ) , a known inhibitor of Plasmodium development [9] , [10] . Based on the characterization of LITAF as a transcription factor in other organisms , we examined the possibility that LL3 may also play a similar role in Anopheles and bind DNA . We used two different PCR-assisted DNA-binding site selection assays to identify DNA fragments able to bind to recombinant LL3 ( Figure 4A ) . Following four rounds of selection for each method , the recovered sequences ( Table S2 ) were then used as input for MEME analysis [16] to identify putative LL3-DNA binding motifs ( Figure S4 ) . Both methods produced multiple putative motifs . For both methods , the most frequently recovered consensus sequence was a GGG[A/T]G motif ( Figures 4B , 4C and S4 ) , providing validation of our approach and suggesting that this is a high affinity DNA-binding site for LL3 . This motif also shares a striking resemblance to the CTCCC motif ( reverse complement of the LL3 motif ) described for LITAF on the TNF-α promoter [3] . An additional , highly degenerate motif was also identified within the affinity-based enrichment ( Figure 4B ) . The two motifs were chosen from those identified by MEME analysis ( Figure S4 ) based on their presence in the SRPN6 promoter and likely role in SRPN6 regulation ( Figures 5 and 6 ) . Having obtained evidence that LL3 binds to specific DNA sequences , we next investigated whether this putative transcription factor is capable of binding to DNA in the SRPN6 promoter . LL3 recognition of SRPN6 promoter sequences would suggest that SRPN6 expression is directly affected by LL3 expression ( Figure 3C ) . One kilobase of the putative SRPN6 promoter was initially divided into five equal fragments and examined by electrophoretic mobility shift assay ( EMSA ) . Two positive fragments were further subdivided to finely map LL3 binding sites . Specific LL3 binding was detected to the −800 to −761 , and −163 to −121 regions ( Figure S5 ) . These ∼40 bp regions were then examined by mutational analysis to further narrow the sequences of LL3-DNA interactions ( Figure S6 ) . The results of these experiments are summarized in Figure 5 . Two 10 bp regions within the −800 to −761 fragment were identified as being critical for LL3-DNA interactions ( Figure S6 ) . These sequences from −800 to −791 ( R1 ) and −770 to −761 ( R2 ) are highly similar and closely resemble the consensus sequence described in Figure 4B ( top ) . Within the −163 to −121 fragment , binding was attributed to a 10 bp region ( R3 ) in which a GGGAG motif was identified similar to that detected in Figure 4B ( bottom ) and Figure 4C . In addition , each of the identified regions share the first 3 bp ( ATG ) , but it is unclear how these residues influence LL3 binding ( Figure 5 ) . This close correlation of the LL3 DNA binding motifs identified by two independent methods ( PCR-assisted selection and EMSA ) strongly reinforces the validity of these results and the likelihood that LL3 directly regulates SRPN6 through interactions with its regulatory regions . Taking advantage of the availability of immunoresponsive cell lines for An . gambiae , we examined the transcriptional response of LL3 to heat-killed Enterobacter cloacae in the hemocyte-like Sua5B cell line . When challenged with heat-killed bacteria , LL3 expression was significantly increased when normalized to basal levels of transcript in the non-induced cDNA sample ( Figure 6A ) . In addition , the levels of SRPN6 transcript are also significantly increased , despite being expressed at much higher levels of basal transcription ( data not shown ) . In view of the reduced SRPN6 mRNA abundance when LL3 is silenced ( Figure 3C ) and the detection of LL3 DNA-binding motifs in the SRPN6 promoter ( Figure 5 ) , we further investigated the possibility that LL3 directly contributes to the regulation of SRPN6 expression . With this aim , we built constructs that placed firefly luciferase coding sequence under the control of the SRPN6 promoter and quantified its expression in Sua5B cells where LL3 can be induced by the addition of bacteria ( Figure 6A ) . Plasmids carrying the wild type SRPN6 promoter or promoters containing mutations in the three putative LL3 DNA-binding motifs ( Figure 6B ) were used for these experiments . Luciferase expression was measured in naïve cells or following induction with heat-killed E . cloacae . A moderate but non-significant increase of luciferase expression after bacteria induction was detected when the gene was driven by the wild type promoter ( Figure 6C ) , similar to the induction levels for the endogenous SRPN6 transcript ( Figure 6A ) . Upon bacteria induction , a significant decrease in luciferase expression was detected in the promoter constructs containing one or multiple mutations of the LL3 binding sites , except for the [−162 to −153] ( R3 ) mutation ( Figure 6C ) . In non-induced samples , only the expression construct containing mutations of all three of the LL3 binding sites ( ALL mutant; [−800 to −790] , [−770 to −760] , and [−162 to −153] ) resulted in a significant decrease , while the other constructs displayed only marginal differences in basal luciferase expression ( Figure 6C ) . Taken together , it appears that the sites −800 to −791 ( R1 ) and −770 to −760 ( R2 ) play an important role in the transcription and induction of SRPN6 , while the site from −162 to −153 ( R3 ) may play a more cooperative role as evidenced by the significant decrease in luciferase expression in the triple mutant construct . Given the complexity of the SRPN6 gene ( it contains an approximate 3 . 5 kb intron following the start codon ) , it is possible that the entire transcriptional machinery was not present in the luciferase constructs . However , the changes in expression upon bacterial challenge were remarkably similar to those of the endogenous SRPN6 transcript ( Figure 6A ) . In summary , LL3 binding sites seem to be more important for activation of SRPN6 expression ( and perhaps of other genes involved in host defense mechanisms ) rather than maintenance of basal expression in cell culture . It is also important to note that the immune response is complex and other factors , in addition to LL3 , may also play a role in SRPN6 gene regulation .
Ookinete invasion of the mosquito midgut represents a critical bottleneck in Plasmodium development . To ensure its transmission , the parasite must overcome large parasite losses to reach the basal lamina and evade components of the mosquito hemolymph as it transitions to a mature oocyst [17] , [18] . Recent advances have increased our understanding of how the development of Plasmodium parasites is restricted in its mosquito host , but our understanding of these mechanisms is incomplete . This report investigates for the first time , possible roles played by a LITAF-like transcription factor in the mosquito anti-Plasmodium response . Originally identified from a P . berghei-infected midgut subtraction library [11] , our observations demonstrate that LL3 expression is induced in response to the physical disruption of the midgut epithelium as a result of P . berghei and P . falciparum ookinete invasion . These results are consistent with previous gene expression analysis for LL3 [19] , and are remarkably similar to the patterns of SRPN6 expression identified in previous experiments [9] and in this report . Consistent with these results , immunolocalization experiments imply that LL3 expression occurs in cells of the midgut epithelium in close proximity to invading ookinetes . These LL3-positive cell clusters are similar to those previously described for SRPN6 [9] , and other markers of invaded cells [20]–[22] , suggesting that LL3 is expressed as a result of ookinete invasion . During the invasion process , ookinetes traverse multiple cells before reaching the basal lamina where they begin the transition to an oocyst [20] , [21] . Meanwhile , the invaded cells undergo a series of morphological and molecular changes that lead to apoptosis and their ultimate removal into the midgut lumen [21] , [22] . These damaged cells are marked by elevated levels of nitric oxide synthase ( NOS ) , an enzyme involved in the production of nitric oxide , that create a highly toxic environment in which the ookinete must reach the basal lamina to survive according to the “time bomb” theory of invasion [21] , [22] . As a result , the rate at which the ookinete crosses the cell could greatly determine invasion success [21] , [22] . Recent work has identified that together with NOS , enzymes that mediate protein nitration within invaded cells are required to effectively label ookinetes for recognition and TEP1-mediated lysis [23] . With a presumed role as a transcription factor , LL3 may be connected to these events by promoting a transcriptional program that leads to parasite recognition by the mosquito complement system , explaining the increased parasite numbers in LL3-silenced mosquitoes . Alternatively , the increased parasite survival associated with LL3-silencing may be attributed to a “late-phase” phenotype as described for components of the STAT pathway [7] . The identification of the mechanisms of LL3 anti-Plasmodium immunity will be a major focus of future experiments . Through the use of PCR-assisted DNA-binding site selection assays , we identified several DNA fragments that are recognized by LL3 . Although there is inherent noise within the experimental system , MEME analysis identified putative motifs that were independently replicated , providing validation of this approach . From both assays , the predominant motif identified was a GGG[A/T]G consensus sequence . Interestingly , this is the reverse complement to the CTCCC motif that LITAF recognizes on the TNF-α promoter and suggests that LL3 binding site recognition is conserved in mosquitoes . Presumably , LL3 influences the regulation of a large repertoire of genes involved in the mosquito innate immune response through interactions with the GGG[A/T]G sequence or other predicted motifs . Bioinformatics approaches to identify putative downstream targets of LL3 in mosquitoes have been further complicated by the short length of the GGG[A/T]G motif , resulting in large numbers of candidate target genes that await further validation . In addition , very little is known regarding the downstream targets of mammalian LITAF , thus providing little information to search for orthologous genes under the control of LL3 in mosquitoes . Identifying the genes under LL3 regulatory control remains a priority for future investigation . From our experiments , it is clear that LL3 has a direct role in at least one previously described component of the mosquito immune response , SRPN6 [9] , [10] . Upon LL3-silencing , we detect a significant decrease in SRPN6 transcript in An . gambiae following P . berghei infection , and have identified LL3 recognition elements in the SRPN6 promoter that directly regulate SRPN6 expression in cultured cells . Annotated as a predicted serine protease inhibitor or serpin , similar serpin family members have been implicated in the down-regulation of immune pathways in An . gambiae through their interaction with a target protease [9] , [20] , [24] , [25] . However , the precise role of SRPN6 in the immune response has yet to be elucidated and is further confounded by the complex phenotype obtained following SRPN6 knockdown in which the developmental success of P . berghei varies on the species and strain of the mosquito host [9] . SRPN6-silencing in susceptible lines of An . gambiae did not impact infection intensity , but implicate SRPN6 function in parasite clearance [9] . Based upon the large increase in oocyst numbers following LL3 knockdown in An . gambiae with P . berghei and P . falciparum , it is clear that the effects of LL3-silencing resonate well beyond the regulation of SRPN6 in the mosquito anti-Plasmodium response . Taken together , we provide the first description of LITAF-like genes in dipteran insects and demonstrate the involvement of at least one member of this class of putative transcription factors as a novel component of the An . gambiae innate immune response . Our findings provide an important starting point for further investigation into the mechanisms of LL3 function and the targets under its regulatory control . New questions regarding the identification of signaling pathways involved in LL3 activation will be addressed and efforts will be made to place LL3 in the overall context of mosquito immunity . Based upon its homology to mammalian LITAF , one may speculate that LL3 influences the expression of a TNF-α-like molecule or other yet unidentified cytokines involved in the mosquito innate immune response . It will also be interesting to examine if LL3 or other LITAF-like genes also influence SRPN6 expression in the mosquito salivary glands , where SRPN6 has also been implicated in limiting Plasmodium sporozoite invasion [10] . In conclusion , these results provide evidence for a new component of the mosquito response to Plasmodium infection . Further work may lead to improved strategies to curtail the transmission of malaria .
This project was carried out in accordance with the recommendations of the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health . The animal protocol was approved by the Animal Care and Use Committee of the Johns Hopkins University ( protocol number M009H58 ) . Anonymous human blood used for parasite cultures and mosquito feeding was obtained under IRB protocol NA 00019050 approved by the Johns Hopkins School of Public Health Ethics Committee . Informed consent is not applicable . The colony of Anopheles gambiae ( Keele strain ) was obtained from Drs . Hilary Hurd and Paul Eggleston at Keele University . Mosquitoes were maintained on 10% sucrose at 27°C and 80% relative humidity with a 14/10 h light/dark cycle . For P . berghei infections , mosquitoes were fed on anaesthetized Swiss Webster mice infected with ANKA 2 . 33 ( a gametocyte-minus clone ) , ANKA-GFP [26] , or MAOP [12] parasites . P . falciparum infections were performed by diluting mature NF54 gametocytes to 0 . 3% gametocytemia and fed using an artificial membrane feeder . Dissections were performed in 1X PBS , and oocysts counts were performed by midgut dissection at 10 d ( P . berghei ) or 7 d ( P . falciparum ) post-infection , stained with 0 . 2% mercurochrome and visualized with a compound microscope . Total RNA was prepared from mosquito cell or tissue samples using TRIzol ( Invitrogen ) , and cDNA was prepared using SuperScriptIII ( Invitrogen ) according to the manufacturer's protocol . Gene expression was analyzed by quantitative real-time PCR using gene-specific primers and Power SYBR Green PCR Master Mix ( Applied Biosystems ) on a StepOnePlus Real-Time PCR System ( Applied Biosystems ) . PCR was performed using an initial denaturation of 95°C for 5 min , followed by 40 cycles of 95°C for 15 sec , 65°C for 30 sec , and 70°C for 30 sec . Measurements were taken during the 70°C extension at each cycle and a melting curve was used following amplification to confirm product specificity . qPCR results were normalized using An . gambiae ribosomal protein S7 as a reference and target gene expression was analyzed according to the 2−ΔΔCt method [27] . Measurements were performed in triplicate and all experiments were replicated with at least two independent biological samples . Gene-specific primers sequences are as follows; rpS7 ( F: 5′-ACCACCATCGAACACAAAGTTGACACT-3′ and R: 5′-CTCCGATCTTTCACATTCCAGTAGCAC-3′ ) , LL3 ( F: 5′-GTACGCACGAAAGTGAAGCACGAAT-3′ and R: 5′- AATGTTTGTACGAGCCAATGAACGTGT-3′ ) , and SRPN6 ( F: 5′-CTCTACTTCAAAGCCAAGTGGAAGACG-3′ and R: 5′-CTGTATCAGGTACATCGTGCTGGTGTC-3′ ) . A 243 bp fragment encompassing the LL3 ORF was amplified from An . gambiae midgut cDNA using the following primers; F: 5′-CACCACTACCATCATAGTGACGAACCCGC-3′ and R: 5′-ATGTTTGTACGAGCCAATGAACGTGTTGC-3′ . Following gel purification with the Qiaex II gel extraction kit ( Qiagen ) , the LL3 PCR fragment was cloned into a pBAD202/D-TOPO vector ( Invitrogen ) according to product specifications and later sequenced to verify the sequence accuracy of the inserted DNA . The resulting pBAD-LL3 plasmid was transformed into the BL21 strain of E . coli for protein expression . Soluble recombinant LL3 was isolated using BugBuster ( Novagen ) , His-purified using Ni-NTA agarose ( Qiagen ) , and eluted with 50 mM NaH2PO4 , 300 mM NaCl2 , 250 mM imidazole , and 0 . 05% Tween . Eluted protein samples were diafiltered using Amicon Ultra columns ( Millipore ) and 1× PBS for buffer exchange . Individual sample preps were aliquoted in small volumes for one-time use and stored at −80°C . To generate polyclonal sera in mice for use in immunofluorescence experiments , a LL3 KLH-conjugated peptide ( TVRTKVKHESTTSTC ) was added to an initial 1∶1 mixture of 1X PBS and Complete Freund's adjuvant ( Sigma ) and immunized by intra-peritoneal injection . Subsequent boosts ( four total ) were performed every two weeks as above using incomplete Freund's adjuvant ( Sigma ) and bleeds were performed to monitor the immune response before the final serum was collected by heart puncture . Midguts from non-infected mosquitoes were dissected in 1X PBS and incubated for 20 min in 1X PBS alone , or with pervanadate treatment as previously described [14] . Immunofluorescence assays were performed as previously described for midgut tissues [20] , stained with ProLong Gold antifade reagent with DAPI ( Invitrogen ) and visualized on a Nikon 90i compound fluorescence microscope . Primary antibody dilutions were made as follows: 1∶500 mouse anti-LL3 , 1∶1 , 000 rabbit anti-Pf aldolase [28] . Secondary Texas Red goat anti-mouse ( Invitrogen ) or Alexa Fluor 488 goat anti-rabbit ( Invitrogen ) antibodies were used at a 1∶1 , 000 dilution . A 467-bp fragment consisting of the LL3 ORF and 3′ UTR was PCR amplified from midgut cDNA using the primers 5′-ATGACTACCATCATAGTGACGAACCC-3′ and 5′-TTACACCATTATTAAATAAATAACACAACTTGAGATG-3′ and subcloned into a pJet1 . 2 vector using the CloneJet PCR cloning kit ( Fermentas ) . To create a template for dsRNA , T7 promoter sequences were added to existing gene specific primers to amplify T7-PCR product templates for LL3 ( F: 5′-TTAATACGACTCACTATAGGGAGAATGACTACCATCATAGTGACGAACCC-3′ and R: 5′- TTAATACGACTCACTATAGGGAGATTACACCATTATTAAATAAATAACACAACTTGAG-3′ ) and the GFP control ( F: 5′- TTAATACGACTCACTATAGGGAGAATGGTGAGCAAGGGCGAGGAGCTGT-3′ and R: 5′- TTAATACGACTCACTATAGGGAGATTACTTGTACAGCTCGTCCATGCC-3′ ) . The resultant T7-PCR templates were PCR purified and concentrated using the DNA Clean and Concentrator ( Zymo Research ) , then used to produce dsRNA using the MEGAscript RNAi kit ( Ambion ) according to the manufacturer's protocol . dsRNA products were re-suspended at a final concentration of 3 µg/µl in 1X PBS and used for injections as previously described [29] . Two days post-injection , surviving mosquitoes were fed on P . berghei- or P . falciparum-infected blood and maintained at 19°C or 25°C respectively . The efficiency of dsRNA-mediated silencing was examined by midgut dissection 24 h post-blood meal and subsequently analyzed by qRT-PCR as described above . To select DNA fragments that bind with affinity to rLL3 , a PCR-assisted DNA-binding site selection was performed as previously described with slight modification [30] , [31] . Using an oligonucleotide with a random 10-bp region ( 5′- CGCGGATCCTGCAGCTCGAGN10GTCGACAAGCTTCTAGAGCA-3′ ) as a template , PCR was performed for 12 cycles ( 1 min at 95°C , 1 min at 55°C , 1 min at 72°C ) using the following forward ( 5′-CGCGGATCCTGCAGCTCGAG-3′ ) and reverse primers ( 5′-TGCTCTAGAAGCTTGTCGAC-3′ ) to amplify a double stranded DNA product . “Cold” selection was performed by incubating the PCR template with 10 µg rLL3 in 15 mM HEPES , 25 mM KCl , and 100 µl Ni-NTA agarose ( Qiagen ) in 1X PBS . The reaction was incubated at 25°C for 30 min then added to a Poly-Prep Chromatography Column ( BioRad ) and washed with 1X PBS . DNA was eluted from the bound His-tagged rLL3 by the addition of elution buffer ( 50 mM NaH2PO4 , 300 mM NaCl , 250 mM imidazole , 0 . 05% Tween ) and used for PCR amplification ( 20 cycles ) of the resulting template for the next round of selection . An alternate “hot” selection , was performed by end-labeling the forward primer with [γ-32P] ATP using T4 polynucleotide kinase ( New England Biolabs ( NEB ) ) and used to PCR amplify a DNA template containing a random 20-bp region ( 5′- CGCGGATCCTGCAGCTCGAGN20GTCGACAAGCTTCTAGAGCA-3′ ) as above with the reverse primer ( 6 cycles ) . Labeled fragments were purified using Micro Bio-Spin columns ( BioRad ) and incubated with ∼1 µg of rLL3 protein for 20 min at room temperature in binding buffer [15 mM HEPES , 25 mM KCl , 2 µg BSA , 2 mM DTT , 10% glycerol , and 100 ng M13 reverse primer ( Invitrogen ) to reduce non-specific binding] . Reaction components were separated on a 7% polyacrylamide/TBE gel at 100 V for ∼90 minutes at 4°C , then dried and exposed to film . “Shifted” complexes were excised and incubated in TE buffer overnight at 25°C and used for PCR amplification ( 20 cycles ) for the next round of selection . For both “cold” and “hot” methods , a total of 4 rounds of selection were performed before cloning the amplified template into a pJet1 . 2 vector ( Fermentas ) for sequencing . The resulting selected DNA fragments were used as input for MEME analysis to generate consensus motifs [16] . To identify regions of the SRPN6 promoter that are capable of binding rLL3 , a 1 kb region of the putative promoter was dissected into five fragments of 200 bp and amplified by PCR using the primers listed in Table S1 . Following PCR purification with the DNA Clean and Concentrator kit ( Zymo Research ) , 1 pM of DNA was radiolabeled with [γ-32P] ATP using T4 polynucleotide kinase ( NEB ) according to the manufacturer's protocol and purified with Micro Bio-Spin columns ( BioRad ) . Radiolabeled fragments were incubated in binding buffer [15 mM HEPES , 25 mM KCl , 2 µg BSA , 2 mM DTT , 10% glycerol , 100 ng M13 reverse primer ( Invitrogen ) , and ∼250 ng of poly dA/dT to reduce non-specific binding] and ∼250 ng rLL3 protein for 20 min in the presence or absence of specific ( self ) or non-specific ( rpS7 ) competitor fragments . EMSA reactions were fractionated on a 7% polyacrylamide/TBE gel at 100 V for ∼90 min at 4°C , then dried and exposed to film . A more detailed analysis of the 200 bp fragments that demonstrated putative binding was performed by dissecting each fragment into five fragments of ∼40 bp . Forward and reverse-complementary oligonucleotides were synthesized corresponding to each region ( Table S1 ) , annealed to form a double-stranded DNA fragment , and radiolabeled with [γ-32P] ATP as previously mentioned . Incubation with rLL3 was performed in the presence of specific ( self ) or non-specific ( AgB2t ) competitor fragments . Supershift assays were performed as described above with the exception that DTT was removed from the binding buffer to not interfere with antibody affinity , since DTT is a strong reducing agent . Anti-His , anti-LL3 , or mouse pre-immune sera were added to the binding reactions at a 1∶1000 dilution , and incubated for 20 min before gel loading . To measure regulation conferred by LL3 binding to the SPRN6 promoter , SRPN6 regulatory regions were cloned into a pGL2-control ( Promega ) luciferase vector as follows . The 130-bp SRPN6 5′UTR was PCR amplified from cDNA with the primers ( restriction sites underlined ) F: 5′-ATAAGATCTGTCTCGAGAGCGTACACCAGCGTAACGG-3′ and R: 5′-ATAAAGCTTTGTGGAGCATTCAACTCCAACGTTCAAC-3′ . Following restriction digestion with BglII and HindIII , the fragment was gel purified using the Gel DNA Recovery kit ( Zymo Research ) and cloned into the corresponding BglII and HindIII sites within the pGL2-control vector ( Promega ) . Subsequently , 1 kb of the SRPN6 promoter was PCR amplified ( using the primers F: 5′- GCAGCCGGTATGGCCGGTTGTGGTTAAATTC-3′ and R: 5′- TGAATGGCTTCGATCGGCGGTGAAAC-3′ ) , and sub-cloned into the pJet1 . 2 vector ( Fermentas ) . The SRPN6 promoter fragment was then digested with BglII and ligated into the corresponding BglII site in the pGL2-SRPN6-5′UTR construct and sequenced to verify the accuracy of the inserted DNA sequence . Using phosphorylated primers , putative LL3 binding sites within the SRPN6 promoter were disrupted by PCR mutagenesis as follows . Primer pairs targeting the putative sites from −800 to −790 ( F: 5′-CCGGCACTAGCTCCaaaaaaaaaaTGTCATTTTGAAGGCGTTAAA-3′ , R: 5′-ATGAAAACGATTCTGTTTCAATGTGTTTACGGTGCAGT-3′ ) , −770 to −760 ( F: 5′- TTTTGAAGGCGTTAAaaaaaaaaaaAGTGTGTTTAAGCTTCCG-3′ , R: 5′-TGACATCATAACCATGGAGCTAGTGCCGGAAGAAAACG-3′ ) , and −162 to −153 ( F: 5′-CGTCCAAGCACTCCAaaaaaaaaaaAGCAGCAGCAGCAGCAGCAGC-3′ , R: 5′- GTAAAAGTGCAAAATTTGCAATCGCAAATGGCACC-3′ ) using Phusion polymerase ( NEB ) and the pGL2-SRPN6 promoter plasmid as a template . Following PCR , the linearized plasmids were ligated and transformed , then sequenced to verify the mutation . Double and triple mutants were sequentially created by a second or third PCR mutagenesis using a mutated −770 to −760 template as outlined above . Luciferase assays were performed by transfecting a 100∶1 ratio of each respective SRPN6 firefly luciferase promoter construct with a Renilla luciferase internal control construct under the copia promoter in Anopheles gambiae Sua5B cells using a standard Effectene ( Qiagen ) protocol . Luciferase expression was measured using the Dual Luciferase Reporter Assay System ( Promega ) in the presence , or absence , of heat-killed E . cloacae following the protocol outlined in Gupta et al . [7] . Each sample was measured in triplicate and experiments were performed in duplicate . | The mosquito innate immune system serves as the primary defense response against invading pathogens , including that of the malaria parasite Plasmodium . The mosquito immune response is remarkably efficient in eliminating the parasite as indicated by the low prevalence of Plasmodium oocysts in wild caught mosquitoes . In an effort to understand the mechanisms of immune response , we report the first evidence of a LPS-induced TNF-α factor ( LITAF ) -like gene family in insects and describe the role of one member , LITAF-like 3 ( LL3 ) , in anti-Plasmodium immunity in the mosquito Anopheles gambiae . Silencing of LL3 greatly increases parasite survival . The gene appears to function as a transcription factor that binds to specific regions of the SRPN6 promoter , a known anti-Plasmodium gene , and modulates its transcript abundance . In summary , LL3 appears to be a novel component of the mosquito innate immune response . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"mosquitoes",
"immunity",
"vector",
"biology",
"innate",
"immunity",
"anopheles",
"biology",
"microbiology",
"host-pathogen",
"interaction",
"parasitology"
] | 2012 | Regulation of Anti-Plasmodium Immunity by a LITAF-like Transcription Factor in the Malaria Vector Anopheles gambiae |
Successful control of falciparum malaria depends greatly on treatment with artemisinin combination therapies . Thus , reports that resistance to artemisinins ( ARTs ) has emerged , and that the prevalence of this resistance is increasing , are alarming . ART resistance has recently been linked to mutations in the K13 propeller protein . We undertook a detailed kinetic analysis of the drug responses of K13 wild-type and mutant isolates of Plasmodium falciparum sourced from a region in Cambodia ( Pailin ) . We demonstrate that ART treatment induces growth retardation and an accumulation of ubiquitinated proteins , indicative of a cellular stress response that engages the ubiquitin/proteasome system . We show that resistant parasites exhibit lower levels of ubiquitinated proteins and delayed onset of cell death , indicating an enhanced cell stress response . We found that the stress response can be targeted by inhibiting the proteasome . Accordingly , clinically used proteasome inhibitors strongly synergize ART activity against both sensitive and resistant parasites , including isogenic lines expressing mutant or wild-type K13 . Synergy is also observed against Plasmodium berghei in vivo . We developed a detailed model of parasite responses that enables us to infer , for the first time , in vivo parasite clearance profiles from in vitro assessments of ART sensitivity . We provide evidence that the clinical marker of resistance ( delayed parasite clearance ) is an indirect measure of drug efficacy because of the persistence of unviable parasites with unchanged morphology in the circulation , and we suggest alternative approaches for the direct measurement of viability . Our model predicts that extending current three-day ART treatment courses to four days , or splitting the doses , will efficiently clear resistant parasite infections . This work provides a rationale for improving the detection of ART resistance in the field and for treatment strategies that can be employed in areas with ART resistance .
Malaria remains a scourge of humanity , affecting hundreds of millions of people and causing ~600 , 000 deaths each year [1] . Infection with Plasmodium falciparum is responsible for the majority of severe malaria cases . During the asexual blood phase of its lifecycle , this protozoan parasite invades , grows , and multiplies within red blood cells ( RBCs ) . The initial stage of intraerythrocytic growth ( 0–~24 h ) , during which the parasite exhibits an unfilled cytoplasm in Giemsa-stained smears ( referred to as “rings” ) , is characterized by a relatively slow metabolism [2] . Ring-stage—infected RBCs are freely circulating and are thus the predominant stage detected in samples taken from the peripheral blood of infected patients . From ~24 h to ~40 h post-invasion ( p . i . ) , in the “trophozoite” ( or growing ) stage , the parasite increases the rate of uptake and digestion of hemoglobin from the host cytoplasm and shows a large increase in metabolic rate . These mature parasites are characterized by the presence of hemozoin , the classic “malaria pigment” that results from hemoglobin digestion . Trophozoites are rarely observed in the circulation of infected patients because of their adherence to endothelial cells and consequent sequestration away from the circulation . Complications associated with cerebral sequestration are responsible for much of the malaria-related mortality and morbidity [3] . From ~40 h p . i . , the parasite undergoes cytokinesis , forming a schizont that can contain up to 32 daughter parasites ( merozoites ) . At ~48 h p . i . , the schizont bursts , releasing the merozoites and heralding a new round of infection . Artemisinin and its derivatives ( collectively referred to as ARTs ) have contributed enormously to decreasing rates of malaria deaths over the last decade . ARTs are among the few antimalarials that are active against ring-stage parasites , thus reducing the parasite burden in P . falciparum infections quickly and providing prompt therapy for severe infections [3] . The ARTs contain an endoperoxide group that is critical for their activity . The mechanism of ART action remains poorly understood , but ARTs are thought to be pro-drugs that need to be activated by opening of the endoperoxide ring , i . e . , splitting the bonded oxygen atoms [4] . This process requires the presence of heme or non-heme iron sources ( and possibly other activators ) [5 , 6] . The activated ART intermediates are thought to react with susceptible ( nucleophilic ) groups within parasite proteins and other cellular components , leading to parasite killing; however , the details remain unclear [7] . A disadvantage of ARTs is their short half-lives in vivo ( ~1–2 h ) . Accordingly , they are co-administered with longer half-life partner drugs in ART combination therapies ( ACTs ) to prevent recrudescence and to slow the emergence of resistance [8] . Current antimalarial control is highly dependent on ACTs , which makes the emergence of ART resistance extremely concerning [9–11] . Decreased sensitivity to ARTs , which manifests as delayed parasite clearance , is now a problem in six Southeast Asian countries and is translating into decreased clinical efficacy in areas with concomitant partner-drug resistance [12 , 13] . Enormous efforts are underway to contain and eliminate ART resistance . Initially , monitoring ART resistance was hampered by the lack of a suitable in vitro correlate [10] . Recently , assays employing short pulses mimicking clinical drug exposure [14–16] , such as the ring-stage survival assay ( RSA ) , have provided a good correlation between reduced in vitro sensitivity to dihydroartemisinin ( DHA ) , the clinically relevant ART derivative , and delayed parasite clearance [15] . Combined with whole genome sequencing , this allowed the identification of mutations in a protein with a Kelch domain ( i . e . , β-propeller tertiary structure , referred to as K13; PF3D7_1343700 ) [17] that are strongly associated with the slow-clearance phenotype . A large Genome-Wide Association Study ( GWAS ) added further support to the suggestion that K13 is the major locus controlling P . falciparum resistance to ARTs [18] . Recent studies using genetic modification of the K13 locus have confirmed a central role for K13 mutations in conferring ART resistance [19 , 20] . Here we present a detailed in vitro investigation of the drug responses of K13 wild-type and mutant isolates of P . falciparum sourced from a region in Cambodia ( Pailin ) with a marked penetrance of ART resistance [10 , 21] . Our results provide insights into the molecular basis of ART action and resistance and point to a class of compounds that could be used to synergize the activity of ARTs against both sensitive and resistant parasites . Modelling of parasite responses suggests alternate therapeutic regimens that should be vigorously pursued and provides tools that will immediately impact the way resistance is measured in the field .
We determined the whole genome sequences of four laboratory-adapted isolates of P . falciparum collected from adult patients enrolled in clinical trials conducted between 2009 and 2010 in the Pailin Referral Hospital in Cambodia [21] . High sequencing coverage with good mapping quality was achieved across all four genomes ( mean sequence coverage of 77x and mean mapping quality >30; see S1 Table ) . A Pairwise Distance matrix ( 5x5 ) built on 26 , 438 SNPs in coding genes was used to generate a Neighbor Joining tree to examine isolate relatedness . The isolates are quite divergent , with ~16 , 000 non-synonymous SNPs between any two isolates ( excluding var , rif , and stevors ) , and they exhibit multiple mutations in many of the drug-resistance—related genes . One strain ( PL2 ) encodes the wild-type K13 genotype , while the others ( PL1 , Y493H; PL5 , C580Y and PL7 , R539T ) represent three K13 mutants that are commonly observed in Cambodia ( S2 Table ) . A recent fine-structure analysis of parasite samples collected as part of the large-scale Tracking Resistance to Artemisinin Collaboration ( TRAC ) study revealed that particular non-synonymous polymorphisms in apicoplast ribosomal protein S10 ( arps10 ) , multidrug resistance protein 2 ( mdr2 ) , ferredoxin ( fd ) , and chloroquine resistance transporter ( pfcrt ) are markers of a genetic background on which K13 mutations are likely to arise , but individually they have little contribution to ART resistance [18 , 22] . Other GWAS have reported SNPs in other genes that show association with the ART resistance phenotype [23–27] . The K13 mutant and wild-type Pailin strains exhibited the expected SNPs at the arps10 ( V127M ) and mdr2 ( T484I ) loci and at one of the pfcrt ( I356T ) loci but showed variable sequences at the other loci ( see S2 Table for a summary of some relevant loci ) . Both K13 mutant and wild-type strains exhibited the fd ( D193Y ) and pfcrt ( N326S ) SNPs that are strongly associated with ART resistance founder populations [18] . The entire genomes for the four Pailin strains have been made available through the European Nucleotide Archive ( ENA ) database under the accession number PRJEB8074 . We previously demonstrated that the sensitivity of laboratory strains of P . falciparum to ARTs exhibits a complex dependence on drug exposure time and concentration [14] . For this work , we used assays comprising very short drug pulses in an effort to mimic in vivo exposure and to maximize the discrimination of subtle differences in parasite responses . We defined parasite viability as the fraction of the parasite population that survives drug exposure and is able to enter the next parasite cycle . We found that the 50% lethal dose ( LD50 ) and the viability at saturating drug concentrations ( the minimum viability , Vmin ) are the most useful measures of cellular cytotoxicity , in agreement with other studies of ARTs and other drugs [15 , 28 , 29] . In the current work , tightly synchronized cultures of the four strains were subjected to 3-h pulses of DHA at different stages of the asexual lifecycle ( Fig 1A and 1B ) . Notably , the PL2 strain exhibits a low Vmin ( 3h ) ( <5% ) across all stages of development ( Fig 1A ) , confirming an ART sensitive phenotype . In contrast , the PL1 , 5 , and 7 strains displayed 5%–60% survival over the first 10–12 h p . i . ( Fig 1A and 1B ) . The youngest ring stages ( 1 . 2 h p . i . ) of the mutant strains exhibited the greatest viability following drug exposure ( Vmin ( 3h ) >40% ) , as previously reported [15] , confirming the resistant phenotype of these strains . Of note , the PL2 and PL7 strains exhibit similar LD50 ( 3h ) values at 6 h p . i . , but PL7 shows a higher capacity to survive drug exposure ( Vmin ( 3h ) = 20% ) ( Fig 1A ) . Thus Vmin ( 3h ) appears to be the more sensitive indicator of the resistance genotype , as suggested previously [14] . We also monitored the drug response of mature parasites during their transition into the next cycle ( Fig 1A and 1B , b panels ) . Continuous in vitro culturing of parasite cultures results in broadening of the age distribution over time . We measured the degree of broadening in this study ( e . g . , the schizont-to-ring transition shown in the b panels in Fig 1 ) . In agreement with a previous study [14] , ~80% of parasites that are synchronized to a 1-h window undergo the schizont-to-ring transition over a period of ~4 h in the next cycle ( i . e . , ~48 h later ) . Late trophozoites and early schizonts ( average ages earlier than 6 h pre-invasion ) of all strains exhibit similar LD50 ( 3h ) values with no detectable survival when exposed to 0 . 7 μM DHA . By contrast late-stage PL1 , 5 , and 7 schizonts ( average ages later than 4 h pre-invasion ) , which form rings during the course of the assay , exhibit increased LD50 ( 3h ) and Vmin ( 3h ) values . This indicates that the mutant strains have an increased ability to survive a clinically relevant drug pulse ( 0 . 7 μM DHA; 3 h ) from -4 to +12 h p . i . , encompassing almost one-third of the intraerythrocytic cycle . The ability of late-stage schizonts to survive exposure is particularly important , considering each surviving parasite forms approximately ten daughter cells . There are subtle differences between the responses of the different K13 mutants . These may be due to additional genetic factors as well as to direct effects of the different mutations , as indicated by a detailed reverse genetics analysis of the contribution of K13 to resistance in recent Cambodian isolates and reference lines [20] . The time points for transition from ring to trophozoites ( i . e . , the 50:50 point; indicated with "T" in Fig 1A and 1B ) and the lifecycle durations for the K13 mutant and wild-type parasites were: PL1 ( 30 h/49 h ) , PL5 ( 30 h/59 h ) , PL7 ( 27 h/57 h ) and PL2 ( 27 h/52 h ) , and 3D7 ( 22 h/41 h ) . Thus , among this small sample size , there does not appear to be a simple correlation between the K13 mutation and the length of the ring stage or of the intraerythrocytic cycle . Our previous work with laboratory strains showed that ART-mediated killing requires that parasites are exposed to a sufficient concentration of activated ART for a sufficient period of time [14] . We defined a semi-empirical cumulative effective dose ( CED ) model that accounts for the complex in vitro dependence of parasite viability on drug concentration and exposure time and permits the interpretation of stage and strain-dependent differences in drug action in terms of easy to understand underlying parameters . Here we analyzed the dependence of DHA action on time of exposure in K13 wild-type ( PL2 ) and K13 mutant ( PL7 , R539T ) field strains ( Figs 1C and S1 ) . A 3-h exposure of PL2 ( 2 h p . i . ) to 1 . 3 μM DHA is sufficient to reduce parasite viability to almost zero ( Fig 1C , red ) . Remarkably , a significant fraction ( 20% ) of PL7 early rings ( Fig 1C , blue ) remain refractory to a 9-h exposure to 1 . 3 μM DHA , even though the LD50 ( 9h ) value ( 20 nM ) suggests potent drug action at longer exposures; this further illustrates that Vmin is more informative than LD50 in revealing the resistance-associated phenotype . Late rings ( 18 h p . i ) and trophozoites ( 34 h p . i . ) from both strains show similar sensitivity to >3 h exposure to DHA , but small differences in LD50 are evident at very short exposure times ( Fig 1C ) . We found that the CED model is able to adequately describe the response of the Cambodian strains to DHA at different stages of development , as well as the dependence of that response on concentration and exposure time . That is , the CED model can be used to generate the fitted curves in Fig 1C and S1 Fig at the very early ring ( 2 h p . i . ) , the late ring ( 18 h p . i . ) and the mid trophozoite ( 34 h p . i . ) stages . In this model , the effective dose ( ED ) is a saturable function of the drug concentration and is defined by Km , the drug concentration resulting in half the effect of the maximally effective dose , EDmax . Parasite viability is then a sigmoidal function of the cumulative ED ( EDcum ) with slope γ and midpoint t50e , sat . The t50e , sat value is the time taken to kill half of the parasites at saturated ( and fixed ) drug concentration . ( See [14] for further explanation of the model ) . The drug response of late rings and mid trophozoites was similar for both wild-type and mutant strains and the analysis indicated similar CED model parameters ( Table 1; Fig 1C ) . Interestingly , early rings from the K13 wild-type and mutant strains , which produce quite a different drug response , exhibit similar Km values ( 14–19 nM; Table 1 ) . This suggests that DHA initiates an effect in both drug-sensitive and drug-resistant strains at a similar concentration , even in the early ring stages . Since the ARTs are pro-drugs that are converted to the active form by reaction with iron or heme [4 , 6] , this suggests that all strains activate the drug with similar efficiency . By contrast , there is a 3 . 5-fold increase in the t50e , sat values for DHA for early ring stage PL7 parasites ( Table 1 ) , and this underlies most of the difference in the response of this strain . As previously demonstrated for laboratory strains [14] , such an increase in t50e , sat manifests as an increase in the lag time for drug action . In other words , there is a longer delay following drug treatment before the onset of the killing of the K13 mutants . Since t50e , sat=ED50cum/EDmax ( where ED50cum is the cumulative ED required to kill 50% of the parasites [14] ) , this indicates that PL7 requires longer exposure to an effective dose to induce killing ( or the maximal effective dose produced is less ) . This difference is expected to be very important in vivo given the very short in vivo half-lives of ARTs . A number of studies have suggested that ARTs can exert cytostatic effects at sub-lethal concentrations [30 , 31] . Here we have used the RNA-binding dye SYTO-61 ( which can readily distinguish parasites of different ages [30] ) to determine whether the parasites that survive DHA exposure exhibit growth retardation . ( See Materials and Methods for labelling strategy ) . We initially examined growth effects in the laboratory strain , 3D7 , at the ring stage of development , where it exhibits 10-fold lower sensitivity to a 3-h DHA pulse compared to the trophozoite stage [14] . We found that mid-ring ( 6 h p . i . ) and later ring ( 18 h p . i . ) stage parasites that survive exposure to a 4-h drug pulse exhibit a dose-dependent decrease in the intensity of the SYTO-61 signal following treatment ( Fig 2A , green curves ) . The IC50 ( 4h ) for the growth retardation effect was similar across the ring stage ( ~10 nM ) and was 5- to 10-fold lower than the corresponding LD50 ( 4h ) value ( i . e . , cytotoxic effect ) . Examination of the population profile of SYTO-61 staining shows an absolute increase in the number of viable parasites with decreased SYTO-61 signals in the drug-treated samples compared to the untreated controls ( Fig 2B , asterisks ) . This indicates that the decrease in the SYTO-61 signal following drug exposure reflects drug-induced growth retardation rather than selective killing of the oldest parasites . Similarly for both the K13 wild-type ( PL2 ) and R539T mutant ( PL7 ) strains , early ring-stage parasites ( 2 h p . i . ) that survive drug exposure exhibit dose-dependent growth retardation ( Fig 2C , green curves ) . Interestingly , the two strains exhibit similar IC50 ( 3h ) values for growth retardation ( 10 nM ) despite their very different sensitivities to killing ( LD50 ( 3h ) = 10 and >1000 nM for PL2 and PL7 , respectively; Fig 2C , compare green and grey curves ) . These IC50 ( 3h ) values remain relatively constant throughout the ring-stage , with the maximum effect evident in late rings ( Fig 2C , right panels ) . We compared the effect of the time of exposure to drug on growth retardation and viability . The response of the PL2 and PL7 strains at late-ring stage to exposure to DHA for a very short period ( 1 . 5 h ) is similar , and mainly comprises growth effects without loss of viability ( Fig 2D ) . Differences in viability manifest at longer exposures ( Fig 2D , 3 and 6h ) . While PL2 parasites succumb to longer drug exposure , the growth-retarded PL7 parasites are able to withstand drug pressure for longer . This demonstrates that it is the ability of the growth-retarded PL7 parasites to withstand subsequent drug pressure that is responsible for their resistance phenotype . Drug-induced growth retardation effects are important as they will influence the stage at which surviving parasites are exposed to recommended ART regimens , which often include daily doses administered 24 h apart . We quantitated the magnitude of this growth effect by subjecting tightly synchronized PL7 parasites ( 1 . 5 h p . i . ) to a 3 . 5-h DHA pulse ( 1 μM ) and periodically examining parasite morphology by Giemsa staining over a period of 40 h , following the drug pulse . At times >30 h , two distinct parasite populations were identified corresponding to those containing hemozoin ( predominantly trophozoite morphologies ) and those with pyknotic or early ring morphologies ( Fig 3A ) . The fraction of parasites exhibiting a trophozoite-like appearance ( 30%; n >100 parasites ) matched the viability as measured by flow cytometric analysis in the cycle following the drug pulse ( 28% ) , indicating that these trophozoites represent the viable population and confirming this as a simple method for measuring loss of viability ( i . e . , parasites rendered incapable of reproduction ) in the same cycle as the drug treatment . A comparison of the sizes ( areas of Giemsa-stained parasites ) of the surviving trophozoites with those from an untreated culture shows that the surviving parasites were , on average , delayed 6 h in their progression through the cycle and exhibited a broader age distribution ( Fig 3A and 3B ) . The above analysis indicated that unviable parasites ( i . e . , incapable of reproduction ) can exhibit normal ring morphologies for some time after exposure to ART . We examined the timing of the appearance of pyknotic forms in Giemsa smears , following a DHA pulse , to determine whether the differential drug response of sensitive and resistant strains also influenced the rate of generation of pyknotic forms . Early ring-stage parasites ( 1 . 5 h p . i . ) were pulsed with 1 μM DHA for 3 . 5 h to generate unviable parasites ( 100% and 72% of PL2 and PL7 parasites , respectively , based on flow cytometric analysis in the cycle following the drug pulse ) , and parasite morphology was quantitated ( from Giemsa smears ) every 2–4 h following the drug pulse . Surprisingly unviable PL2 and PL7 parasites retained a ring-like morphology 13 h after the drug pulse , with pyknotic forms comprising <10% of the unviable population ( Fig 3C and 3D ) . A significant fraction of unviable parasites from both strains ( 12% and 28% for PL2 and PL7 , respectively ) exhibited ring-like morphologies 35 h following the drug pulse . Interestingly , the half-time for adoption of pyknotic morphology was significantly longer for PL7 ( 32 h ) than for PL2 ( 23 h ) . This indicates that unviable resistant parasites retain a ring-like morphology for longer following DHA treatment . In vivo , these unviable K13 mutant parasites may remain in the blood stream for longer . The growth retardation caused by exposure to DHA is reminiscent of the cell stress response observed in other organisms . In other systems , cellular insults can result in protein unfolding , culminating in polyubiquitination of proteins and their destruction via the proteasome . Interestingly , a very recent analysis of the transcriptomes of parasite samples collected as part of the TRAC study provided evidence for up-regulation of protein homeostasis genes ( such as the ubiquitin-proteasome pathways ) that correlates with delayed clinical clearance of parasites [32] . To determine the level of protein damage following ART treatment , we sought to determine the level of ubiquitinated proteins in parasite extracts . We found that we were not able to reliably measure the parasite-associated signal above the host RBC background at the very early ring stage ( mean fold change = 0 . 9 ± 0 . 2; n = 5 ) . Therefore we examined effects in trophozoites ( which have much higher levels of protein ubiquitination ) where K13 wild-type and mutant trophozoite stage parasites show less dramatic , but still measureable differences in response to very short pulses of DHA ( see Fig 1C , 34 h p . i ) . We found that this differential response was more pronounced when parasites were treated with the less potent parent drug , artemisinin ( qinghaosu; QHS ) . We found that a very short pulse ( 90 min ) of QHS killed PL7 much less efficiently than PL2 and 3D7 ( LD50 ( 1 . 5h ) values of 857 , 208 , and 153 nM , respectively . To determine the effect of QHS treatment on cell stress levels , infected RBCs were saponin-lysed to release the soluble RBC cell contents and parasite extracts were subjected to SDS-PAGE and probed with an antibody that recognizes ubiquitin ( Fig 4A ) . A profile of ubiquitination was observed similar to that reported previously [33] , with protein ubiquitination in trophozoites notably higher than in uninfected RBC ghosts . The level of protein ubiquitination increased significantly upon ART treatment ( 90 min pulse of 1 μM QHS ) , consistent with engagement of the ubiquitin-proteasome system . The level of protein ubiquitination was higher in ART-treated 3D7 and PL2 parasites than in PL7 parasites ( Fig 4A ) , consistent with the resistant parasites experiencing a lower level of cellular stress . By contrast , a 90 min pulse treatment with 20 nM WR99210 ( a concentration sufficient to cause 100% killing ) had no effect on the level of ubiquitination ( Fig 4A , right panel ) . A quantitative analysis of several experiments is presented in Fig 4B . Given the evidence for accumulation of ubiquitinated proteins following ART treatment , we examined the effects of inhibitors of the proteasome; a proteinase complex that plays a critical role in degrading unfolded proteins . Proteasome complexes are present in both the host and parasite cytoplasm , though selective inhibition of the host proteasome does not affect parasite growth or replication [34 , 35] . Epoxomicin is a well-characterized and highly specific proteasome inhibitor [36] and has activity against the P . falciparum proteasome [37] . Epoxomicin showed activity against all stages of the field strains , when used alone , with maximal potency against early ring-stages ( see y-axis intercepts in right panels in Figs 5 and S2 ) . This activity was independent of K13 genotype . We examined the interaction between DHA and epoxomicin by examining the effect of epoxomicin on the dose response profile of DHA , and by measuring the isobologram for the pair of drugs [38] at the 50% lethal dose level . We initially examined 3D7 parasites . Unlike the Pailin strains , early ring-stage 3D7 parasites ( 2 h p . i . ) exhibit ART hypersensitivity [14] and we observed no interaction between DHA and epoxomicin at this stage ( Fig 5A , top ) panel . In contrast , a sub-lethal concentration of epoxomicin ( 18 nM ) enhances the potency of DHA ~10-fold against the ring-stage of 3D7 ( Fig 5A , middle panel ) ; this is illustrated by the concave shape of the LD50 isobologram . This suggests that the proteasome inhibitor overcomes the cell defense systems that protect the mid-ring stage of 3D7 . We next examined the ability of epoxomicin to synergize the action of DHA against the K13 mutant isolate , PL1 . A pronounced synergistic interaction is evident at early ring and ring stages ( Fig 5B ) . Notably , some synergism is also evident in the trophozoite stage ( Fig 5B , bottom panel ) . This is in stark contrast to the effect of hemoglobinase inhibitors , which produce strong antagonistic interactions in the trophozoite stage [30] . We observed a similar synergistic interaction with two other K13 mutant strains ( PL5 and 7 ) , with pronounced synergism at early-ring and ring stages ( S2B and S2C Fig ) . The synergistic effect was less pronounced but still evident in the Pailin K13 wild-type isolate ( PL2 ) ( S2A Fig ) . Very recently , modification of the P . falciparum K13 locus in defined genetic backgrounds was used to demonstrate a central role for K13 mutations in conferring ART resistance [20] . These studies included a laboratory-adapted K13 mutant isolate from Cambodia ( Cam3 . II_ R539T ) and a reverted transfectant in the same line , in which the K13 wild-type genotype has been restored . We examined the sensitivity of these parasites to DHA at different stages of intraerythrocytic development ( S3 Fig ) . Cam3 . II exhibited marked resistance to DHA in the very early ring-stage with a Vmin ( 3h , 1 μM ) value of 69% and an LD50 value of >>1 μM ( S3 Fig ) , while the Cam3 . II_rev line showed markedly enhanced sensitivity with a Vmin ( 3h , 1 μM ) value of 8% and an LD50 value of 47 nM , in good agreement with the recent report [20] . We examined the interaction of epoxomicin with DHA against Cam3 . II_R539T and Cam3 . II_rev . Epoxomicin exhibited very strong synergism with DHA in the very early ring-stage of the Cam3 . II_R539T isolate ( S3 Fig ) , consistent with the proteasome inhibitor overcoming the K13-mediated resistance mechanism . However , synergism was also observed in the ring and trophozoite stages of Cam3 . II and in all stages of the revertant line ( S3 Fig ) . This confirms that proteasome inhibitors can enhance the activity of ARTs against both sensitive and resistant parasites . Proteasome inhibitors are used clinically in humans to treat myeloma [26] . We examined the effect of two of these compounds , Carfilzomib , an epoxyketone , and Bortezomib , a peptide boronate [39] . Like epoxomicin , Carfilzomib was potent against all strains and stages examined and exhibited strong synergism with DHA , particularly in the less sensitive , very early ring stage ( S4 Fig ) . Similarly , the clinically used proteasome inhibitor Bortezomib exhibited strong synergism with DHA ( S5 Fig ) . These results confirm that the proteasome is involved in the parasite's response to DHA and that inhibiting its activity enhances the level of killing of the parasite . Proteasome inhibitors such as Carfilzomib have previously been tested for in vivo activity against a murine model [40] . While the activity of Carfilzomib alone against P . berghei is low , we were interested to determine whether sub-lethal concentrations of Carfilzomib might synergize the activity of DHA in the P . berghei mouse model . In agreement with a previous report [40] , we found that treatment with up to 1 mg/kg of Carfilzomib monotherapy had no toxic effects , but also had no beneficial effects in reducing parasite burden ( Fig 6A ) . We found that DHA treatment alone ( initiated at ~1% parasitaemia ) at a dose of 5 or 10 mg/kg/day gave a moderate decrease in parasite burden ( Fig 6B and 6C ) , while 15 or 20 mg/kg was sufficient to abrogate the parasite burden ( S6A and S6B Fig ) . By contrast , a combination of Carfilzomib ( 0 . 5 or 1 mg/kg/day ) and DHA ( 5 mg/kg/day ) was associated with a significant reduction in the parasite growth ( Fig 6B , red and blue curves ) while a combination of Carfilzomib ( 0 . 5 or 1 mg/kg ) and DHA ( 10 mg/kg ) almost completely abrogated the parasite burden ( Fig 6C , red and blue curves ) . Of particular interest is the observation that a combination of DHA ( 5 mg/kg ) + Carfilzomib ( 1 mg/kg ) ( S6C Fig , green curve ) or DHA ( 10 mg/kg ) + Carfilzomib ( 0 . 5 and 1 mg/kg ) ( S6C Fig , blue and purple curves ) largely abrogated the parasites in the circulating reticulocytes , the blood cell preferentially parasitized by P . berghei . This further confirms the role of the proteasome in protecting malaria parasites against the toxic effects of DHA and points to a possible means of synergizing the activity of ARTs in vivo , using repositioned proteasome inhibitors . The availability of the data from our extensive kinetic analysis of the K13 wild-type and mutant Pailin strains offers the possibility of modelling the drug response of resistant and sensitive parasites at different exposure times and concentrations in vitro and also of extending this analysis to infer behavior in vivo [14] . To do this , we extended the CED mathematical model to take into account the age- and exposure-time—dependence of the drug responses of the sensitive and resistant strains , as well as drug-induced growth retardation and population broadening . We provide an Excel spreadsheet that presents the Mathematical Model in a user-friendly format ( S1 Spreadsheet ) . We have provided a presentation in Prezi that explain the steps involved in using the spreadsheet ( https://prezi . com/9xo9b8igvjzl/ ) . We applied the model to predict parasite clearance rates in vivo during a three-day course of ART monotherapy . We assumed DHA is applied at a rate of 2 mg/kg at 0 , 24 , and 48 h , i . e . , in a typical regimen [41] . We took into account the age distribution of parasites at the time of patient presentation ( which varies depending on disease severity [32 , 42 , 43] ) , the age-dependence of drug action ( CED parameters , from Table 1 ) , the broadening of the age distribution with time ( S1 Appendix ) , and drug-induced growth effects ( from Figs 2 and 3 ) . This simulation evaluates the in vivo consequences of the different in vitro drug responses of ring-stage K13 wild-type and mutant parasites ( see S1 Appendix ) . Strikingly , the first and third ART doses given to a hypothetical patient with a PL7-like ( resistant ) infection result in <10-fold reduction of viable parasites ( Fig 7 ) . In contrast , the PL2 ( sensitive ) strain shows a 50-fold reduction in parasite burden at the corresponding times ( Fig 7A , orange curves ) . As a result , the parasite burden in a PL7-like infection is ~50-fold higher after a three-day treatment . As a consequence of drug-induced synchronization and growth retardation , the parasite age distribution at 72 h consists predominantly of late rings ( S7 Fig ) . Importantly , administration of an additional dose at 72 h is predicted to decrease the parasite load of a PL7-like infection at 96 h to the level observed with a three-day treatment of a PL2-like sensitive infection ( Fig 7A , asterisks ) . In consequence , we predict that a four-day course of ACT will significantly reduce the incidence of treatment failure in areas with ART resistance . In field studies , Giemsa smears are used to monitor the effectiveness of drug treatment . This approach , however , only detects circulating rings and not mature , sequestered parasites . Our simulation incorporates sequestration and shows that the density of circulating viable parasites will exhibit time-dependent fluctuations over many orders of magnitude , particularly in the period 24–48 h after treatment ( Fig 7A , grey curves ) . These fluctuations are not evident in real patient-averaged data [41] , as shown in Fig 7B ( symbols ) , nor in data from individual patients [42] . We posit that this discrepancy arises because our initial simulation assumes that unviable parasites are immediately removed from the circulation . Splenic clearance represents the major in vivo route for removing killed rings [44 , 45] , with unviable parasites persisting for >60 d after ART chemotherapy in splenectomized patients [46] . Our analysis in Fig 3C and 3D shows that unviable rings retain their ring-like morphologies for many hours , indicating splenic clearance would similarly occur over a period of hours . Estimates of the half-life for splenic clearance of dead parasites range from 3 to 6 h [47 , 48] . Incorporating a clearance half-life of 5 h for removal of unviable parasites ( Fig 7B , dashed lines ) produces clearance profiles that resemble those measured in patients exhibiting delayed clearance from Pailin ( blue symbols ) , but predicts that both the mutant and wild-type strains would be removed from the circulation at similar rates , despite clear differences in killing ( Fig 7A , orange curves ) . This strongly suggests the existence of additional strain-dependent factors involved in the clearance of unviable parasites . A possible explanation for a strain-dependent effect is that unviable parasites persist in the circulation ( with a ring-like morphology ) before changing their physico-mechanical properties sufficiently to initiate splenic clearance . This is consistent with the delayed appearance of pyknotic forms following treatment of cultures with DHA ( Fig 3C and 3D ) , which indicates that unviable parasites retain ring-like morphologies for an extended period . It is also of interest that the half-time to pyknosis is particularly extended in PL7 ( K13 mutant ) parasites . Indeed , incorporation of an additional strain-dependent term , as well as maintaining a strain-independent splenic clearance rate , permits improved predictions of the observed parasite clearance curves ( Fig 7B , solid curves ) . This analysis has implications for monitoring parasite clearance times in the field . We anticipate that killed rings that retain a ring-like morphology would persist in circulation and be counted in Giemsa smears . Moreover , unviable resistant parasites would persist for longer in the circulation . Our simulations indicate that in clinical practice , the ring-stage parasites that are detected in Giemsa smears , after ART treatment , likely comprise mainly unviable parasites ( Fig 7C , dashed curves ) . A significant fraction of these will have been rendered unviable after the first dose ( dotted lines ) , especially in fast clearing , sensitive strains . Consequently , the relationship between the in vitro marker of resistance ( decreased parasite killing ) and the clinical marker of resistance ( delayed parasite clearance ) is more complex than previously appreciated . To examine the potential effect on the interpretation of field studies , we simulated a scenario to examine the effect of a split-dose ART treatment on parasite clearance . A recent clinical trial reported that a split-dose ART treatment regimen did not improve parasite clearance times for malaria infections with either ART-sensitive or-resistant P . falciparum [49] . Indeed , the simulations ( see S1 Appendix , S4A Fig ) show that splitting the dose will have only a very small effect on the parasite clearance curves . By contrast , there is a very large effect on the number of viable parasites . The simulation predicts that a split-dose regimen should reduce the load of a resistant ( PL7-like ) infection to a level well below that observed in a sensitive ( PL2-like ) infection subjected to a standard treatment . Direct quantitation of circulating viable parasites during drug treatment would complement the current parasite clearance approach for classifying ART resistance in the field . Our simulations indicate that the fraction of circulating parasites that are viable 3 h after commencement of treatment is independent of the typical variation of serum DHA concentrations ( Cmax = 0 . 5 to 20 μM [9 , 49] ) and represents a useful parameter for characterizing the in vivo response of a particular strain ( S8 Fig ) . Within this short period of time , there is sufficient difference in loss of parasite viability to distinguish between sensitive and resistant infection , without the complicating effects of significant loss of viable parasites due to sequestration , or from splenic clearance of unviable parasites , which complicates the interpretation at longer times ( S8 Fig ) .
It is generally accepted that ARTs are pro-drugs . That is , they are administered in an inactive form and are activated by reductive cleavage of the endoperoxide ring ( see reviews [4–6 , 50] ) . The resulting free radicals are thought to react with susceptible groups within a range of parasite proteins and other components , leading to cellular damage and killing . When applied in vitro as clinically relevant short pulses , ARTs are significantly more active against trophozoite-stage parasites than against the mid-ring stage [14] . This likely reflects , in part , the higher availability of iron-containing activators ( as a result of hemoglobin degradation ) at the trophozoite stage . However , the rate and efficiency of parasite killing will also depend on the ability of the parasite to defend itself against cellular damage . In this work , we have undertaken a careful analysis of the response of K13 wild-type and mutant parasites to ART at different stages and at different exposure times with a view to understanding and overcoming the resistance mechanism . We found that at low concentrations of DHA , we were able to distinguish cytostatic effects ( growth inhibition ) from cytotoxic activity ( which renders the parasites incapable of reproducing ) . Interestingly , while different stages and strains of P . falciparum exhibit very different levels of sensitivity to killing by DHA , the IC50 values for induction of the growth effects are similar . This is consistent with the suggestion that cytostatic effects are initiated as soon as the toxic insult is detected , with downstream killing , if and when the cell defense systems are overwhelmed . The observed growth retardation is reminiscent of the cellular stress responses reported in other organisms . For example , oxidative and non-oxidative stress events activate an unfolded protein response , leading to shut-down of protein translation and other metabolic pathways [51 , 52] . While P . falciparum appears to lack the genes for a classical unfolded protein response [53] , it possesses a functional ubiquitin-proteasome system [33 , 54] and has been shown to undergo eIF2-α–mediated arrest of protein translation , leading to stalling of growth [55] . Consistent with this , we found that the level of ubiquitinated proteins is increased upon exposure to an ART insult , indicating that activated ART damages proteins and initiates a stress response that engages the ubiquitin-proteasome system . By contrast , exposure to a lethal pulse of an anti-folate inhibitor ( WR99210 ) had no effect on the level of ubiquitination . Because ARTs are very short-lived in vivo , the growth stasis that is induced by ART exposure would buy time for the proteasome to degrade ubiquitinated proteins , enabling survival until the ART concentration has declined . Resistance could arise as a result of decreased ART activation or via the mitigation of downstream damage . ART-sensitive and-resistant Pailin strains show similar Km values for ART-induced killing and similar IC50 values for growth inhibition , indicating similar rates of ART activation . By contrast the extended lag phase before onset of killing and the delayed conversion to the pyknotic state exhibited by K13 mutants , as well as the lower levels of ubiquitinated proteins and the synergism with proteasome inhibitors , are consistent with the suggestion that an enhanced cellular stress response underlies resistance . We present a possible model for ART action and the cell stress response in Fig 8 , in which death occurs when the level of damage overwhelms the parasite's proteasome system . As predicted by this model , we found that proteasome inhibitors , such as epoxomicin , Carlifzomib , and Bortezomib , markedly synergize the action of DHA . In the laboratory strain 3D7 this effect is particularly marked in the mid-ring stage of development , when the parasite shows low sensitivity to DHA , but is not evident in the very early ring stage , when the parasite exhibits ART hypersensitivity . In the K13 mutant strains , PL1 , 5 , and 7 , the synergism is particularly marked at the very early ring stage when these parasites are especially resistant to DHA . In an effort to distinguish the role of the K13 gene product from other contributing genetic differences , we examined the level of synergism in a Cambodian K13 mutant ( Cam 3 . II ) and a genetically matched K13 wild-type ( reverted ) transfectant . The revertant shows very early ring stage sensitivity that is similar to that that observed for the Pailin wild-type strain ( PL2 ) . Epoxomicin markedly synergized the activity of DHA against the very early ring stage of the K13 mutant , but also increased its activity against the revertant . This suggests that a proteasome-engaging cell stress response is involved in protecting both sensitive and resistant parasites from the action of ARTs , but that this response is more effective in the K13 mutant parasites . Our data suggest that proteasome inhibitors could provide a synergistic combination with ARTs that would enhance their effectiveness against both K13-mutant and wild-type parasites . K13 shares some sequence similarity with KEAP1 and KLHL8 , which are involved in E3 ubiquitin ligase complexes that regulate the cytoprotective and developmental responses in mammalian systems [56] . A recent transcriptomic analysis showed up-regulation of the unfolded protein response in K13 mutant parasites , including genes involved in protein folding , unfolded protein binding , protein export , post-translational translocation , signal recognition particle , endoplasmic reticulum retention sequences , the proteasome , and the phagosome [32] . Taken together with our data , this strongly indicates a role for enhanced proteostasis mechanisms as the basis for ART resistance in P . falciparum . The transcriptomic analysis of Mok et al . also suggested that , at a population level , resistant parasites exhibit decelerated development at the ring stage . We found that parasites ( e . g . PL1 and PL2 ) that exhibit similar lifecycle profiles ( as judged by Giemsa analysis ) can have very different responses to ARTs . This indicates that while decelerated development may be a contributor , other factors are also important . Further work is required to fully elucidate the role of K13 in enhancing the cell stress response . Importantly , our work suggests that a proteasome inhibitor could be used to synergize the activity of ARTs in vivo , and potentially to overcome resistance . Carfilzomib and Bortezomib are FDA-approved for the treatment of multiple myeloma [57] and inhibitors that specifically target the plasmodial proteasome have been identified [39 , 58 , 59] . These compounds show low toxicity in human cell lines [59] and limited toxicity in mice [40 , 58] . In agreement with previous reports , we find that Carfilzomib shows only weak antimalarial activity , when used alone , against the ring-stage of a mouse model of malaria [40] . In contrast we observed marked synergism of the action of DHA by Carfilzomib , against P . berghei in vivo , particularly in reducing the parasite burden in reticulocytes , the preferred host blood cell . While further work is needed to determine the efficacy of proteasome inhibitors as ART-synergizing-agents in patients , this work offers a potential avenue to overcome ART resistance . We analyzed the concentration and exposure-time—dependence of the response of K13 mutant and wild-type parasites in terms of the CED model , enabling for the first time prediction of in vivo parasite clearance profiles from in vitro assessments of ART sensitivity . Our modelling indicates that slower in vivo clearance of resistant strains may reflect both decreased killing of parasites and a slower rate of clearance of parasites that have been rendered unviable . Prolonged circulation of unviable ( but morphologically unchanged ) ring-stage parasites will complicate the analysis of Giemsa-stained peripheral blood smears . This should be considered in the analysis of parasite clearance curves in surveillance studies . While current methods for monitoring clearance curves are adequate for detecting reduced sensitivity of infecting parasites , they largely reflect parasite killing during the first treatment dose and may not be suitable for evaluating the effectiveness of new antimalarials or alternative treatment regimens . For example , our simulations predicted that splitting the ART dose would substantively decrease the level of viable parasites but would have no effect on parasite clearance curves . This suggests that a recent clinical study of a split-dose regimen [49] may have underestimated its effectiveness . Methods enabling direct monitoring of parasite viability after ART treatment are needed . Our simulations suggest that a suitable parameter to measure is the fraction of circulating viable parasites at a time point 3 h after the commencement of treatment . Our data suggest that culturing drug-exposed ring-stage parasites in a drug-free environment for 35 h will enable the ready distinction of the viable ( trophozoites ) from non-viable ( rings and pyknotic ) parasites and readily distinguish K13 wild-type from mutant parasites . We anticipate that this approach can be made field-adaptable by culturing washed blood collected from patients 3 h after treatment . Assessment of parasites with a trophozoite-like morphology after 30–40 h in culture would provide a robust and direct measure of the likelihood of treatment failure . Another important implication of our findings is that extended ACT treatment courses could reduce the viable parasite load to curative levels in ART-resistant falciparum malaria , since further parasite maturation on the fourth day of treatment will render them much more sensitive to ART . This is consistent with recent trial data [13] showing 97 . 7% efficacy of a six-day treatment course , and adds further support to calls for urgent testing of extended treatments in affected areas . The implementation of new treatment regimens could help battle ART resistance . This is critical , given that a recent modelling study suggest that even a 30% ACT failure rate worldwide would result in more than 116 , 000 additional deaths per year , and US$385 million of annual productivity losses [60] .
P . falciparum isolates were collected from adult patients enrolled in clinical trials conducted between 2009 and 2010 in Pailin Referral Hospital in western Cambodia as described previously [21] . Strains were adapted to cultivation in vitro in RPMI supplemented with glutamine and 10% human serum . K13 propeller genotyping of strains were performed as previously described [21] . Parasite lines were expanded and aliquots frozen . All analyses were performed within 4 wk of thawing . All strains are independent isolates based on typing of MSP1 , MSP2 , and GLURP , and all exhibit a PfCRT mutant genotype ( CVMNT ) , as typical for this region [21 , 61] . A slow-clearance isolate from Pursat province in Cambodia ( RF967/ Cam3 . II ) , harboring the R539T mutation , and a cloned reverted line ( Cam3 . IIrev ) , carrying the wild-type allele , were generated as described elsewhere [20] . In vitro culturing , including generation of very tightly synchronized cultures , was carried out as previously described [14 , 28] . All data presented pertains to cultures synchronized to a 1-h window . The parasite ages defined correspond to the average post-invasion ( p . i . ) age at the beginning of each drug pulse . Parasite age distributions during the schizont-to-ring transition were quantitated from Giemsa smears by periodically examining the number of schizonts and rings of untreated cultures during the transition , as previously described [14] . Genomic DNA was extracted using the ISOLATE II Genomic DNA kit ( Bioline , United Kingdom ) . Genomic DNA ( 200 ng ) was sheared using an ultrasonicator , and Illumina TruSeq Nano DNA library preparation was carried out according to the Manufacturer’s instructions . The libraries were pooled and run with paired-end 300 bp reads on a MiSeq platform over 600 cycles of sequencing using the MiSeq Reagent kit v3 ( Illumina , United States ) . All raw reads have been submitted to the European Nucleotide Archive ( ENA ) under the accession number PRJEB8074 . Following adapter trimming , the raw reads were mapped to the reference 3D7 genome ( version 3 . 0 ) using BWA mem and filtered for mapping quality score of at least 30 . Duplicate reads were marked and removed using SAMtools and Picard , and the reads were realigned around INDELS using Genome Analysis Tool Kit ( GATK ) RealignerTargetCreator ( Broad Institute , US ) . Following recalibration of base quality scores , SNPs were called using GATK UnifiedGenotyper , and hard-filtering of SNPs was performed to obtain high-quality variants . SNPs that failed any of the following cut-off filters were removed from the analysis: depth of read <5 , variant quality as function of depth QD <2 . 0 , strand bias ( P ) <10-6 , mapping quality <40 , MappingQualityRankSum <−12 . 5 , ReadPosRankSum <−8 . 0 . Genetic variants were annotated using snpEFF . Drug pulse assays used to estimate LD50 and Vmin have been described previously [14 , 28] . Parasite viability following a drug pulse is defined as the fraction of the parasite population that survives drug exposure and is able to enter the next parasite cycle . Viability was determined by measuring the parasitemia in the parasite cycle following the drug pulse . For this , parasites were fluorescently labelled with the RNA-binding dye SYTO-61 and the parasitemia quantitated by flow cytometry [62] . Viability was calculated in relation to parasitemia in the "untreated parasite" control ( parasites not exposed to drug ) and "kill" control cultures . The latter refer to parasites maintained under constant drug pressure ( >100 times the LD50 ( 48 h ) ) for 48–96 h to ensure quantitative killing of parasites . The sensitivity limit for viability using this assay is 5% . LD50 is the drug concentration producing 50% viability . Vmin is defined as the viability at saturating drug concentration and was established by examining viability at the highest drug concentration employed in a particular assay . SYTO-61 labelling of parasites was also used to measure growth effects related to drug treatment . Such measurements require labelling to be performed when the untreated parasite control culture has progressed to mid-–late trophozoite stage in the cycle following the drug pulse . This ensures that the SYTO-61 signal from viable parasites exhibiting growth effects show a measurable decrease compared to the no drug control . The SYTO-61 frequency histogram at a particular drug concentration was corrected for the presence of unviable/dead parasites by subtracting the appropriate amount of the "kill" control histogram based on the measured viability at that drug concentration . Corrected SYTO-61 signals refer to the median value of the viable parasite population and were calculated from the corrected histograms , and are only reported for sample measurements exhibiting the following criteria: parasitemia >0 . 3% , number of parasites measured >500 , and fraction of total parasites measured that are viable >0 . 7 . Trophozoite-infected RBCs ( 25–35 h ) or uninfected RBC ( 3% hematocrit ) were incubated with 6 μM QHS or 20 nM WR99210 for 90 min at 37°C . Cells were pelleted , washed three times in PBS supplemented with anti-protease mixture ( APM ) , 20 mM N-ethylmaleimide ( NEM ) , 2 mM PMSF , 0 . 5 mM EDTA , and complete mini EDTA-free protease inhibitor mixture ( Roche ) . Cell pellets were resuspended in 10 volumes of 0 . 15% ( w/v ) saponin in PBS for 10 min on ice . Cells were pelleted and washed twice with PBS + APM . Parasite pellets were subjected to SDS-PAGE ( 4%–12% acrylamide; Life Technologies ) , transferred to nitrocellulose ( iBlot , Life Technologies ) , and probed with polyclonal rabbit anti-ubiquitin IgG ( Z 0458 , Dako , 1:100 dilution in PBS ) , followed by goat anti-rabbit IgG coupled to horseradish peroxidase . Chemiluminescence was detected using a LAS-3000 Imaging System . Membranes were stripped with 0 . 2 M glycine , 0 . 1% ( w/v ) SDS , 0 . 01% ( v/v ) Tween-20 , pH 2 . 2 , and re-probed with rabbit anti-PfGAPDH [63] . Initial control experiments compared the signal in uninfected RBCs and very early ring-stage—infected ( 1–2 h p . i . ) RBCs ( 5% parasitemia ) . Densitometric analysis of the region of the gel between 80 and 200 kDa was performed using ImageJ software . The data were corrected for background and for loading based on the PfGAPDH signal . Mice were housed with strict temperature control ( 21°C ) and under a 12:12 light:dark cycle . All mouse experiments were approved by the Animal Ethics Committee at Macquarie University ( ARA 2012/018 ) and conformed to the National Health and Medical Research Council guidelines . For rodent malaria infection , 250 μl of thawed P . berghei ANKA parasitized blood were intraperitoneally injected onto SJL donor mice . Once the donor mice reached 5%–15% parasitemia , the mice were exsanguinated and the blood was diluted in Kreb’s buffered saline [64] at a dose 1x106 parasitized RBCs . The parasitized RBCs were injected into the peritoneal cavity of BALB/c female mice . After inoculation , the mice were monitored daily or twice daily using tail bleed smearing or flow cytometry using JC-1 dye gated on TER119 , cd71 , and cd45 [65] . For the drug administration , infected BALB/c mice ( ~21 g ) were treated with intravenous doses of Carfilzomib ( 0 . 5 , 1 , and 1 . 5 mg/kg ) , intra-peritoneal doses of DHA ( 5 , 10 , 15 , and 20 mg/kg ) , a combination of Carfilzomib and DHA ( 5 or 10 mg/kg of Carfilzomib , with respectively 5 and 10 mg/kg of DHA ) , or vehicle ( 10 mM citrate buffer , 20% Kleptose for Carfilzomib and 60% dimethyl sulfoxide , 40% Polysorbate 80 for DHA ) . The mice were drug-injected for three consecutive days , starting day 3 post-inoculation . A Student’s one tailed t test was performed and corrected with a Bonferroni procedure for multiple testing to investigate the statistical differences between drug treatments . Parasite viability ( V ) as a function of drug concentration ( C ) and drug exposure time ( te ) were analyzed according to the CED model [14]: V ( C , te ) =[1+ ( Ctet50e , sat ( Km+C ) ) γ]−1 where Km is the drug concentration resulting in half the maximum effective dose , γ is the slope of the sigmoidal function , and t50e , sat is the minimum time required to kill 50% of the parasites ( at saturating drug concentrations ) . The model was fitted to the data , using Microsoft Excel with the Solver add-in . The method employed for simulating parasite load during a three-day course of ART monotherapy is presented in S1 Appendix . An Excel spreadsheet for the simulation of parasite clearance is also supplied ( S1 Spreadsheet ) . | Resistance to artemisinin antimalarials , some of the most effective antimalarial drugs , has emerged in Southeast Asia , jeopardizing malaria control . We have undertaken a detailed study of artemisinin-sensitive and-resistant strains of Plasmodium falciparum , the parasite responsible for malaria , taken directly from the field in a region where resistance is developing . We compared these strains to lab strains engineered with either mutant or wild-type resistance alleles . We demonstrate that in sensitive P . falciparum , artemisinin induces growth retardation and accumulation of ubiquitinated proteins , indicating that the drugs activate the cellular stress response . Resistant parasites , on the other hand , exhibit reduced protein ubiquitination and delayed onset of cell death following drug exposure . We show that proteasome inhibitors strongly synergize artemisinin activity , offering a means of overcoming artemisinin resistance . We have developed a detailed model of parasite responses and have modelled in vivo clearance profiles . Our data indicate that extending artemisinin treatment from the standard three-day treatment to a four-day treatment will clear resistant parasites , thus preserving the use of this critical therapy in areas experiencing artemisinin resistance . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2015 | Targeting the Cell Stress Response of Plasmodium falciparum to Overcome Artemisinin Resistance |
Plasmacytoid dendritic cells ( pDC ) have been shown to efficiently sense HCV- or HIV-infected cells , using a virion-free pathway . Here , we demonstrate for classical swine fever virus , a member of the Flaviviridae , that this process is much more efficient in terms of interferon-alpha induction when compared to direct stimulation by virus particles . By employment of virus replicon particles or infectious RNA which can replicate but not form de novo virions , we exclude a transfer of virus from the donor cell to the pDC . pDC activation by infected cells was mediated by a contact-dependent RNA transfer to pDC , which was sensitive to a TLR7 inhibitor . This was inhibited by drugs affecting the cytoskeleton and membrane cholesterol . We further demonstrate that a unique viral protein with ribonuclease activity , the viral Erns protein of pestiviruses , efficiently prevented this process . This required intact ribonuclease function in intracellular compartments . We propose that this pathway of activation could be of particular importance for viruses which tend to be mostly cell-associated , cause persistent infection , and are non-cytopathogenic .
Although representing a rare cell type of the immune system , plasmacytoid dendritic cells ( pDC ) are the most important source of systemic interferon ( IFN ) type I in the early phase of many virus infections , and as such a critical early alarm system against viruses [1] , [2] . This is based on the ability to produce around 1000 times more IFN type I than any other cell type [1] . Accordingly , pDC possess the necessary cell biological features such as Toll-like receptor ( TLR ) 7 and TLR9 and constitutive high levels of IFN regulator factor ( IRF ) -7 to sense viruses with high efficiency [2] . While it is evident from the literature that TLR7 is the most important sensor of RNA viruses and TLR9 for DNA viruses , it is less clear how the viral nucleic acids have access to these compartments and how the encapsulated viral nucleic acids get in contact with the TLR's . Recently , a novel process of pDC stimulation by infected cells independent of viral particles and their uptake by pDC has been described for hepatitis C virus ( HCV ) in which pDC sense infected cells in a cell contact- and TLR7-dependent manner [3] , [4] . This process has been described to be more effective than stimulation of pDC with cell-free virions . Stimulation of pDC by infected cells has also been reported for HIV [5] . However , this process was blocked by neutralizing anti-envelope antibodies [5] implying a different mechanism of RNA transfer to the TLR7 compartment as observed with HCV where stimulation is not blocked by virus neutralization [3] . For the latter virus , lipid rafts and tetraspanin-enriched membrane microdomains have been described to be involved in infected cell sensing by pDC [4] . Considering the potential importance of this mechanism of pDC activation , we initiated this study with a focus on another member of the Flaviviridae , classical swine fever virus ( CSFV ) belonging to the pestivirus genus . CSFV is the causative agent of a viral hemorrhagic fever in pigs with disease characteristics resembling dengue hemorrhagic fever if pigs are infected with highly virulent CSFV strains . However , with low virulent strains chronic disease and persistent infections are observed [6] . CSFV , in contrast to HCV , has a particular tropism for cells of the macrophage and DC lineage and most efficiently infects pDC [7]–[9] . Our results demonstrate that the basic characteristics of pDC stimulation by infected cells resemble those of HCV . In addition , using such cultures the present study identified a striking function of the enigmatic viral Erns , an essential structural protein with ribonuclease ( RNase ) activity [10] . The pestivirus genus is composed of major veterinary pathogens , the most important being CSFV and bovine viral diarrhea virus ( BVDV ) . Interestingly , Erns is unique to pestiviruses . Despite its function as structural protein , Erns exists as soluble form secreted from infected cells and has been proposed to be involved in immune evasion of pestiviruses ( for review , see [11] ) . Removal of the RNase activity was demonstrated to result in virus attenuation [12] , [13] and abrogation of the capacity of pestiviruses to establish immune-tolerance and persistent infections after infection of fetuses [14] , [15] . However , it has been difficult to understand the mechanism of immune evasion using in vitro studies . Recombinant Erns degrades synthetic single-stranded and double-stranded RNA added to the cultures [16]–[18] but pestiviruses with or without RNase activity do not induce IFN type I in cell culture and replicate to the same titers as their wild type counterpart . In this study we have identified how Erns potently counteracts IFN-α induction in pDC . It represents the first example of a viral protein that prevents the stimulation of pDC by infected cells , and thus represents a novel pathway of viral evasion of the type I IFN system . Furthermore , it underlines the importance of stimulation of pDC by infected cells , rather than virions .
In accordance to previous studies [7] , [8] , CSFV as well as virus replicon particles ( VRP ) lacking the Erns gene ( VRPΔErns ) were poor stimulators of pDC , inducing between 0 and 550 IFN-α units per ml , dependent on the experiment . Interestingly , stimulation of pDC by co-culture with CSFV-infected or VRPΔErns -infected SK-6 cells induced up to 100-fold more IFN-α compared with direct infection of the pDC , with an optimum at 40'000 to 80'000 infected SK-6 cells per 2×105 CD172a+ enriched pDC ( Figure 1A and B ) . While no significant difference between direct CSFV and VRPΔErns stimulation was observed , CSFV-infected SK-6 cells stimulated an average of 5 . 1 more IFN-α when compared to direct stimulation by CSFV ( Figure 1C and D ) . This difference was even more evident when direct simulation with VRPΔErns was compared to stimulation by VRPΔErns-infected cells ( Figure 1E ) . Interestingly , VRPΔErns-infected SK6 cells were in average around 8 times more stimulatory than CSFV-infected SK6 cells ( Figure 1F ) . In accordance to previous studies demonstrating that pDC were the only cell type able to respond to CSFV by IFN-α production [19]–[21] , pDC were the only source of IFN-α following stimulation with infected cells , as demonstrated by intracellular IFN-α staining which was only found in the CD4highCD172a+ pDC population . Furthermore , purified monocytes did not produce IFN-α in response to any of the stimuli tested ( Supplementary Figure 1 ) . The above results suggested that infected SK-6 cells would transfer viral RNA to pDC resulting in pDC activation . Considering the fact that VRP deliver self-replicating RNA which replicates for many days in SK-6 cells [22] , we tested if functional replicon RNA was transferred between SK-6 cells and pDC by determining the expression of the viral NS3 protein in pDC . NS3 is generated by post-translational processing of the CSFV precursor polyprotein . Detectable amounts of NS3 in cells can only be obtained with replication competent pestivirus genomes , i . e . full-length genomes and replicons . As shown in Figure 2 , after co-culture of pDC with VRPΔErns-infected SK-6 cells for 22 h , approximately 12–14% pDC expressed NS3 , indicating either a transfer of intact full-length replicon RNA or viral NS3 protein between the cells . Interestingly , co-culture of CSFV-infected cells with pDC resulted in a higher degree of infection ( 94% ) compared to direct infection by the virus ( 65% ) . Another observation was that , when infectious CSFV or VRP were present , the percentage of NS3-expressing pDC was higher when compared to the monocytes that were co-purified using the CD172a selection . It is also noteworthy that NS3+ monocytes were found after co-culture with VRP-infected SK-6 cells . The results presented in Figure 2 also highlight that there is no correlation between infectious titers , percentage of viral protein expressing pDC and IFN-α responses . An over 20 times higher IFN-α response was found when pDC were stimulated with VRP-infected SK-6 cells in which infectious virus was barely detectable . We confirmed these results by employing RT-PCR to quantify the number of viral genome copies in these cultures . pDC were stimulated either directly with VRPΔErns or with VRPΔErns-infected SK-6 cells , and then re-sorted from these cultures by CD172a MACS sorting . This demonstrated that around 0 . 1% of the total viral genome present in infected SK-6 cells was transferred to pDC/monocytes . Direct infection of enriched pDC was ∼6 times more efficient in delivering RNA but induced much lower IFN-α levels ( Table 1 ) . In order to rule out a role for free virions in the induction of IFN-α , we measured the infectivity in VRPΔErns-infected SK-6 at the time of co-culture with pDC and found titers below 102 TCID50/ml . This corresponded to MOI of less than 10−4 TCID50/enriched pDC , suggesting that stimulation of pDC does not occur by infection of pDC with VRP . The lack of involvement of free virions in pDC stimulation by infected cells was further confirmed by addition of a neutralizing monoclonal antibody against the main glycoprotein E2 of CSFV or by addition of a porcine polyclonal hyper immune serum obtained from an infected pig which contained antibodies against both glycoproteins of the virus , E2 and Erns . The antibody preparations completely blocked direct activation of pDC by cell-free virions ( Figure 3A and B ) , but were unable to inhibit activation by infected cells , independent whether the stimulation used CSFV- or VRPΔErns-infected SK-6 cells ( Figure 3C and D ) . With CSFV-infected SK-6 cells , the antibodies even enhanced pDC stimulation . Antibodies from naïve animals or an irrelevant monoclonal antibody had no effect ( data not shown ) . Similarly , the neutralizing antibodies blocked infection of pDC after direct stimulation but were unable to inhibit the expression of NS3 in pDC after stimulation by infected cells ( supplementary Figure S2 ) . We further confirmed the absence of any virus particle in pDC stimulation by employing RNA-transfected SK-6 cells to stimulate pDC . To this end , SK-6 cells were transfected with in vitro transcribed RNA synthesized with plasmids encoding the Erns-deleted genome of CSFV ( pA187-ΔErns , ΔErns RNA ) or a genome of CSFV devoid of all structural proteins ( pA187Δ-Apa , ΔApa RNA ) . The results shown in Figure 3E demonstrate that CSFV RNA-transfected SK-6 cells can activate pDC , although the levels of IFN-α were low , compared to VRP-infected SK-6 cells . This was explainable by the relatively low transfection efficiency of 15–20% in terms of NS3+ expression ( data not shown ) . We also generated VRP with a deletion of E2 instead of Erns . SK-6 cells infected with such VRPΔE2 were more potent in inducing IFN-α in pDC than VRPΔE2 virions . Interestingly , SK-6 cells infected with VRPΔErns were more stimulatory than SK-6 cells infected with VRPΔE2 ( Figure 3F ) . In order to determine the role played by TLR7 in sensing CSFV-infected cells , we used the immunoregulatory sequence 661 ( IRS661 ) representing an oligodeoxynucleotide inhibitor of TLR7 which had been previously established for the human , murine and porcine immune systems [23] , [24] . IRS661 at a concentration of 0 . 7 µM efficiently inhibited CSFV- and VRP-induced pDC activation ( Figure 4A ) . With infected cells inducing much higher levels of IFN-α , the reduction caused by IRS661 was still over 80% with the highest inhibitor concentration ( Figure 4B ) . A striking observation was that in all experiments VRPΔErns-infected SK-6 cells were clearly more efficient at inducing IFN-α compared to CSFV-infected SK-6 cells ( Figure 1F ) . Considering that VRPΔErns lacks the Erns gene , we postulated a role for this viral protein in inhibition of this novel type of IFN-α induction and tested this by comparing SK-6 cells stably expressing Erns to the parent wild type SK-6 cells . Indeed , the SK-6 ( Erns ) cells infected with VRP or CSFV were very inefficient at inducing IFN-α in contact with pDC ( Figure 5A and B ) . This was not a result of different susceptibility to the viruses as the infection rate of the SK-6 ( Erns ) was 99% , similar to CSFV ( Figure 6F ) . As expected , Erns expressed by the SK-6 ( Erns ) trans-complemented VRPΔErns to generate infectious particles resulting in a higher degree of NS3+ pDC ( supplementary Figure S3 ) . Considering that this must be associated with a higher viral RNA load in pDC , these results indicate a potent function of Erns in preventing the stimulation of pDC by infected cells . Erns has been shown to be mainly associated with intracellular membranes in particular of the ER , with almost no cell surface expression [25] . However , considering that approximately 16% is secreted to the extracellular space [26] , [27] , we tested if Erns secreted by SK-6 ( Erns ) cells was responsible for the observed inhibition . To this end , supernatants of SK-6 ( Erns ) or SK-6 were added to co-cultures of VRPΔErns-infected SK-6 cells and pDC . As shown in Figure 5C , there was no evidence for any suppressive effect of soluble Erns in the supernatants . We next determined the role of Npro , a well established type I IFN antagonist of CSFV in non-pDC and pDC targeting IRF3 and IRF7 transcription factors [8] , [19] , [28] . To this end , we compared the ability of SK-6 cells infected with VRPΔErns and VRPΔErns D136N expressing a non-functional mutant of Npro [22] , to activate pDC . Our results demonstrated that only Erns , but not Npro prevents the activation of pDC by infected cells ( Figure 5D ) . In order to confirm this function of Erns , we compared the effect of SK-6 and SK-6 ( Erns ) cells infected with two other RNA viruses , the picornavirus foot-and-mouth disease virus ( FMDV ) and the coronavirus transmissible gastroenteritis virus ( TGEV ) , on pDC . When the stimulation used FMDV- or TGEV-infected SK-6 and SK-6 ( Erns ) cells for comparison , the responses with infected SK-6 ( Erns ) cells were 3 to 6-fold lower compared to SK-6 cells ( Figure 5 E and F ) . This was not caused by an inhibitory effect of Erns on virus replication ( Supplementary Figure S4 ) . In order to exclude a potential “toxicity” derived from Erns expressing cells we also tested the responses of pDC to CpG when stimulated in co-cultures with SK-6 and SK-6 ( Erns ) cells , and found similar levels of IFN-α ( Supplementary Figure S5 ) . In order to test the requirement of RNase activity for the above Erns function , we constructed an RNase-negative mutant of CSFV by deleting the histidin codon at position 346 [13] . The mutation abolished the RNase activity ( Figure 6A ) , confirming previously published data [13] . Interestingly , RNase activity of Erns was detectable in cell extracts only ( Figure 6A ) . While SK-6 cells infected with wild type CSFV induced under 500 U/ml of IFN-α , cells infected with the CSFV-ErnsΔ346 mutant induced approximately 10-fold higher responses ( Figure 6B ) . Furthermore , direct stimulation of pDC by virus also dramatically increased when the CSFV-ErnsΔ346 mutant was employed . Nevertheless , the levels of IFN-α remained clearly under those induced by infected cells . For further confirmation that the inhibitory effect of Erns in pDC depends on the RNase-activity of Erns , we constructed lentivirus-transduced SK-6 cell lines expressing the parent Erns [SK-6LV ( Erns ) ] or an RNase-inactive mutant Erns [SK-6LV ( ErnsΔ346 ) ] . As expected , the deletion of the histidine at position 346 completely abolished the RNase activity of Erns ( Figure 6C ) . Again , RNase activity of the parent Erns was detectable only in the cell extracts . As expected , only SK-6 expressing RNase active Erns prevented IFN-α induction by VRP infection of the cell lines ( Figure 6D ) . All SK-6 cell lines expressing both wild-type and mutant Erns had comparable levels of Erns , demonstrating that the striking differences observed in RNase and inhibitory activity for activation of pDC were not caused by lack of Erns expression . In addition , the levels of Erns in these cells were below those found after natural infection by wild-type CSFV ( Figure 6E ) . Furthermore , all Erns-expressing cell lines were found to be fully susceptible to VRPΔErns infection as determined by NS3 expression ( Figure 6F ) . The observed higher levels of NS3 expression in Erns expressing cells was a consequence of trans-complementation of VRPΔErns and generation of infectious virions in these cultures . Considering that macrophages ( MΦ ) and endothelial cells ( EDC ) represent important target cells for CSFV , we also tested if infected MΦ and EDC , similar to SK-6 cells , were able to activate pDC . Indeed , VRPΔErns-infected and CSFV-infected MΦ were able to activate pDC while no responses were detectable with CSFV alone ( Figure 7A ) . VRPΔErns were relatively inefficient at infecting MΦ ( 26% NS3+ versus 97% NS3+ with CSFV ) , which explains the lower IFN-α responses when compared to CSFV . The highest levels of IFN-α were induced by MΦ infected with the CSFV-ErnsΔ346 mutant ( 83% NS3+ MΦ ) confirming the functioning of Erns also in MΦ ( Figure 7A ) . Neither the VRPΔErns , nor WT CSFV or the CSFV-ErnsΔ346 mutant were able to induce IFN-α in MΦ ( data not shown ) . We also further confirmed our findings using the immortalized porcine EDC line PEDSV . 15 . VRPΔErns-infected and CSFV-ErnsΔ346-infected EDC were more potent at activating pDC when compared to CSFV-infected EDC ( Figure 7B ) . In contrast to MΦ , the rate of endothelial cells infection was comparable with the viruses and VRPΔErns ( VRPΔErns: 80% NS3+ , CSFV: 92% NS3+ and CSFV-ErnsΔ346 mutant 92% NS3+ ) . Based on the above results we postulated that viral RNA is transported from the infected cells to the TLR7 compartment of pDC in a manner avoiding contact of the RNA with the extracellular space . Consequently , we were interested to investigate if membrane vesicles transporting viral RNA from the infected cell to the pDC could be involved in pDC activation by infected cells . To this end , we co-cultured pDC with VRP-infected SK-6 cells in transwell culture dishes using 0 . 4 or 1 µm pore sizes . Only pDC with direct contact to infected SK-6 cells produced large quantities of IFN-α ( Figure 8A and B ) . Notably , pDC responses to the CpG control were in the same level of magnitude if the pDC were cultured in the insert or in the well of the plate ( data not shown ) . When the same experiments were performed with CSFV-infected SK-6 cells , low IFN-α responses were also observed in all transwell conditions ( Figure 8B ) . Considering that CSFV virions are only 60 nm , this response was probably mediated by direct stimulation of pDC with virions passing the membranes . Similar to VRP-infected or CSFV-infected SK-6 cells , infected MΦ and EDC were unable to induce IFN-α when separated from pDC using a transwell culture system ( data not shown ) .
During a virus infection , pDC will not only encounter virions but also virus-infected cells . The ability to sense the latter has the advantage of being able to sense infection before or without release of virions and also to better sense viruses which are particularly cell-associated and tend to cause persistent virus infections , such as HIV , HCV and pestiviruses . The present study underlines the importance of this by demonstrating that pDC stimulation by infected cells can be much more efficient than stimulation by virions . This is also emphasized by the identification of a viral protein that appears to have evolved to efficiently inhibit this pathway . After HIV [5] , HCV and Venezuelan equine encephalitis virus [3] , CSFV is now the fourth virus for which this mechanism of pDC stimulation has been found to be more potent than direct stimulation of pDC by virions . Although free CSFV was also able to stimulate pDC , the IFN-α responses were inferior to those induced by cell-free virions . Interestingly , also with FMDV , a virus which does not or very inefficiently stimulate pDC [31] , infected cells were able to stimulate pDC . The basic characteristics of pDC stimulation by CSFV-infected cells are similar to those observed with HCV . It represents a TLR7-dependent process which cannot be blocked by neutralizing antibodies and does not require expression of viral glycoproteins [3] . This is in contrast to the situation for HIV [5] . Our data indicate that this pathway of pDC activation is dependent on cell contact , intact actin filaments , microtubules as well as cell membrane cholesterol indicating a role for lipid rafts , although future studies are required to directly demonstrate the functional interaction of these cellular components with viral proteins . We have also identified that the pestiviral Erns potently prevents pDC activation by infected cells . Erns possesses several remarkable features , of which the RNase activity is of particular interest , considering that it is expressed by an RNA virus . Erns has structural similarities with plant T2 RNases which have their optimal catalytic activity at an acidic pH [32] with a preference for cleaving single-stranded RNA [33] , [34] . This would point on an activity within the endosomal compartment , which is also supported by our data indicating that genomic RNA does not appear to be degraded . The protein also has an unusual membrane anchor composed of an amphipathic helix without a typical membrane anchor [27] , [35] , but with a retention signal ensuring its association with intracellular membrane compartments [25] . Whether this enables accumulation in the endosomal system with appropriate orientation needs to be investigated . We postulate that viral RNA may have access to the endosomal system via the autophagy process . These endosomes could be transferred to pDC by a pathway to be defined and then fuse with the TLR7 compartment . Alternatively , both cytoplasmic viral RNA and Erns could be autophagocytosed after transfer to pDC followed by fusion of autophagosomes with TLR7-containing endosomes [36] . Only at this acidic location the RNAse function would be highly active and rapidly degrade the viral RNA to reduce TLR7 triggering . Consequently , one of the first questions to be addressed in future studies is whether Erns is transported from SK-6 cells to pDC to degrade RNA in the TLR7 compartment of pDC . Certainly , Erns can act in pDC as our data is showing that CSFV with Erns lacking RNase activity induce much higher IFN-α responses compared to wild type CSFV when pDC are directly stimulated by virions . Strikingly , Erns can exert its inhibitory function when expressed independently of the viral context in the infected cells that stimulate the pDC . In addition , our results suggest that Erns functions by using intracellular rather than extracellular pathways , since SK-6 ( Erns ) supernatants did not suppress pDC activation by VRPΔErns-infected SK-6 cells . This is contrary to a role proposed for secreted Erns by several authors . Based on the observation that a minor part of the protein was found to be secreted from infected cells or cells expressing Erns [25]–[27] , recombinant Erns was tested and found to degrade synthetic single-stranded and double-stranded RNA added to the cultures [16]–[18] . Considering that pDC are by far the most potent producers of IFN type I and of crucial importance in linking innate to adaptive immunity , our data shed light into one of the most fascinating aspects of pestivirus biology . These viruses cause persistent infections , in both cattle and pigs . When bovine fetuses are infected transplacentally by BVDV between the 2nd and 4th month of pregnancy , a time point at which the fetuses are not yet immunocompetent [37] , persistently infected calves are born which are fully immunotolerant to the virus and cannot mount any adaptive immune responses against the virus . These persistently infected calves play a major role in the epidemiology of the disease by shedding virus . Similar observations have been reported for pigs after infection of pregnant sows with low virulent strains of CSFV before eradication of CSFV from most European countries and the U . S . A . [38]–[44] . This contrasts with the pathogenesis of highly virulent strains of CSFV which induce acute disease with high mortality which is associated with high systemic levels of IFN-α [45] . This appears contradictory to the present report but compared to other viruses the ability of CSFV to activate pDC is weak [9] . Even if pDC are stimulated by cells infected with wild type CSFV , IFN type I responses remain relatively weak compared to influenza virus . Our concept to explain this apparent contradiction to the in vivo situation with virulent strains of CSFV is based on the prominent tropism of CSFV for MΦ and DC , and its localization predominantly in lymphoid tissue . Highly virulent isolates of CSFV cause a very rapid and strong viremia , in which the virus reaches simultaneously all primary and secondary lymphoid tissues where relatively high numbers of pDC are localized . This situation could explain why such viruses are able to induce a potent IFN-α response despite the activities of Npro and Erns [9] . Interestingly , it has been demonstrated that the virus not only needs a functional Npro , but also an Erns with active RNase to establish persistent infections in cattle [14] . Npro induces the degradation of IRF3 and thereby efficiently prevents IFN type I induction in all host cells including conventional DC , which have been induced to express IRF7 by IFN type I pre-treatment [19] , [28] . However , Npro can only partially prevent IFN-α responses in pDC [8] and is unable to stop the much more potent activation of pDC by infected cells ( this study ) . We thus propose that Erns has evolved to prevent this pathway of innate immune system activation , which is much more potent and therefore likely to be essential for the virus to be able to establish persistent infections , representing a main survival strategy of pestiviruses [46] .
Bleeding and care of donor pigs was carried out in accordance with EU standards and National laws ( Tierschutzgesetz SR455 ) . Specifically , approval of the protocol employed was obtained by the Animal Welfare Committee of the Canton of Bern , Switzerland ( animal license BE26/11 ) . The porcine kidney cell lines SK-6 [47] , PK-15 ( LGC Standards-ATCC , Molsheim , France ) and the porcine immortalized endothelial cells PEDSV . 15 [48] ( obtained from Dr . Jörg Seebach , University of Geneva , Switzerland ) were propagated in Earle's minimal essential medium ( MEM ) substituted with 7% horse serum and in Dulbecco's minimal essential medium ( DMEM ) supplemented with 5% horse serum , nonessential amino acids and 1 mM Na-pyruvate , respectively . SK-6 cells stably expressing Erns of CSFV strains Alfort/187 , termed SK-6 ( Erns ) were generated as described previously [49] . Baby Hamster Kidney ( BHK ) 21 cells were grown in Glasgow's minimum essential medium ( Life Technologies ) supplemented with 5% v/v fetal bovine serum ( FBS , Biowest , Nuaillé , France ) . Peripheral blood mononuclear cells ( PBMC ) were obtained from blood of specific-pathogen-free pigs using ficoll-paque density centrifugation ( 1 . 077 g/L , Amersham Pharmacia Biotech ) . pDC were enriched as described previously [31] by cell sorting of CD172a+ PBMCs using the magnetic cell sorting system ( MACS ) with LD columns ( Miltenyi Biotec GmbH , Germany ) . This permits a 10-fold enrichment of functional pDC to 2–5% . In some experiments , pDC were enriched by a first depletion of CD14+ monocytes followed by CD172a enrichment [24] , permitting a pDC enrichment of around 5–10% . Enriched pDC were cultured in DMEM with Glutamax , 20 µm β-mercaptoethanol ( Life Technologies ) . 10% FBS was only added to when indicated . Porcine monocytes were isolated by CD14+ selection using MACS with LS columns . Porcine MΦ were generated from CD14+ PBMCs as previously described using a 3-day culture in DMEM supplemented with 10% autologous porcine serum [50] , [51] . CSFV strain vA187-1 was derived from the full-length cDNA clone pA187-1 [52] . Plasmid pA187-1 carries a full-length cDNA copy of the CSFV strain Alfort/187 and served as basis for all viruses and replicon cDNA constructs . The vA187-ErnsΔ346 virus ( referred to as CSFV-ErnsΔ346 mutant ) , with a histidine deletion in Erns at position 346 of the viral polyprotein resulting in loss of RNase activity [13] was rescued by standard procedure [52] , [53] from plasmid pA187-ErnsΔ346 . This latter construct was generated from pA187-1 using PCR-based site-directed mutagenesis with oligonucleotide primers encompassing the deletion and PfuUltra DNA polymerase ( Agilent ) , employing standard cloning techniques as previously described [54] . The plasmids pA187-ΔErns carrying an in frame deletion of the Erns gene in the pA187-1 backbone , and pA187-D136N-ΔErns carrying the same deletion and expressing a non-functional D136N mutant of Npro were described elsewhere [22] , [49] . Plasmid pA187-E2del373 carrying an in frame deletion of the complete E2 gene in the pA187-1 backbone was also described previously [22] , [49] , [53] . Plasmid pA187-ΔApa encoding a CSFV replicon with a deletion of the structural protein genes C , E1 , Erns and most of the E2 gene ( i . e . the codons encoding amino acids 96 to 962 of the polyprotein ) was described earlier [53] . In vitro transcribed replicon RNA was produced using SrfI-linearized pA187-ΔErns , pA187-D136N-ΔErns , pA187-E2del373 or pA187-ΔApa as described [22] , [49] , [53] . VRPΔErns carrying a genome with a complete deletion of the Erns gene were described previously , and produced by transfection of SK-6 ( Erns ) cells with A187-ΔErns or A187-D136N-ΔErns replicon RNA . The SK-6 ( Erns ) cells express the Erns protein required for the generation of VRPΔErns by trans-complementation [22] , [49] . Similarly , VRPΔE2 carrying the pA187-E2del373-derived replicon with a complete E2 deletion were rescued by transfection of SK-6 ( E2p7 ) cell line expressing the CSFV E2 and p7 proteins as described previously [22] , [49] , [53] . TGEV ( TGEV; strain Perdue 115 ) was propagated in PK15 cells [55] . The FMDV type O UK/2001 isolate was grown in ( BHK21 ) cells as described previously [56] . All virus titers were determined on SK-6 cells , PK-15 cells or BHK21 cells ( for CSFV , TGEV and FMDV , respectively ) by standard endpoint dilution and were expressed as 50% tissue culture infectious doses ( TCID50 ) per ml . CSFV or VRP RNA was quantified using a published real-time RT-PCR [57] . Briefly , RNA was extracted using the Trizol method and RT-PCR performed with the SuperScript III Platinum One-Step qRT-PCR System ( Life Technologies ) using 7500 Real-time PCR System , Applied Biosystems . To determine the absolute number of RNA copies , vA187-1 RNA transcripts generated in vitro were employed . A lentivirus ( LV ) expression system using plasmids obtained from the laboratory of Dr . Didier Trono ( http://tronolab . epfl . ch/ Ecole Polytechnique Federale de Lausanne , Switzerland ) or through Addgene ( Cambridge MA , USA ) [58] , [59] was employed . For cloning of Erns and its RNase-knock-out mutant variant into the lentiviral transfer plasmid pWPT-GFP ( Addgene ) the pA187-1 and pA187-ErnsΔ346 plasmids were used as template for PCR amplification using primers to insert the MIuI and Sall restriction sites excising the GFP in the pWPT-GFP vector . The PCR products were first inserted into the pJET vector ( Fermentas ) . The cloned Erns genes of pJET-Erns and pJET-ErnsΔ346 were verified by nucleotide sequencing and then excised with MIuI and Sall and ligated into the pWPT-GFP vector employing standard techniques and Stbl2 bacteria , resulting in pWPT-Erns and pWPT-ErnsΔ346 respectively . All primer sequences and construction details can be obtained on request . In order to generate lentiviruses , HEK293T cells were transfected with the envelope plasmid ( pMD2 . G ) , the packaging plasmid ( pCMV-R8 . 74 ) and the pWPT-Erns or pWPT-ErnsΔ346 plasmid using standard calcium phosphate precipitation . Medium was changed after overnight incubation at 37°C and the supernatant harvested after 48 h , centrifuged ( 350× g , 10 min ) and filtered . The virus was purified and enriched by centrifugation on a 20% sucrose cushion at 100'000× g for 90 min at 4°C . The cells were transduced twice with 1∶100 dilutions of the purified lentiviruses in 1 ml serum free medium of a T25 cell culture flask followed by culture overnight at 37°C and medium change between the transductions . Transduction efficiency was found to be over 90% in terms of detectable anti-Erns expression by flow cytometry and found to remain stable over at least three passages . For pDC phenotyping and isolation , monoclonal antibodies ( mAb ) against CD172a ( mAb 74-22-15A ) , CD14 ( CAM36A ) and CD4 ( mAb 74-12-4 and PT90A ) were used . Hybridomas for mAb 74-22-15A and 74-12-4 were kindly donated by Armin Saalmüller ( University of Veterinary Medicine , Vienna , Austria ) . The mAb PT90A and CAM36A were purchased from VMRD ( Pullman , WA ) . The hybridomas for mAb HC/TC26 [60] and C16 [61] binding the CSFV glycoprotein E2 and nonstructural protein NS3 respectively were kindly provided by Irene Greiser-Wilke , Hannover Veterinary School , Hannover , Germany . Erns expression was demonstrated with mAb 140 . 1 ( used at a 1∶200 dilution , Prionics , Switzerland ) as described previously [62] . For NS3 and Erns detection the cells were fixed and permeabilized with FIX&PERM solution ( An der Grub Bio Research GmbH ) . As fluochromes , isotype-specific fluorescein isothiocyanate ( FITC ) , R-phycoerythrin ( RPE ) ( Southern Biotechnology Associates ) , RPE-Cy5 ( Dako ) and APC ( Becton Dickinson , Basel , Switzerland ) conjugates were used as described previously [63] . For stimulation of enriched pDC by cells infected with CSFV and VRP , SK-6 cells were infected at an multiplicity of infection ( MOI ) of 5 TCID50/cell , cultured for 24 h and then washed four times to remove the inoculums . Infectivity was verified and found to be above 90% NS3+ cells following infection with CSFV or VRP at these conditions . The SK-6 cells were then added at 40'000 cells/well for 96 well plate ( if not indicated otherwise ) or at 200'000 cells/well for 24-well plates . Freshly isolated CD172a+ enriched pDC were then added to the cultures at 200'000 cells/well of 96-well plates or 1×106 per well of 24-well plates . After another 22 h , supernatants were isolated for IFN-α ELISA and the cells for NS3 expression in some experiments . For stimulation of pDC by cells infected with FMDV or TGEV virus the cells were infected at the indicated MOIs , incubated for 90 min at 37°C and washed 4 times before addition of enriched pDC . For FMDV BHK21 cells were employed , for TGEV PK15 cells . In some experiments we employed 24-well plate transwell inserts with 0 . 4 µm or 1 . 0 µm pore sizes ( Corning , Sigma Chemicals , Buchs Switzerland and Becton Dickinson , Basel Switzerland ) . All cultures with enriched pDC were done at 39°C , 6% CO2 . As a positive control for pDC stimulation , direct CpG D32 at 10 µg/ml [63] was used . A 50-mer RNA oligonucleotide probe complementary to nucleotides 12242-12193 of the vA187-1 genome sequence ( GenBank accession number X87939 . 1 ) and carrying a Dyomics 781 modification at the 5′ end ( Dy-781-O1-RNA ) was synthesized by Dr . Fabian Axthelm ( Microsynth AG , Balgach , Switzerland ) . The Dy-781-O1-RNA probe was mixed at 40 nM final concentration with MEM containing 3×10−3 U RNase A/ml as digestion control , with 50 mM TrisHCl pH 7 . 4 as negative control , and with the samples to be tested for RNase activity , and incubated for 1 h at 37°C . The treated probes were mixed with 2 volumes of 97% Formamide ( Sigma ) and separated on a 10% polyacrylamide and 35% urea gel in 133 mM TrisHCl , 45 . 5 mM boric acid and 3 . 2 mM EDTA . Image acquisition was performed with the Odyssey Infrared Imaging System ( LI-COR ) . As TLR7 inhibitor IRS661 ( 5′-TGCTTGCAAGCTTGCAAGCA-3′ ) and a control oligonucleotide ( Ctrl-ODN; 5′-TCCTGCAGGTTAAGT-3′ ) was used [23] . IRS661 and Ctrl-ODN were purchased from Eurofins MWG Operon ( Ebersberg , Germany ) . The impact of various metabolic inhibitors on SK-6 cells was tested by addition of the inhibitor for the last 2 h of infection , in order to avoid interference with infection . Before addition of pDC , the inhibitors were removed and the cells washed three times . The following final concentrations were used: 1 µM latranculin B , 5 µM nocodazole or 20 mM methyl-β-cyclodextrin ( MβCD ) . All chemicals were purchased from Sigma Chemicals . IFN-α was quantified by enzyme-linked immunosorbent assay ( ELISA ) using the mAbs K9 and F17 ( kindly provided by Dr . B . Charley , INRA , Jouy-en-Josas , France ) as described previously [31] . For detection of IFN-α by intracellular staining mAb F17 with a previously published protocol was employed [63] . P values were calculated by an unpaired t-test using the GraphPad Prism Software . | Plasmacytoid dendritic cells ( pDC ) represent the most potent producers of interferon type I and are therefore of major importance in antiviral defences . A TLR7-dependent induction of interferon-α in pDC by infected cells in the absence of virions has been demonstrated for hepatitis C virus . Here , we show that this pathway is also very efficient for classical swine fever virus , a pestivirus that is also a member of the Flaviviridae . Our data indicate a transfer of RNA from the donor cell to pDC in a cell-contact-dependent manner requiring intact lipid rafts and cytoskeleton of the donor cell . Importantly , we demonstrate that the enigmatic viral Erns protein unique to pestiviruses efficiently prevents this pathway of pDC activation . This novel function of Erns is dependent on its RNase activity within intracellular compartments . The present study underlines the importance of pDC activation by infected cells and identifies a novel pathway of virus escaping the interferon system . Considering that Erns is required for pestiviruses to establish persistent infection of foetuses after transplacental virus transmission resulting in the development of immunotolerant animals , this report also points on a possible role of pDC in preventing immunotolerance after viral infection of foetuses . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"medicine",
"infectious",
"diseases",
"veterinary",
"diseases",
"immune",
"cells",
"immunity",
"veterinary",
"virology",
"innate",
"immunity",
"antigen-presenting",
"cells",
"immunology",
"biology",
"viral",
"diseases",
"immunomodulation",
"veterinary",
"science"
] | 2013 | Efficient Sensing of Infected Cells in Absence of Virus Particles by Blasmacytoid Dendritic Cells Is Blocked by the Viral Ribonuclease Erns |
The blood–brain barrier ( BBB ) , which forms the interface between the blood and the cerebral parenchyma , has been shown to be disrupted during retroviral-associated neuromyelopathies . Human T Lymphotropic Virus ( HTLV-1 ) Associated Myelopathy/Tropical Spastic Paraparesis ( HAM/TSP ) is a slowly progressive neurodegenerative disease associated with BBB breakdown . The BBB is composed of three cell types: endothelial cells , pericytes and astrocytes . Although astrocytes have been shown to be infected by HTLV-1 , until now , little was known about the susceptibility of BBB endothelial cells to HTLV-1 infection and the impact of such an infection on BBB function . We first demonstrated that human cerebral endothelial cells express the receptors for HTLV-1 ( GLUT-1 , Neuropilin-1 and heparan sulfate proteoglycans ) , both in vitro , in a human cerebral endothelial cell line , and ex vivo , on spinal cord autopsy sections from HAM/TSP and non-infected control cases . In situ hybridization revealed HTLV-1 transcripts associated with the vasculature in HAM/TSP . We were able to confirm that the endothelial cells could be productively infected in vitro by HTLV-1 and that blocking of either HSPGs , Neuropilin 1 or Glut1 inhibits this process . The expression of the tight-junction proteins within the HTLV-1 infected endothelial cells was altered . These cells were no longer able to form a functional barrier , since BBB permeability and lymphocyte passage through the monolayer of endothelial cells were increased . This work constitutes the first report of susceptibility of human cerebral endothelial cells to HTLV-1 infection , with implications for HTLV-1 passage through the BBB and subsequent deregulation of the central nervous system homeostasis . We propose that the susceptibility of cerebral endothelial cells to retroviral infection and subsequent BBB dysfunction is an important aspect of HAM/TSP pathogenesis and should be considered in the design of future therapeutics strategies .
The Blood-Brain barrier ( BBB ) constitutes the interface between the blood and the central nervous system ( CNS ) . It is composed of astrocytes , pericytes and brain microvascular endothelial cells . This latter cell type forms the major structural and functional element of the BBB , with endothelial cells sealed together with Tight Junctions ( TJs ) . Under physiological conditions , the BBB maintains CNS homeostasis and selectively regulates intracellular and paracellular passage of ions , molecules and cells [1] . BBB integrity is compromised during retroviral infection; for example , BBB breakdown has been reported during Human Immunodeficiency Virus Type 1 ( HIV ) infection , especially during HIV-related encephalitis and HIV-associated dementia [2] . One to three percent of the 20 million people infected worldwide by the retrovirus HTLV-1 ( for human T-lymphotropic virus type 1 ) develop HTLV-Associated Myelopathy/Tropical Spastic Paraparesis ( HAM/TSP ) [3] . This is a slowly progressive paraplegia of the lower extremities , involving demyelination and neuronal degeneration mainly in the thoracic spinal cord . BBB disruption has been attested in HAM/TSP patients by several lines of evidence , such as fibrinogen leakage and IgG deposits in CNS parenchyma [4] as well as lymphocyte passage through brain endothelium [4]–[7] . As previously shown , BBB disruption is associated with alterations in tight junctions between endothelial cells in the vasculature of a HAM/TSP patient [8] . The mechanisms of BBB disruption during retroviral-associated pathologies are not yet fully understood . Most studies focus on the effect of soluble molecules secreted by infected lymphocytes on BBB functions and intercellular TJ organization . In the case of HIV infection , the viral protein Tat has been shown to induce an inflammatory process in brain endothelial cells , or endothelial cell apoptosis [9] , and to be able to disrupt the intercellular TJs [10] . In the context of HTLV-1 infection , we recently demonstrated that proinflammatory cytokines , such as IL-1α and TNFα , secreted by infected lymphocytes , are sufficient to disrupt TJs between human brain endothelial cells and induce permeability changes [8] . Alternative mechanisms could contribute to BBB dysfunction associated with HTLV-1 infection . Although neurological disease in mice infected with the PVC-211 Murine Leukemia Virus has been associated with infection of brain endothelial cells [11] , the infection of brain endothelial cells by human retroviral agents and its role in BBB breakdown is still a matter for debate . In the case of HIV infection , a number of earlier studies reported infection of endothelial cells in adult brain tissue [12]–[14] , based upon morphological appearance and vascular localization of cells found positive by immunocytochemistry , in situ hybridization or PCR-in situ hybridization for viral transcripts . Conflicting results were obtained in vitro from brain-derived endothelial cells ( for review , see [15] ) . In the case of HTLV-1 , no evidence for infection of human brain endothelial cells has been reported so far , most likely due to the rarity of material from patients with HAM/TSP , and the low level of HTLV-1 expression in tissues . Although an increased adherence of T lymphocytes from HAM/TSP patients to human brain endothelial cells has been observed [16] , the main data concern extra-neural endothelial cells: it has been demonstrated in vitro that human venous endothelial cells derived from umbilical cords are susceptible to HTLV-1 infection [17] , [18] , and that HTLV-1 proviral DNA could be detected in dermal endothelial cells ex vivo [19] . In this study , we investigated the susceptibility of human brain endothelial cells to HTLV-1 infection , and its possible consequences on BBB integrity , both in vitro , in a human brain endothelial cell line , and ex vivo on spinal cord autopsy sections from HAM/TSP patients . We found that human brain endothelial cells can be productively infected in vitro by HTLV-1 , with consequent alterations in the BBB , evidenced by increased lymphocyte migration and passage of small molecules through endothelium . These data provide a basis for and transient BBB alterations that may be observed during BBB pathogenesis .
Three cellular components have been identified as forming part of the HTLV-1-entry complex: heparan sulfate proteoglycans ( HSPGs ) [20] , [21] , Neuropilin-1 [22] , a co-receptor for VEGF165 and semaphorin 3a , and the glucose transporter Glut-1 [23] . The expression of HSPGs in BBB endothelial cells has previously been reported; in vivo , HSPGs are ubiquitously expressed at the cell surface or throughout the extracellular matrix of all mammalian tissues [24]; in particular , HSPGs have been previously detected in cerebral blood vessels [25] . In this study , we examined the expression of Glut-1 and Neuropilin-1 in the endothelial cells that form the BBB in situ . As HAM/TSP is characterized by lymphocyte infiltration and inflammation mainly within the thoracic spinal cord , we focused our investigation on this region . In tissue sections derived from the thoracic spinal cord of uninfected control cases , Glut-1 expression was diffuse and patchy on fibres throughout the grey matter , intense on fibers in the dorsal root and also detected in the meninges surrounding the cord ( Fig . 1A ) . Glut-1 was expressed prominently on blood-vessels within both the white and the grey matter ( Fig . 1A and B ) . The Neuropilin-1 ( NP-1 ) was expressed diffusely in the posterior , lateral and anterior columns , and in the dorsolateral fasciculus/dorsal root ( Fig . 1D ) . Cellular expression was particularly noted at the apex of the posterior column , and on motor neurons in the anterior column ( data not shown ) . Significantly , NP-1 was highly expressed on blood vessels , and within the meninges in all segments of the thoracic spinal cord ( Fig . 1D and E ) . Vascular endothelial cell expression of Glut-1 and NP-1 was confirmed by double immunolabeling with Factor VIII , a specific marker for endothelial cells ( Fig . 1C and F ) . We then determined whether expression of these receptors by endothelial cells was conserved in HAM/TSP . In tissue sections of the thoracic spinal cord derived from HAM/TSP patients , Glut-1 immunoreactivity was detected in blood vessels , in the absence ( Fig . 2A ) or presence of cell infiltrates ( Fig . 2B and C ) . Neuropilin-1 could also be detected in blood vessels in sections from HAM/TSP patients ( data not shown ) . Since microvascular endothelial cells that constitute the BBB express HTLV-1 receptors , we examined whether the infection of these cells by HTLV-1 could be detected in situ , by performing in situ hybridization for a viral mRNA ( the messenger that encodes the viral transactivator Tax ) on the spinal cord sections . Cellular infiltrates were positive for viral Tax mRNA ( data not shown ) . However , we focused our analyses on spinal cord regions where the infiltrates were absent , to prevent the signal within the infected lymphocytes from masking the signal from resident cells within the CNS parenchyma . Since astrocytes are known to be targets of HTLV-1 infection [26] , [27] , the detection of a positive signal in several GFAP immunoreactive cells constituted a suitable positive control ( Fig . 2D ) , as shown in previous studies [4] . We also encountered rare positive signals associated with vascular structures ( Fig . 2E ) . This observation suggested the possibility of infection of cerebral endothelial cells by HTLV-1 . Although viral transcripts were found to be associated with blood vessels , this observation could not be taken as definitive evidence for the infection of endothelial cells forming the BBB specifically as other cell types such as pericytes are closely associated with endothelial cells . We therefore took advantage of a human-derived brain endothelial cell line , hCMEC/D3 , that had been previously reported to retain many BBB characteristics [28] . We first investigated the expression of the 3 ( co- ) receptors for HTLV-1 entry by flow cytometry analysis . Glut-1 expression was highly detected in permeabilized hCMEC/D3 cells . In order to detect expression of this protein at the cell surface , immunostaining was performed on fixed but non-permeabilized cells . The expression of Glut-1 at the cell surface was detected in 31% of the hCMEC/D3 cells ( Fig . 3A ) . Similarly , cell surface expression of NP-1 and HSPGs were detected in 82% and 59% of the hCMEC/D3 cells respectively ( Fig . 3B and C ) . In order to determine if the expression of surface receptors on endothelial cells allows HTLV-1 entry , we investigated whether hCMEC/D3 cells could be infected by HTLV Env-pseudotyped LacZ vectors . β-Gal production was detected upon infection of hCMEC/D3 cells by A-MLV as well as by H-MLV pseudotypes . Addition of serum from an uninfected donor did not reduce infection by either pseudotype . In contrast , addition of serum from an HTLV-1-infected HAM/TSP patient abolished the infection by H-MLV , while high level of A-MLV pseudotype infection was still observed ( Fig . 3D ) . These results indicate that hCMEC/D3 cells allow HTLV Env-mediated entry . The ability of a retrovirus such as HTLV-1 to enter a particular cell type is usually correlated with the ability of target cells to fuse and form syncytia with infected T lymphocytes . We therefore quantified the number of syncytia in hCMEC/D3 and lymphocyte cocultures . Whereas hCMEC/D3 did not fuse with C81-66 lymphocytes ( HTLV-1-infected cells that do not express Env glycoproteins ) , numerous syncytia were observed in co-cultures with infected MT2 lymphocytes at 24 hours . The number of nuclei per syncytium ranged from 3 to 11 ( Fig . 4A ) . Syncytia were found immunoreactive for viral p24 ( Fig . 4C ) . In order to ascertain the endothelial origin of the syncytia , we prestained the hCMEC/D3 cells with a vital fluorescent molecule ( CellTracker Red CMTPX ) . Syncytia were shown to be fluorescently labeled ( Fig . 4D ) . Addition of serum from an uninfected patient did not prevent the formation of syncytia . In contrast , a dramatic reduction in the number and the size of syncytia was observed when serum from an HAM/TSP patient was added to the medium ( Fig . 4A ) , showing that the fusion is HTLV-mediated . Similarly , syncytia formation could also be inhibited by addition of VEGF165 , a physiological ligand of NRP-1 that has been shown to inhibit the binding of the HTLV-1 Env proteins to target cells [22] , of dextran sulfate , an inhibitor of HSPG-mediated HTLV-1 entry [20] , or of an antibody directed against the glucose transporter Glut-1 ( Fig . 4B ) . It is worth noting that no significant alterations in syncytia formation were observed following addition of an irrelevant isotype-matched antibody to the culture medium . These data suggest that HTLV-1 entry into hCMEC/D3 cells is dependent on the interactions between viral envelope proteins and the three putative cellular receptors for HTLV-1 infection ( heparan sulfate proteoglycans , neuropilin-1 and Glut-1 ) . As HTLV-1 can enter hCMEC/D3 cells , we then determined whether this event allows a productive infection . Endothelial cells were co-cultivated with irradiated lymphocytes . The irradiation dose was lethal for these cells , as confirmed by Trypan blue staining which indicated that 100% of the lymphocytes were dead by day 8 post-irradiation ( data not shown ) . HTLV p19 was detected in the supernatants of co-cultures but decreased during the 10 first days: this correlated with lymphocyte and syncytia cell death ( data not shown ) . At 10 days post coculture , no syncytium could be observed . From day 13 post-contact , p19 production to reach about 510 pg/day at day 22 , indicating that human brain endothelial cells produced viral proteins ( Fig . 5A ) . HTLV-1 productive infection could be prevented by addition of AZT , an inhibitor for the reverse transcriptase , to the culture medium . The percentage of positive endothelial cells for viral p24 increased in parallel to p19 levels detected in the supernatant reach maximal infectivity rates at day 22 , with 18% of the endothelial cells positive for p24 ( Fig . 5B ) . No cytopathic effect of HTLV-1 infection endothelial cells was observed . At day 22 post co-culture ( 2 days after renewal of the medium ) , the supernatant was collected and ultracentrifuged . The pellet was resuspended and added to culture medium of reporter 293T-LTR-GFP cells . After 6 days of culture , fluorescent cells were visualized , demonstrating the presence of Tax protein within these reporter cells . Furthermore the fluorescent cells could form syncytia , and expressed the viral envelope protein ( Fig . 5C ) . In addition , the number of the fluorescent cells was dramatically reduced ( up to 37% ) in the presence of AZT indicating that the detected fluorescence signal in 293T-LTR-GFP cells was specific to de novo infection of reporter cells . These results indicate that cerebral endothelial cells produced infectious viral particles . Lastly , we assessed the impact of HTLV-1 infection of endothelial cells on BBB function , by evaluating the paracellular permeability of hCMEC/D3 cell monolayers and the transmigration of lymphocytes through the barrier in an infected and non-infected context . hCMEC/D3 cells , incubated with irradiated C81-66 ( HTLV-1-infected lymphocytes that are not productively infected ) or with MT2 cells 15 days prior to the experiment were seeded on filters and allowed to reach confluence . The paracellular permeability of monolayers of hCMEC/D3 cells infected with HTLV-1 was much higher than the permeability of hCMEC/D3 cells previously cocultured with control C81-66 T-lymphocytes ( Fig . 6A ) . Similarly , transmigration of uninfected T-lymphocytes ( CEM and Jurkat ) across monolayers of hCMEC/D3 cells infected with HTLV-1 was increased compared to that across a monolayer of uninfected hCMEC/D3 cells ( Fig . 6B ) . In addition , the migration of HTLV-1-infected lymphocytic cell lines through hCMEC/D3 cells was increased compared to control lymphocytes as previously reported [8] . However , no differences in HTLV-1-infected lymphocyte transmigration across HTLV-1-infected or non-infected hCMEC/D3 cells were observed . These data indicate that HTLV-1-infection of endothelial cells alters classical BBB functions . We then analyzed the expression levels of proteins that constitute the tight junctions by Western blot . As indicated in Fig . 6C , ZO-1 levels were dramatically reduced in HTLV-1-infected hCMEC/D3 cells whereas , in the case of occludin , the expression levels of the 55 kDa isoform , but not those of the 60 kDa isoform were decreased . Since we have previously shown that the Myosin Light Chain Kinase ( MLCK ) is important in tight junction regulation of endothelial cells incubated with HTLV-1 infected lymphocytes , we determined whether the inhibition of MLCK activity could prevent long-term barrier impairment of HTLV-1 infected endothelial cells ( Fig . 6D ) . Treatment of the endothelial cells for 24 h with the MLCK inhibitor ML7 failed to restore low paracellular permeability in HTLV-1 infected hCMEC/D3 cells . The molecular mechanisms of TJ disruption in endothelial cells induced by direct HTLV-1 infection appear to be different to those induced by HTLV-1-infected lymphocytes , as previously demonstrated [8] .
The BBB constitutes an interface between the bloodstream and CNS parenchyma , and regulates the intracellular and paracellular passage of molecules and cells between the two compartments [1] . Disruption of the BBB in HAM/TSP is strongly suggested by observations of perivascular cuffing with lymphocyte and macrophage infiltrates [4] , [5] , [7] , and confirmed by reports of fibrinogen and IgG deposits in the CNS parenchyma of HAM/TSP patients [4] . BBB breakdown is an important step in HAM/TSP pathogenesis , especially by facilitating migration of lymphocytes into the CNS . Infiltrated lymphocytes are believed to cause demyelination and axonal degeneration that are hallmarks of HTLV-1-associated neuropathology [29] . The mechanisms underlying BBB alteration during HAM/TSP are not yet well determined . In an in vitro model of the BBB , composed of human cerebral microvascular endothelial cells [28] , we previously demonstrated the importance of proinflammatory cytokines secreted by infected lymphocytes in the early stages of BBB disruption [8] , [30] . In the present study we investigated whether BBB endothelial cells were susceptible to HTLV-1 infection and the impact of such an infection on BBB integrity . BBB dysfunction associated with retroviral infections has been previously described . In a murine model , the infection of brain endothelial cells has been reported both in vitro and in vivo in the case of PVC-211 murine leukemia virus ( a neuropathogenic variant of the Friend MuLV ) , with a direct correlation between the replication efficiency of a virus in brain endothelial cells in vitro and its ability to cause neurological disease in vivo [11] . Moreover , in the case of infection by the Feline Immunodeficiency Virus , infection of brain endothelial cells has been proposed to represent one of the ways of viral entry into the CNS [31] . Meanwhile , infection of brain endothelial cells by human retroviral agents is still a matter of debate . In the case of HIV infection , previous studies have reported infection of endothelial cells in adult brain tissue [12]–[14] , but these have been mainly based upon interpretation of the morphological appearance and vascular localization of cells found positive by immunocytochemistry , in situ hybridization or PCR-in situ hybridization , whereas conflicting results have been obtained in vitro from brain-derived endothelial cells ( for review , see [15] ) . Up until now , no study has focused on the subject of human endothelial cell infection by HTLV-1 . Recently three membrane proteins have been described as components of the receptor for HTLV-1 entry: the glucose transporter Glut-1 [23] , and a receptor for VEGF , Neuropilin-1 [22] and heparan sulfate-proteoglycans [21] , [32] . Vascular expression of Glut-1 [33] , [34] and Neuropilin-1 [35] , [36] has been previously reported within the CNS , under normal and pathological conditions . We have shown that the HTLV-1 receptors are expressed on blood vessels of the adult human thoracic spinal cord , a region which is characterized by BBB impairment , lymphocyte infiltration and inflammation during HAM/TSP . Expression of these proteins was found in endothelial cells within the spinal cord of HAM/TSP patients , irrespective of the extent of lymphocyte infiltration . We could also detect expression of these receptors at the cell surface of a human cerebral endothelial cell line , hCMEC/D3 . We then looked for HTLV-1 infected endothelial cells by in situ hybridization using DNA probe directed against transactivator Tax transcripts on spinal cord sections of a HAM/TSP patient . Astrocytes are reportedly susceptible to HTLV-1 infection in vitro [26] , [27] . Astrocytic infection with HTLV-1 , also reported using this technique in CNS tissues ( Ozden et al . 2002 ) , served as a positive control in our samples . HTLV-1 transcripts were found only very rarely associated with the vasculature . The difficulty in detecting the virus within endothelial cells in situ may be in part due to the specific immune response developed against HTLV-1 in HAM/TSP patients [37] . HTLV-1 infected endothelial cells may be detected and lysed by cytotoxic lymphocytes , thereby constituting a new mechanism for transient BBB disruption , which remains to be explored further . Moreover , the tissue sections studied are from a patient with a rapid progressing HAM/TSP [4] . Additional studies in other HAM/TSP patients , with slower disease progression , might facilitate the detection of HTLV-1-infected endothelial cells . As the ex vivo detection was difficult to ascertain , we studied the susceptibility of an in vitro model , hCMEC/D3 , to HTLV-1 infection . We confirmed that the expression of the receptors at the cell surface of hCMC/D3 cells allowed viral entry . In fact , it has been demonstrated that cholesterol , a central component of lipid rafts , was necessary for HTLV-1 infection , especially on post-binding entry steps [38] . Thus we examined whether the spatial organization of viral receptors on the membrane of cerebral endothelial cells would enable infection with HTLV . Using LacZ-reporting viral particles pseudotyped with the HTLV envelope [23] , we demonstrated that the envelope recognized and fused correctly with the membrane of endothelial cells . Expression of functional receptors at the cell surface are usually also demonstrated via the ability of target cells to form syncytia with infected lymphocytes [39] . Syncytia were indeed formed during co-cultures of infected lymphocytes and hCMEC/D3 , and this could be prevented by addition of serum from an HAM/TSP patient , confirming the role of viral envelope in the process of cell-cell fusion . Similarly , the formation of syncytia could be prevented by addition of binding inhibitors directed to the previously described HTLV-1 receptors , demonstrating that it is a proper receptor-mediated entry . The death of cells forming syncytia was observed after one week in culture , as reported for other cell types [40] , but p19 detection in the supernatant at later stages confirmed the persistent infection of endothelial cells . The production of viral proteins corresponds to that of the infectious virus , as shown using 293T-LTR-GFP reporter cells assays . Of note , direct infection of cerebral endothelial cells may constitute a new mechanism of entry for the retrovirus into the CNS . In fact , endothelial cells and astrocytes are tightly associated at the BBB; infected endothelial cells could then transmit infectious particles to astrocytes , which have been shown to be susceptible to HTLV-1 infection both in vitro and in vivo [4] , [26] , [27] , [41] . Finally , we analyzed the impact that infection of endothelial cells had on BBB integrity . At day 20 following co-culture with irradiated lymphocytes , the endothelial cells showed no cytopathic effect and no syncytia was observed . We demonstrated that HTLV-infected hCMEC/D3 cells could no longer form confluent monolayers resistant to molecular diffusion and cellular migration . Indeed , BBB dysfunction results in enhanced transendothelial migration of both HTLV-1-infected and uninfected lymphocytes . Both lymphocytes carrying HTLV-1 and tight junction opening could facilitate the spread of the virus and cytotoxic lymphocytes into the CNS . The mechanisms for such an alteration are not completely understood and may be multifactorial . We previously demonstrated that inflammatory cytokines produces by HTLV-1 infected lymphocytes induced BBB disruption , by increasing the expression and activity of the Myosin Light Chain Kinase ( MLCK ) , which is transcriptionally regulated by the NF-κB pathway [8] . Since Tax expression induces cytokine secretion [42] , [43] , [44] , disruption of the BBB could be a consequence of the autocrine or paracrine effects of proinflammatory cytokines secreted by HTLV-1 infected endothelial cells . This could explain why a low infection rate ( <20% ) of the cells is sufficient to significantly alter BBB-related functions . However , ML-7 treatment ( an inhibitor of MLCK activity ) of virally infected endothelial cells could not prevent such a disruption . This is consistent with the observations of McKenzie and Ridley who recently showed that MLCK activity is not required for long term modification of TJ protein expression , mediated by TNFα [45] . Further investigations could focus on possible interactions between viral Tax protein and proteins that constitute TJs . For example , the protein ZO-1 bears a PDZ domain [46] and Tax a PDZ binding domain [47] and suggest that these proteins interact with each other . This interaction could induce relocalization of ZO-1 and disorganize the TJ , or the protein could directly target the proteasome , as Tax does for the Retinoblastoma protein [48] . In conclusion , we have shown that endothelial cells , which constitute the BBB , are susceptible to infection by HTLV-1 . This represents a new mechanism for BBB disruption in HAM/TSP , either directly as the expression of Tax induces a loss of BBB functions , or indirectly as the infected endothelial cells will be targeted by the immune system . Moreover , as the infection is productive , endothelial cells could allow entry of the virus into the CNS and facilitate the infection of astrocytes within the CNS parenchyma . Some may argue that the disruption of the BBB due to the infection of the endothelial cells seem to be a minor event in the natural course of HAM/TSP pathogenesis when compared to the proinflammatory cytokines [8] . However , considering this possibility should not be neglected for the design of potential new treatments . For example , the inhibition of the VEGF has been proposed previously as a way to prevent lymphocyte migration throughout the vasculature [49] , [50] . Our results suggest that such a treatment should be envisioned with caution in the context of BBB , as it could increase the availability of the Neuropilin-1 to the virus and thereby , could facilitate infection of the BBB by HTLV-1 .
The human Cerebral Microvascular Endothelial Cell line , hCMEC/D3 , was immortalized after transduction with lentiviral vectors encoding the catalytic subunit of human telomerase hTERT and SV40 T antigen , as described previously [28] . hCMEC/D3 cells were grown in Endothelial Growth Medium-2 ( EGM-2MV , Clonetics , Cambrex Biosciences , Workingham , UK ) without hydrocortisone , on Biocoat tissue culture flasks ( BD Biosciences , Bedford , MA ) . MT-2 and C81-66 were used as HTLV-1-infected T cell lines . These are cell-lines derived from human umbilical cord blood T cells , following culture with irradiated cells from at ATLL patient . Both cell lines express the viral transactivator Tax-1 , although C81-66 does not produce any viral particles . CEM and Jurkat were used as uninfected control T-cell lines . Non-adherent cell lines were grown in RPMI 1640 medium ( Gibco BRL , Gaithersburg , MD ) supplemented with 1 mM glutamine and 10% FCS . The HEK epithelial cells containing an integrated HTLV-1 long terminal repeat ( LTR ) coupled to a green fluorescent protein ( GFP ) reporter gene ( called herein 293T-LTR-GFP ) [51] were grown in Dulbecco's modified Eagle's medium ( Gibco BRL ) supplemented with glutamine ( 1 mM ) , 100 U/ml penicillin , and 10% FCS . Frozen tissue autopsy sections from a HAM/TSP patient were obtained as previously described [4] . For one experiment , we used paraffin embedded material from the thoracic spinal cord of a HAM/TSP patient from Chile , whose case has been previously described [52] . Frozen spinal cord tissues from uninfected control patients were obtained from the UK Multiple Sclerosis Tissue Bank and as previously described [53] . All tissue specimens were obtained in accordance with the respective hospital and national regulations and ethical rules . Fixed frozen ( 4% paraformaldehyde , PFA ) and snap frozen blocks of tissues were sectioned serially at 10 µm using a cryostat; sections were then air-dried , and fixed in methanol before processing for immunohistochemistry . Sections were either labeled by immunoperoxidase technique ( 3-step procedure , DAKO , Glostrup , Denmark ) or alcaline phosphatase technique ( Vectastain , Vector Laboratories , CA , USA ) . Endogenous peroxidase was blocked with 2 , 5% hydrogen peroxidase in methanol , and non-specific labeling blocked with normal sera corresponding to the secondary antibody species . Sections were dehydrated in graded alcohols , cleared in xylene and coverslipped using permount . The primary antibodies used were a mouse antibody to Glut-1 ( MAB 1418 , R&D systems , Minneapolis , MN , USA ) , to NP-1 ( clone A-12 , Santa Cruz Biotechnology , Santa Cruz , CA , USA ) , or to Factor VIII ( DAKO ) . Cultures were fixed with 4% PFA . Staining with primary antibody was performed after incubation for 30 minutes with 10% normal goat serum and 0 . 05% saponin diluted in PBS . The following primary antibodies were used: mouse antibodies to HTLV-1-p24 ( ab9081 , Abcam , Cambridge , UK ) , to Glut-1 ( R&D systems ) , to NP-1 ( Santa Cruz Biotechnology ) . Specific secondary antibodies were coupled with Fluorescein ( Vector Laboratories ) . After washes , preparations were mounted in DAPI-containing Vectashield medium ( Vector laboratories ) . ISH was performed on serial sections of frozen tissues by means of 32P antisense and sense riboprobes corresponding to the complete tax mRNA , as described elsewhere [54] . Sections were stained prior to ISH by immunoperoxidase using a rabbit polyclonal antibody against Glial Fribrillary Acidic Protein ( anti-GFAP , DAKO ) . Cells were analyzed for surface or intracellular receptor expression , with or without permeabilization with triton . Cells were resuspended in PBS-EDTA and incubated at 4°C for 30 min with a primary antibody to NP-1 , Glut-1 , HSPGs ( clone F69-3G10 , Seikagaku Corp . , Tokyo , Japan ) or HTLV-1 p24 . Secondary antibody staining was performed by incubating the cells with fluorescein-isothiocyanate ( FITC ) -labeled antibody ( Vector Laboratories ) at 4°C for 30 min . Cells were washed twice with phosphate-buffered saline ( PBS ) and analyzed by using a FACScan flow cytometer and Cellquest software ( Becton Dickinson , San Jose , CA , USA ) . Replication-defective LacZ retroviral vectors pseudotyped with either the HTLV ( H-MLV ) , or amphotropic murine leukemia virus ( A-MLV ) envelope proteins were produced by transfection of 293T cells with Gag/Pol , Env , and LacZ plasmids as described elsewhere [23] . Target cells were plated on 24-well plates ( 5 . 104 ) for 24 h and supernatants from H-MLV ( 1/5 ) , or A-MLV ( 1/100 ) pseudotype-producing cells were added . Infectivity was assessed 48 h later by measuring the level of lacZ activity with the β-Gal reporter gene assay kit ( Roche , France ) . Syncytia were obtained after 24 h coculture of hCMEC/D3 cells and MT-2 infected lymphocytes ( 1∶1 ratio ) . The p24 staining was performed by an indirect immunofluorescence assay using the mouse anti-HTLV-1 p24 antibody on PFA fixed cells . To ascertain the endothelial origin of the syncytia , the hCMEC/D3 cells were prelabeled with CellTracker™ Red CMTPX ( Molecular Probes , Invitrogen , Carlsbad , CA , USA ) . The preparations were visualized with a Zeiss Axiovert apparatus ( Iena , Germany ) or Leica DMRB ( Wetzlar , Germany ) . The syncytia formation was inhibited by addition of dextran sulfate ( 100 µg/mL , Sigma Aldrich , St Louis , MO , USA ) , VEGF165 ( 50 ng/mL , R&D systems ) or rabbit polyclonal antibody directed to Glut-1 ( ab15309 , Abcam; anti Cbl-b antibody H454 was used as an irrelevant antibody ) . The extent of syncytia formation was assessed after 24 h in coculture by counting their numbers , as well as nuclei per syncytia , using Giemsa staining . All microscopic fields from 16 mm diameter coverslips were evaluated , from three different cultures . For inhibition experiments , serum from a HAM/TSP patient or control serum ( lacking HTLV-1 antibodies detected by Western blot assay ) was added at the beginning of the coculture . Chronically infected cells ( MT-2 lymphocytes or C81-66 cells as control as they can not transmit infection ) were irradiated at 10 Gy and washed twice with PBS to eliminate free radicals . hCMEC/D3 monolayers were cocultivated overnight at 37°C with the irradiated lymphocytes at a 1∶1 ratio and extensively washed three times with medium lacking serum . Human endothelial cell cultures were then maintained in normal culture medium . Aliquots of culture medium were collected at different time to detect viral proteins . HTLV-1 p19 was detected in culture media using the HTLV p19 antigen ELISA assay ( Zeptometrix , Buffalo , NY , USA ) . In order to prevent the viral infection , 25 µM of AZT , an inhibitor of the reverse transcriptase , were added to the culture medium . Secondary infection was performed as described previously [55] . Briefly , HCMEC/D3 cells were cocultured with irradiated MT-2 lymphocytes ( or control C81-66 cells ) for 15 days . The cells were then extensively washed and the medium replaced . Forty-eight hours later , the growth medium was collected and clarified by low-speed centrifugation ( 2 , 500 rpm for 5 min ) then filtered through a 45-µm filter . The resulting media was then layered on a 20% glycerol gradient and the virus in it was pelleted by centrifugation in a SW28 rotor at 22 , 000 rpm for 2 h . The pellet was then resuspended in 200 µl of serum-free DMEM . 293T-LTR GFP indicator cells were incubated with 200 µl of the virus suspension in a total volume of 2 ml of serum-free DMEM for 2 h . Complete medium was then added and changed twice a week . One week later , the cells were fixed in 4% PFA , and visualized with a Zeiss Axiovert apparatus . The specificity of the signal was confirmed by addition of AZT into the culture medium . TJ proteins expression in HCMEC/D3 cells ( cocultured for 15 days with MT-2 or C81-66 irradiated lymphocytes ) was investigated by immunoblot analysis . Cells were washed twice with PBS , lysed in appropriate buffer ( 50 mM Tris-HCl , pH 7 . 4 , 120 mM NaCl , 5 mM EDTA , 0 . 5% Nonidet P-40 , 0 . 2 mM Na3VO4 , 1 mM dithiothreitol , 1 mM phenylmethylsulfonyl fluoride ) in the presence of a cocktail of protease inhibitors ( Roche Applied Science , Indianapolis , IN , USA ) , and incubated on ice . Protein concentration was determined by the Bradford method ( Bio-Rad , Hercules , CA , USA ) . Samples were loaded into 4–20% Tris/Gly gels ( NOVEX , Invitrogen ) , subjected to SDS-PAGE , and transferred onto a nitrocellulose membrane ( Immobilon-P , Millipore , Billerica , MA , USA ) . Following incubation with specific antibodies ( rabbit anti ZO-1 and Occludin from Zymed or mouse anti-HTLV-p24 from AbCam ) and extensive washing in PBS-Tween 0 . 05% , membranes were incubated with horseradish peroxidase-conjugated secondary antibodies ( Vector Laboratories ) and developed using either the SuperSignal WestPico or SuperSignal West Femto Chemiluminescent substrate kit ( Pierce , Rockford , IL , USA ) . To ensure equal amount of protein loaded per well , membranes were stripped with the Re-blot Plus Kit ( Chemicon International , Temecula , CA , USA ) and reprobed with a specific anti β-tubulin antibody ( Santa Cruz Biotechnology ) . Permeability of hCMEC/D3 cell monolayers was measured using a method adapted from Dehouck et al . [56] and Romero et al . [30] on Transwell-ClearTM filters ( polyester , 12 mm diameter , pore size 3 µm , Costar , Brumath , France ) . Briefly , 105 cells/well were seeded on filters previously coated with rat-tail collagen I ( BD Biosciences ) and bovine plasma fibronectin ( Sigma Aldrich ) . At confluence , hydrocortisone was added to EBM-2 medium as recommended by the manufacturer . After 24 hours , cells were used for experiments . Coculture experiments were set up by adding of 105 lymphocytes to the endothelial monolayer in the presence or absence of inhibitors . For the permeability test , the culture medium was replaced by DMEM without phenol red . FITC-labelled dextran ( molecular weight 70 kDa , Sigma Aldrich ) was added to the upper compartment and inserts were transferred sequentially at 5 minutes intervals from well to well for 30 min . The quantity of FITC-dextran that had diffused through the monolayer into the lower compartment at each time point was determined using a fluorescence multiwell plate reader ( Wallac VictorTM 1420 , PerkinElmer , Wellesley , MA , USA ) . The permeability coefficients of the endothelial monolayers were then calculated as previously described [30] . Lymphocytes were labeled with CellTracker™ Green BODIPY® ( Molecular Probes ) according to manufacturer's instructions . Labeled lymphocytes were added to the upper chamber of Transwell-ClearTM insert filters ( polyester , 12 mm diameter , pore size 3 µm , Costar ) containing confluent hCMEC/D3 monolayers . After 24 hours at 37°C , the monolayer was extensively washed with PBS/EDTA , in order to collect lymphocytes adherent to each side of the membrane . Cells were lyzed using water . Fluorescence intensity was determined using a fluorescence multiwell plate reader . | The blood–brain barrier ( BBB ) forms the interface between the blood and the central nervous system ( CNS ) . BBB disruption is considered to be a key event in the pathogenesis of retroviral-associated neurological diseases . The present paper deals with the susceptibility of the endothelial cells ( i . e . , one of the main cellular components of BBB ) to retroviral infection , and with the impact of infection in BBB function . This study focuses on the Human T-Lymphotropic Virus ( HTLV-1 ) , which infects 20 million people worldwide , and is the etiological agent of a neurodegenerative disease called HTLV-1 Associated Myelopathy/Tropical Spastic Paraparesis ( HAM/TSP ) . We first demonstrated that the cerebral endothelial cells express the receptors for the retrovirus in vitro , and on spinal cord autopsy sections from non-infected and HAM/TSP patients . We found on these latter that vascular-like structures were infected and confirmed in vitro that the endothelial cells could be productively infected by HTLV-1 . We demonstrated that such an infection impairs BBB properties in vitro , as well as tight junctions , that are cell adhesion structures . This study is the first to demonstrate the impact of HTLV-1 infection on human BBB integrity; such a susceptibility has to be considered in the design of future therapeutics strategies . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"infectious",
"diseases/infectious",
"diseases",
"of",
"the",
"nervous",
"system",
"infectious",
"diseases/viral",
"infections",
"virology/host",
"invasion",
"and",
"cell",
"entry"
] | 2008 | Alteration of Blood–Brain Barrier Integrity by Retroviral Infection |
Wnt/β-catenin signaling is an ancient pathway in metazoans and controls various developmental processes , in particular the establishment and patterning of the embryonic primary axis . In vertebrates , a graded Wnt activity from posterior to anterior endows cells with positional information in the central nervous system . Recent studies in hemichordates support a conserved role for Wnt/β-catenin in ectoderm antero-posterior patterning at the base of the deuterostomes . Ascidians are marine invertebrates and the closest relatives of vertebrates . By combining gain- and loss-of-function approaches , we have determined the role of Wnt/β-catenin in patterning the three ectoderm derivatives of the ascidian Ciona intestinalis , central nervous system , peripheral nervous system and epidermis . Activating Wnt/β-catenin signaling from gastrulation led to a dramatic transformation of the ectoderm with a loss of anterior identities and a reciprocal anterior extension of posterior identities , consistent with studies in other metazoans . Surprisingly , inhibiting Wnt signaling did not produce a reciprocal anteriorization of the embryo with a loss of more posterior identities like in vertebrates and hemichordate . Epidermis patterning was overall unchanged . Only the identity of two discrete regions of the central nervous system , the anteriormost and the posteriormost regions , were under the control of Wnt . Finally , the caudal peripheral nervous system , while being initially Wnt dependent , formed normally . Our results show that the Ciona embryonic ectoderm responds to Wnt activation in a manner that is compatible with the proposed function for this pathway at the base of the deuterostomes . However , possibly because of its fast and divergent mode of development that includes extensive use of maternal determinants , the overall antero-posterior patterning of the Ciona ectoderm is Wnt independent , and Wnt/β-catenin signaling controls the formation of some sub-domains . Our results thus indicate that there has likely been a drift in the developmental systems controlling ectoderm patterning in the lineage leading to ascidians .
Ascidians belong to the tunicates , the sister group of the vertebrates . Together with the cephalochordate ( amphioxus ) and vertebrate phyla they form the super-phylum of chordates whose specific body plan includes a notochord and a dorsal hollow neural tube during embryonic life . Comparative developmental studies between these three phyla is essential for elaborating evolutionary scenarios explaining the emergence of chordates and their diversification [1–4] . Ascidians are particularly puzzling organisms since they took a significantly different evolutionary path from other chordates resulting in divergent morphological , embryological and genomic features . Their development is fast and stereotyped with very few cells; and ascidian genomes have undergone compaction and extensive rearrangements when compared to vertebrates and amphioxus . This raises the question of whether developmental mechanisms controlling typical chordate structure formation are conserved between ascidians and other chordates . In particular , ascidian embryos are the emblematic examples for the concept of mosaic development . However , it is well known that cell-cell communication is involved in cell fate determination , yet possibly at only short distances ( i . e . neighboring cells ) [5–7] . Wnt signaling , one of the pathways present in animals , allows cells to communicate through the secretion of the Wnt ligands that bind to their cognate Frizzled receptors . It is involved in a wide range of biological processes during embryogenesis and adult homeostasis [8 , 9] . The canonical Wnt pathway ( cWnt ) that uses the protein β-catenin as a central mediator to control target gene transcription is extensively involved in axis formation during the development of many metazoans [10 , 11] . Three discrete developmental processes contribute to antero-posterior ( AP ) axis formation in bilaterians: germ layer specification , AP patterning and posterior growth . During cleavage/blastula stages , nuclear accumulation of β-catenin in the vegetal hemisphere specifies endomesoderm in several phyla ( nemerteans , echinoderms , hemichordates and ascidians ) [12–15] . A similar function for the specification of the endoderm at the oral pole of cnidarians suggests that this constitutes an ancient function at the base of metazoans [16 , 17] . In vertebrates , a posterior to anterior gradient of activity provides cells with positional information and patterns the central nervous system ( CNS ) [18–21] . A recent study in hemichordates demonstrated that this function for cWnt is conserved at the base of the deuterostomes [22] . Finally , in both insects and vertebrates , Wnt signaling controls body elongation during posterior growth [23] . Posterior growth does not exist in ascidians since the embryo elongation at the improperly named tailbud stages occurs through cell division and rearrangement without proper addition of new tissue from a growth zone [24] . Embryonic axes are determined very early and can be identified in the fertilized egg before first cleavage [5] . cWnt participates in endomesoderm formation along the animal-vegetal axis [13 , 25] . The AP axis is orthogonal and determined following ooplasmic movements that localize asymmetric cleavage determinants to the posterior . A consequence for AP patterning is that anterior ( so called a-line ) and posterior ( b-line ) ectoderm precursors have intrinsically different potentials in response to neural induction as soon as they arise at the 8-cell stage [26] . Interestingly , not only the CNS but also the epidermis is patterned along the AP axis; and this patterning also involves signals from vegetal tissues [27] . The transcription factor FoxA-a is the anterior determinant that establishes the early a-line versus b-line potentials [28 , 29] . Since direct transcriptional FoxA-a targets are Wnt antagonists–a-line expressed Sfrp1/5 and Ror genes–there is a potential role for Wnt signaling in ectoderm AP patterning . Moreover , the AP identity of the two sensory pigment cells within the CNS is controlled by Wnt signaling [30] . However , a global function for Wnt in ectoderm AP patterning has not been investigated; and this is the topic of the present manuscript . Sequencing and annotation of the ascidian Ciona robusta ( formerly known as C . intestinalis type A ) has revealed a complement for Wnt signaling compatible with a functional pathway [31–34] . A recent phylogenetic analysis has shown that the ten Wnt ligands found in the Ciona genome correspond to 10 out of the 13 families present at the base of chordates , with the loss of Wnt1 , Wnt4 and Wnt8 [35] . Their spatio-temporal expression has been described throughout embryogenesis for eight of them , but only a few show a restricted pattern ( Wnt3 , Wnt5 , Wnt6 and Wnt7 ) [35–39] . In particular , they do not display a staggered expression in the posterior of the embryo as observed for many metazoans including vertebrates and amphioxus ( reviewed in [10] , [35] ) . The only possible similarity would be the expression of the four above mentioned ligands in caudal muscle at cleavage/gastrula stages and the epidermal expression of Wnt5 at the posterior tip of the forming tail . At the opposite pole of the embryo , the Wnt antagonists , Sfrp1/5 and Ror genes , are expressed in the anterior ectoderm as described above [28 , 29 , 36 , 37] . The C . robusta genome contains five Frizzled receptors [33] . The expression pattern is known for three of them ( Frizzled1/2/7 and Frizzled5/8 are maternally and ubiquitously expressed; Frizzled4 is expressed in the ectoderm from the 16-cell stage and later in various discrete regions ) , but does not allow us to predict where and when Wnt signaling is active [36] . In the present study , we have combined ectopic activation and down-regulation of the cWnt pathway to assess the effects on AP patterning of the C . intestinalis embryonic ectoderm . Activating cWnt from gastrulation leads to a loss of anterior ectoderm that is converted in posterior ectoderm . By contrast , inhibiting cWnt has varying effects depending on the ectoderm derivatives . Epidermis AP patterning is unchanged . The CNS is largely unaffected , except for its anterior and posterior ends , suggesting a function of cWnt in refining a global AP pattern that is defined by other means . Finally , the early definition of the caudal peripheral nervous system ( PNS ) requires cWnt signaling but redundant mechanisms allow proper differentiation of this tissue . Consequently , while the Ciona ectoderm displays a sensitivity to cWnt activation that is compatible with the expected function for cWnt at the base of deuterostomes , cWnt is only marginally required for ectoderm AP patterning .
LiCl or small molecule inhibitors of Gsk3β have been previously used in ascidian embryos to activate the cWnt pathway [13 , 25 , 40] . We have further developed such treatments using two distinct inhibitors , 1-azakenpaullone and BIO [41 , 42] . These inhibitors were tested at two doses ( 5 and 10 μM for 1-azakenpaullone; 1 and 2 . 5 μM for BIO ) . The results presented here correspond to the highest dose for each molecule , conditions leading to fully penetrant and identical phenotypes for both inhibitors . As expected , early treatments led to ectopic endoderm formation as revealed by staining for endoderm specific endogenous alkaline phosphatase activity ( S1 Fig ) . Treatments starting at the 32-cell stage or later produced embryos with a dramatically abnormal morphology but without ectopic endoderm , allowing us to determine effects on ectoderm patterning without interfering with germ layer formation . A previous report has suggested that activating the cWnt pathway interferes with epidermal sensory neuron ( ESN ) formation along the AP axis in the tail [40] . We reproduced the reported results: a loss of anterior ventral ESN formation ( revealed by the expression of Etr at late tailbud stages ) when the treatment was initiated at early neurula stages ( stage 14 ) and an absence of effect when the treatment was initiated at initial tailbud stages ( stage 17 ) ( Fig 1D–1G ) . However , when the treatment was initiated at the onset of gastrulation ( stage 10 ) , we observed ectopic ESNs located in the ventral trunk midline ( Fig 1B and 1C ) . Caudal ESNs are known to arise from a neurogenic territory characterized by the expression of Klf1/2/4 [43] . The presence of ectopic ESNs in the trunk upon cWnt activation was accompanied by the ectopic expression of Klf1/2/4 in the ventral trunk midline ( Fig 1I–1L ) , suggesting that these ectopic ESNs arise from an ectopic neurogenic territory . Interestingly , Klf1/2/4 ectopic expression was also observed in treatments starting at stage 14 while Etr expression was repressed . To further investigate the apparent posteriorization of the ectoderm following cWnt activation , we determined the expression of Zf115 , a gene with a marked restricted expression in the tail epidermis [44] . Zf115 was ectopically expressed in the entire trunk epidermis for both of our early treatments ( Fig 1N–1Q ) but not for the latest treatment ( S2 Fig ) . To further delineate the sensitivity of the ectoderm to cWnt activation , we performed 30 min treatments at various developmental stages and assessed the expression of both Etr and Zf115 . Ectopic expression of Etr in the ventral trunk was observed when such short treatments were performed during gastrulation ( stages 10 to 13 ) , but not later ( S2 Fig ) . The loss of anterior ventral tail Etr expression described above was not observed in the pulse treatments suggesting a longer exposure time may be required . Ectopic Zf115 expression in the trunk was observed for all pulse treatments with a reducing effect as the treatment was delayed: treatment at the onset of gastrulation led to an ectopic expression in the entire trunk while later treatments led to an extension limited to the posterior trunk ( S2 Fig ) . Above results suggest that cWnt activation converts trunk ectoderm into tail ectoderm with a maximum sensitivity during early gastrulation . In this section , we will determine what are the effects of activating cWnt from gastrulation on all three ectoderm derivatives: the epidermis , the peripheral and the central nervous system . We have thus examined by in situ hybridization the expression of a panel of AP markers for the ectoderm at early tailbud stages ( stage 19/21 ) following 1-azakenpaullone or BIO treatment from initial gastrula ( stage 10 ) . Both drugs led to similar effects ( Figs 2 and S3 ) . Interestingly , while we observed a dose response on the morphology of embryos treated with 1-azakenpaullone , the effect on the AP markers examined remained consistent for all doses ( S4 Fig ) . We did not observe a graded effect similar to what was observed when the treatment was staggered over timed intervals ( S2 Fig ) . To verify the specificity towards Wnt/β-catenin signaling of the above results , we overexpressed Wnt5 , a ligand normally restricted to the posterior ectoderm [36] , throughout the ectoderm using DNA electroporation . This led to ectopic expression of the tail midline markers Msxb , Klf1/2/4 and Nkx-C in the ventral trunk epidermis ( S5 Fig ) . However , the embryo morphology was severely affected , making gene expression analysis tedious . We turned to overexpression of ΔN-β-catenin , a version of β-catenin that is deleted from the N-terminal domain ( containing Gsk3β phosphorylation sites ) and that behaves as a dominant active form [40] . We could reproduce the results obtained using Gsk3 inhibitor treatments: ectopic expression of Six1/2 , Six3/6 , Msxb and Klf1/2/4 ( Figs 3B , 3D and 3L and S5 ) , and repression of the epidermal expression of Hox1 , Islet and Ror-a ( Fig 3F , 3H and 3J ) . The CNS expression of Hox1 was not affected since we targeted the ectoderm using the promoter of the Fucosyl transferase gene [40]; CNS Hox1 positive cells originate from vegetal lineages and do not express this gene . These observations strengthen our findings that AP patterning defects result from direct action of Wnt/β-catenin signaling . In previous results , we observed that activating cWnt led to the formation of an ectopic neurogenic territory in the ventral trunk epidermis . It is known that Bmp signaling is required to specify the tail ventral midline and that Bmp signaling is active throughout the ventral epidermis , both in the trunk and the tail [43 , 46] . We thus expressed Noggin , a secreted Bmp antagonist , together with ΔN-β-catenin . As predicted , ectopic expression of Klf1/2/4 and Msxb was suppressed ( Figs 4E and S6 ) . As previously reported , when Bmp signaling was activated , Klf1/2/4 and Msxb were ectopically expressed throughout the tail epidermis ( Figs 4C and S6 ) [43 , 46] . Activation of cWnt signaling in addition to Bmp led to ectopic activation of both genes in the trunk epidermis as well ( Figs 4F and S6 ) , suggesting that combining cWnt and Bmp signals is sufficient to launch the tail ventral neurogenic program ( Fig 4G ) . We have used the overexpression of two different proteins to block Wnt signaling . TcfΔC is a dominant negative form of the transcription factor Tcf that normally regulates transcription , together with β-catenin , downstream of the binding of a Wnt ligand to a Frizzled receptor . It contains a C-terminal deletion that eliminates the DNA binding domain of Tcf and has been previously used in Ciona to inhibit β-catenin nuclear activity during endomesoderm formation [25 , 47] . Sfrp1/5 is a naturally secreted antagonist of Wnt signaling that acts by sequestering Wnt ligands and thus preventing them from binding to Frizzled receptors [48] . Both molecules were overexpressed in the entire ectoderm using the promoters of the Friend of gata ( Fog ) or Fucosyl transferase ( Ft ) genes [40 , 49] . The following combinations produced the strongest phenotypes and were used in subsequent experiments: pFog>TcfΔC and pFt>Sfrp1/5 . We first determined the efficiency of our constructs by testing their ability to counteract the effect of Gsk3 inhibitor treatment . Overexpression of TcfΔC was sufficient to suppress the ectopic activation of both Six3/6 and Klf1/2/4 triggered by 1-azakenpaullone treatment ( Fig 5A–5D ) . We could not perform the same assay for Sfrp1/5 since it acts upstream of Gsk3 inhibitors in the cWnt pathway . However , its overexpression had similar effects as TcfΔC overexpression did . Given the robust phenotypes on epidermal expression following cWnt activation , we expected a strong reciprocal effect for Wnt inhibition: loss of posterior markers and posterior extension of anterior marker expression . We were surprised to see that epidermal expression of Islet , Ror-a , Hox1 and Cdx was unchanged ( Fig 5E , 5F , 5I and 5J ) , with possibly a weak reduction in levels of expression in the most affected embryos as depicted for Cdx on Fig 5J . The only clear difference we could detect was a repression of the epidermal expression of Hox12 using both constructs ( Fig 5Kii and 5Kiii ) , but we did not detect a concomitant posterior extension of Cdx into the tail tip ( Fig 5J ) . Consequently , with the exception of Hox12 and possibly the tail tip , epidermis AP patterning is largely unchanged following Wnt signaling inhibition . We have determined , at tailbud stages , the expression of CNS genes whose expression was modified following cWnt activation: Six1/2 , Six3/6 , Hox1 and Hox12 . Both Six1/2 and Six3/6 were robustly repressed by pFog>TcfΔC , but only slightly downregulated by pFt>Sfrp1/5 in a minority of embryos ( Fig 5G and 5H ) . This suggests that both genes could be positively regulated by cWnt signaling although in a ligand independent manner . Anterior tail nerve cord Hox1 expression was unchanged ( Fig 5I ) . Expression of Hox12 in the posterior of the tail nerve cord was downregulated by pFt>Sfrp1/5 but unaffected by pFog>TcfΔC ( Fig 5K ) . This difference possibly stems from the embryonic origin of the Hox12 expressing cells that may be of vegetal origin ( A-line ) . Consequently , they do not express the promoters used and as such , do not express the transgenes . Since Sfrp1/5 is a secreted molecule it can prevent these cells from receiving Wnt signals . In summary , CNS patterning is regulated by Wnt signaling at only the anteriormost and the posteriormost regions of the axis . The above results prompted us to test the effects of Wnt signaling modulations on the early formation of the CNS at the neural plate stage ( stages 13/14 ) ( Fig 6 ) . Etr , whose expression is found in the CNS precursors ( rows I to IV according to [50] ) and in the palp forming region at the medial anterior neural plate border ( rows V and VI; Fig 6Ai and 6D ) , displayed a loss of expression in this latter territory following cWnt activation ( Fig 6Aii ) . This corroborates previous results obtained at later stages with the markers Ror-a , Otx and Islet . However , the loss of this marker does not correspond to a conversion into more posterior neural tissue since the expression of Six3/6 , which is immediately posterior to the anterior neural plate border , was unchanged ( row IV; Fig 6B ) . Moreover , Ap2-like2 expression was also unchanged and did not extend posteriorly ( Fig 6C ) suggesting that a conversion into epidermis had not occurred . When Wnt signaling was inhibited , Etr was ectopically expressed laterally in the anterior neural plate border ( Fig 6Aiii and 6Aiv ) . These observations suggest that Wnt signaling regulates medio-lateral patterning of the anterior neural plate border . Importantly , as development proceeds , the medial part of the anterior neural plate border stained by Etr will form the very anterior palps region while the lateral part , Etr negative , will form the region immediately posterior containing anterior apical trunk ESNs ( aATENs ) ( Fig 6D and 6E ) [51 , 52] . We next tested the requirement of Wnt signaling for tail PNS formation . In a reciprocal manner to their activation following cWnt activation ( S5 Fig ) , the genes Msxb , Klf1/2/4 and Nkx-C were strongly downregulated following either TcfΔC or Sfrp1/5 overexpression ( Figs 7C , 7D and S7 ) . Of these , Msxb was the most affected gene and displayed a complete loss of expression by in situ hybridization for the strongest phenotypes ( Figs 7C , 7D and S8 ) . We next assessed ESN formation using Etr as a marker . To avoid confusion with CNS staining , we only scored ventral ESNs . The number and location of ESNs is stochastically determined and so varies from embryo to embryo [43] . For the control embryos electroporated with pFog>Venus , we counted 6 . 8 ESNs on average ( n = 44 embryos ) . The numbers for the experimental embryos were as follows: 6 . 8 ESNs for pFt>Sfrp1/5 ( n = 46 ) and 5 . 8 for pFog>TcfΔC ( n = 42 ) . This suggests that Wnt signaling is not essential for tail PNS formation . A possible explanation comes from the observation of Achaete-scute a-like2 , a transcription factor expressed in the tail neurogenic midlines [45] . Contrary to the other three genes examined , Achaete-scute a-like2 expression was unchanged following either activation or inhibition of Wnt signaling ( Fig 7E–7H ) . Achaete-scute a-like2 could thus compensate the downregulation of other transcription factors and allow tail PNS formation when Wnt signaling is blocked .
Data from cWnt activation together with the expression of Wnt5 and Wnt6 posteriorly and of Wnt antagonists ( Sfrp1/5 , Ror-a and Ror-b ) anteriorly fit with a global view of graded Wnt activity from posterior to anterior . These data are in agreement with the proposed ancient role for cWnt signaling in patterning the AP axis during early embryonic development , at least at the base of the deuterostomes [22] . In particular , the ascidian epidermis that contains neurogenic domains ( forming ESNs ) is highly regionalized along the AP axis . While we are not aware of similar organization in vertebrates , with the exception of specific regions such as the amphibian cement gland , similarities might be drawn with the hemichordate neurogenic ectoderm whose AP pattern is regulated by Wnt signaling [22] . Our results from cWnt activation are at first glance similar to what has been observed in other metazoans–repression of anterior identities and promotion of more posterior identities . We would like to discuss these observations by combining cWnt inhibition data and by restricting our comparisons to deuterostomes ( Fig 9 ) . The marked difference between activation ( dramatic posteriorization phenotypes ) and inhibition ( discrete and limited phenotypes ) of cWnt signaling was rather puzzling . An obvious explanation could be the incomplete inhibition of the pathway . TcfΔC has been previously used to inhibit endomesoderm formation in Ciona [25] and we have shown that it can suppress the action of the Gsk3 inhibitor 1-azakenpaullone ( Fig 5 ) ; and Sfrp1/5 led to similar effects in the experiments presented here . The activation data that we have presented ( Figs 1 and S2 ) show that the ectoderm is responsive to cWnt signaling for a prolonged period of time during gastrulation and neurulation , and cWnt signaling might consequently be ongoing during this time window ( only around 4hrs in Ciona developing at 18°C [61] ) . We thus tested various combinations of the two ectodermal drivers ( pFog and pFt ) , a strong ubiquitous driver ( pEf1α [62 , 63] ) as well as combinations of TcfΔC and Sfrp1/5 ( S8 Fig ) . This did not lead to a dramatically stronger repression of the genes that we have tested . While we cannot rule out that cWnt is active in the ectoderm before our earliest driver ( pFog: 16-cell stage ) , we conclude that a partial inhibition of cWnt is not the most likely explanation for the modest phenotypes that we have observed . In addition to Wnt signaling , several pathways ( Fgf , retinoic acid , Shh ) participate in patterning the CNS of the vertebrate embryo [56 , 64–66] . In Ciona , while retinoic acid regulates AP identity of both the CNS and the epidermis at the level of the anterior tail , Fgf regulates tail tip identity of the epidermis but also CNS patterning at various places ( tail tip , posterior sensory vesicle , pigment cells , anterior neural plate border ) [24 , 40 , 67–70] . Fgf signaling is thus likely to interact with cWnt and may act as a redundant signal that compensates for the loss of Wnt signaling in our experiments; we aim at testing their respective functions in future experiments . A major outcome of cWnt signaling is the regulation of gene expression and transcriptional reporters containing Tcf binding sites have been used as proxies to determine cWnt activity . We have used a reporter previously described in Ciona [40] and found the same global conclusions: reporter activity could be detected in endomesoderm derivatives and in the neurogenic tail ventral midline ( S1 Table ) . We also detected activity in the posterior dorsal midline . In addition , our quantification of reporter activity showed that while endomesodermal activity was detected in a large majority of the embryos , epidermal activity was found , at best , in around 10% of the embryos . Furthermore , this reporter was not active in the CNS regions where we functionally uncovered a role for Wnt signaling . This suggests that cWnt activity in the ectoderm may be very low or possibly at levels undetectable by the reporter used , or that this reporter may not be a faithful readout of cWnt in the ectoderm . Finally , a major explanation for the modest roles of cWnt in ectoderm AP patterning is likely to stem from the mosaic development of ascidians . In particular , it is well known that the binary AP difference in the ectoderm occurs as early as the 8-cell stage between the trunk and the tail ectoderm precursors , and that FoxA-a acts as an anterior determinant [26 , 28 , 71 , 72] . cWnt might thus be involved , possibly together or redundantly with other signals , in refining this basic pattern . For example , both Wnt5 and Wnt6 are expressed posteriorly and could participate in the definition of the posteriormost CNS and caudal PNS . Wnt6 is also expressed transiently in the anterior neural plate border similarly to Six3/6 and could play a role in the patterning of this region of the embryo [36 , 37] . Further combinatorial and targeted experiments will be required to definitively determine the precise function of Wnt signaling in ectoderm patterning . While cWnt is not essential for caudal PNS formation , we have uncovered two distinct functions . First , cWnt appears to interact with Bmp signaling to position within the embryo the ventral neurogenic midline , by regulating the expression of the gene Msxb ( S6 Fig ) . This is not the only mechanism involved since the expression of Achaete-scute a-like2 , another early midline gene , is Wnt independent . It would be interesting to uncover and compare the mechanisms that initiate the transcription of both genes in the tail ventral ectoderm through the study of their cis-regulatory DNAs . The timed activation of cWnt allowed us to uncover a later function for cWnt that is independent of the posteriorization; cWnt repressed ESN formation ( Fig 1 ) . It is well known that Notch signaling regulates the number of ESNs that form in the caudal midlines and launches a proneural transcriptional cascade [43 , 73–75] . It will thus be important to determine whether cWnt interacts with this GRN and at which level .
Ripe adults of Ciona intestinalis ( formerly referred to Ciona intestinalis type B [32] ) were provided by the Centre de Ressources Biologiques Marines in Roscoff ( EMBRC-France ) . Embryo obtention and electroporation were performed as described [72]: 50 μg of each plasmid DNA were used in a 350 μl electroporation volume placed in a 4 mm cuvette and a single pulse of 25V for 32 ms was applied using an ECM830 electroporator ( BTX , Harvard Bioscience ) . Stock solutions of 1-azakenpaullone ( 191500 , Calbiochem , Merck ) and BIO ( 361550 , Calbiochem , Merck ) were prepared at 10 mM in DMSO . Dilutions were made in sea water just before use at the concentration indicated in the text . Embryo staging and neural plate description were performed according to [50 , 61] . We have used several previously reported constructs: pFog>Noggin , pFog>Admp and pFog>Venus [43] , pFt>ΔN-β-catenin and p12xTcf>nlsLacZ [40] . The other constructs were generated using dedicated Gateway vectors [76] . The activity of the following promoters has been previously described: pFog ( pan-ectodermal from the 16-cell stage ) [49] , pFt ( pan-ectodermal from the 64-cell stage ) [40] and pEf1α ( ubiquitous from early gastrula stages ) [62 , 63] . While the first two were available in Gateway vectors , the last one was introduced following PCR amplification ( Forward primer: AAAAAGCAGGCTTTGCTTTACCATCGCGTGACG , reverse primer: AGAAAGCTGGGTTTTGGAAGGTTGGGGTTAACC ) using pSPCiEF1α>Cas9 [77] as a template . We have used entry clones containing the coding sequence of ΔN-β-catenin ( generated by a BP reaction from pFt>ΔN-β-catenin [40] ) , TcfΔC ( generated by PCR from pRN3-TcfΔC [25] . Forward primer: AAAAAGCAGGCTCAGAAAAAATGCCTCAGTTAAACTCGGA , reverse primer: AGAAAGCTGGGTTCATGGCCGACTTGGTTTG ) , Sfrp1/5 ( generated by RT-PCR from initial tailbud stages C . robusta RNA . Forward primer: CAGAAAAAATGGGATCGTGGATAAAAGGA , reverse primer: TTATCTCCCAGCAGAACCAGTG ) and Wnt5 [78] ( clone cien109569 ) . Whole mount in situ hybridization and X-gal staining ( detection of β-galactosidase activity following p12xTcf>nlsLacZ electroporation ) were performed as previously described [26 , 79] . Dig-labeled probes were synthesized from C . robusta clones described in previous publications [26 , 80 , 81] , obtained from cDNA libraries [78 , 82] or generated by cloning RT-PCR products ( from initial tailbud stages embryonic RNA ) into pGEM-T Easy ( Promega ) ( S2 Table ) . Effects on gene expression were analyzed for each marker on 15–40 embryos for inhibitor treatments and 40–70 electroporated embryos ( the number of independent experiments is indicated in the figure legend ) . Embryos treated with DMSO or electroporated with pFog>Venus were used as controls . Colorimetric detection of endogenous alkaline phosphatase activity was adapted from [13]: embryos were fixed 10 min at room temperature in sea water containing 5% formaldehyde , washed twice 10 min in TMNTw ( 100 mM NaCl , 50 mM MgCl2 , 100 mM Tris pH 9 . 5 , 0 . 1% Tween20 ) and stained in TMNTw containing 3 . 3 μl/ml of NBT ( 50 mg/ml ) and 1 . 75 μl/ml of BCIP ( 50 mg/ml ) . All pictures were taken from embryos in PBTw using a Zeiss Discovery V20 dissecting scope equipped with an AxioCam ERc5s digital camera . Image panels and figures were constructed with Adobe Photoshop and Adobe Illustrator . The genes described in this study are represented by the following gene models in the KH2012 C . robusta assembly: genes whose expression has been analyzed by in situ hybridization ( see S2 Table ) , Fog ( KH . C10 . 574 ) , Ft ( KH . C11 . 299 ) , Ef1α ( KH . C14 . 52 ) , Noggin ( KH . C12 . 562 ) , Admp ( KH . C2 . 421 ) , β-catenin ( KH . C9 . 53 ) Tcf ( KH . C6 . 71 ) , Sfrp1/5 ( KH . L171 . 5 ) , and Wnt5 ( KH . L152 . 45 ) . | The Wnt/β-catenin pathway is a system of cell-cell communication . It has an ancient origin in animals and plays multiple roles during embryogenesis and adult life . In particular , it is involved in determining , in the vertebrate embryo , the identity of the different parts of the body and their relative positions along the antero-posterior axis . We have investigated in an ascidian ( or sea squirt ) species , a marine invertebrate that is closely related to vertebrates , whether this pathway had a similar role . Like in vertebrates , activating Wnt/β-catenin led to a posteriorization of the embryo with a loss of anterior structures . By contrast , unlike vertebrates , ascidian embryos formed rather normally following Wnt/β-catenin inactivation . Since hemichordates ( or acorn worms ) , earlier divergent invertebrates , use Wnt/β-catenin in a manner comparable to vertebrates , it is in the ascidian lineage that changes have occurred . Consequently , ascidians build an antero-posterior axis , very similarly organized to that of vertebrates , but in a different way . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"skin",
"medicine",
"and",
"health",
"sciences",
"integumentary",
"system",
"in",
"situ",
"hybridization",
"molecular",
"probe",
"techniques",
"nervous",
"system",
"vertebrates",
"animals",
"developmental",
"biology",
"epidermis",
"molecular",
"biology",
"techniques",
"embryos",
"sea",
"squirts",
"research",
"and",
"analysis",
"methods",
"ectoderm",
"embryology",
"probe",
"hybridization",
"molecular",
"biology",
"signal",
"transduction",
"eukaryota",
"anatomy",
"cell",
"biology",
"central",
"nervous",
"system",
"biology",
"and",
"life",
"sciences",
"wnt",
"signaling",
"cascade",
"cell",
"signaling",
"organisms",
"signaling",
"cascades"
] | 2019 | Antero-posterior ectoderm patterning by canonical Wnt signaling during ascidian development |
Networks are a way to represent interactions among one ( e . g . , social networks ) or more ( e . g . , plant-pollinator networks ) classes of nodes . The ability to predict likely , but unobserved , interactions has generated a great deal of interest , and is sometimes referred to as the link prediction problem . However , most studies of link prediction have focused on social networks , and have assumed a completely censused network . In biological networks , it is unlikely that all interactions are censused , and ignoring incomplete detection of interactions may lead to biased or incorrect conclusions . Previous attempts to predict network interactions have relied on known properties of network structure , making the approach sensitive to observation errors . This is an obvious shortcoming , as networks are dynamic , and sometimes not well sampled , leading to incomplete detection of links . Here , we develop an algorithm to predict missing links based on conditional probability estimation and associated , node-level features . We validate this algorithm on simulated data , and then apply it to a desert small mammal host-parasite network . Our approach achieves high accuracy on simulated and observed data , providing a simple method to accurately predict missing links in networks without relying on prior knowledge about network structure .
Complex interactions between host and parasite species can be described as a network , with host and parasite species as two distinct node types connected by links that represent associations between a given parasite and host species . Understanding the structure [1] and stability [2 , 3] of host-parasite networks is important for establishing drivers of host-parasite interactions , parasite specificity , and the consequences of host extinctions on parasite diversity . Recently , authors have applied concepts and tools from community ecology and graph theory to host-parasite interactions [4–7] in an effort to understand how host and parasite communities interact , including investigations into how host community diversity influences disease transmission [8] , how parasites interact within infected hosts [9] , and how host functional and phylogenetic similarity promote parasite sharing [10 , 11] . Additional research has focused on topological measures of host-parasite networks—such as nestedness [12] and modularity [13]—which attempt to quantify the formation of patterns of interactions between host and parasite species . These patterns may influence network stability [2] and resilience [3] . Identifying the factors influencing the formation of these patterns is an important nascent area of research . There is little consensus about whether various reported topological patterns are common [14–16] , which may be a result of the influence of sampling effort and the effect of incomplete detection on measures of topological network structure [17] . Specifically , the detection of patterns in most studies is predicated on having completely sampled the network of host-parasite interactions . That is , all interactions between host and parasite species are assumed to have been documented in the course of the study . However , such exhaustive sampling is rare at best , as logistical constraints often limit detection of all interactions . Moreover , the total number of potential host-parasite interactions increases as a product of the number of host and parasite species , creating a large number of opportunities for a missed detection of a host-parasite interaction . It is unlikely that studies of ecological networks are recording all of the potential interactions between species , as even long term data have been unable to detect a large number ( nearly 50% of plant-pollinator interactions ) of species interactions [18] . Incomplete sampling compromises inference of network structure and stability , and may undermine studies of parasite specificity and measures of parasite species richness for a given host species . Despite this complication , there is a body of research aimed at predicting host-parasite interactions . This work is of clear importance to wildlife and human health—as the it is possible to identify potential spillover events [19–21]—and to a general understanding of the traits associated with parasite specialization . To this end , current approaches examine parasite species independent of the network within which they are embedded , using host traits to predict likely interactions . Two such efforts attempted to predict the fish host community parasitized by helminth parasites [22 , 23] . However , approaches to date have not explicitly considered how the distribution of host and parasite traits , or the complex interactions at the host-parasite network level could influence predictability of host-parasite interactions . By considering all potential interactions simultaneously , it is possible to find the most probable interactions given the entire network , rooting the problem of predicting likely host-parasite interactions within a body of theory from the study of complex networks [24 , 25] . Here , we address this problem by developing and testing a method capable of determining the number of likely unobserved host-parasite interactions , and accurately predicting the most likely , but undetected , host-parasite interactions in the network . This is not a new problem , as computer scientists have struggled with the link prediction problem for decades , most notably in studies of social networks [26–28] . We focus on link prediction in bipartite networks , with a specific application to ecological networks . Previous work in link prediction for bipartite networks has required information on traits of both node classes ( e . g . , host and parasite species ) , as well as knowledge of network topology ( e . g . , degree distribution ) [29] . Here , we develop a highly accurate link prediction method based on trait matching between host and parasite species . That is , we make no assumption about network topology , but predict bipartite interactions using only trait data on host and parasite species . We examine the performance of our algorithm on simulated data extensively , and then test the algorithm on an ecological host-parasite network of small mammals and their resident parasite communities in a New Mexican desert ecosystem .
We propose an approach to identifying cryptic associations in host-parasite networks based on numerical estimation of conditional density functions . We represent the connections between hosts and parasites as a sparse bipartite graph ( H , P , E ) with vertex sets H ( host species ) and P ( parasite species ) and edges E , such that an edge connects Hi and Pj if species j parasitizes species i . If there is an edge between Hi and Pj , we write yi , j = 1 whether the edge has been observed or not; otherwise yi , j = 0 . Not all edges have been observed and not all possible edges exist . Thus , E consists of both observed edges Eo and unobserved edges Eu = E∖Eo and is itself a subset of the possible edges E ˜ = H × P . Attached to each host and parasite species are vectors of features h and p , respectively . Thus , edge ( Hi , Pj ) has the combined feature set xi , j = ( hi , pj ) . To identify cryptic links in Eu , we seek a ranking of edges according to their probability . The probability that there is an edge between two vertices given its feature set is written P ( y = 1|x ) . From Bayes’ theorem , we have P ( y = 1 | x ) = f 1 ( x ) P ( y = 1 ) f ( x ) where f1 is the conditional probability of feature set xi , j given that yi , j = 1 , P ( y = 1 ) is the connectance of the graph , and f is the density of all possible combined feature sets . That is , f1 is the probability density of features when a link exists between host and parasite , and f is the density of features for all possible host-parasite combinations . The model assumes that the observation process ( probability of detection ) is either constant or random with respect to host and parasite features . Extensions of this model could address this assumption through the incorporation of features related to sampling probabilities or the use of model simulations directly incorporating the observation process . Since we seek only a rank ordering , we ignore P ( y = 1 ) which is simply a normalizing constant , and estimate q = f1/f . Estimating q is a density-ratio estimation problem [30] . The plug-in approach we propose , which we call plug-and-play , is to separately estimate f1 and f from the features of Eo and E ˜ and to take the quotient as required for evaluating any given host-parasite pair , i . e . , q ^ = f 1 ^ / f ^ . In practice , we use the kernel density estimator npudens in the np package [31] and the “normal-reference” bandwidth . This nonparametric approach to density-ratio estimation was chosen because it generally performs very well , particularly when the feature set contains a combination of binary and continuous features [32] . The estimated probabilities of all edges in E ˜ \ E o are then evaluated and ordered . That is , the model outputs the probability of each edge E ˜ \ E o , which can then be ranked by the most probable undetected edge in the set of cryptic links Eu . The AUC ( area under the receiver operating characteristic ) statistic can be calculated by comparing the observed labels and the estimated probabilities . If probabilities need to be translated into binary states , we begin with the most likely cryptic link , and re-label unobserved edges in order until a stopping criterion is met . Host-parasite networks were simulated as follows . First , we generated a number ( typically n = 5 ) trait values for both host and parasite species by drawing random numbers from a beta distribution , with the two shape parameters ( α and β ) drawn from a uniform distribution bounded between 0 . 5 and 1 . 5 . The beta distribution was chosen for its flexibility and generality to many ecological and epidemiological problems [33 , 34] , as it is bound between 0 and 1 , can take a variety of shapes , and is easily extensible ( e . g . , beta-binomial modeling; [35] ) . Then , the probability that host i interacts with parasite j was given as the outer product of host h and parasite p trait vectors , calculated as the row-wise product of host and parasite trait matrices , where rows correspond to either host or parasite species and columns are traits . This forms a matrix of h rows and p columns . This matrix ( M ) was scaled to the unit interval by dividing each value by the maximum value observed . Interactions were assigned probabilistically by conducting single binomial trials with probability Mi , j . This process was performed iteratively until a specified connectance value was reached ( c = c* ) . h = [ h 1 , h 2 , … h i ] p = [ p 1 , p 2 , … p j] M = h × p while ( c < c* ) M i , j = 1 if M i , j > U ( 0 , 1 ) - if M i , j < U ( 0 , 1 ) To determine how well the plug-and-play model performed , we tested the predictive accuracy of the model on simulated data . We trained models on 80% of the simulated data , and predicted on the remaining 20% test set , i . e . , a setup that assumes only 80% of host-parasite associations to have been sampled . ( This criterion is relaxed in the Supplemental Materials where we show how the fraction of the network used for model training influenced predictive accuracy; S1 Fig ) . The AUC statistic was uesd as a measure of predictive accuracy , and examined how model performance was influenced by interaction matrix size , the fraction of realized links ( i . e . , connectance ) , the number of traits used to predict species interactions , and the inclusion of binary ( e . g . , thresholded at the mean ) and uninformative ( e . g . , standard normal variates ) traits ( see Supplemental Materials for more information ) . Unless otherwise stated , species interaction matrices were created and predicted using five host and parasite traits each , and a connectance ( c ) of 0 . 2 , which reflects observations of empirical host-parasite networks [36] . First , we determined the predictive accuracy of our model on 1000 randomly generated species interaction networks . To examine the influence of interaction matrix size , we varied host and parasite species richness from 10 to 30 , and simulated 50 networks for each host and parasite richness combination . The influence of connectance was examined by creating 1000 species interaction networks with 30 host species and 20 parasite species for each value along a gradient of connectance values from 0 . 05 to 0 . 35 . To examine the influence of host and parasite trait number , we simulated 1000 species interaction networks for each host and parasite trait number combination between 1 and 20 ( total of 20 , 000 networks ) . The influence of training the model on binary trait data was examined by creating 1000 species interaction networks created using 20 host and parasite traits , and varying the fraction of those 20 traits that were binary from 5% ( 1 trait was binary ) to 100% ( all traits were binary ) . To determine if the inclusion of random , uninformative traits influenced predictive power , we simulated 1000 species interactions networks with 10 host and parasite traits , and included between 1 and 50 random host and parasite traits ( 50 , 000 total species interaction networks ) . Lastly , we tested predictive accuracy when the model was trained only on random traits by creating species interaction matrices ( 1000 per treatment ) and then shuffling trait values . The plug-and-play model was able to predict links on simulated bipartite networks with high accuracy ( S2 Fig ) . Further , accuracy was not appreciably reduced by matrix size ( S3 Fig ) , incorporation of binary variables ( S4 Fig ) , number of host and parasite traits ( S5 Fig ) , connectance ( S6 Fig ) , or the incorporation of random variables ( S7 and S8 Figs ) . Specifically , we found that more than three host and parasite traits were needed to have a mean AUC value of 0 . 9 , and training on only a single host and parasite trait resulted in moderate predictive accuracy ( A U C ¯ = 0 . 72 ) . We applied the plug-and-play algorithm to data on parasites of small mammals sampled as part of the Sevilleta Long-Term Ecological Research project . We aggregated data from 1992 to 1997 from six sites in three nearby habitats into one interaction matrix . Details of animal sampling and processing are reported elsewhere [4 , 37] . Hosts with fewer than five captures over the six year sampling effort were excluded from analysis , resulting in a total of 22 small mammal host species and 87 parasite species , including both macroparasites ( e . g . , helminths ) and microparasites ( e . g . , coccidians ) . Host trait data were obtained from Pantheria [38] , supplemented with published literature sources ( see Supplemental Table A1 of [4] for more information ) . Host trait data included life history traits ( Table 1 ) , and phylogenetic information . Phylogenetic relationships were estimated using the first five axes of a principal coordinates analysis ( PCoA ) on the phylogenetic distance matrix obtained using the mammal supertree [39] and the ape R package [40] . Together , these first five PCoA vectors captured 95% of the variance in the eigenvalues , suggesting that most of the information in the phylogeny was captured in these five vectors . Host life history traits included host diet breadth , body mass , home range size , maximum age , and species abundance ( Table 1 ) . Parasite trait data included three variables representing the life history and transmission modes of parasites; parasite type ( arthropod , protozoan , or helminth ) , parasite genus ( genus ) , and location ( intracellular or extracellular ) . Some host trait data was unavailable , and we imputed the unavailable data using the randomForest R package [41] . This procedure imputes missing data by first replacing missing values with column averages , and then iteratively updating imputed values based on proximity of observations to one another in the random forest model . Variable importance was determined by permuting each predictor variable 500 times , and determining the reduction in model performance as a result for each permutation . Model accuracy ( AUC ) was determined through 5-fold cross validation . The final model was trained on all available data . We then determined the number of likely missing links from the host-parasite network , and sequentially added the most likely links as predicted by our trained model . We used the Abundance-based Coverage Estimator ( ACE; [42] ) , commonly used for species richness estimation , to estimate the number of missing links . ACE is a non-parametric species richness estimator typically applied to communities of free-living organisms ( [43 , 44] ) and has previously been demonstrated to perform well for many different coverage levels and survey designs ( [45] ) . We treat links between known hosts and parasites to be equivalent to organisms in the traditional context , which allows us to estimate the likely number of links missing from the network . At each link addition , we calculated properties of the network to observe how network structure changed with link addition . Some stuctural properties change obviously and deterministically with link addition ( e . g . , mean degree and connectance ) , which we ignore . Rather , we focused on stochastic aspects of network structure , including measures previously related to network stability ( nestedness; [3 , 14] ) , aggegration of parasite species among host species ( togetherness and variance-to-mean ratio; [46] ) , and measures of interaction clustering or host-parasite co-occurrence ( C-score; [47] ) . The resulting changes to network metrics with model-predicted link addition were compared with changes in network metrics if links were added randomly . Nestedness , quantified as the NODF metric [48] , measures the tendency of hosts with few parasites to harbor nested subsets of the parasite communities of parasite species-rich hosts , and has previously been related to network structural stability [3] . Nestedness was quantified relative a null model , as aspects of matrix size and fill alter the raw measure . Further , the use of the standard score ( z-score ) allows a quantification of the magnitude of divergence from a null expectation , which is commonly used for significance testing . Thus , this approach allows us to determine changes in the magnitude of nestedness with link addition relative to a null expectation . We used the sequential swap algorithm to randomize matrix interactions [49] , and compared the empirical network to 1000 null networks after each link addition . Togetherness measures the tendency of host species to share parasites , with large values suggesting ecological similarity between hosts may be more important than competition in driving community structure , and small values suggesting the opposite ( [12 , 50] ) . The variance-to-mean ratio is an index of aggregation traditionally used in studies of single species parasite distributions [46 , 51] , where larger values indicate more skewed or aggregated parasite burdens . Here , we use it to express the skew in parasite species richness for a range of host species . Originally used to infer interspecific competition , the C-score ( or checkerboard score; calculated here as the mean pairwise score for all host species ) is more generally a measure of non-independence in species interaction patterns , with large values indicating that species occupy different habitats ( [47] ) . These interaction differences could be a result of interspecific competition , dispersal limitation , or differences in host habitat utilization . In terms of host-parasite networks , this would correspond to parasite communities with little overlap in host use , such that parasite communities are clumped across the range of potential host species .
The plug-and-play algorithm recovered the Sevilleta small mammal-parasite interaction network structure with high accuracy ( AUC = 0 . 82 ) when trained on all available data , and performed fairly well during 5-fold cross validation , with a mean AUC from 500 training/test data splits of 0 . 63 , and a maximum observed AUC of 0 . 81 . We permuted predictor variables to obtain measures of variable importance , which suggested that host litter size , parasite genus , and host diet breadth were the most important variables to model performance ( Fig 1 ) . Meanwhile , some covariates had a negative effect on the model , resulting in improvement in predictive accuracy with randomization . These included coarse , low-variance variables such as habitat breadth and trophic status , as well as potentially important variables such as parasite type ( e . g . , helminths ) , and host body mass . Predictive model accuracy is predicated on the network being fully sampled , such that predicted links that are not observed in the empirical network are treated as errors , and reduce accuracy . We predicted that between 110 and 157 links were missing from the empirical network , changing the connectance from 0 . 12 to between 0 . 18 and 0 . 21 . We then sequentially added the most probable links , based on model-predicted suitability scores ( Fig 2 ) , plug-and-play examine how network properties changed . Measures of network structure fluctuated with link additions ( Fig 3 ) . Specifically , nestedness , quantified as the z-score in NODF values relative to null models , fluctuated from -4 . 6 to -0 . 6 . Since these z-scores can be used for significance testing , this suggests that the addition of missing links can change the ability to detect fundamental network properties . Further , togetherness , variance-to-mean ratio , and C-score all declined more strongly with the addition of predicted missing links compared to the addition of random links . Further , togetherness actually increased when link addition was random .
Here , we present , validate , and test a link prediction algorithm that does not require information on network structure for training , extending the problem of link prediction in social networks to bipartite networks . This is important , as network structure is often dynamic , and generalizing link prediction to novel or changing networks is necessary for some applications ( e . g . , forecasting the most probable prey items or parasites of a novel host species to the network ) . Our approach allows for the ranking of node characteristics , which can enhance our understanding of what determines the likelihood of species interactions , and for the prediction of cryptic interactions , which can influence network structure . In our small mammal-parasite network , we determined that host litter size , parasite genus , and host diet breadth were the top three most important predictors of host-parasite interactions . Host litter size was the most important interaction predictor , suggesting the importance of host life history traits . Because host litter size is linked to other aspects of host biology known to alter parasite burdens , such as host metabolic rate [52] , we suspect that the importance of litter size in this analysis may reflect an aspect of the host species’ pace of life [53 , 54] . The second most important variable to our predictive model was parasite type ( i . e . , arthropod , helminth , or protozoa ) , which accounts for unmeasured differences among parasite species in their transmission or host preferences . Lastly , host habitat breadth , which can influence contact rates with parasites was an important variable in our model . Interestingly , despite the previously documented importance of host phylogenetic distance in predicting parasite community similarity [10] , we found no evidence that host phylogeny improved predictive accuracy in this system . The inclusion of some covariates actively detracted from model performance , a phenomenon not observed in simulated data . This is likely a result of the low information content of these variables , or could signal the influence of variable interactions on model predictive accuracy . Our algorithm predicted that between 110 and 157 links were missing from the network . When these links were added based on their suitability score , several network properties changed , including nestedness , togetherness , variance-to-mean ratio , and checkerboard score . While the ability to detect nestedness fluctuated with link addition , the other three metrics of network interaction patterns demonstrated consistent declining trends . This suggests that the interaction patterns became less clumped ( as indicated by the checkerboard score ) , parasite communities became less dissimilar ( as indicated by togetherness ) , and less aggregated ( as indicated by variance-to-mean ratio ) . Taken together , these findings suggest link addition was not confined to species that already had many links , otherwise the variance-to-mean ratio wouldn’t have been reduced . Instead , the addition of missing links reduced overdispersion commonly observed in many host-parasite networks ( including in Fig 2 ) . Ecologists have long recognized the issue of incomplete sampling leading to imperfect detection [55] , but only recently have studies of ecological networks addressed this issue [2 , 17 , 56] . Here , we present an algorithm capable of accurately reconstructing a network using information on interactor traits , and predicting interaction likelihoods . This overcomes the problem of imperfect detection , and allows for the forecasting of the most probable links in ecological networks . Other approaches for the link prediction problem in bipartite networks exist . For instance , recent Bayesian approaches have used occupancy models [17] and Dirichlet network distributions [57] . However , these approaches are largely used to address slightly different problems . The first is an attempt to combine occupancy models with metacommunity analysis , predicting missing links as a means to correct error , and not for the sake of predicting unknown links . The second was developed to predict links in integer-based directed networks , and was developed under the assumption that nodes have repeated and directed interactions , such as a network of email correspondence among a group of people . Extensions of this approach could potentially support binary bipartite networks as we have examined . Another approach , the matching-centrality method [29] , allows for the accurate forecasting of unobserved links in both unipartite and bipartite networks . Our approach differs in that we rely solely on trait matching between bipartite interactors to predict interaction probability , meaning that the algorithm is insensitive to network structure ( allowing for increased flexibility ) . Lastly , by relying on host and parasite traits , our approach may provide insight into what host traits , parasite traits , or trait combinations promote the likelihood of a host-parasite interaction , and further provides a way to quantify the relative importance of host and parasite traits to interaction patterns . Extensions of our current approach could disentangle the effect of disproportionate sampling effort , as well as other host and parasite traits , to provide a more complete understanding of what controls host-parasite interactions . This trait-based approach can be applied to other bipartite networks ( e . g . , plant-pollinator ) , as well as to spatial networks ( e . g . , metapopulations ) . The incorporation of missing links into networks that change seasonally or are logistically difficult to sample provides a more accurate description of network interactions . Further , the incorporation of these interactions may change basic network properties in non-random ways . The functional consequences for revising our understanding of ecological networks are not currently known . | The majority of host-parasite associations are poorly understood or not known at all because the number of associations is so vast . Further , interactions may shift seasonally , or as a function of changing host densities . Consequently , host-parasite networks may be poorly characterized since effects of cryptic host-parasite associations on network structure are unknown . To address this , we developed theory and applied it to empirical data to test the ability of a simple algorithm to predict interactions between hosts and parasites . The algorithm uses host and parasite trait data to train predictive probabilistic models of host-parasite interaction . We tested the accuracy of our approach using simulated networks that vary greatly in their properties , demonstrating high accuracy and robustness . We then applied this algorithm to data on a small mammal host-parasite network , estimated model accuracy , identified host and parasite traits important to prediction , and quantified expected changes to structural properties of the network as a result of link relabeling . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"medicine",
"and",
"health",
"sciences",
"ecology",
"and",
"environmental",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"sociology",
"social",
"sciences",
"parasitic",
"diseases",
"simulation",
"and",
"modeling",
"mathematics",
"forecasting",
"statistics",
"(mathematics)",
"network",
"analysis",
"social",
"networks",
"research",
"and",
"analysis",
"methods",
"computer",
"and",
"information",
"sciences",
"parasitism",
"mathematical",
"and",
"statistical",
"techniques",
"pathogenesis",
"community",
"ecology",
"ecology",
"host-pathogen",
"interactions",
"trophic",
"interactions",
"biology",
"and",
"life",
"sciences",
"species",
"interactions",
"physical",
"sciences",
"statistical",
"methods"
] | 2017 | Predicting cryptic links in host-parasite networks |
Schistosoma mansoni infection is proven to be a major health problem of preschool-age children in sub-Saharan Africa , yet this age category is not part of the schistosomiasis control program . The objective of this study was to compare the impact of single and double dose praziquantel ( PZQ ) treatment on cure rates ( CRs ) , egg reduction rates ( ERRs ) and re-infection rates 8 months later , in children aged 1-5 years living along Lake Victoria , Uganda . Infected children ( n= 1017 ) were randomized to receive either a single or double dose of PZQ . Initially all children were treated with a single standard oral dose 40 mg/kg body weight of PZQ . Two weeks later a second dose was administered to children in the double dose treatment arm . Side effects were monitored at 30 minutes to 24 hours after each treatment . Efficacy in terms of CRs and ERRs for the two treatments was assessed and compared 1 month after the second treatment . Re-infection with S . mansoni was assessed in the same children 8 months following the second treatment . CRs were non-significantly higher in children treated with two 40 mg/kg PZQ doses ( 85 . 5%; 290/339 ) compared to a single dose ( 83 . 2%; 297/357 ) . ERRs were significantly higher in the double dose with 99 . 3 ( 95%CI: 99 . 2-99 . 5 ) compared with 98 . 9 ( 95%CI: 98 . 7-99 . 1 ) using a single dose , ( P = 0 . 01 ) . Side effects occurred more frequently during the first round of drug administration and were mild and short-lived; these included vomiting , abdominal pain and bloody diarrhea . Overall re-infection rate 8 months post treatment was 44 . 5% . PZQ is efficacious and relatively safe to use in preschool-age children but there is still an unmet need to improve its formulation to suit small children . Two PZQ doses lead to significant reduction in egg excretion compared to a single dose . Re-infection rates with S . mansoni 8 months post treatment is the same among children irrespective of the treatment regimen .
Schistosomiasis is a parasitic water borne neglected tropical disease [1 , 2] of considerable public health relevance in the tropics and subtropics [3 , 4] . It is estimated that 200 million people are infected with schistosomiasis worldwide with 85% of the disease burden found in Sub-Saharan Africa [5] . In Uganda , approximately 20 million people are at risk of being infected with intestinal schistosomiasis caused by S . mansoni and 4 million individuals are estimated to be infected [6] . Currently , the focus of schistosomiasis control by preventive chemotherapy is on school-age children of 6–19 years as this category has the highest risk of infection [7 , 8] . Mass drug administration ( MDA ) with PZQ for schistosomiasis is the main approach adopted by the Uganda national schistosomiasis control program to reduce related morbidity in school-age children and adults . A number of studies on children under 6 years of age have , however , revealed that these children are at higher risk of schistosomiasis infection than previously thought [9–12] . Young children living on lakeshores or irrigated land actively get infected with schistosome parasites usually through bathing , playing or swimming in schistosome-infested waters . These children may also get exposed passively to infective water when bathed with lake water , which are carried back home [10] . Preschool-age children are not yet targeted in schistosomiasis preventive chemotherapy campaigns . The exclusion of preschool-age children ( <6 years of age ) from mass treatment programmes for control of schistosomiasis has been justified by the limited documentation on the safety of PZQ in this age group [13 , 14] . Although PZQ is recommended for individuals aged 4 years and above [8] , there is no suitable formulation of PZQ for young children [15] . Meanwhile , studies conducted in some schistosomiasis endemic countries ( Mali , Niger , Sudan , Uganda and Zimbabwe ) have documented that it is safe and efficacious to treat children ( <6 years ) with PZQ [16 , 17] . Based on the same findings , it has been recommended that young children who live in schistosomiasis endemic areas should be considered for treatment with PZQ during childhood [16] . PZQ is a large white tablet containing 600mg of active ingredient [18] and is , despite its large size and bitter taste , the drug still being used to treat human schistosomiasis on a large scale [19–21] . The inactive distomer ( S-PZQ ) contributes to the bitter taste and doubles the size of the tablets [21] . It is the drug of choice because it is effective on all schistosome species , cheap , safe and has minimal side effects [20 , 22] . The standard recommended dose is a single oral treatment of 40 mg/kg body weight [8 , 23 , 24] sufficient to achieve CRs of 60–90% [25 , 26] . Upon oral intake , PZQ penetrates the tegument and rapidly moves through the tissues of schistosomes causing muscle contraction and damage [25 , 27] . PZQ attacks only the mature schistosome worms but not the immature stages [20] and has therefore no effect on recent infections [28 , 29] . Consequently it is recommended to administer a second dose of PZQ two weeks after a first dose in order to kill the worms which were immature during the first treatment and hence , increase CRs . This study was designed to compare the efficacy of PZQ in terms of CRs and ERRs using single and double dose regimens and its effect on S . mansoni re-infection 8 months post treatment in children aged 1–5 years living along Lake Victoria in eastern Uganda .
The study was carried out in 33 sites ( 31 shoreline communities and 2 islands ) along Lake Victoria in eastern Uganda ( Fig 1 ) ; where different epidemiological studies reported high prevalence of S . mansoni infections [11 , 30–33] . Malacological surveys have also confirmed that Biomphalaria choanomphala , one of the main freshwater snail intermediate host of S . mansoni , exists in high numbers along Lake Victoria shorelines in Uganda [34] . The major activities carried out by these communities are fishing and subsistence farming . The lake is the major source of water for domestic use and the communities are characterized by poor sanitation . We designed a randomized clinical trial ( ClinicalTrial . gov , identifier: NCT01901484 ) with two treatment arms ( single versus spaced double dose ) of PZQ with follow-up assessments at month 1 and 8 months post treatment ( Fig 2 ) . Following a baseline parasitological survey , all the children tested positive for S . mansoni eggs and who provided three stool specimens and informed consent from their parents were recruited into the study . Recruited children were then individually randomized ( by an independent statistician ) to single and double dose PZQ treatment groups by computer random number generation in Stata/IC 12 . 0 . Baseline characteristics of the two treatment groups are summarized in Table 1 . Age , sex , distribution of intensity levels ( light , moderate and heavy ) and sampled sites were comparable between the two treatment groups . Children were orally administered a single dose of PZQ ( 40 mg/kg of body weight ) at baseline and another similar dose two weeks later to those in the double dose treatment group . Eligibility was limited to preschool-age-aged children ( 1–5 years ) . Follow-up surveys were carried out 1 month and 8 months after the second treatment with PZQ to assess efficacy and re-infection , respectively . The study compared the efficacy of two different PZQ treatment strategies; single and spaced double dose . It was assumed that there would be a significant difference in the number of children cured between the two treatments after 1 month . It was expected that 1 month after treatment parasite infection would be reduced to 30% and 20% with the single dose and double dose , respectively [35 , 36] . Using the formula of comparison of two proportions [37] with 90% power and considering a 10% difference between the two proportions to be at the 95% significance level , the total sample size calculated is n = 784 children . Each intervention group will therefore be n = 392 children . In order to account for a 20% loss to follow-up; the sample was increased to 490 children in each treatment arm . Ethical approval and clearance were obtained from Makerere University , Ethics Committee and the Uganda National Council for Science and Technology ( Ref . No . UNCST-HS1274 ) , respectively . The trial was registered with ClinicalTrial . gov , identifier: NCT01901484 . The aim and procedure of the study were explained to the parents/guardians of the recruited children in the local language ( Lusoga ) , and consent obtained before PZQ chemotherapy . Participation was voluntary and the parents/guardians had the right to withdraw their child/children at any time point from the study . Treatment was administered by trained medical personnel blinded to the children’s intervention group in the presence of each child’s caretaker . Children were closely monitored by trained and authorized staff for 30 minutes after drug administration and the caretakers were asked to report unusual behavior of their children within 24 hours . In case of vomiting and diarrhea , sachets of oral rehydration salt were given to the caretakers together with advice on how to dissolve the salt and administer the solution to the children . Participant children were asked to provide one stool sample on three consecutive days . Multiple stool collections were proposed due to day-to-day variation in egg counts of S . mansoni [38 , 39] . Two slides of 41 . 7mg Kato-Katz [40] thick smears were prepared from each stool specimen and examined under a microscope ( 10x ) to determine S . mansoni egg counts . For quality control , 10% of the prepared slides were randomly selected and re-examined by an independent expert microscopist . In case of discrepancies , the slides were reread and consensus obtained . Three stool specimens were collected again from each child both at a follow-up of 1-month and 8 months after the second treatment to determine PZQ efficacy and re-infection rates , respectively . Similar parasitological procedures used at baseline were applied in the follow-up surveys . Following parasitological screening , egg positive children were randomly assigned to two treatment groups; one receiving a single and the other a double dose . Children were weighed using a portable digital scale ( Seca 771 , Seca GmbH , Hamburg , Germany; accuracy: 0 . 1 kg ) and offered a standard oral dose of PZQ ( 40 mg/kg body weight ) . The brand used was PZQ CIPLA 600 mg ( Kundaim , India ) . The required dose can be administered relatively accurately because each PZQ tablet ( 600mg ) is scored with 3 lines and can thus be subdivided into 4 equal parts of 150 mg [41] . The tablets were crushed [42] and mixed with drinking water to facilitate oral uptake in small children [15] . Before drug administration , children were given a piece of bread , and orange juice was provided after drug administration to minimize gastrointestinal side effects [43–45] and mask the bitter taste of PZQ [19] , respectively . Two weeks after the first treatment , children belonging to the double dose group were given a second dose of PZQ ( 40 mg/kg of body weight ) . In case of vomiting , the children were offered a second dose . Treated children remained under supervision for a period of 30 minutes to monitor any immediate reactions as a result of drug administration and the caregivers were further encouraged to report any treatment-related symptoms observed in the children 24 hours post-treatment . In case of adverse events , the affected children would be referred to the nearest local health services . Data were double entered and cross-checked using EpiData version 3 . 1 ( The EpiData Association , Odense , Denmark ) and Microsoft Excel spread sheet software . Statistical analysis was carried out in Stata/IC release 12 . 0 ( StataCorp; College Station , TX , USA ) . Intensity of infection ( expressed in eggs per gram of stool , EPG ) was calculated by multiplying the mean for the six slides by a factor of 24 . Infection intensities were classified according to WHO guidelines [8]; as light: 1-99EPG , moderate:100-399EPG and heavy: ≥400EPG . The effects of treatment on geometric mean intensity ( GMI ) of infection were assessed by determining the change in GMI egg counts among all the children at 1 month post treatment , irrespective of whether they remained egg positive or became egg free . At-test for log transformed EPG at baseline and 1 month post treatment was used . Test of proportions of interest were calculated and comparisons made using the Pearson chi-square ( χ2 ) test . The parasitological CRs for the two treatment regimens were calculated as the proportion of children with no egg excretion after treatment among those with eggs in their stool at baseline . CRs obtained with the two treatment regimens were compared using χ2 . ERRs were determined by comparing the GMI egg output at baseline and 1 month after the second treatment ( 1–[GMI after treatment/GMI before treatment] ) x 100 ) . The proportions and 95% confidence intervals ( 95% CI ) of children who were egg-negative 1 month post treatment but became positive at the 8 month follow-up were generated to compare re-infection between children who had received single and double dose treatments . Univariate and multivariate logistic regression analyses were used to assess relationships of sex , age group , intensity of infection and treatment dose with CR of S . mansoni infections using χ2 test . Adjusted odd ratios at 95% confidence interval were used [46] . P-values <0 . 05 were considered to indicate statistical significance .
During the baseline parasitological survey , 1 , 203 children were found infected with S . mansoni , but only 1 , 017 were randomized to treatment due to lack of informed consent or less than three stool specimens in 186 children ( Fig 2 ) . Before treatment another 292 were lost resulting in 725 receiving treatment ( 375 and 350 in the single and double dose group , respectively ) . The children that remained for PZQ treatment were comparable to the children lost to follow-up with respect to age , sex , and S . mansoni intensity of infection ( Table 2 ) . At the 1 month follow-up , 696 children were investigated for infection , 357 from the single dose group and 339 from the double dose group . The final cohort at 8 months follow-up consisted of 463 children; 236 and 227 children in the single and double dose group , respectively ( Fig 2 ) . A total of 554 children did not complete the study due to refusal of caregivers , absenteeism on the treatment/survey day or inaccessibility of habitation during the rainy season . Table 3 summarizes CRs of children who received a single and double dose of PZQ ( 40 mg/kg ) . At 1 month post-treatment follow-up , 84 . 3% children ( 587/696 ) were tested egg negative for S . mansoni and therefore considered cured . The results show that more children were cured with a double dose 85 . 5% ( 290/339 ) compared to 83 . 2% ( 297/357 ) cured by a single dose although this difference was not statistically significant ( P = 0 . 39 ) . There was a significant difference ( P = 0 . 01 ) in CRs between double ( 81 . 8% ) and single dose ( 62 . 0% ) interventions of the moderately infected children . Children in the age group 12–24 months were all cured by either a single dose ( 25/25 ) or double dose ( 33/33 ) , while the lowest CRs ( single dose: 75 . 8%; double dose: 80 . 0% ) were observed in the oldest age group of 49–60 months . Table 4 shows the results of the univariate and multivariate regression analyses on the relationships of CR with sex , age group , intensity of infection and treatment dose . Children had equal chances of being cured by either one or two doses of praziquantel ( OR: 1 . 6; 95% 0 . 9–2 . 5; P = 0 . 07 ) . There was no difference in CRs between girls and boys in any of the treatment arms . Older children , 49–60 months old , were more likely to remain infected after treatment with praziquantel ( OR: 0 . 3; 95% CI 0 . 2–0 . 6; P <0 . 01 ) than their younger counter parts . Children with moderate and heavy intensity infections at baseline were more difficult to cure from infection ( OR: 0 . 1; 95% CI 0 . 1–0 . 2; P<0 . 01 ) than those with light-intensity infection . Table 5 shows a comparison of ERRs between single and two doses assessed 1-month after PZQ intervention . The highest ERR of 99 . 3% ( 95% CI:99 . 2–99 . 5 ) was observed in the double dose treatment in which GMI of infection reduced from 105 . 2 EPG at baseline to 0 . 70 EPG at 1 month after chemotherapy . In contrast , ERR in the single dose group was 98 . 9 ( 95% CI: 98 . 7–99 . 1 ) , with S . mansoni GMI reduced from 79 . 5 EPG at baseline to 0 . 86 EPG 1 month after treatment . Difference in rates of egg reduction between single and double dose treatment regimen was significant ( P<0 . 01 ) . Overall intensity of infection at baseline was significantly higher ( P = 0 . 008 ) in children in the double dose treatment arm . There was no significant difference ( P = 0 . 94 ) in intensity of infection among children who received either a single or two praziquantel doses 1 month post treatment . Table 6 shows the prevalence of re-infection in children . A total of 206 out of 463 ( 44 . 5% ) children that were egg negative at 1-month follow-up were tested egg positive for S . mansoni 8 months later . There was no significant difference ( P = 0 . 22 ) in re-infection rate comparing single dose children 92/236 ( 39 . 0%; 95% CI: 32 . 7–45 . 3 ) and double dose children 114/227 ( 50 . 2%; 95% CI: 43 . 7–56 . 8 ) . Re-infection was non-significantly higher in boys than in girls . Re-infection intensity at 8 months post treatment was 4 . 0 EPG ( 95% CI: 2 . 9–5 . 8 ) and 7 . 1 EPG ( 95% CI: 5 . 1–9 . 8 ) for children who received a single and double dose , respectively; significantly higher ( P = 0 . 03 ) among children who received a double dose treatment . Majority of infections at all 3 assessment points were of light intensity; at baseline ( 58% ) , 1-month ( 83% ) and 8 months ( 65% ) follow-ups ( Fig 3 ) . There was a marked reduction in heavy infection after treatment but it rose again to half its original level 8 months later ( Fig 3 ) . Abdominal discomfort and diarrhoea were observed and reported within 30 minutes to 24 hours after drug administration . The symptoms were , however mild and short-lived , many subsided in less than three hours . Abdominal pain and diarrhoea were mostly reported in children aged 3–5 years . These side effects were mainly observed during the first PZQ administration with a single dose and not after administering the second dose . No severe adverse events were reported .
This study compared the efficacy and effect of one versus two spaced standard doses of PZQ ( 40 mg/kg body weight ) in preschool-age children . Generally , the study revealed that PZQ in its current formulation is acceptable and efficacious in treating the young infected children . The overall CR ( 84 . 3% ) and ERR ( 99 . 1% ) obtained in this study are in the range of CRs/ERRs found in earlier trials [25 , 47 , 48] using 40 mg/kg PZQ dose . Although PZQ efficacy obtained in this study is satisfactory , it should be noted , however , that splitting or crushing tablets leads to dose inaccuracy due to reduced drug weight and poor absorption of the drug [49 , 50] . The split and crushed tablets administered to the children in the present study could have lost their actual weight thus reduced dosage leading to therapeutic failure and hence final inaccurate CRs . This entails in not having proper PZQ formulations for children less than 6 years of age [15] . Treatment with PZQ leads to reduction in number of schistosome eggs via reduction in worm loads . This study presents significant ERRs in children who received two spaced standard doses of PZQ treatments ( 99 . 3% ( 95% CI: 99 . 2–99 . 5 ) compared to those that received a single treatment ( 98 . 9% ( 95% CI: 98 . 7–99 . 1 ) . A similar trend was reported by other researchers comparing single and double dose PZQ efficacy in preschool-age children , school-age age children and adults [36 , 43 , 51 , 52] . As mentioned earlier , PZQ is effective against mature schistosome worms but not juveniles [29 , 53 , 54] . It can be argued that the second treatment given two weeks later was able to target the now fully developed egg-laying adult worms which escaped drug action in the first treatment resulting into fewer eggs being laid and hence a higher ERR with the double regimen . This could be the case especially in the heavily infected children . CR is also influenced by the infection worm load/intensity of S . mansoni infection prior to treatment [55 , 56] . Indeed , in this study , there was a notable association between S . mansoni infection intensity at baseline and CR; children with light ( 1–99 EPG ) infection intensity had the highest cure rate despite of whether they received a single or double dose . However , the importance of a second dose was appreciated as the intensity of infection increased due to the fact that children who were moderately ( 100–399 EPG ) and heavily ( 400 and more EPG ) infected had a higher CR with a double dose than with a single dose . Other studies [36 , 51 , 52 , 56] have reported similar findings of increased CRs using two doses of PZQ . A CR difference between the two treatment regimens is also demonstrated in the varying age groups; children in age group 12–24 months were all cured with either single or double dose , while the rest of the age groups 25–36 , 37–48 and 49–60 months who usually have higher infection intensities tend to have a higher CR with a double dose ( CR: 87 . 3% , 85 . 4% , 75 . 2% ) than with a single dose ( 81 . 9% , 83 . 3% and 71 . 5% ) , respectively . Older children aged 49–60 months are more exposed to schistosome infected water [10] and thus get more worm loads difficult to clear with a single PZQ intervention . They might require two doses . On the other hand , young children who are usually less exposed are lightly infected and therefore benefitted more from only a single treatment . Total prevalence of re-infection 8 months post-treatment was 44 . 5% . It has been hypothesized that , after repeated rounds of infections and PZQ chemotherapy humans slowly acquire protective immunity to S . mansoni leading to partial resistance to re-infection [57 , 58] . Treatment with PZQ boosts adult worm immunoglobulin E ( IgE ) antibodies which are associated with resistance to re-infection [59 , 60] . Based on this hypothesis children aged 1–5 should experience high re-infection because they have not been exposed to S . mansoni long enough to produce schistosome resistant antigens as acquired immunity against schistosomes develops slowly over several years [61–63] . We had expected though that children receiving double treatments would experience a lower re-infection rate and/or intensity than those receiving only a single treatment because double treatments could delay the susceptibility to re-infection . However , this was not the case and was similar to other studies in Uganda performed in school-age children and adults [36] . On the other hand baseline infection intensity has been significantly associated with S . mansoni re-infections [64] . It is not surprising , therefore , that children in the double treatment arm who had heavy infection distribution at baseline turned out to be more re-infected 8 months post treatment . It has also been pointed out that increased resistance to re-infection is induced by repeated rounds of PZQ treatment and this was not the case with children in this study who were being treated for the first time [57] . Children in the double dose regime experienced more ( 50 . 2% ) re-infection than those who received a single dose ( 39 . 0% ) although this difference was not significant . However , the intensity of infection 8 months post-treatment was significantly higher ( P = 0 . 03 ) in children who received two doses of PZQ . This means that at 8 months post treatment the double dose did not offer any further protection against schistosomes . It is also noted that incidentally at baseline , children in the double dose treatment arm were significantly more intensely infected that those in the single dose arm . Probably , double dose children had more residue immature worms that survived both treatments and these fully matured in the 8 months period thus laying more eggs . New re-infections as well as old ones in the double dose arm could therefore have led to the increased intensity compared to the children in the single treatment arm who originally had lighter infections . Mild and short lived side effects due to PZQ administration that are reported in this study may be attributed to the low S . mansoni infection intensities in the children aged 1–5 years . Previous studies show that the frequency and severity of side effects is proportional to the intensity of infection [65 , 66] and that bloody diarrhoea commonly correlates to pre-treatment egg load [67] . The bloody diarrhoea is due to the dying schistosomes which are attacked within 15 minutes after oral intake of PZQ; the PZQ penetrates the worm skin and rapidly moves through tissues causing muscle contraction and bleeding [25 , 27 , 68] . The vomiting was probably mainly caused by the children’s unfavourable reaction to the bitter taste of PZQ [69 , 70] which irritates the mouth and throat [69] . Many children will tend to reject the drug upon administration and end up gagging because it is very difficult to mask the unpleasant taste of crushed tablets . Another explanation for the few registered side effects observed in this study is the food ( piece of bread and soft drink ) that was given to the children before and after administering the PZQ , respectively . Other studies have also associated pre-treatment snack with reduced side effects of PZQ [11 , 43–45] . However , more side effects were observed at the first treatment ( because of the initial heavier worm burden ) than two weeks later ( low intensity ) when a second dose was administered . There was a huge loss to follow-up mainly due to absenteeism and migration . However , the number of children that were lost in both treatment groups were comparable; 273 children lost in the single dose and 281 children lost in the double dose group . A second dose of PZQ administered to children below 6 years does not add value to the CR or reduce the rate of re-infection with schistosomiasis . However , since these children live in endemic communities where transmission is permanent and the rate of schistosome infections is high , a second dose of PZQ could be useful as it leads to significant higher ERRs compared to a single dose indicating a reduction in worm burden and consequently reduction in subsequent morbidity [71 , 72] . It is evident that preschool-age children can safely and effectively be cured by a standard oral dose of PZQ without serious side effects . This age group is , therefore , recommended for inclusion in the on-going Uganda schistosomiasis control programme . However , this would only be possible if a paediatric formulation of PZQ is in place as crushing of tablets and providing juice to mask the bitter taste will not be logistically possible in the current set up of the programme . | Intestinal schistosomiasis is a parasitic water borne neglected tropical disease of considerable public health relevance in the tropics and subtropics . In Uganda , approximately 20 million people are at risk of being infected with S . mansoni , which causes intestinal schistosomiasis , and 4 million individuals are estimated to be infected . Currently , schistosomiasis control using PZQ chemotherapy is mainly focusing on school-age children ( age 6–19 years ) and to those adults considered to have the highest risk of infection . So far preschool-age children are not targeted in schistosomiasis preventive chemotherapy campaigns due to the limited documentation on the safety of PZQ in this age group . However , a number of studies have revealed that children less than 6 years of age get infected with S . mansoni usually through bathing , playing or swimming in schistosome-infested waters and are at higher risk of infection than previously thought . This study compared efficacy of PZQ treatment in terms of CRs and ERRs using a single and double dose among children aged 1–5 years , living along Lake Victoria in eastern Uganda . The effect of the two different regimes on S . mansoni re-infection 8 months post treatment was also assessed . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [] | 2015 | Single Versus Double Dose Praziquantel Comparison on Efficacy and Schistosoma mansoni Re-Infection in Preschool-Age Children in Uganda: A Randomized Controlled Trial |
In past years , much attention has focused on the gene networks that regulate early developmental processes , but less attention has been paid to how multiple networks and processes are temporally coordinated . Recently the discovery of the transcriptional activator Zelda ( Zld ) , which binds to CAGGTAG and related sequences present in the enhancers of many early-activated genes in Drosophila , hinted at a mechanism for how batteries of genes could be simultaneously activated . Here we use genome-wide binding and expression assays to identify Zld target genes in the early embryo with the goal of unraveling the gene circuitry regulated by Zld . We found that Zld binds to genes involved in early developmental processes such as cellularization , sex determination , neurogenesis , and pattern formation . In the absence of Zld , many target genes failed to be activated , while others , particularly the patterning genes , exhibited delayed transcriptional activation , some of which also showed weak and/or sporadic expression . These effects disrupted the normal sequence of patterning-gene interactions and resulted in highly altered spatial expression patterns , demonstrating the significance of a timing mechanism in early development . In addition , we observed prevalent overlap between Zld-bound regions and genomic “hotspot” regions , which are bound by many developmental transcription factors , especially the patterning factors . This , along with the finding that the most over-represented motif in hotspots , CAGGTA , is the Zld binding site , implicates Zld in promoting hotspot formation . We propose that Zld promotes timely and robust transcriptional activation of early-gene networks so that developmental events are coordinated and cell fates are established properly in the cellular blastoderm embryo .
Early development consists of a highly choreographed series of events controlled by temporally and spatially regulated batteries of genes . Although the sequence and nature of the events may vary between organisms , features such as the maternal-to-zygotic transition ( MZT ) where control of development is transferred from maternal to zygotic genes , and the establishment of gene networks initiated by master regulators , are common to all zygotes , pointing to their essential roles in embryogenesis . One of the best-studied developmental systems is the Drosophila embryo where transcription factor hierarchies act to pattern and subdivide the embryo along the anteroposterior ( AP ) and dorsoventral ( DV ) body axes . Only three hours ( hrs ) after fertilization at the height of the MZT , most of the ∼6000 cells in the embryo have acquired their positional information and cell fates . At this time , the embryo has also completed cellularization , whereby each nucleus of the syncytial blastoderm becomes enclosed by cell membrane [1] , and the processes of sex determination and dosage compensation are underway [2] . Although much attention has focused on the gene networks that regulate these processes , less is known about how they are coordinated to occur in a temporally organized manner . The recent discovery of the transcription factor Zld raised the possibility that a single factor could coordinately activate the early zygotic genome [3] . Expression profiling studies of early embryos lacking maternal expression of zld ( henceforth referred to as zld− ) revealed that 70% of the genes normally activated between 1–2 hrs of development were down-regulated , including many genes required for cellularization , sex determination , and dorsal patterning [3] . However , other early genes displayed more subtle changes in the absence of zld . For example , activation of the ventral gene sna was temporally delayed , but appeared to recover by nuclear cycle ( nc ) 14 [3] . Thus , Zld appeared to regulate early zygotic genes in different ways - some are completely dependent on Zld for activation , while others depend on Zld for proper timing of expression . Zld binds in vitro to CAGGTAG and related motifs referred to as TAGteam sites [3] , which were first identified as conserved sequences over-represented in the regulatory regions of pre-cellular blastoderm genes [2] , [4] . Indeed , the TAGteam sites are located upstream and often close to the transcription start site ( TSS ) of genes down-regulated in zld− [3] . However , many genes with upstream TAGteam sites were unaffected in our profiling studies . They may not be expressed at 1–2 hrs , or they could have a maternal component masking the effect of Zld on their zygotic expression , or like sna , they may have gone undetected in the profiling analysis due to more subtle effects in zld− . Therefore , Zld may play a more extensive role in regulating early developmental genes than previously suggested . To further investigate Zld targets , and possible mechanisms of their coordinated expression , we analyzed Zld binding across the genome in pre-cellular blastoderm embryos . These results , combined with our expression profiling studies , uncovered many new Zld targets , and demonstrated that Zld is responsible for timing the activation of genes across all three patterning systems , DV , AP and terminal . Our expression assays further showed that proper transcriptional onset is critical for the cascade of cross-regulatory interactions among patterning genes , and that changes in timing can lead to profound changes in positional information throughout the blastoderm . We found a remarkable overlap between Zld-bound regions and HOT ( high occupancy transcription factor binding ) regions , or hotspots , reported by the modENCODE consortium [5] . The observation that Zld can be visualized in nuclei before other known transcription factors , and that the most over-represented motif in hotspots is the Zld binding site , hints at a role for Zld in marking , establishing , or maintaining hotspots .
Zygotic gene activation begins with a subset of genes transcribed between 1–2 hrs after fertilization ( see time line in Figure 1A ) [6] . A second more dramatic wave of transcription occurs between 2–3 hrs while the embryo undergoes cellularization in nc 14 [4] , [7] . To identify the genes directly bound by Zld , we performed chromatin immunoprecipitation followed by microarray analysis ( ChIP-chip ) . We first prepared polyclonal antibodies against the C-terminal region of Zld known to bind DNA [3] . This antibody recognized a protein of ∼180 kD , the predicted size of Zld ( Figure 1B ) , and stained whole mount wild-type embryos ( Figure 1C–1E ) but not zld− embryos ( Figure 1F ) . Interestingly , Zld can be detected in nuclei as early as nc 2 ( Figure 1C ) , in contrast to other maternal factors such as Bicoid ( Bcd ) and Dorsal ( Dl ) , which do not appear in nuclei before nc 9 and 10 , respectively [8]–[11] . Zld protein levels appear to increase substantially during the second hour of development ( Figure 1B , 1D ) [12] , coincident with the activation of the zygotic genome [6] . We used the Zld antibody to immunoprecipitate chromatin from 1–2 hr embryos , and hybridized labeled DNA to high-resolution NimbleGen tiling arrays ( see Materials and Methods ) . Our genome-wide binding data indicated that Zld behaves similarly to other transcription factors in early Drosophila embryos binding thousands of genomic regions [13] , [14] , but showing a stronger tendency to bind close to the TSS . Specifically , Zld binds 2626 regions ( p<0 . 05; see Materials and Methods ) , and 83% of these ( 2180 ) lie within 2 kilobases ( kb ) of their TSS . Binding near the TSS may be a distinguishing feature of Zld , and may provide a hint regarding its function . In comparison , only 43% of the Bcd-bound regions are within 2 kb of the TSS [13] . We examined the enrichment and placement of TAGteam sites within Zld-bound regions ( Figure 2A; see Materials and Methods ) . Of the five TAGteam sites defined initially by ten Bosch et al . [2] , CAGGTAG was the most enriched site ( 9 fold enriched ) , followed by CAGGTAa ( 4 . 5 fold ) , tAGGTAG ( 3 fold ) , CAGGTAt ( 2 fold ) , and CAGGcAG ( 1 . 8 fold; Figure 2A ) . In addition , two derivatives ( tAGGTAa and CAGGcAa ) , which are located upstream of the nullo gene ( a likely Zld target [3] ) , were enriched 1 . 9 fold and 1 . 6 fold , respectively ( Figure 2A ) . Note that the enriched sequences tend to be located in the center of the bound regions ( Figure 2A ) , suggesting that they represent binding of Zld to TAGteam sites in vivo . We generated a position weight matrix ( PWM ) using all sites in the bound regions that corresponded to the seven TAGteam sequences above ( Figure 2B; see Materials and Methods ) . To identify additional TAGteam sites that might bind Zld , we used the PWM to scan the bound regions ( p≤0 . 0003; see Materials and Methods ) . Only eight heptamers were recovered , including the seven aforementioned TAGteam sites , plus an eighth site , CAGGTAc . This site was found to be enriched 2 . 3 fold in the bound regions ( Figure 2A ) . We emphasize that the PWM is derived from Zld ChIP data , and together with in vitro binding results for all eight sites ( Figure 2D ) , these data define a Zld consensus site . In total , 1240 ( 47% ) of the bound regions contain at least one of the eight sites . Of these , more than half ( 712 ) contain CAGGTAG or CAGGTAA , which are also highly conserved among the sequenced Drosophila genomes when located in bound regions versus non-bound regions , and also when compared to other sequences within the bound regions ( Figure S1A ) . To address whether Zld might bind to a secondary site , as has been found for other transcription factors [15] , we searched for enriched heptamer sequences in Zld-bound regions relative to the genome ( see Materials and Methods ) . Four types of heptamers were recovered ( Figure S1B–S1D ) . The top-ranked enriched site recovered was CAGGTAG ( Figure S1B , S1C ) , validating this approach . Also enriched were TATCGAT and related sequences ( TAT sites; see PWM in Figure 2C ) . 695 ( 26 . 5% ) of the bound regions contain at least one of these sites , however , these regions have on average lower binding scores than CAGGTAG-containing regions ( Figure S1E ) . The TATCGAT site is less conserved than CAGGTAG ( Figure S1A ) , and oligonucleotides containing a TATCGAT site do not bind Zld in vitro ( Figure 2D ) . The remaining two enriched sequences are simple repeats , CTCTCTC and C/GTCACAC ( Figure S1B–S1D ) , and were not further analyzed . However , we noticed that all four enriched sequence types are similar to the motifs found over-represented in hotspots [5] . To examine the relationship between Zld binding and hotspots , we first calculated the percentage of hotspots that contained a Zld-bound region . 48%–64% of hotspots were bound by Zld depending on the transcription factor complexity ( 8–13 factors bound ) , which is striking considering that hotspots were defined using the binding profiles of 41 transcription factors , some of which were from late-staged embryos [5] . We next calculated the average Zld ChIP/input ratio in hotspot windows , which were ranked according to their transcription factor complexity scores [5] . The average Zld ChIP/input ratio increased with increasing complexity of the hotspots , ranging from 4 . 8 in the highest complexity window to 2 . 7 in the lowest ( Figure 2E , blue line ) . In contrast , the background ( see Materials and Methods ) was close to 1 ( p<0 . 0001; Figure 2E , green line ) . To evaluate the regulatory role of Zld binding , we used the genomic tiling arrays to compare expression profiles of wild-type and zld− embryos ( see Materials and Methods ) . We reasoned that by 1 ) interrogating all transcripts with a greater number of probes per gene on the tiling arrays ( to increase statistical power ) , 2 ) using competitive array hybridization to detect more subtle expression differences ( by the ability to normalize data on the same array ) , and 3 ) extending the profiling to include a later time point ( 2–3 hrs ) with both tiling and gene expression arrays , we would capture more Zld target genes . In addition , the tiling arrays allowed direct visual comparison between the Zld ChIP data and transcription profiles , as well as profiles from published datasets such as the hotspots [5] . Figure 3 shows browser views of the sc/sisB , zen , and sna genomic regions . RNA expression of sc/sisB and zen was greatly reduced in zld− at both time points ( Figure 3 , top ) , while sna was only slightly reduced , consistent with our previous observation that sna expression was delayed but recovered in zld− [3] . Zld-bound peaks ( ChIP/input ) were seen over well-defined enhancers ( Figure 3 , red boxes; REDfly [16] ) , which for sc/sisB and zen contain clusters of TAGteam sites ( purple lines ) known to be necessary for enhancer-driven expression [2] . At the sna locus Zld binding is associated with the primary and shadow enhancers [17] , both of which are required for robust expression [17] . Note the significant overlap of Zld-bound peaks and hotspots ( Figure 3 and Figure S1F ) . Comparison of the tiling and gene expression array datasets indicated that the tiling arrays are more sensitive , as has been previously noted [18] , yielding three times more down-regulated genes ( summarized in Table S1 ) . 77% ( 1–2 hrs ) and 82% ( 2–3 hrs ) of the genes in the gene-array dataset were included in the tiling dataset , and importantly , two-thirds of these genes were associated with Zld-bound regions ( Table S1 ) . Many genes that were considered unaffected in our previous analysis [3] are now observed to be down-regulated in zld− , such as the gap gene gt ( see Table S2 for a comparison of the different array datasets at each time point for a subset of pre-blastoderm genes ) . Comparing the two time points , many more genes ( 531 ) came to be expressed in 2–3 hr wild-type embryos , consistent with the burst in transcriptional activity known to occur at this time [4] , [7] . About half of these genes ( 44% ) were down-regulated in zld− and 19% of these were bound by Zld , indicating that Zld activates many of the newly transcribed genes , both directly and indirectly . To further explore the relationship between the position of Zld binding relative to the TSS and effect on gene expression , for each bound region we plotted its location relative to the TSS against the fold change in expression of that gene in wild-type versus zld− . Zld-bound regions showed a tendency to be close to the TSS of genes regardless of whether they were down-regulated in zld− or not; in fact many unaffected genes were bound by Zld within 2 kb of the TSS ( Figure S2A ) . However , we observed a correlation between the location of Zld binding and the level of wild-type expression . Genes considered expressed were more likely to be bound by Zld within 2 kb than genes considered not expressed ( Figure S2B ) . Moreover , the higher the level of expression , the more likely Zld binds near the TSS ( Figure S2B ) , suggesting that such binding is important for transcriptional activation by Zld ( see Discussion ) . As a first step to gain insight into the regulatory networks established by Zld , we performed Gene Ontology ( GO ) analysis on the genes associated with Zld-bound regions . First , we ranked the bound regions according to binding strength from highest to lowest in ten non-overlapping windows , and then analyzed the GO terms of the genes closest to those bound regions . Several groups of early genes were enriched ( Figure S3A , S3B ) . For example , all of the zygotic genes involved in X-chromosome counting/sex determination ( sisA , sc/sisB , os/sisC , dpn , run ) [2] appeared in the top 10% window ( ChIP profiles shown in Figure S4; run in Figure 4 ) , making this GO term highly enriched ( Figure S3B ) . sc/sisB is also known to function in proneural development [19] , and interestingly many genes involved in neurogenesis were enriched in highly bound regions and strongly down-regulated in zld− ( Figure S4B , Table S2 ) , defining another battery of Zld target genes . Also strongly bound were genes involved in cellularization , such as Sry-α , nullo , and slam ( Figure S5 , Table S2 ) , along with cell cycle regulators such as frs that are involved in the nc 14 lengthening , which is concurrent with cellularization [20]–[25] . Many of the enriched GO terms were associated with DV , AP , and terminal patterning ( Figure S3A , S3B ) . This was expected for the DV genes expressed in the dorsal region such as zen and dpp since they are abolished in zld− [3] . What was unexpected was strong Zld binding to genes activated by Dpp/Smads such as Ance and C15 ( Figure S6 ) , because they are downstream in the DV hierarchy and not expressed until nc 14 [26] . This suggests a feed forward loop whereby Zld directly regulates both dpp and its targets . Also unexpected was the strong binding to the ventrally expressed DV genes , which are activated by the Dl morphogen ( Figure 3 and Figure 4 ) . sna and twi are high-level Dl targets expressed in the ventral-most region , the mesoderm , while rho , brk , and sog are lower-level targets expressed in domains of increasing width in the lateral region , the neuroectoderm [27] . However , unlike the dorsally expressed genes , these genes were not significantly affected in the 1–2 hr profiling studies ( Table S2 ) . This seeming contradiction prompted us to investigate their expression patterns by in situ hybridization , which provides higher spatial and temporal resolution . Closer inspection of twi , brk , sog , and rho revealed that their expression was delayed in zld− ( Figure 4 ) , as we noted previously for sna [3] , suggesting that Zld acts in combination with Dl to ensure precise temporal activation of these genes . Like sna [3] , twi expression recovered and appeared normal by nc 14 ( Figure 4B ) , but the lateral stripes of rho , brk , and sog narrowed to about 5–6 cells wide ( arrows in Figure 4D , 4F , 4H ) , the region of intermediate levels of Dl and where the gradient is steepest [9]–[11] . In addition , within the narrow domain , expression was weaker and/or sporadic . Further analysis quantifying the number of nuclei within the rho and sog domains that contained nascent transcripts showed on average 40% of nuclei lacked detectable signal in zld− embryos ( Figure 4J , 4L , 4M ) . In contrast , the wild-type expression domains are more uniform with less than 10% inactive nuclei ( Figure 4I , 4K , 4M ) . Interestingly , Zld binds to both primary and shadow enhancers of sog ( Figure 4O ) and brk ( Figure S1F ) . The presence of a shadow enhancer is thought to increase the potential for reproducible and robust transcriptional activation [17] . In the GO analysis , the pair-rule gene category was associated with one of the highest enrichment scores ( Figure S3A , S3B ) . Zld-bound peaks were distributed across several of the known “stripe elements” as well as near the TSS of the primary pair-rule genes ( Figure 5I–5K and Figure S7 ) . Since their expression levels were only mildly reduced in zld− ( Table S2 ) , we looked closely at the expression patterns . In wild-type embryos , eve , ftz , hairy , and runt were initially expressed in broad domains as early as nc 10 , which refine into the respective seven-stripe patterns by the end of nc 14 ( Figure 5A , 5C , 5E , 5G ) [28] . In zld− , not only was activation delayed by two nuclear cycles , but the stripe patterns were dramatically altered in nc 14 ( Figure 5B , 5D , 5F , 5H ) . Since pair-rule stripes are formed by localized gap repressors acting on stripe enhancers [29] , [30] , we next examined gap gene expression . In wild-type embryos , gt ( Figure 6A ) and tll ( Figure 6G ) transcripts were detected at nc 10 , while kni , Kr and hb transcripts were not observed until nc 11–12 ( Figure 6 , data not shown for hb ) . This varied activation foreshadows the appearance of the gap protein gradients [31] . In zld− embryos initial transcription of all five gap genes was delayed by 1–2 nc ( Figure 6 , data not shown for hb ) . In addition , their patterns were significantly disrupted , which can be explained in part by miscued gap gene interactions . For example , in wild-type embryos , mutual repression between Gt and Kr is known to establish their complementary domains [32] , [33] . The overlap of gt and Kr transcripts in the head region of early zld− mutants , which is never seen in wild-type , may be a consequence of delayed Gt repression ( Figure 6B , top ) , but as Gt accumulates , anterior Kr expression disappears ( Figure 6F , bottom ) . Another example involves Hb repressor function . Hb and Gt set the anterior Kr border , while Hb and Kr establish the anterior border of the posterior gt domain [32] . In zld− , the hb domain was reduced in size ( Figure 6J ) , possibly due to lack of activation in regions of low-level Bcd , and consequently the anterior border of the Kr central domain ( Figure 6F , bottom ) and the gt and kni posterior domains shift anteriorly ( Figure 6B , 6D , bottom ) . The shift in the posterior border of the Kr domain ( Figure 6F , bottom ) is likely due to expanded gt ( Figure 6B ) . Tll is a strong repressor of gap genes [34] , hence the ectopic expression of kni can likewise be explained by the delay in tll expression in zld− ( Figure 6H ) . The posterior tll domain , which expands anteriorly along the ventral surface ( Figure 6H ) , could cause ventral repression of Kr ( Figure 6F ) , as well as the changes observed in the posterior domains of gt , kni , hb , ( Figure 6B , 6D , 6J , bottom ) and the pair-rule genes ( Figure 5 ) . In summary , although many of the observed defects in the gap and pair-rule patterns in zld− are due to delayed and mis-localized gap repressor activity , Zld binding to the gap and pair-rule enhancers ( Figure 5 , Figure 6 , and Figure S7 ) points to a direct role for Zld in activating these genes . The ubiquitous nature of Zld binding to patterning genes prompted us to search for overlap between Zld-bound regions and regions bound by AP [13] , [14] and DV [35] transcription factors . 62% of the Bcd-bound regions , and 70% of the Dl-bound regions , which overlap extensively with Twi- and Sna-bound regions ( referred to as DTS , [35] ) , are bound by Zld ( data not shown ) . In contrast , only about 30% of the regions bound by Bcd are bound by DTS , and vice versa . We next looked at the genes bound by Zld and the other factors , and whether those genes were expressed in early embryos . Zld binds to 72% of the Bcd targets , 70% of the Cad targets , and 80% of the Tll targets ( Figure 7 and Figure S8 ) . Less but considerable overlap ( about 50% ) was observed between Zld targets and gap gene ( Hb , Gt , Kr , Kni ) targets ( Figure S8 ) . We further distinguished target genes by whether they were expressed at 2–4 hrs ( defined as bound by RNA polymerase II ( pol II ) by MacArthur et al . [14] ) . About half of the AP factor-bound target genes were expressed . For each factor , expressed target genes were more likely to be bound by Zld than the non-expressed targets ( Figure 7A and Figure S8A ) . Similarly , Zld binds 59% of the DTS target genes ( 351 genes; Figure 7B ) , but this increases to 76% for the expressed targets , and decreases to 35% for non-expressed targets ( Figure 7A , 7C ) . Thus , the AP and DV target genes that are also bound by Zld are more likely to be expressed in the blastoderm embryo , indicating that Zld binding may promote transcriptional activity .
Our Zld binding analyses indicate that there are at least eight TAGteam sites . CAGGTAG and CAGGTAA were the most over-represented and the most highly conserved in the Zld-bound regions ( Figure 2A and Figure S1A ) . About half of Zld binding is TAGteam site dependent , and all of the sex determination , cellularization , and patterning genes we studied have TAGteam sites in their enhancers and in many cases near the TSS . Curiously , within the CAGGTAG site is CAGGTA , a motif found strongly enriched in hotspots [5] . Likewise , our TATCGAT , CT-repeat , and CAC-related sites ( Figure S1B–S1D ) are similar to additional motifs found in hotspots: GTATCGAT , CTCTCTCTCT , and CTCACACG , respectively , which were proposed by modENCODE to be “candidate drivers” of hotspot formation [5] . TATCGAT is contained within the DRE ( DNA replication related element ) octamer site , TATCGATA , which is found near the TSS of genes involved in DNA replication [36] . Additionally , TATCGATA is similar to the BEAF-32 insulator site [37] . The CT-repeat site is also associated with an insulator sequence , the Trl/GAF motif [37] . It is unclear how Zld interacts with the non-TAGteam sequences since TATCGAT , for example , does not appear to bind Zld in vitro ( Figure 2D ) . It is possible that the enrichment of these sites in Zld-bound regions is due to recruitment of Zld by components of complexes that directly interact with these sequences , or to opportunistic Zld interactions . Thus , at least for hotspots with the CAGGTA motif , which was discovered in the hotspots with highest complexity ( bound by 12–14 factors ) [5] ) , it is possible that Zld binding is involved in their establishment [5] . The idea of an “initial step in the cascade of zygotic gene interactions that control development” was first proposed by Edgar and Schubiger [6] , and the idea of a “timer” in early development that functions alongside the spatially restricted morphogens was proposed by ten Bosch et al . [2] and De Renzis et al . [4] for CAGGTAG sites . Our combined results on Zld extend both of these ideas . Zld protein accumulates to high levels by one hour of development ( Figure 1B , 1D ) , which coincides with the onset of zygotic genome activation [6] . Within a two-hour period , the embryo cellularizes , determines X-chromosome dosage , patterns its body plan , and gets ready for gastrulation . By virtue of a single factor these processes are coordinately activated . One can predict that increasing Zld levels in early embryos would advance timing of activation . Our initial attempts to increase Zld protein levels by adding copies of Zld rescue constructs did not yield higher Zld protein levels ( data not shown ) , indicating that Zld levels may be tightly regulated . However , ten Bosch et al . [2] showed that doubling the number of TAGteam sites in the zen enhancer led to precocious expression , supporting the idea that Zld acts to time zygotic gene activation . In the absence of Zld , all direct targets are either: 1 ) not expressed , 2 ) delayed but recover , or 3 ) delayed but do not recover fully . For example , genes involved in sex determination , cellularization , dorsal patterning , and proneural development are strongly down-regulated in zld− and never recover ( Table S2; Figures S4 , S5 , S6 ) . In contrast , genes involved in AP and ventral patterning were not significantly down-regulated , and how they recovered depended on how they responded to other factors . The high-level Dl targets sna and twi recovered by nc 14 , but the lower-level targets sog , brk , and rho did not recover their normal patterns in zld−; instead they were expressed sporadically in a narrow domain with great variability among embryos ( Figure 4 ) . It appears that intermediate levels of Dl are no longer sufficient for robust and faithful target-gene expression , and lower levels cannot activate them at all; thus , the Dl gradient cannot be interpreted without Zld . These effects are likely due to the lack of direct Zld input , as mutation of the TAGteam sites in the sog primary enhancer caused a similar narrowing of the reporter expression domain [38] . Indirect effects of delayed twi expression may also contribute , since mutation of Twi binding sites in the rho enhancer also resulted in a narrower domain [39] . These observations suggest that Zld not only acts as a timer for Dl target-gene activation , but also potentiates Dl morphogenetic activity over a broad range in the neuroectoderm in order to establish multiple threshold responses . Along the AP axis , Zld may function in a similar way with Bcd . In zld− , the hb border shifts anteriorly ( Figure 6 ) , indicating that in regions of low-level Bcd , Zld enhances the sensitivity of target genes to morphogen concentrations . These results imply that Zld may promote transcription by acting synergistically with the patterning morphogens . It is important to note that the observed delay in expression does not necessarily mean the gene is activated later , but that without the synergy factor , there are not enough detectable transcripts at the time when assayed . Sporadic expression may reflect a similar situation . Beyond the role of Zld in timing transcriptional initiation is a more elaborate timing mechanism , exemplified by the sequential appearance of the gap genes . How does Zld achieve differential activation of target genes within a network ? A simple model would suggest that the activation of Zld target genes correlates with the strength of Zld binding to their regulatory elements . We noticed that the earlier activated genes in the segmentation network had higher binding scores than those activated later . gt , tll , and all of the primary pair-rule genes , which are abundantly expressed by nc 10 , had higher binding scores ( Figure 5 , Figure 6 ) than kni and Kr , which become abundant later in nc 11 and nc 12 , respectively . Later-acting genes such as secondary pair-rule genes , segment polarity genes , and the homeotic genes were bound , but had lower binding scores ( Figure S7 and data not shown ) . Such a mechanism where timing of activation is dependent on strength of binding was shown for the Pha-4 transcription factor in C . elegans pharyngeal development . Pha-4 regulates a wide array of genes expressed at different stages , and the onset of target-gene expression depends on the affinity of Pha-4 binding sites in the regulatory regions of those target genes [40] , [41] . An intriguing possibility for the early Drosophila embryo is that as Zld levels rise in the first hour of development a “temporal” concentration gradient is formed such that interaction with higher affinity binding sites would occur before that with lower affinity sites , thus differentially activating target genes . A second timing mechanism is provided by the intrinsic properties of the regulatory motifs established by Zld . Our data revealed that Zld functions in several coherent feed forward loops , for example , binding both the XSE ( X-chromosome signal element ) genes ( such as sisA ) and Sxl , dpp and its targets , and twi and rho ( Figure S9 ) . Embedded in this type of motif is a mechanism of temporal control since a delay in the activation of the third gene in the loop occurs because of its dependence on accumulation of the second gene product [42] , [43] . For example , the activation of Sxl is 2–3 nc later than that of the XSE genes [2] . In addition , experiments that abolished the TAGteam sites in the SxlPe enhancer caused a 3 nc delay in reporter expression , demonstrating a direct role for these sites , and hence Zld , in timing transcriptional activation [2] . Zld also functions in an incoherent feed forward loop whereby one branch of the loop has the opposite sign [43] . Zld promotes transcription of both the pair-rule and gap genes , while gap proteins repress pair-rule genes ( Figure S9 ) . The primary pair-rule gene transcripts are easily detectable by nc 10 ( Figure 5 ) , even before some of the gap genes , giving a new perspective on the canonical segmentation gene hierarchy in which the pair-rule genes are downstream of the gaps . Early strong activation of the pair-rule genes may be essential to guarantee transcriptional activation before repressor gradients overwhelm the AP axis . We can extract clues from our results about how Zld may function on a mechanistic level . First , Zld appears in zygotic nuclei very early ( Figure 1C ) , before Bcd and Dl , possibly binding to target genes first . Second , loss of Zld results in delayed transcriptional activation and , in many cases , weak and/or sporadic expression ( Figure 4 , Figure 5 , Figure 6 ) . Third , Zld binding is frequently found at early enhancers ( both primary and shadow ) , as well as close to the TSS of genes , hinting at a role in recruitment of the transcriptional machinery . Fourth , Zld binding coincides with hotspots ( Figure 3 and Figure S1 ) , which were found to correlate with regions of nucleosome depletion [5] . Together these observations suggest that Zld increases the transcriptional activity , or expressivity , of target genes . Mechanistically , Zld binding could facilitate either the access of other factors ( both activators and repressors ) to DNA or the interaction of these factors with the transcriptional machinery , an idea put forth by Bradley et al . [44] after observing a correlation between the evolutionary turnover of the CAGGTAG site along with the patterning factor binding sites . An alternative mechanism to ensure robust and coordinated early embryonic expression is pol II pausing ( reviewed in [45] ) . Many Zld target genes such as sog were shown to exhibit polymerase pausing [46] . The delayed and sporadic expression in zld− could be explained by lack of paused pol II . It is evident from our results that Zld coordinates the onset of transcriptional activity of the early gene networks during the MZT ( Figure S9 ) . Considering that Zld is also expressed at later times in development [47] , we predict that Zld will act similarly to increase expressivity of genes in networks that function , for example , in central nervous system development in mid-stage embryos and imaginal disc patterning in larval development . In these processes , similar to the MZT , a simple strategy may be used to collectively activate and temporally control batteries of genes required for establishing the proper gene circuitry .
The yw strain was used to obtain wild-type embryos , and the zld294 allele was used to obtain zld− germline clones as previously described [3] . Rabbit polyclonal antibodies against Zld were generated using the C-terminal part of Zld ( amino acids 1240–1470 ) [3] , containing a cluster of four zinc fingers fused to GST . For chromatin immunoprecipitation experiments , the antibodies were purified from the serum bleeds by antigen affinity chromatography [48] against purified recombinant-Zld protein coupled to an affinity column . For Western blotting , 50 appropriately staged Drosophila embryos were homogenized in SDS Laemmli loading buffer and briefly centrifuged . The protein concentration of the supernatant was determined ( Bradford ) , and equal amounts of protein were loaded in each lane of a 6% SDS-PAGE gel ( 40 µg per lane ) . The blotting and transfer were performed according to standard procedures [48] . Embryos were fixed and hybridized as previously described [3] using digoxygenin-UTP ( Roche Biochemicals ) labeled RNA probes synthesized from subcloned cDNA sequences or genomic intronic DNA sequences ( for sog and rho ) . Antibody staining was performed by incubating fixed embryos with rat anti-Zld antibodies ( 1∶200 dilution ) followed by incubation with AlexaFluor 488 donkey anti-rat IgG ( 1∶500 dilution ) secondary antibodies ( Invitrogen ) . Embryos were visualized by fluorescence microscopy using an FX-A Nikon microscope and by Nomarski optics using a Zeiss Axiophot microscope . Flourescent in situ hybridization ( FISH ) was performed as previously described [17] using intronic probes for sog and rho , sheep anti-DIG antibodies ( Roche Biochemicals ) , and AlexaFluor 555 donkey anti-sheep IgG secondary antibodies ( Invitrogen ) . Images were acquired as previously described [17] using a Leica TCS SP5 confocal microscope ( 40× oil immersion objective ) with 1024×1024 resolution and approximately 250 nm/pixel . More than 15 Z-sections from nc 14 embryos were taken at 0 . 5 µm intervals to capture as many nascent transcripts in nuclei as possible . The Z-sections containing pixel intensities higher than the median intensity of all pixels were selected for analysis . For each position in the X–Y plane , the pixel with the strongest intensity across all Z-sections was defined as the intensity value for that X–Y position . All of the Z-sections from the DAPI channel were processed by Helicon Focus ( HelicoSoft ) to generate clear images of nuclei , which were identified by customized Matlab scripts . Every FISH signal was assigned to the closest nucleus , only when the distance between a FISH signal and the center of a nucleus was smaller than 1 . 5× the radius of the nucleus . The assigned nuclei were considered as with expression and pseudo-colored . Electrophoretic mobility shift assays ( EMSA , or gel shift assays ) were performed as previously described [3] . The following oligonucleotide sequences were derived from genomic DNA sequences . Each is 21 nucleotides in length and contains a TAGteam or TATCGAT site plus surrounding sequences: zen1: CACTATTTAGGTAGACACTGT , zen2: TGGGTTTCAGGTAGGTGAATA , zen3: ATAAACACAGGCAGCTGGTGC , eve3: ACAATTGCAGGTAAGTAGAGC , nullo-1: AAAGGATCAGGTACCCGGGGT , nullo-2: GTCGGAGCAGGCAACGGGCAT , sema1: TCGTCGGTAGGTAAAAGTTGT , tat: CCAGCCGCAGGTATTTAGTTC , zen1m: CACTATTTGAATAGACACTGT , tatcgat: TCACTACTATCGATGACGATG ( TAGteam or TATCGAT sites are underlined ) . A Drosophila melanogaster tiled genomic microarray set was designed by J . R . M . in concert with the bioinformatics team at Roche NimbleGen using Genome Release 5 . This array set , which comprises two HD2 ( 2 . 1 million feature ) microarrays , utilizes 50-mer oligonucleotide probes with up to 100 close matches per sequence tolerated , and a median probe spacing of 33 bp . For our design , we chose to tolerate a large number of close matches in order to include on the array more heterochromatin and repeat regions , including transposon sequences . The Design Names of the arrays are as follows: 081229_Dm_JM_ChIP_1_HX1 and 081229_Dm_JM_ChIP_2_HX1 . Total RNA was extracted from three independent collections of 2–3 hr yw and zld− Drosophila embryos by TRIzol ( Invitrogen ) . cDNA was prepared using the GeneChip HT One-Cycle cDNA Synthesis Kit ( Manufactured by Invitrogen for Affymetrix ) and labeled with the BioArray™ HighYield™ RNA Transcript Labeling Kit ( Enzo ) . Labeled probes were hybridized to Drosophila Genome 2 Affymetrix arrays and processed by a GeneChip Fluidics Station 400 . Data were acquired by the GeneChip Scanner 3000 and processed/normalized by the Affymetrix GeneChip Operating Software ( GCOS ) . Genes were identified as present when at least two of the three replicates had present ( P ) assignment ( p<0 . 05 ) . t-test analysis was performed on the data from the three biological replicates . The fold change of each gene was determined by the ratio of yw mean/zld− mean . Double-stranded cDNA was generated from total RNA isolated ( TRIzol , Invitrogen ) from staged ( 1–2 hrs and 2–3 hrs at 25°C , verified by DAPI staining of a portion of the collected embryos ) wild-type and zld− embryos , then amplified/labeled using either Cy3- or Cy5-coupled random nonomers , respectively . 15 ug of each labeled cDNA were used for competitive hybridization , coupling the same staged wild-type and zld− samples in order to facilitate direct comparison ( see Roche NimbleGen Gene Expression Protocol [http://www . nimblegen . com/products/lit/expression_userguide_v5p0 . pdf] for specific labeling and processing details ) . After hybridization for 16–20 hrs , the arrays were washed , dried and then scanned on an Axon GenePix 4000B microarray scanner from Molecular Devices . Intensity readings of probes were corrected according to their GC content using a set of random probes on the arrays , and normalized by the Lowess normalization method [49] . After normalization , a median filter was applied ( using a sliding window of three probes with the center probe given the median value ) . The expression level of each exon was calculated by taking the median of all the probes covering the exon , and the expression level of each RNA isoform was calculated by averaging the expression levels of all the exons of the isoform without weighting . Using the “fdrtool package” in R [50] , we set the background threshold as 5% FDR , which was previously shown to be an appropriate cutoff for tiling array data [51] . Genes were considered as expressed ( present ) if more than 50% of the probe signals were higher than the threshold , otherwise they were considered as not expressed ( absent ) . Using this approach , we concluded that 49 . 5% of the genome was represented as RNA in 1–2 hr embryos , which is similar to findings from other studies [4] , [52] , [53] . Since , in most cases , each gene was represented by multiple probes , and thus was interrogated multiple times in one experiment , we applied the t-test to obtain p-values for each gene in the yw and zld− . samples . The fold change of each transcript was determined as the ratio of yw/zld− . 1–2 hr yw embryos were fixed in 2% formaldehyde for 20 min . Nuclei were harvested and sonicated to release and shear genomic DNA . The chromatin was immunoprecipitated by purified anti-Zld antibody and protein A beads . The chromatin immunoprecipitated ( ChIPed ) DNA samples were amplified by GenomePlex WGA1 ( Sigma ) followed by GenomePlex WGA3 ( Sigma ) twice . ChIPed DNA was labeled/amplified with Cy3-coupled random nonomers and the corresponding input DNA was labeled with Cy5-coupled random nonomers; 34 µg of each labeled DNA was combined and co-hybridized to the Drosophila tiling array set ( as described above ) according to the Roche NimbleGen ChIP-chip Protocol ( http://www . nimblegen . com/products/lit/chip_userguide_v6p1 . pdf ) . After hybridization for 16–20 hrs , the arrays were washed , dried and then scanned on an Axon GenePix 4000B microarray scanner from Molecular Devices . For each dataset , the ratio of ChIP to input intensities was obtained , log2 transformed , and standardized to z scores . The datasets of the dye swap replicates were averaged to eliminate dye bias . ChIP enrichment scores were calculated using the R package “Ringo” [54] as the sum of probe levels minus the threshold with settings of 0 . 05 p-value and 10 minimum probes . A higher score indicates a higher ChIP/input ratio in the region . Data was visualized using the Integrated Genome Browser [55] , and a median filter was applied to the ChIP/input ratio for the visualization . Validation was performed by qPCR to assess the enrichment of the enhancer region of sc/sisB , a known target of Zld , and the coding region of CG18125 , a negative control , in ChIPed versus input DNA . The sc/sisB region showed 23 . 53 fold enrichment over CG18125 ( data not shown ) . The enrichment index for a specific DNA motif was calculated as the density of the motif in Zld-bound regions divided by the density of that motif in the Drosophila genome . Enrichment indices of sites were calculated in 100 bp non-overlapping windows across the 5 kb flanking regions from the center of Zld-bound regions . The background was estimated as the average of enrichment indices of 20 random heptamers . We took seven CAGGTAG related heptamers , including five that were previously identified by ten Bosch et al . [2]: CAGGTAG , TAGGTAG , CAGGTAA , CAGGCAG and CAGGTAT , and 2 newly discovered related heptamers from the Zld-bound region upstream of nullo: CAGGCAA and CAGGTAC , and then weighted every heptamer according to their enrichment indices to generate a primary PWM using the “Biostrings” package [56] and visualized by the “seqLogo” package [57] in R . The enrichment analysis with this PWM defined sequences with a p-value larger than 0 . 0005 as not enriched in Zld-bound regions . We then generated a new PWM by using the enriched sequences ( p≤0 . 0005 ) with their surrounding nucleotides . We repeated the enrichment test using this new PWM ( Figure 2B ) and eight TAGteam motifs ( p≤0 . 0003 ) . Enrichment indices of all possible heptamers were calculated for Zld-bound regions using a 500 bp window centered around the middle of Zld-bound peaks . Using an enrichment score of 3 . 5 as the cutoff , 11 heptamers were recovered ( Figure S1B ) . The highest-ranking heptamer was CAGGTAG . Heptamers could be separated into two groups with different core sequences , CAGGTA and TATCGA ( heptamers containing simple repeats were not further analyzed ) . The CAGGTA group was reminiscent of TAGteam sites . All of the Zld-bound regions that contain at least one of the new enriched sites that have the TATCGA core were analyzed by MEME4 . 4 . 0 [58] and a PWM was generated . The distance from the center of every Zld-bound region to every annotated TSS was calculated ( according to Drosophila melanogaster genome release 5 . 29 ) . The closest TSS ( for both strands ) was assigned to each Zld-bound region . Thus each bound region typically has two assigned genes . Hotspot data was obtained from Roy et al . [5] . Only the regions bound by eight or more transcription factors were analyzed in this study . Hotspots were ranked by their complexity scores [5] . In the 100 non-overlapping hotspot windows , Zld binding was calculated as the average of ChIP/input ratios of all probes located in the hotspot regions . To estimate the background of Zld binding in hotspots , the ChIP/input ratio for all probes on the tiling arrays were randomly shuffled 20 times , and the average of the ChIP/input ratio from the shuffled probes corresponding to a hotspot was calculated . Target gene lists for AP factors ( Bcd , Cad , Hb , Gt , Kr , Kni and Tll ) were obtained from MacArthur et al . [14] . The target-gene list for Dl , Twi , and Sna ( DTS ) was obtained from Zeitlinger , et al . [35] . We limited the search to target genes that lie within 2 kb of the factor-bound regions . We identified the target genes bound by both Zld and each AP factor , Zld and DTS , and Zld and Mad , and distinguished whether or not they were expressed at 2–4 hrs ( defined as bound by pol II , [14] ) . Pair-wise heat maps were generated to represent the overlapping percentages between factors . GO term analysis was performed using the Database for Annotation , Visualization and Integrated Discovery ( DAVID ) v6 . 7 [59] , [60; http://david . abcc . ncifcrf . gov/] on the genes associated with Zld-bound regions . We limited this analysis to genes that are downstream of a Zld-bound region , except in cases where there was no downstream gene then we used the upstream gene . | Development of a fertilized egg into a multicellular organism comprises a series of precisely timed events initially controlled by factors deposited into the egg . Some of these factors are localized to specific regions of the embryo and instruct cells to adopt certain fates . In this way , these “morphogen” factors lend pattern to the body plan so that different appendages are formed in the right places . In contrast , other factors are evenly distributed throughout the embryo and regulate processes concerning all cells . For example , in Drosophila , Zld is a ubiquitous factor that collectively activates batteries of genes essential for further development . Here we show that Zld also functions alongside the spatial morphogens to ensure timely and robust activation of their target genes . In the absence of Zld , activation of these genes is delayed , which derails the proper order of gene interactions and ultimately disrupts gene expression patterns . Our results demonstrate the significance of a timing mechanism in coordinating regulatory gene networks during early development , and they bring a new perspective to classical concepts of how spatial regulation can be achieved . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"systems",
"biology",
"developmental",
"biology",
"cell",
"fate",
"determination",
"embryology",
"organism",
"development",
"molecular",
"development",
"genetics",
"biology",
"genomics",
"morphogenesis",
"pattern",
"formation",
"genetics",
"and",
"genomics",
"cell",
"differentiation"
] | 2011 | Temporal Coordination of Gene Networks by Zelda in the Early Drosophila Embryo |
Dysregulated immune responses may contribute to the clinical complications that occur in some patients with dengue . In Vietnamese pediatric dengue cases randomized to early prednisolone therapy , 81 gene-transcripts ( 0 . 2% of the 47 , 231 evaluated ) were differentially abundant in whole-blood between high-dose ( 2 mg/kg ) prednisolone and placebo-treated patients two days after commencing therapy . Prominent among the 81 transcripts were those associated with T and NK cell cytolytic functions . Additionally , prednisolone therapy was not associated with changes in plasma cytokine levels . The inability of prednisolone treatment to markedly attenuate the host immune response is instructive for planning future therapeutic strategies for dengue .
Dengue is an acute , mosquito-borne illness caused by any of the four types of dengue virus ( DENV1-4 ) . There are an estimated 390 million symptomatic and asymptomatic infections per year [1] . The clinical evolution is variable , ranging from non-specific febrile illness to severe and sometimes fatal disease . One of the commonest complications observed is a transient vasculopathy , manifesting as increased vascular permeability with altered haemostasis , typically 3–6 days after fever onset . Dysregulated host immune responses , particularly those associated with secondary infections , are widely held to contribute mechanistically to the vasculopathy that characterizes severe dengue [2] . No specific therapies or licensed vaccines are currently available and management relies on assiduous supportive care . Synthetic glucocorticoids are frequently employed as adjunctive therapy in disease states where the host immune response is thought to be a significant contributor to disease pathogenesis . We recently performed a randomized , controlled trial of early oral prednisolone therapy in 225 confirmed pediatric dengue cases [3] . Although the trial was primarily designed to assess safety we were unable to detect any reduction in the severity of plasma leakage or other recognised complications of dengue . We report here on immunological correlates of prednisolone therapy in this trial with a view to understanding the lack of clinical benefit imparted by prednisolone and to guide future intervention strategies for dengue .
A randomized , placebo-controlled double-blind trial assessing the safety of early oral corticosteroid therapy in dengue patients was conducted at the Hospital for Tropical Diseases , Ho Chi Minh City , Vietnam between August 2009 and January 2011 as approval from the Ethical Committee of the Ministry of Health of Vietnam ( 2407/QĐ-BYT ) and the Oxford Tropical Research Ethics Committee ( OxTREC 33-08 ) and has been reported elsewhere [3] . The trial registration number is ISRCTN39575233 . Briefly , patients aged from 5–20 years with fever for less than 72 hrs and a positive dengue NS1 rapid test were randomly allocated to oral treatment with high-dose prednisolone ( 2 mg/kg ) , low-dose prednisolone ( 0 . 5 mg/kg ) or identical placebo for 3 days provided the patient or their parent/guardian gave written informed consent and children 12–17 years gave assent; all patients recovered fully and we found no association between treatment allocation and any of the predefined clinical , haematological or virological endpoints . The research blood specimens that form the basis of the work described here were collected as part of the trial protocol at pre-specified time-points: enrolment ( pre-treatment ) ; 2 days post initiation of treatment; and at follow-up in late convalescence ( median 29 ( IQR 27 , 30 ) days after enrolment ) . To facilitate interpretation of the findings with respect to defervescence , the first day that the temperature fell to 37 . 5°C or less and remained below this level for 48 hours or until discharge was taken as the day of deferescence; fever day 0 was defined as the calendar day of defervescence , with days before this point numbered consecutively as fever days −1 , − 2 , − 3 respectively . Eleven cytokines ( IL-1β , IL-2 , IL-4 , IL-5 , IL-6 , IL-10 , IL-12p70 , IL-13 , IFNγ , TNFα and IP10 ) were quantified using a multiplex biometric immunoassay following the instructions of the manufacturer ( Bio-Plex Precision Pro Assays , Human cytokine 10-Plex , Bio-Rad Inc . , USA ) . The limits of detection were as follows: 0 . 23 ( IL-1β ) , 0 . 84 ( IL-2 ) , 0 . 14 ( IL-4 ) , 1 . 5 ( IL-5 ) , 1 . 23 ( IL-6 ) , 0 . 96 ( IL-10 ) , 0 . 2 ( IL-12p70 ) , 1 . 19 ( IL-13 ) , 0 . 34 ( IFN-γ ) , 0 . 14 ( TNF-α ) and 10 ( IP10 ) , all pg/ml . The gene expression microarray assay , and the procedures for normalization and analysis of the microarray data were conducted as described elsewhere [4] . Samples from the first 123 consecutive patients enrolled in the study were used as the “discovery” cohort . A fluidigm system ( Fluidigm Corp . , USA ) was used for realtime PCR ( RT-PCR ) validation of those differentially abundant transcripts identified in the gene expression microarray , using samples from the whole patient cohort and following the manufacturer's instructions . The delta Ct value for each gene was calculated by subtracting the Ct value of the gene of interest from the Ct value of 18S , the house-keeping gene . All group comparisons of microarray data were based on ANOVA with correction for multiple testing with the Benjamini-Hochberg method , as implemented in the GeneSpring Software ( Silicon Genetics ) . A fold change of 1 . 5 was defined as the cut-off for screening significant entities . We used multivariable linear regression modeling for all comparisons of PCR results , expressed as delta Ct values , between the treatment arms . For the overall comparison across treatment arms , a linear trend test was used . In view of the likely evolution of gene expression during the illness episode , and the known associations of many of the genes of interest to immune parameters , we adjusted for day of illness and the absolute neutrophil and lymphocyte counts . Additionally we adjusted for the pre-treatment value when examining within-patient changes in delta Ct over time . P values for testing of multiple PCR results were corrected using the Benjamini-Hochberg method . The relative expression ratios of the genes between the treatment arms were estimated using the delta delta Ct formula: R = 2−ΔΔCt with corresponding 95% confidence intervals based on the bootstrap . Genes with significantly different relative expression ratios between treatment arms ( i . e . adjusted p<0 . 05 and 95%CI not including 1 ) were considered up or down regulated as appropriate . Multivariable linear regression was used for all comparisons of log-transformed values of cytokine concentrations adjusting for day of illness and pre-treatment values and linear trend tests performed as described above . All the analyses were corrected for multiple tests using the Benjamini-Hochberg method . All analyses other than those pertaining to the microarray data were performed using R , version R2 . 13 . 2 ( R Foundation for Statistical Computing , Vienna , Austria ) .
Baseline characteristics for the first 123 dengue patients consecutively enrolled in the trial who formed the discovery cohort are described in Table 1 together with similar information for the 223 patients used for the PCR validation; clinical and laboratory features were similar across the three intervention arms . At enrolment , none of the 47 , 231 transcripts evaluated in the microarray were differentially abundant between groups of patients allocated to different treatment arms . Similarly , there were no transcripts differentially abundant between treatment groups in late convalescence , or between placebo and low-dose prednisolone patients 2 days after commencing treatment . By contrast , 81 differentially abundant transcripts ( 25 that were more abundant and 56 less abundant ) , representing 67 genes , were detected when comparing whole-blood gene expression profiles between the high-dose prednisolone and placebo groups 2 days after commencing treatment . RT-PCR validation measurements were targeted to a subgroup of 31 ( 46% ) of the 67 genes on the basis of their having plausible roles in immune function . RT-PCR validation was performed with 600 whole blood RNA samples collected at baseline ( N = 208 ) , 2 days after starting treatment ( N = 200 ) , and in late-convalescence ( N = 192 ) from 223 ( 99% ) of the 225 patients in the trial . With the exception of one target ( CLIC3 ) , RT-PCR measurements were entirely concordant with the microarray findings ( Figure 1 ) . Annotation of RT-PCR validated gene elements enabled functional grouping according to their recognized roles in NK and T cell cytolytic function , T cell activation and innate immune responses ( Figure 1 ) . These analyses were extended by investigating a prednisolone dose-response ( placebo , low-dose , high dose ) relationship . After adjustment for the a priori defined variables ( day of illness , absolute neutrophil and lymphocyte counts , baseline transcript abundance ) , a highly significant prednisolone dose-related effect on gene transcript abundance was observed for all 31 transcripts ( Table S1 ) . Collectively , these results define a discrete gene transcript profile that is associated with high-dose prednisolone therapy in dengue patients . Concentrations of 11 cytokines and chemokines were measured in 636 serial plasma samples from 222 patients at the three time-points . Although cytokine/chemokine concentrations were within the detectable range in 98% of samples tested , their levels were not significantly elevated during the acute phase of illness compared to follow-up , and there were no significant differences between treatment groups 2 days after starting therapy ( Table 2 ) .
The current study was linked to a randomized controlled trial of early prednisolone therapy for dengue that demonstrated the safety of prednisolone but did not provide evidence of improved clinical or laboratory outcomes for patients [3] . Here we provide insights into these trial findings by identifying a surprisingly small prednisolone-associated footprint ( just 81 transcripts differentially abundant from 47 , 231 evaluated ) on the whole-blood gene expression profile that manifests during DENV infection . Furthermore , acute-phase plasma cytokine concentrations were not measurably attenuated by prednisolone treatment . The limited immunomodulation achieved by prednisolone is consistent with it having negligible measurable benefits in the clinical trial in which this current study was nested . Dysregulated immune responses are widely believed to contribute to the pathogenesis of severe dengue [2] , and hence corticosteroid therapy has been trialed in several small studies of patients with shock due to vascular leakage [5] , [6] . No benefit was demonstrated in those studies , but treatment was initiated when shock was already established , when it is probably too late to modulate the host immune response . In our recent intervention study prednisolone was commenced during the early febrile phase , but we were still unable to demonstrate any amelioration in the severity of plasma leakage . Against a backdrop where the immunomodulatory actions of corticosteroids are well established [7] , [8] , it is surprising that we did not observe stronger signals of immune-modulation . The absence of a measurable impact of prednisolone on plasma cytokine concentrations two days after enrolment reflects the absence of any significant elevation of the 11 cytokines/chemokines in the acute phase compared to convalescence . This is at odds with a body of literature indicating elevated plasma/serum cytokine concentrations are a prominent feature in the host response [9] . However a limitation of this current study , unlike previous work [10] , [11] , was that we did not measure cytokine/chemokine concentrations in serial daily plasma specimens and therefore we may have missed transient changes in particular markers . Changes in the whole-blood host gene expression profile occurred in patients that had received 2 mg/kg prednisolone ( but not 0 . 5 mg/kg ) compared to placebo-treated patients . This is to our knowledge the first ex vivo and “global’ investigation of the effect of prednisolone on the host immune response during an infectious disease . A striking finding , that was independent of the blood lymphocyte count , was the prednisolone-associated under-abundance of transcripts representing granzyme B ( GZMB ) , granzyme H ( GZMH ) , granulysin ( GNLY ) , perforin ( PRF1 ) , Ksp37 ( FGFBP2 ) and cathepsin W ( CTSW ) , each of which is associated with the secretory and cytolytic activities of T and NK cells . Taken together , the diminished abundance of transcripts encoding T and NK effector proteins might suggest impaired anti-viral cytolytic responses during high-dose prednisolone therapy . However prolonged viremia levels were not observed in prednisolone-treated patients [3] . It is plausible that the lower transcript abundance of these elements in whole blood is not biologically significant in terms of resolving infection and that other components of the immune response , such as antibodies or complement , are more important and/or compensatory . Transcripts from neutrophils , independent of the absolute neutrophil count , were more abundant in high-dose prednisolone treated patients . This included three neutrophil markers ( IL1R2 , S100A12 and ORM1 ) , all known to be corticosteroid induced [12]–[14]; interestingly these markers were also independently identified as being elevated at a similar timepoint in relation to defervescence ( fever day −2 to −3 ) in a previous study of 35 Vietnamese children who subsequently developed DSS ( [4] ) . Further studies will be required to understand if prednisolone exacerbates the functional neutrophil response in addition to its well-described demargination effects [15] [16] . There are limitations to our study . We did not measure complement activation yet there is good evidence that complement is consumed and split products generated in the course of dengue and that this might be important to pathogenesis [17] . Our sampling schedule may have missed transient but important differences in cellular gene expression signals between patients in different treatment arms . Prednisolone may also have actions in tissues that are not revealed in whole blood . Nevertheless the results of this study , coupled with the findings from the clinical trial [3] , suggest that early prednisolone therapy has little impact on the host immune response or the clinical evolution of dengue . One possible explanation is that early prednisolone therapy is “too little , too late” to attenuate the infection-driven processes that lead to the altered capillary permeability , thrombocytopenia , and haemostatic derangements . We can only speculate that even earlier treatment , or higher dose therapy , might have led to a greater prednisolone impact on the immune response and clinical/laboratory phenotype . Notwithstanding these limitations , the results described here underscore the challenge of modulating an immune response that has been driven by days of DENV replication in host tissues . More fundamentally , these results are a reminder that although immune-driven pathophysiological changes are good candidates to explain capillary permeability , the precise causal mechanisms remain poorly understood . This represents a major knowledge gap in our understanding of disease pathogenesis that also undermines development of specific therapies . | Dengue is an acute , mosquito-borne febrile illness and around 390 million cases occur annually in more than 100 countries . A host pro-inflammatory immune response is widely believed to contribute to the clinical complications that occur in some patients with dengue . Synthetic glucocorticoids , which are immunomodulatory agents commonly used in medicine , have been suggested as a therapy for dengue . We recently performed a randomized , controlled trial of early oral glucocorticoid therapy in 225 dengue cases in Vietnam , comparing a three day regimen of high ( 2 mg/kg ) or low ( 0 . 5 mg/kg ) dose prednisolone with placebo . Here , we report on immunological changes occurring during prednisolone therapy with a view to understanding the lack of clinical benefit by glucocorticoid therapy and to guide future intervention strategies for dengue . In whole-blood gene expression arrays we found 81 transcripts from 64 genes differentially abundant between high-dose prednisolone and placebo treated patients . Prominent were the genes associated with T and NK cell cytolytic functions . These results are a reminder that the mechanisms causally behind some of the complications of dengue ( e . g . altered capillary permeability ) are very poorly understood and represent a major knowledge gap in our understanding of disease pathogenesis that also undermines attempts to improve clinical management . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [] | 2013 | Corticosteroids for Dengue – Why Don't They Work? |
Schistosomiasis is one of the most significant diseases in tropical countries and affects almost 200 million people worldwide . The application of molluscicides to eliminate the parasite's intermediate host , Biomphalaria glabrata , from infected water supplies is one strategy currently being used to control the disease . Previous studies have shown a potent molluscicidal activity of crude extracts from Piper species , with extracts from Piper tuberculatum being among the most active . The molluscicidal activity of P . tuberculatum was monitored on methanolic extracts from different organs ( roots , leaves , fruit and stems ) . The compounds responsible for the molluscicidal activity were identified using 1H NMR and ESIMS data and multivariate analyses , including principal component analysis and partial least squares . These results indicated that the high molluscicidal activity displayed by root extracts ( LC50 20 . 28 µg/ml ) was due to the presence of piplartine , a well-known biologically-active amide . Piplartine was isolated from P . tuberculatum root extracts , and the molluscicidal activity of this compound on adults and embryos of B . glabrata was determined . The compound displayed potent activity against all developmental stages of B . glabrata . Next , the environmental toxicity of piplartine was evaluated using the microcrustacean Daphnia similis ( LC50 7 . 32 µg/ml ) and the fish Danio rerio ( 1 . 69 µg/ml ) . The toxicity to these organisms was less compared with the toxicity of niclosamide , a commercial molluscicide . The development of a new , natural molluscicide is highly desirable , particularly because the commercially available molluscicide niclosamide is highly toxic to some organisms in the environment ( LC50 0 . 25 µg/ml to D . similis and 0 . 12 µg/ml to D . rerio ) . Thus , piplartine is a potential candidate for a natural molluscicide that has been extracted from a tropical plant species and showed less toxic to environment .
Schistosomiasis is a tropical disease caused by parasitic worms of the genus Schistosoma and is found predominantly in areas without sanitization or clean water , including regions of Africa , South Asia and Central and South America . Presently , this disease affects an estimated 200 million people worldwide [1] , [2] . In the Americas , the only human schistosome is Schistosoma mansoni , which uses mollusks of the genus Biomphalaria as its intermediate host . One strategy used to control schistosomiasis is the management of snail populations in lakes and rivers using synthetic molluscicides . Presently , niclosamide ( Bayluscide , Bayer , Leverkusen , Germany ) is the only commercially available molluscicide that has been recommended by the World Health Organization ( WHO ) for large-scale use in Schistosomiasis Control Programs [3] . However , niclosamide is also toxic to non-target organisms . Furthermore , the application of niclosamide is costly , and this drug does not prevent recolonization of sites by surviving snails , which could lead to the selection of molluscicide-resistant populations [4]–[6] . Due to these disadvantages , the WHO is eager to find alternative drugs to facilitate schistosomiasis control; among these efforts is ongoing research on plant molluscicides , which have been considered , in several cases , as potential candidates due to their accessibility , structural diversity , low cost and possible rapid biodegradation [7] . Members of the Piperaceae family have been widely studied as a source of secondary metabolites with biological activity; among these species , Piper tuberculatum extracts , or their isolated compounds , have shown a diverse range of biological activities , such as insecticidal and fungicidal properties [8]–[11] . In a previous study , P . tuberculatum crude extracts showed molluscicidal activity against B . glabrata adult snails [12] . Additionally , many researchers have emphasized that the amides present in P . tuberculatum could be responsible for the antifungal , antitumor , antiparasitic and antiproliferative activities assigned to this species [10] , [13]–[16] . In this study , the primary compound responsible for the molluscicidal activity attributed to P . tuberculatum crude extracts from different organs was identified by 1H NMR and ESIMS data , combined with principal component analysis ( PCA ) . Additionally , partial least squares ( PLS ) analysis was performed to provide quantitative analysis and to confirm the pattern visualized in the PCA . The amides piplartine , piperine , piperlonguminine and pellitorine isolated from different organs were evaluated for molluscicidal activity on B . glabrata adults and embryos . The results obtained associating the multivariated analysis ( PCA and PLS ) with chemical composition and molluscicide activity revealed piplartine as principal amide responsible for the activity in P . tuberculatum . The acute toxicity of piplartine was also evaluated using validated ecotoxicological assays in the daphnid Daphnia similis and the fish Danio rerio .
This study was performed in strict accordance with the recommendations by the Aquatic ecotoxicology – Acute toxicity – Test with fish according to ABNT NBR ( Brazilian Assocn . of Tech . Stds . ) 15088 ( norms related to evaluation of the acute toxicity in Danio rerio and Pimephales promelas of samples from effluents , superficial or subterranean water supplies and chemical substances soluble or dispersed in water ) . The protocol was approved by Comissão de Ética no uso de animais do Instituto Butantan ( CEUAIB ) , São Paulo , Brazil ( Permit Number: CEUAIB 434/07 ) . P . tuberculatum Jacq . was collected from the Chemistry Institute at University of São Paulo , and the botanical classification was performed by Dr . Elsie Franklin Guimarães ( Instituto de Pesquisas Jardim Botânico do Rio de Janeiro ) . A voucher specimen ( Kato-169 ) was deposited in the herbarium of the same institute . The roots , stems , leaves and fruits of P . tuberculatum were dried in an oven at 45°C . The organs were then ground , and the powdered materials were extracted with methanol at room temperature ( 25–27°C ) three times and filtered . Extracts were evaporated to dryness under vacuum in a rotaevaporator and stored . A stock solution containing 1 , 000 µg/ml of each extract was prepared by suspending 10 mg of extract in 0 . 1 ml of 99 . 9% dimethylsulphoxide ( DMSO; Aldrich , Milwaukee , Wisconsin , USA ) and bringing the volume to 100 ml with dechlorinated water . Stock solutions were diluted with dechlorinated water for use in assay solutions . NMR analysis was performed using 20 mg of P . tuberculatum extracts obtained from different organs of the plant . The samples were dissolved in 800 µl CDCl3 ( 99 . 8% , Cambridge Isotopes Laboratories TM ) containing 0 . 05% of tetramethylsilane as an internal standard . The 1H NMR spectra were obtained with a Bruker DPX 200 MHz 5 mm probe . Each spectrum consisted of 256 scans and 300 k data points , with a pulse width of 8 . 0 µs ( 30° ) and relaxation delay of 2 . 0 s . All spectra data were subjected to Fourier transformation using the program MestReC ( version 4 . 8 . 6 . 0 , Mestrelab ) and had line broadening of 0 . 4 Hz . Spectra signals were integrated in regions of equal width ( 0 . 02 ppm ) corresponding to the region δ 0 . 5–10 . 00 . The signals corresponding to each amides were assigned based on published data [10] . ESIMS analyses were performed in a Quattro II triple quadrupole mass spectrometer ( MS ) ( Micromass , Manchester , UK ) . First , the samples were prepared by dissolving the crude extract in MeOH at a concentration of 1 mg/ml . The electrospray positive ionization mode was employed with a capillary voltage of 4 . 5 kV , skimmer of 50 V and nitrogen gas flow of 250 and 30 l/h . Samples were injected directly into the MS using a mobile phase flow of 50 µl/min ( MeOH∶H2O 1∶1 ) , and the data were processed with MassLynx ( Micromass ) version 3 . 2 . The molecular mass to charge ratios ( m/z ) of each amides were determined calculating the molecular formulae of each compound according to previous studies [10] . The quasi-molecular ions detected for piplartine ( C17H19NO5 , MW 317 ) were corresponding to its sodium adduct [M+Na]+ 340 , [M+Na+1]+ 341 and its fragments at m/z 221 and 222 . The m/z detected for pellitorine was 224 and 225 corresponding to its molecular mass ( C14H25NO ) . The piperine , dihydropiperine and dihydropiperlonguminine were detected by m/z 286 , 288 and 276 corresponding to their molecular mass C17H19NO3 , C17H19NO3 and C16H21NO3 , respectively . The PCA and PLS analysis were performed using 1H NMR , ESIMS and molluscicide activity data of methanolic extracts from different organs of P . tuberculatum . To minimize the potential lack of reproducibility that is associated with both the headspace generation process and the response of the mass detector , the signals generated ( raw data ) were subjected to area normalization , in which the area under the curve becomes equal for all spectra [17] . The same normalization process was applied to 1H NMR to reduce systematic variations due to intensity scaling effects resulting from variations in the total concentrations of solutes between samples . LC50 values ( µg/ml ) were submitted to the standard score process , in which the mean is subtracted from the variable values , and the resultant values are divided by the standard deviation . To perform the PCA and PLS analyses , each variable ( i . e . , each 1H NMR integrated region and intensity of m/z mass to charge ratios in the mass spectra of each sample ) was subtracted by the variable mean; this process ensured that all results would be interpretable in terms of variation from the mean . Leave-one-out cross validation was used to determine the robustness of the generated PLS model . The amides pellitorine , piperlonguminine and piperine were purified as previously described [10] . Methanolic extracts from different organs of P . tuberculatum were submitted to successive column chromatography using silica gel and a gradient of solvents at increasing polarity . The NMR data indicated that the composition of root crude extract was accounted for 92% piplartine; thus , this extract was submitted to recrystallization in MeOH to obtain pure piplartine . Consistent with common recrystallization protocols , 200 mg of crude extract from roots was dissolved in 3 ml of hot MeOH and recrystallized , yielding 140 mg of pure piplartine . Piplartine was identified using 1H NMR analysis ( Bruker DPX 200 MHz ) in CDCl3 ( 99 . 8% , Cambridge Isotopes Laboratories , Inc . ) and compared with authentic sample available [10] . Tests were performed according to the methodology recommended by the WHO [5] , [7] . Adults and egg masses of B . glabrata ( Say , 1818 ) were obtained from a Belo Horizonte population ( MG , Brazil ) and reared under laboratory conditions for several years , with fresh lettuce ad libitum to maintenance and a balanced ration during the assay . In all assays , both positive and negative controls were used to examine the susceptibility of the organisms under the assay conditions . The commercially available molluscicide niclosamide was used in the positive control group; the negative control group received dechlorinated tap water containing 1% DMSO . Snails with 10–18 mm of shell diameter were exposed to P . tuberculatum extracts and amides ( piplartine , piperlonguminine , pellitorine and piperine ) for 24 h at 25°C±2°C . After exposure , the snails were washed and observed daily for 7 days , and the death rate was recorded . For P . tuberculatum root , stem , leaf and fruit extracts , concentrations less than 1000 µg/ml were evaluated , and amides were evaluated at concentrations less than 20 µg/ml . The LC90 and LC50 values were determined . Ten animals were used per concentration and experiments were repeated three times . Plastic sheets served as the substrate for oviposition , and small circles with one egg mass attached were excised . Five egg masses at the blastula , gastrula , trocophore and veliger stages [18] were exposed to piplartine , pellitorine , piperlonguminine or piperine at concentrations below 20 µg/ml for 24 h to determine the LC90 and LC50 values . Following the exposure , the egg masses were washed and observed for mortality and malformations daily for 7 days using stereomicroscopy . Assays were repeated three times with approximately 100 embryos for each concentration .
A preliminary exploratory analysis was performed using PCA with 395 values of ion abundances from mass spectra data and 367 values of integrated areas from 1H NMR . The data from this analysis clustered into groups according to the organ of P . tuberculatum from which the extract was obtained . These data were then labeled according to their respective LC50 values . The scores plot generated from the ESIMS data ( Figure S1 ) revealed a clear difference between roots and other organs ( fruit , leaf and stem ) , which are grouped on the left side of the first principal component ( PC1 ) and account for 81% of the total variance . The corresponding loadings plot shows that quasi-molecular ions with a m/z of 221 , 222 , 340 and 341 contribute significantly to this factor ( Figure S1 ) . These quasi-molecular ions correspond to the sodium adduct and ion fragments of piplartine ( Figure S2 ) . Additionally , the second principal component ( PC2 ) explained 16% of the total variance and , together with the loadings plot , was used to assign quasi-molecular ions to pellitorine ( m/z of 224 and 225 ) , piperine ( m/z of 286 ) and dihydropiperine ( m/z of 288 ) , which are characteristic components of the fruit . The amide dihydropiperlonguminine ( m/z of 276 ) is responsible for the spatial separation between leaves and stems on the graph [10] . Scores and loadings plots from the PCA generated from integrated 1H NMR data ( Figure S3 ) revealed that PC1 ( which explained 89% of the total variance ) discriminated between root extracts ( the most active extract against B . glabrata; shown on the right side of the scores plot ) , fruit and stem extracts ( grouped into the second quadrant , near the center ) and leaf extracts ( the least active , shown in the third quadrant ) . Importantly , the two first PCs explain nearly 100% of the total variance , and the separation of the roots from the other organ groups is largely due to methoxyl signals ( δ 3 . 9 ) from piplartine ( Figure S4 ) . The fruits and stems were grouped according to the signals corresponding to the presence of pellitorine ( δ 0 . 82–0 . 92 and δ 1 . 24–1 . 30 ) . The PLS results from the ESIMS data generated significant coefficients of determination and values of internal prediction ( 0 . 98 and 0 . 83 , respectively ) ; however , only four samples were analyzed in this study . Figure 1 shows the experimental LC50 values and the predictions of the model . From the scores and correlation loadings plot in Figure S5 , it was determined that quasi-molecular ions with m/z ratios of 221 , 222 , 340 and 341 ( highly represented in root extracts ) inversely correlate with their LC50 values , which indicate that these quasi-molecular ions are the major contributors to root extract activity . PLS analysis of the 1H NMR data provides the coefficient of determination and values of internal prediction ( 0 . 98 and 0 . 83 , respectively ) , which can be visualized on a plot of measured and predicted values of LC50 ( Figure 2 ) . This result , shown in the PLS analysis using ESIMS data , corroborates with the initial findings obtained by the PCA analysis . Using the scores plot and correlation loadings plot ( Figure S6 ) , was determined that the integrated regions with values of δ 0 . 86 , 0 . 88 , 0 . 9 and 3 . 89 have a negative correlation to their LC50 values; therefore , extracts with greater values of integrated regions ( namely , root extracts ) are more active . Amides were isolated as crystals , and these amides were identified as piplartine , piperlonguminine , piperine and pellitorine by comparing their spectroscopic data with the literature [10] . Piplartine , pellitorine , piperlonguminine and piperine were initially evaluated at 20 µg/ml against B . glabrata adults . While pellitorine , piperlonguminine and piperine were not active at this concentration ( Table 2 ) , piplartine caused 100% mortality following 24 h of treatment , and the same result was observed at 10 , 9 and 8 µg/ml . Concentrations of 7 , 6 and 5 µg/ml caused 86 . 6 , 70 . 0 and 46 . 6% mortality , respectively , after 7 days . Thus , the LC90 and LC50 were 6 . 94 and 4 . 19 µg/ml , respectively ( Table 3 ) . The number of dead snails was not higher than 3 . 3% in the negative control group , which was included in the final statistical analysis . Piplartine , pellitorine , piperlonguminine and piperine were evaluated at a concentration of 20 µg/ml against the blastula , gastrula , trocophore and veliger stages . Piplartine was the only amide that caused 100% mortality to embryos at all stages . Thus , piplartine was evaluated in a dose-response assay; interestingly , sensitivity was inversely correlated with the developmental stage: the LC100 was 1 . 2 ppm for the blastula stage ( Figure 3 ) , 2 . 2 µg/ml for the gastrula , 3 . 6 µg/ml for the trocophore and 5 . 0 µg/ml for the veliger stage . The mortality rate did not exceed 2% in the embryonic negative control group for any stage . Given the effectiveness of piplartine as a molluscicide and ovicide , the acute toxicity of the compound to D . similis and D . rerio was investigated ( Table 4 ) . Piplartine was nearly five times more toxic to D . rerio than to D . similis . Lethality and immobilization were the endpoints applied to estimate LC50 to D . rerio and D . similis , respectively . General abnormalities were also recorded during the D . rerio experiments , such as erratic swimming , extended abdomen , body hemorrhaging , red pigmented spots , exophthalmia and abnormal head shape ( Figure 4 ) . These effects were transient and only occurred during the 48 h exposure period .
Former studies indicated the root extract of P . tuberculatum as most potent among the extract from different parts of this plant . The multivariated analysis using NMR and MS data indicated the influence of the compounds present in the extracts on the molluscicidal activity . The abundance of quasi-molecular ions with m/z ratios of 340 and 221 ( ESIMS data ) and a δ 3 . 89 signal ( 1H NMR spectra ) corresponding to the amide piplartine , noted for its molluscicidal activity . Indeed , piplartine has a wide range of biological activities , including cytotoxicity against cultured tumor cells and antiproliferative , anti-platelet aggregation , antifungal , insecticidal , trypanocidal , leishmanicidal and schistosomicidal properties [14] , [16] , [23] . Piplartine exhibited molluscicidal and ovicidal activities at a concentration lower than the concentration recommended by the WHO for a molluscicide candidate ( activity at less than 20 ppm ) . The amide was approximately seven times more toxic to embryos than to adult snails ( LC90 of 0 . 99 µg/ml and 6 . 94 µg/ml , respectively ) ; additionally , embryos at the blastula stage were the most sensitive to piplartine , followed by the gastrula , trocophore and veliger stages . Embryos in the early stages of development are mitotically very active and are expected to exhibit a higher sensitivity to chemical compounds [12] , [24] , [25] . In addition , embryos exposed to concentrations below the LC100 had malformations , particularly when exposed to the compound in the initial stages of development , namely , the blastula and gastrula stages . The death of the embryos is likely related to the induction of embryonic malformations because embryos with such malformations generally show delayed embryonic development and die during the spawning stage [26] , [27] ( Figure 3B ) . Despite its potential molluscicidal activity , piplartine is classified as a category 2 toxin to D . similis and D . rerio according to the Global Harmonization System [28] and a category 3 toxin ( LD50 32 . 3±3 . 4 µg/ml ) to Artemia salina [29]; the compound is , however , substantially less toxic than niclosamide ( category 1 ) to D . similis and D . rerio organisms . These results implicate piplartine as a potential natural molluscicide that acts by interfering with the life cycle of the parasitic trematode S . mansoni by eradicating the parasite's intermediate host . Piplartine not only efficiently kills adults of B . glabrata at low concentrations ( LC50 4 . 19 µg/ml ) but also leads to the lethality of the embryos inside the eggs , minimizing recolonization of the environment by the mollusks . | Schistosomiasis is a disease caused by parasitic worms of several species of genus Schistosoma that affects almost 200 million people mostly common in Asia , Africa and South America . The transmission is carried out by the parasitic larvae hosted in fresh water snails of the genus Biomphalaria . Considering the socioeconomic importance of this disease , the management of the snail population in the lakes and fresh water sources is one strategy to control the schistosomiasis . Nowadays , one synthetic compound , niclosamide , is available , but it is considered toxic to other organisms in the environment . Thus in this work piplartine was evaluated as a new active natural molluscicide extracted from a tropical plant . In addition a fish Danio rerio and a microcrustacean Daphnia similis were used as model organisms to evaluate the environmental toxicity risk of piplartine that was less toxic compared to niclosamide in the experimental conditions . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"chemistry",
"biology"
] | 2013 | Schistosomiasis Control Using Piplartine against Biomphalaria glabrata at Different Developmental Stages |
Fitness is a parameter used to quantify how well an organism adapts to its environment; in the present study , fitness is a measure of how well strains of human immunodeficiency virus type 1 ( HIV-1 ) replicate in tissue culture . When HIV-1 develops resistance in vitro or in vivo to antiretroviral drugs such as reverse transcriptase or protease inhibitors , its fitness is often impaired . Here , we have investigated whether the development of resistance in vitro to a small molecule CCR5 inhibitor , AD101 , has an associated fitness cost . To do this , we developed a growth-competition assay involving dual infections with molecularly cloned viruses that are essentially isogenic outside the env genes under study . Real-time TaqMan quantitative PCR ( QPCR ) was used to quantify each competing virus individually via probes specific to different , phenotypically silent target sequences engineered within their vif genes . Head-to-head competition assays of env clones derived from the AD101 escape mutant isolate , the inhibitor-sensitive parental virus , and a passage control virus showed that AD101 resistance was not associated with a fitness loss . This observation is consistent with the retention of the resistant phenotype when the escape mutant was cultured for a total of 20 passages in the absence of the selecting compound . Amino acid substitutions in the V3 region of gp120 that confer complete AD101 resistance cause a fitness loss when introduced into an AD101-sensitive , parental clone; however , in the resistant isolate , changes elsewhere in env that occurred prior to the substitutions within V3 appear to compensate for the adverse effect of the V3 changes on replicative capacity . These in vitro studies may have implications for the development and management of resistance to other CCR5 inhibitors that are being evaluated clinically for the treatment of HIV-1 infection .
The relative replication ability ( fitness ) of a human immunodeficiency virus type 1 ( HIV-1 ) quasispecies is governed by how individual clones fluctuate in dominance as they adapt to the host environment [1] . The relative fitness of two viruses in vitro is best estimated by head-to-head competition experiments [2] . Additional selection pressures ( e . g . , immune responses ) influence HIV-1 replication in vivo , but the fitness of HIV-1 in peripheral blood mononuclear cell ( PBMC ) cultures increased with the extent of viral diversity within a cohort of infected people , and isolates from long-term non-progressors were less fit than ones from rapid progressors [3] . A structured treatment interruption clinical trial showed that HIV-1 fitness also influences the magnitude of viremia rebound and the set point [4] . When HIV-1 develops resistance to the reverse transcriptase and protease inhibitors , its fitness is typically impaired [2 , 5] , which helps explain how beneficial effects of therapy can occur even when HIV-1 replication is incompletely suppressed and highly resistant variants are present [6–9] . Resistance to a fusion inhibitor , enfuvirtide ( T-20 ) , has been associated with an in vitro fitness reduction in some [10 , 11] , but not all , studies [12 , 13] . The instability of resistance when T-20 is discontinued suggests that resistance mutations impair fitness in vivo [14] . The CCR5 inhibitors are a new class of compounds for treating HIV-1 infection and include maraviroc ( UK-427 , 857 ) and vicriviroc ( SCH-D ) , which are now in phase II/III trials . Resistance to these inhibitors , as to any other antiviral agent [15 , 16] , will inevitably develop during therapy . We have generated several CCR5 inhibitor–resistant isolates and clones in cell culture systems [17–19] . Our best-characterized variants were derived from the HIV-1 primary isolate CC1/85 under the selection pressure of AD101 , a precursor of vicriviroc . AD101 resistance is conferred by four amino acid substitutions in the gp120 V3 region; the resistant viruses continue to enter primary CD4+ T cells via CCR5 by utilizing the AD101-CCR5 complex [17 , 19 , 20] . Resistance of CC1/85 to maraviroc occurs via a similar mechanism [21] . We have investigated whether resistance to AD101 in vitro carries a fitness cost by using a new dual-infection growth-competition assay . This system employs molecularly cloned viruses of defined phenotypes that are essentially isogenic outside the env genes under study , but which can be quantified individually using TaqMan quantitative PCR ( QPCR ) . We found that the AD101-resistant Envs were no less fit than those from the parental isolate . Accordingly , the escape mutant remained AD101-resistant during 20 passages in culture in the absence of the selecting compound .
We generated clonal , Env-chimeric parental and AD101-resistant viruses by a standard method in which the env gene of the pNL4–3 infectious molecular clone [22] is replaced by one of interest [17] . We previously used this system to make Env-chimeric viruses containing env genes from parental and AD101-resistant variants of the CC1/85 primary isolate; the chimeras possess the co-receptor usage and entry inhibitor sensitivity properties conferred by the inserted env gene [17] . We assessed the replication kinetics of a representative clone of each virus by measuring the rate of p24 antigen production in separate cultures of the same PBMC preparations . In the absence of AD101 , the wild-type clone CC1/85 cl . 7 and the AD101-resistant variant CC101 . 19 cl . 7 replicated at comparable rates and to similar extents ( Figure 1 ) . The resistant virus also replicated comparably ( ~25 ng/ml of p24 after 8 to 9 d ) in the presence of 1 μM AD101 , a concentration that completely inhibited p24 production by CC1/85 cl . 7 ( Figure 1 ) . The high-level AD101 resistance of CC101 . 19 cl . 7 was not , therefore , associated with any apparent reduction in replication efficiency . However , comparing replication rates in different mono-infection cultures does not always allow the identification of significant differences in replicative fitness [2 , 23 , 24] . Better discrimination can be achieved when both viruses replicate in the same culture . We therefore developed an assay that could accurately quantify the replication of two different clonal , Env-chimeric viruses in the same culture . The main challenge in determining the outcome of dual-infection growth-competition experiments is to devise a strategy to detect the two competing viruses without introducing any changes in other regions of the HIV-1 genome that could potentially affect fitness . We therefore created sequence tags for virus detection by altering 11 nucleotides between positions 5466 and 5493 of pNL4–3 , a region containing only the vif open reading frame . The changes create a stretch of synonymous mutations that do not affect the Vif protein sequence , do not overlap the central termination sequence [25] , and do not significantly interfere with RNA secondary structure . Probes were designed for specific annealing to either the wild-type ( vifX ) or the silent-mutated ( vifY ) vif sequences under the conditions of a TaqMan QPCR assay ( Figure 2A ) . A standard plasmid carrying tandem copies of both the vifX and vifY vif sequences was constructed to serve as a template for standardizing QPCR results when both TaqMan probes target their specific sequences . Seven 10-fold serial dilutions of a known molar concentration of this plasmid were used to generate standard curves in every QPCR experiment . The relationship between the plasmid copy number and the QPCR threshold cycle ( CT ) value was log-linear , and the plasmid standard was efficiently detected , the sensitivity limit being <50 copies of plasmid DNA per reaction ( Figure 2B; unpublished data ) . The average PCR efficiency for detection of the vifX and the vifY sequences was 91% and 95% for the six representative vifX and vifY multiplexed assays shown in Figure 2B , respectively . We inserted the env genes of the parental clone CC1/85 cl . 7 and its AD101-resistant variant CC101 . 19 cl . 7 into the vifY version of pNL4–3 ( Table 1 ) . These vifY-containing clones were used as references against which all the vifX-containing viruses were competed . We always report fitness differences in terms of the replication of the vifX virus relative to the vifY reference . A plasmid expressing NL4–3 env and containing the vifY sequence was also made to serve as a reference virus in some experiments ( Table 1 ) . We tested the assay's specificity by using the vifX and vifY probes to simultaneously detect their target sequences in DNA amplified from cells infected with only one virus . Separate PBMC cultures were infected with either the vifX or vifY versions of the CC1/85 cl . 7 or CC101 . 19 cl . 7 clones , and the vifX and vifY sequences were detected simultaneously using the multiplexed QPCR assay ( Figure 2C ) . Only the vif allele present in the virus used to inoculate each culture was ever detected; the average DNA copy number per QPCR reaction derived from the cognate probe was typically >104 , but from the non-cognate probe , <50 ( Figure 2C ) . We routinely use single-infection cultures as specificity controls and have never detected the vif allele not known to be present in these cultures ( unpublished data ) . In each set of competition assays , the two viruses were added to PBMC cultures at the same multiplicity of infection ( MOI ) , 0 . 0001 . In an otherwise identical culture , each virus was also added at a higher MOI , 0 . 0005 . In additional cultures , one virus was present at an MOI of either 0 . 0001 or 0 . 0005 , the other in 10-fold excess . The resulting average copy numbers detected per QPCR reaction were then used to determine the proportion of each virus present in each test culture on day 10 ( NL4–3 ) or day 14 ( CC1/85 cl . 7 or CC101 . 19 cl . 7 ) ( see Materials and Methods ) . The relative proportions of each virus at each condition were determined and the fitness difference ( WD ) value was calculated for each MOI , as described elsewhere [3] . Under these experimental conditions , WD is an approximation of the fold difference in the replication ability of the two viruses added to a dual-infection culture . The WD value is reported as the fitness of the virus bearing the vifX allele relative to a reference virus containing vifY . To determine if the vif tag could influence viral fitness and affect the outcome of head-to-head competitions in dual-infection cultures , we established competitions with two viruses with the same Env protein ( NL4–3 , CC1/85 cl . 7 or CC101 . 19 cl . 7 ) but a different vif allele ( i . e . , vifX or vifY ) . The WD values were always ~1; that is , there was no consistent pattern of victory or defeat ( Figure 2D ) . In addition , when vif was sequenced from cultures mono-infected for 10 d with the modified ( vifY ) version of NL4–3 , it was unchanged ( unpublished data ) . Hence , as intended , the vif tag does not affect fitness . The extent of assay-to-assay variation in this type of experiment also indicates that WD values differing by <3-fold are not meaningful ( Figure 2D; unpublished data ) ; conversely , WD values >3 or <0 . 33 indicate that two viruses present in direct-competition assays have different replication capacities . Hence , we have developed a sensitive and specific assay to quantify the replication of two different clonal viruses in the same growth-competition culture . We used the assay to determine whether there was a difference in replicative fitness between AD101-resistant and AD101-sensitive viruses . We first measured the replication of R5 clonal , Env-chimeric viruses ( Table 1 ) in PBMC mono-cultures for 14 d , using the TaqMan assay as an endpoint ( Figure 3 ) . Every virus yielded a reproducible copy number when tested at an MOI of 0 . 0005 . All three AD101-sensitive , parental CC1/85 clones were inhibited when 1 μM AD101 was added 1 h prior to infection , the average copy number reduction ranging from 33- to 480-fold for the different clones . The AD101-sensitive CCcon . 20 cl . 11 passage control clone was also strongly inhibited by AD101 , with a 990-fold reduction in copy number . In contrast , the clones from the CC101 . 19 isolate were AD101 resistant; there were no significant differences in copy numbers between the AD101-treated and control cultures , even at inhibitor concentrations of 20 μM , and at either MOI ( Figure 3; unpublished data ) . The experimentally mutated clone CC1/85 cl . 7 ( V3 ) , which contains four amino acid changes in V3 conferring complete AD101 resistance [17] , was also strongly AD101 resistant ( Figure 3 ) . The CC1/85 cl . 7 ( HP ) clone was inhibited by AD101 , but the copy number reduction was only 9-fold , substantially less than the 480-fold reduction seen with the parental CC1/85 clone from which CC1/85 cl . 7 ( HP ) was derived by introducing the H308P substitution . The H308P polymorphism , also present in CC1/85 cl . 8 , was very rapidly selected for by AD101 and confers partial resistance in a conventional , PBMC-based assay ( i . e . , a 5- to 10-fold shift in IC50 ) [17] . Thus , the pattern of resistance of the various clones in the QPCR-based assay qualitatively reproduces the phenotypes the infectious , replication-competent viruses display in a standard virus production assay [17] . Variations in how viruses replicate in single-infection cultures are not always true indications of replicative fitness differences [2 , 23 , 24] . We therefore performed competitive fitness assays in dual-infected cultures . Each of the clonal viruses listed in Table 1 that bore a vifX allele was tested in a competitive dual-infection experiment against the vifY-bearing versions of both the AD101-sensitive virus CC1/85 cl . 7 and the AD101-resistant virus CC101 . 19 cl . 7 . The competition cultures were again established at two MOIs ( 0 . 0001 and 0 . 0005 ) and at three ratios at each MOI ( 1:1 , 1:10 , 10:1 ) . The fractional proportions of each virus present in each culture were calculated , with the averages over multiple experiments used to calculate WD values for the two test viruses at each MOI [3] . In general , the AD101-sensitive clone CC1/85 cl . 6 and the engineered , partially AD101-resistant clone CC1/85 cl . 7 ( HP ) substantially out-competed each reference virus , replicating 10- to 86-fold better in the dual-infection cultures ( Figure 4A and 4B ) . In contrast , the AD101-resistant clone CC101 . 19 cl . 3 and the engineered , AD101-resistant clone CC1/85 cl . 7 ( V3 ) were consistently less fit than either reference virus , replicating 2 . 4- to 71-fold less well under all the various conditions described in Figure 4A and 4B . CC1/85 cl . 7 ( V3 ) did not replicate detectably in the presence of the CC101 . 19 cl . 7 reference virus in any experiment , so no WD could be calculated for this pairing; however , based on the known parameters of the assay , the inference is that CC1/85 cl . 7 ( V3 ) is >1 , 000-fold less fit than CC101 . 19 cl . 7 . The relative fitness of the other AD101-resistant and AD101-sensitive clones all fell within a spectrum , with some patterns evident . For example , the AD101-sensitive clone CC1/85 cl . 8 and the AD101-resistant clones CC101 . 19 cl . 7 and CC101 . 19 cl . 15 were generally fitter than CC1/85 cl . 7 ( parental ) and CCcon . 20 cl . 11 ( passage control ) ( Figure 4A and 4B ) . There was no correlation between the copy numbers produced after 14 d in single-infection cultures and the WD values determined from dual-competition cultures ( compare Figure 3 with Figure 4A and 4B ) . The paradoxical nature of this finding reinforces the unreliability of judging HIV-1 fitness based solely on single-infection cultures [2 , 23 , 24] . Because similar patterns of relative fitness differences were seen regardless of the reference virus or MOI , we averaged the fitness differences derived for each test virus in the various individual cultures to derive a replicative fitness rank order relative to an arbitrary reference point ( Figure 4C ) . The resulting fitness differences spanned an ~1 , 050-fold range; the two extremes represent viruses engineered from CC1/85 cl . 7 to contain either one ( CC1/85 cl . 7 ( HP ) , the most fit ) or four amino acid changes in V3 ( CC1/85 cl . 7 ( V3 ) , the least fit ) . The fitness spectra of the various naturally occurring AD101-sensitive and AD101-resistant clones overlapped within this range . In theory , at high MOIs , recombination between a vifX and a vifY virus with the site of recombination between the vif gene and the env gene could confound our results . If this happened , competitions in which the observed WD value was very high or very low would be poorly reproducible . We derived the data in Figure 4A and 4B by determining the average relative virus proportions from at least three experiments before calculating the WD value , thereby minimizing the influence of outliers . In Figure 4D we reanalyzed the data for one set of competitions ( CC1/85 cl . 6 vifX versus CC1/85 cl . 7 vifY ) by calculating the WD values within each individual experiment . The average WD values shown in Figure 4D were 31 and 63 at MOIs of 0 . 0001 and 0 . 0005 , respectively , and the WD values derived from Figure 4A and 4B were 10 and 63 at these respective MOIs . The similarity in outcome between the two approaches suggests that outliers do not contribute significantly to the observed fitness differences , implying that recombination is not occurring in the cultures . The above experiments suggest that AD101 resistance is not necessarily associated with a replicative fitness reduction ( Figure 4 ) . However , all of these data sets were generated with clonal , recombinant viruses that are identical outside env , and not with the original uncloned isolates that emerged in the resistance-selection experiments . Any analysis of these isolates is complicated by the likelihood that additional sequence changes occurred outside env while HIV-1 adapted to prolonged culture in PBMCs . Although we thought it unlikely that clonal bias had obscured any underlying fitness differences , we tested the original resistant isolates to gain further insights into the relative fitness of the AD101 escape mutants . The copy numbers obtained for each isolate after 14 d of PBMC culture were similar when the vifX probe was used ( Figure 5A ) , but no copies were detectable with the vifY probe ( unpublished data ) . Thus , these isolates can be used as test viruses in competitions against vifY-tagged references . As expected , replication of the CC1/85 isolate , but not the CC101 . 19 isolate , was impaired by 1 μM AD101 ( Figure 5A ) . The average copy numbers calculated for both isolates were somewhat lower than those seen with the corresponding clones under similar conditions ( compare Figure 5A to Figure 3 ) . This is not surprising since the latter are based on the genomic backbone of NL4–3 , a clone adapted for optimal growth in cell culture [22] . Having shown that the QPCR assay can be used to study the replication of the isolates on which the clones were based , we examined the fitness of the CC1/85 and CC101 . 19 isolates relative to the vifY-tagged reference clones CC1/85 cl . 7 and CC101 . 19 cl . 7 in head-to-head competitions without AD101 . The outcome of these assays allowed us to determine the comparative fitness of the two isolates . The replicative fitness of each isolate was always less than that of the corresponding reference clone ( WD < 1 , Figure 5B ) . This is likely to be a consequence of using pNL4–3 as a background vector into which env genes are inserted to make chimeric viruses . NL4–3 has probably acquired several mutations outside env that improve its fitness in vitro , compared to primary isolates like CC1/85 [22] . When CC1/85 cl . 7 was used as the reference virus , the AD101-resistant isolate CC101 . 19 was found to be 89- and 390-fold more fit than the parental CC1/85 isolate at the MOIs of 0 . 0005 and 0 . 0001 , respectively . With CC101 . 19 cl . 7 as reference , CC101 . 19 was 85- and 42-fold more fit than CC1/85 at the same two MOIs . The average of these four fitness difference estimates is ~150 . Hence , if any sequence changes outside env did arise during culture of CC101 . 19 , they improved its fitness . A conservative conclusion is that the development of AD101 resistance by CC101 . 19 has not come at the price of a dramatic fitness loss . If the acquisition of AD101 resistance caused a replicative fitness loss in PBMC cultures , the resulting AD101-resistant isolate would revert to sensitivity when the cultures were continued without the inhibitor . Because the above studies , using both clones and isolates , implied that AD101 resistance did not create an unfit virus , we hypothesized that any such reversion would be slow , and might not even occur . We reported that when the AD101-resistant isolate generated after 22 passages in the presence of AD101 ( CC101 . 22 ) was cultured in PBMCs without AD101 for nine additional passages ( CC101 . 22R9 ) , it remained highly AD101 resistant [19] . We therefore returned the CC101 . 22R9 isolate to culture in PBMCs for 11 additional passages , hence 20 in total , without AD101 . The AD101 sensitivity profiles of the CC1/85 , CC101 . 19 , and CC101 . 22R9 isolates , and ones receiving 10 , 15 , or 20 passages without AD101 ( CC101 . 22R10 , CC101 . 22R15 , and CC101 . 22R20 , respectively ) , were determined ( Figure 6 ) . All the isolates from the AD101-free culture remained completely resistant to AD101 concentrations as high as 5 μM , whereas the parental CC1/85 isolate was inhibited in the 1–10 nM range . Hence the AD101-resistant phenotype is highly stable , which is consistent with the conclusion that it is not associated with any significant decrease in replicative fitness in PBMC cultures . Finally , we sought to determine whether AD101 affected the replicative fitness of AD101-resistant viruses; these viruses are clearly drug-resistant and not drug-dependent , but does a dose of the drug give them a boost ? The standard QPCR assay system cannot be used to determine the relative fitness of a single virus in the presence and absence of an inhibitor . Instead , we used a vifY-tagged version of the X4 clone NL4–3 as a reference virus that is unaffected by AD101 , a CCR5 inhibitor , because it uses CXCR4 for entry . The vifY-tagged NL4–3 virus was competed against the AD101-resistant vifX-tagged CC101 . 19 cl . 7 clone in cultures either lacking or containing AD101 at a concentration ( 20 μM ) that saturates CCR5 ( Figure 7A ) . Because NL4–3 replicates very efficiently in PBMC culture , we added AMD3100 ( 20 nM ) , a CXCR4 inhibitor specific to this virus ( i . e . , one that would not affect the R5 virus CC101 . 19 cl . 7 ) , to reduce its replication rate to a level comparable to that of its competitor in the dual-infection cultures . This AMD3100 concentration had no effect on the copy numbers of CC101 . 19 cl . 7 determined in single-infection cultures , nor did AD101 affect NL4–3 replication ( unpublished data ) . Under these conditions , in the absence of AD101 , the fitness of CC101 . 19 cl . 7 was indistinguishable ( WD ~1 . 5 ) from that of NL4–3 at either MOI , indicating that AMD3100 had indeed adjusted the replication rate of NL4–3 to the intended degree ( Figure 7A; unpublished data ) . In the presence of 20 μM AD101 , the corresponding fitness differences were 12 and 49 at MOIs of 0 . 0001 and 0 . 0005 . Thus , at these MOIs , CC101 . 19 cl . 7 is 8 . 1- to 32-fold fitter when AD101 ( 20 μM ) is present ( Table 2 ) ; AD101 therefore modestly enhances the replicative fitness of the CC101 . 19 cl . 7 clone . To see whether the enhancing effect of AD101 was unique to CC101 . 19 cl . 7 , dual-infection experiments were performed in which the vifX vif-bearing clones CC101 . 19 cl . 3 and CC101 . 19 cl . 15 were competed against vifY-tagged CC101 . 19 cl . 7 , with AD101 again either absent or present at 20 μM . We showed above that CC101 . 19 cl . 15 was fitter than CC101 . 19 cl . 7 in the absence of AD101 , whereas CC101 . 19 cl . 3 was less fit ( Figure 4A and 4B ) . When AD101 was present , the existing fitness differences between CC101 . 19 cl . 15 or CC101 . 19 cl . 3 and CC101 . 19 cl . 7 were only minimally altered . For example , the fitness of CC101 . 19 cl . 15 relative to CC101 . 19 cl . 7 was increased by 1 . 9-fold at an MOI of 0 . 0001 , but at an MOI of 0 . 0005 it was decreased by 1 . 7-fold; such changes are insignificant ( Figure 7B; Table 2 ) . Similarly , when AD101 was present , the fitness of CC101 . 19 cl . 3 relative to the CC101 . 19 cl . 7 reference virus was increased by 2 . 8- and 1 . 3-fold at the low and high MOIs . The AD101-resistant CC1/85 cl . 7 ( V3 ) virus was found not to replicate detectably when in competition with the CC101 . 19 cl . 7 reference virus , whether AD101 was present or not; no fitness differences could therefore be calculated for this pairing ( unpublished data ) . Overall , we conclude that AD101 did not significantly affect pre-existing fitness differences between the various CC101 . 19-derived clones . A corollary is that , like CC101 . 19 cl . 7 , the fitness of the AD101-resistant clones CC101 . 19 cl . 15 or CC101 . 19 cl . 3 is also increased when AD101 is present during a 14-d PBMC culture ( summarized in Table 2 ) . Thus , all the AD101-resistant viruses we tested are not just resistant to the selecting compound , they actually replicate more efficiently in its presence , albeit only to a modest extent that may not always be apparent in conventional PBMC assays .
We have shown that an R5 primary isolate , CC1/85 , becomes highly resistant when serially passaged in PBMC culture with increasing concentrations of AD101 [19] . A resistant isolate from passage 19 , CC101 . 19 , remained CCR5-dependent for replication in PBMC culture . Env sequence changes were necessary and sufficient to confer AD101 resistance , which developed in two stages: a single amino acid polymorphism in V3 , H308P , was selected during the earliest passages with AD101 and conferred modest ( 5- to 10-fold ) resistance; three de novo mutations , also in V3 ( K305R , A316V , G321E ) , arose later to produce a highly ( >20 , 000-fold ) resistant virus [17] . The H308P polymorphism increases the efficiency with which HIV-1 can use low CCR5 levels for entry , whereas the later changes create gp120 proteins that can recognize the inhibitor-bound conformation of CCR5 [17 , 19 , 20] . When we commenced this study , several lines of evidence suggested the AD101 escape mutants might not have reduced fitness relative to the parental and passage control isolates: ( i ) The resistant phenotype as well as the underlying genetic changes were stable , in the short-term , after the inhibitor-selection pressure was withdrawn; ( ii ) the replication rates of the CC1/85 and CC101 . 19 clones in primary CD4+ T cells were similar; and ( iii ) the CC101 . 19 clones did not require AD101 for replication [17 , 19] . Our initial experiments supported these findings ( Figure 1 ) . We therefore used the dual-competition assay to compare three env clones from both the parental CC1/85 and the AD101-resistant CC101 . 19 isolates in the vifX background , with vifY-tagged reference viruses containing env genes from either CC1/85 cl . 7 or CC101 . 19 cl . 7 . Three other env genes inserted into vifX-tagged viruses were similarly studied: a CC1/85 env engineered to contain the V3 H308P change ( CC1/85 cl . 7 ( HP ) ) ; a derivative of the same env containing all four V3 changes that confer complete AD101 resistance ( CC1/85 cl . 7 ( V3 ) ) ; and an env cloned from the 20th-passage control isolate ( CCcon . 20 cl . 11 ) ( Table 1 ) . Using WD values averaged over MOIs and reference viruses , with WD = 1 representing an arbitrary reference point for comparisons , the three env clones from CC1/85 spanned an ~50-fold range of fitness differences , as did the three CC101 . 19 clones . A critical point is that these ranges overlapped each other , implying that AD101 resistance in vitro caused no intrinsic fitness loss ( Figure 4C ) . The two extremes in the fitness range are represented by the two engineered viruses . The single H308P change allowing CC1/85 cl . 7 to use free CCR5 more efficiently confers an ~50-fold fitness increase when AD101 is absent . Despite this , the minor H308P variant did not expand in frequency in the passage control culture , for reasons that have yet to be understood [17 , 20] . The three additional V3 changes that then permit use of the AD101-bound form of CCR5 produce a virus ~20-fold less fit than the parental CC1/85 cl . 7 clone . Compared , then , to the partially resistant single mutant , the three later-arising substitutions responsible for complete resistance cause an ~1 , 000-fold fitness loss . We note , however , that these engineered viruses are isogenic elsewhere in env , whereas the original uncloned isolates and the derivative clones are not . Hence , the fitness loss associated with the three later-arising V3 substitutions may be compensated for by additional changes elsewhere in gp120 ( or even in gp41 ) ; we previously noted the possible importance of such compensatory changes [17 , 18] . However , it is unlikely that any compensatory changes arose directly from a selective pressure for increased fitness , because V3 is the only env region undergoing selection after the fourth passage with AD101 [17] . Rather , we believe the V3 changes arose in an Env context that was selected for early in the escape process , and in that particular environment they cause no fitness reduction . These scenarios contrast markedly with the fitness effects associated with the pathways to protease and reverse transcriptase inhibitor resistance; there , primary resistance mutations affecting the action of the drug typically confer a fitness loss , but are compensated for by secondary mutations that eventually restore fitness [2] . We passaged a control isolate in PBMCs in the absence of AD101 [36] . The resulting 19th passage isolate , CCcon . 19 , and a related clone , CCcon . 20 cl . 11 , became more sensitive to soluble CD4 without any change in AD101 sensitivity , due to slowly accumulating sequence changes in the V1/V2 region [36] . We hypothesized that these passage control viruses had mutated to improve their fitness in the absence of counter-selection by the neutralizing antibodies that are present in vivo [36] . We now show that the fitness of CCcon . 20 cl . 11 falls within the range spanned by the various CC1/85 and CC101 . 19 clones . As the V1/V2 changes accumulated slowly , any fitness differences they do impart may be too minor to be quantified , and/or that more CCcon . 20 env clones would need to be studied to detect them . Genetic and phenotypic changes during prolonged cultures of primary isolates in PBMCs have now been described by others [21 , 37] . We have studied the fitness of HIV-1 clones that are identical outside env , whereas the uncloned isolates from our original resistance-selection experiments may have had other sequence changes influencing their phenotypes . Analysis of these isolates is complicated by any sequence changes arising as the virus adapts to extended culture in PBMCs , rather than due to the effect of the CCR5 inhibitor . Even so , we considered that studying the isolates could still be informative about gross fitness changes , and might alleviate concerns about whether the limited number of clones available for study creates any clonal bias . Compared to either reference virus ( CC1/85 cl . 7 or CC101 . 19 cl . 7 ) , the AD101-resistant CC101 . 19 isolate had , on average , a WD value ~150-fold higher than the parental CC1/85 isolate ( Figure 5B ) . Even though we cannot exclude an influence of the culture process on fitness , it does seem reasonable to conclude that AD101 resistance is not associated with a fitness loss at the level of either the uncloned isolates or the derivative clones . A corollary of this conclusion is that AD101 resistance should remain stable during extended in vitro passage in the absence of the inhibitor . We showed that CC101 . 22 , a resistant isolate from passage 22 of the original AD101 selection culture , retained its phenotype during nine additional passages in the absence of AD101 [19] . Moreover , the four V3 changes responsible for AD101 resistance also remained stable over this time [17] . We have now cultured this isolate without AD101 for an additional 11 passages , 20 in all , a total comparable with the 22 passages the virus had received with AD101 present . Isolates from these cultures remained highly AD101 resistant ( Figure 6 ) , strongly supporting the conclusion that complete resistance to AD101 does not confer a significant fitness loss . In contrast , another CCR5 inhibitor–resistant isolate , generated under the selection pressure of SCH-C in PM1 cells and derived from the subtype G strain JV1083 , did gradually revert to sensitivity when it was passaged for prolonged periods without SCH-C . The genetic pathway to reversion was similar , but not identical , to the pathway to resistance , with V3 changes predominating in both instances ( J . Riley , L . Wojcik , W . Huang , S . Xu , S . Kuhmann , et al . , unpublished data ) . In another study , two maraviroc-resistant isolates derived from the primary isolates RU570 and CC1/85 were generated in PBMCs [21] . The resistance of both viruses was associated with V3 changes , but even when the same CC1/85 isolate was used in different experiments , different genetic changes conferred resistance [21] . After 20 passages without maraviroc , the resistant isolates partially ( CC1/85 ) or completely ( RU570 ) reverted to sensitivity [21] . Overall , HIV-1 can follow different genetic routes to resist small molecule CCR5 inhibitors , pathways perhaps associated with varying degrees of fitness loss . How resistance to small molecule CCR5 inhibitors is manifested can be both assay- and cell type–dependent [20] . When the replication-competent Env-chimera CC101 . 19 cl . 7 was tested in a PBMC-based replication assay , the AD101-related compound vicriviroc caused no inhibition , but rather modestly but consistently enhanced p24 production . In contrast , when the same env gene was used in a single round , Env-pseudotype assay involving PBMCs , vicriviroc was partially inhibitory , the extent of inhibition plateauing at ~25% . When the same Env-pseudotype virus was studied in U87-CD4/CCR5 cells , an inhibition plateau was again observed , but now at ~90% inhibition . We have concluded that CC101 . 19 cl . 7 can enter cells by using both the inhibitor-bound and inhibitor-free forms of CCR5 , and that the height of the plateau , when it occurs , is a measure of the relative efficiency with which the two CCR5 configurations are used , a parameter influenced by the cell type [20] . Similar conclusions have been drawn regarding maraviroc resistance [21] . To gain further insights , we investigated whether entry via the AD101-bound form of CCR5 influences the fitness of the AD101-resistant clones , using a saturating AD101 concentration ( 20 μM ) to ensure that no free receptor remained on the target cells . All three CC101 . 19-derived clones were fitter in the presence of AD101 than they were in its absence ( Figure 7; Table 2 ) . This is mechanistically informative , because HIV-1 entry is not enhanced by vicriviroc or AD101 in single-round assays ( [20]; unpublished data ) . As the fitness assays involve 14 d of replication in PBMC culture , it seems likely that the fitness increase conferred by AD101 , like the replication enhancement caused by vicriviroc , only arises during multiple replication cycles; we are now investigating why this occurs . One possible factor is that CCR5 ligands like AD101 and vicriviroc can upregulate both CCR5 and its chemokine ligands MIP-1β and RANTES in PBMC cultures , events irrelevant for single-cycle entry assays [20] . The interplay between the various CCR5 ligands ( chemokines , small molecule inhibitors , and gp120 ) during a multi-cycle replication process is likely to be complex . Two small molecule CCR5 inhibitors , maraviroc and vicriviroc , are now in advanced clinical trials as therapeutics , and cause significant ( ~1 . 5 log10 ) viral load reductions [38 , 39] . As with all HIV-1 therapies , resistance development must be anticipated [15 , 16] . The existing classes of antiretrovirals sometimes provide continued therapeutic benefit even when resistance arises , and the drug-sensitive phenotype reemerges when the selecting drug is withdrawn . These events are hallmarks of a fitness cost to resistance , suggesting that fitness decreases measured in vitro do have clinical relevance [2 , 14 , 27] . There are real limitations to in vitro assays when predicting what might happen to HIV-1 in vivo , particularly for studies involving the envelope glycoproteins [16] . These proteins face multiple evolutionary pressures in vivo , conferred by neutralizing antibodies , changes in the numbers and types of target cells , alterations in the number and nature of their co-receptors , and the presence of both natural ( chemokine ) and unnatural ( drugs , when present ) co-receptor ligands . Most of these selection pressures are absent in vitro , or else present as uncontrolled variables ( e . g . , production of chemokine ligands in PBMC cultures noted above ) . There are changes in the in vitro sensitivity of CCR5 inhibitor-resistant viruses to certain neutralizing monoclonal antibodies ( P . Pugach and J . Moore , unpublished data ) , but here again the in vivo situation will be more complex . If , however , our central observation—that some CCR5 inhibitor-resistant viruses are no less fit than parental strains—is directly relevant to what might happen in vivo , the implication is that such viruses would persist for prolonged periods after CCR5 inhibitor therapy is discontinued . Although no inhibitor would then be present to occupy CCR5 , the resistant viruses are not “drug-dependent” , but can use the free receptor . One theoretical concern is that the alteration in how resistant viruses utilize CCR5 could enable them to use CCR5 conformational variants differently or more efficiently , opening up additional target cells to infection . HIV-1 variants with higher CCR5 affinities and lower sensitivities to CCR5 ligands can arise naturally during HIV-1 infection , because of the drop in the average levels of CCR5 available on target cells [40–44]; CCR5 inhibitor-resistant variants might , in principle , have a similar advantage over wild-type viruses under certain conditions . Nonetheless , naturally occurring viruses with complete resistance to CCR5 inhibitors have rarely been observed ex vivo or in vivo . Thus , whatever replication advantage might be conferred by Env configuration changes allowing use of the CCR5-inhibitor complex , other counter-selection pressures presumably prevent this from occurring naturally . Ex vivo analyses of viruses derived from long-term clinical studies of CCR5 inhibitors , together with the general experience gained from these trials , might clarify some of the above issues .
AD101 ( SCH-350581 ) was provided by Julie Strizki ( Schering-Plough Research Institute , Kenilworth , New Jersey , United States ) . The small molecule CXCR4 inhibitor AMD3100 was obtained from the National Institutes of Health ( NIH ) AIDS Research and Reference Reagent Program . Viruses used in the fitness studies include the parental R5 isolate ( CC1/85 ) , the AD101-resistant isolate ( CC101 . 19 ) , and three clonal , NL4–3/Env-chimeric infectious viruses derived from each of these isolates ( Table 1 ) . Mutants of the one of the CC1/85 clones with varying degrees of resistance include a virus with one amino acid change in V3 ( H308P ) conferring partial AD101 resistance , and one with all four V3 changes ( K305R , H308P , A316V , G321E ) necessary and sufficient for full resistance [17]; these clonal viruses are designated “CC1/85 cl . 7 ( HP ) ” and “CC1/85 cl . 7 ( V3 ) , ” respectively . Their construction and properties have been described [17 , 20] . A clone from the passage 20 control isolate ( CCcon . 20 cl . 11 ) [36] was also studied . Clonal proviruses containing the mutated vif sequence were constructed as below . Infectious virus stocks were prepared by transient transfection of 293T cells with pNL4–3/env plasmids using Lipofectamine 2000 ( Invitrogen , http://www . invitrogen . com ) according to the manufacturer's instructions , as described [17 , 20 , 36] . Stocks of the CC1/85 , CC101 . 19 , and CC101 . 22R9 isolates were prepared as described [19] . All infectious stocks were stored in aliquots at −80 °C . The titers ( 50% tissue culture infectious dose [TCID50] ) of all stocks were determined in PBMC culture by standard methods [45] . CD8-depleted PBMC cultures were purified and stimulated as described [17 , 18 , 20] . PBMCs from between two to four random donors were mixed in equal proportions to reduce donor-dependent effects on viral replication . After stimulation , PBMCs were maintained in lymphocyte medium ( RPMI 1640 + 10% FBS + 2 mM L-glutamine + 100 U/ml IL-2 ) . The CC101 . 22R9 isolate [19] was returned to culture for 11 additional passages in activated PBMC culture without AD101 . Passages were performed weekly by adding a 5-ml aliquot of culture supernatant and cells from the previous passage to 15 ml of freshly stimulated PBMCs at 2 × 106 cells/ml . The remaining supernatant was filtered and frozen in 1-ml aliquots at −80 °C for drug sensitivity testing . To determine the growth kinetics , 100 TCID50 of the test virus were used to inoculate 2 × 105 cells in 200 μl of lymphocyte medium ( MOI = 0 . 0005 ) in replicate wells of a 96-well plate . After incubation at 37 °C for the indicated times , duplicate wells were harvested and virus was inactivated by addition of 1% Empigen BB detergent ( Sigma-Aldrich , http://www . sigmaaldrich . com ) . All samples from a single experiment were tested simultaneously for p24 antigen using an in-house ELISA [46] . Replication rates were determined by plotting the increase in p24 antigen over time . The sensitivities of isolates CC1/85 , CC101 . 19 , CC101 . 22R9 , CC101 . 22R10 , CC101 . 22R15 , and CC101 . 22R20 to inhibition by AD101 were assayed as described previously [17] . Briefly , PBMC cultures were infected in 96-well plates , as described above , in the presence of varying AD101 concentrations , then p24 concentrations were measured after 7 d . The first step in the construction of the mutated pNL4–3 vif tracking vector , pNL4–3 vifY , involved subcloning of the region of pNL4–3 located between the SphI ( position 1448 ) and EcoRI ( position 5744 ) restriction sites into pNEB193 ( New England Biolabs , http://www . neb . com ) . The resulting pNEB-NL4–3 was subjected to site-directed mutagenesis using the Stratagene ( http://www . stratagene . com ) Quickchange I kit and the following primers , as specified by the manufacturer: HIV Vif Sense ( 5′-CCA TAG AAT GGA GGA AAA AGA GAT ATA GC-3′ ) , HIV Vif AntiS ( 5′-GTT GCA GAA TTC TTA TTA TGG C-3′ ) , HIV VifY B sense ( 5′-AGC TTG CAA TAT CTA GCG TTG GCA GCA TTA ATA AAA CCA AAA CAG-3′ ) , and HIV VifY B AntiS ( 5′-CAA CGA TAG ATA TTG CAA GCT TCC TAC CTT GTT ATG TCC TGC-3′ ) to form pNEB-NL4–3-vifY . The mutated SphI to EcoRI fragment was sub-cloned back into pNL4–3 to form pNL4–3-vifY and sequenced using the Vif Seq S primer ( 5′-TGG CAA GTA GAC AGG ATG AGG A-3′ ) . To construct the CC1/85 cl . 7 and CC101 . 19 cl . 7 reference proviruses with the vifY allele , the env sequence was removed from the appropriate pNL4–3-env plasmid using the EcoRI and XhoI restriction sites , then ligated into the corresponding sites in pNL4–3-vifY . We verified that the vifY sequence in vif did not revert to the wild-type ( vifX ) form under the conditions of growth-competition assays ( unpublished data ) . The QPCR standard was created by first cloning the wild-type NL4–3 vif ( VifX ) sequence into the pCR2 . 1 TOPO vector ( Invitrogen ) , using the HIV Vif Sense and HIV Vif AntiS primers described above , to form pCR2 . 1-vifX . The wild-type vif sequence was then sub-cloned from pCR2 . 1-vifX into pNEB-NL4–3-vifY using the EcoRI restriction site . The plasmid , pVifStd , contained tandem copies of both the vifX and vifY alleles . The concentration of this standard was quantified by UV absorbance spectrophotometry . Ten-fold serial dilutions were used as templates to generate standard curves in the real-time TaqMan PCR assays . These were performed in 48-well plates seeded with 2 × 106 stimulated PBMCs in 0 . 8 ml . The two viruses under evaluation were added to the target cells at individual MOIs of 0 . 0001 or 0 . 0005 , which are generally accepted to be low enough to prevent recombination [35] . Three competitions were established for each pair at each MOI , using different ratios of the input viruses ( 1:1 , 1:10 , and 10:1 ) . When appropriate , inhibitors were incubated with target cells for 1 h before virus addition . To limit the inherent variability of PBMC replication assays , each data point was derived from duplicate cultures on the same plate , and all experiments were performed at least thrice . Additional controls included in each experiment entailed mono-infections of PBMCs with each virus separately at different MOIs ( low or high ) in the presence or absence of various AD101 concentrations . Competition cultures involving X4 and R5 viruses were maintained for 10 and 14 d , respectively . The cells were then harvested , washed once with phosphate-buffered saline ( PBS ) , and pelleted for DNA extraction using the QIAamp DNA Blood Mini Kit , according to the manufacturer's instructions ( Qiagen , http://www . qiagen . com ) . Competition experiments were analyzed using a multiplexed TaqMan PCR assay . First , 5 μl of extracted DNA were subjected to a brief , external PCR amplification reaction in a final volume of 25 μl , using the Vif subtype B S ( 5′-TGG CAG GTG ATG ATT GTG TGG CA-3′ ) and Vif subtype B AntiS ( 5′-GGT CTT CTG GGG CTT GTT CCA TCT-3′ ) primers and AccuPrime SuperMix II , as specified by the manufacturer ( Invitrogen ) . DNA amplifications were performed under the following cycling conditions: 1 cycle at 94 °C for 2 min; 10 cycles at 94 °C for 30 s , 55 °C for 30 s , and 68 °C for 45 s; and 1 cycle at 68 °C for 2 min . The reaction product was then purified using the QIAquick PCR Purification Kit ( Qiagen ) . The TaqMan assay utilizes probes to differentiate between the two forms of the NL4–3 backbone ( vifX and vifY ) . The probes are labeled with different fluorescent markers at their 5′-ends . In solution , prior to binding of the primers and probes to their target sequences , and during annealing , the quenching agent on the opposite end of the probes dampens the fluorophores' signals . However , during the PCR extension step , the probe becomes vulnerable to the 5′ to 3′ exonuclease activity of Taq DNA polymerase; freed from proximity to the quencher , the fluorophore can now emit a fluorescence signal that is amplified logarithmically by successive rounds of PCR and monitored in real-time . The number of PCR cycles required to reach a given threshold fluorescence value is inversely proportional to the logarithm of the input quantity of dsDNA containing the probe sequence . By quantifying the fluorescence output from both probes , each competing viral genome can be quantified simultaneously from a single sample . In the assay , the Brilliant QPCR Master Mix ( Stratagene ) was used with the primers Vif beta S ( 5′-AGT TAG TCC TAG GTG TGA-3′ ) and Vif beta AS ( 5′-TCC ATC TGT CCT CTG TCA-3′ ) and the reference dye ROX , according to the manufacturer's specifications . Included in the reaction were the vifX and the vifY probes ( sequences are provided in Figure 2A ) that were labeled with Cy5 and Black Hole Quencher 2 ( vifX ) , and with FAM and Black Hole Quencher 1 ( vifY ) at the 5′- and 3′-ends , respectively . The reactions were run on a Stratagene Mx4000 machine using the cycling conditions: 95 °C for 10 min , then 95 °C for 30 s and 55 °C for 1 min for 40 cycles . All reactions were performed in triplicate , including the seven serial dilutions of the standard DNA template , ( ranging from 5 × 107 to 5 × 101 copies ) , as well as a negative control ( no-template ) . The mean values of the measured numbers of copies per reaction were then determined using the Stratagene Mx4000 software version 4 . 20 , and used for further analysis . The slope and correlation coefficient of each standard curve were calculated based on the average threshold cycle ( CT ) values measured in triplicate for each dilution point . The PCR efficiency , E , was computed as ( 10−1/s − 1 ) × 100% , where s is the slope of the generated standard curve . Fitness differences ( WD ) were calculated for each clonal virus based on its relative production in head-to-head competition , as previously described [3] . Initially , the copy numbers per QPCR reaction of the vifX and vifY alleles in each infection well were determined from the triplicate QPCR reactions for each infection condition; copy numbers for the duplicate infection conditions were then averaged . The copy numbers of the 10:1 and 1:10 infections were weighted by their initial proportions in the inoculum , and the weighted copy numbers from the three conditions ( 1:1 , 10:1 , and 1:10 ) were averaged to give a single copy number per experimental condition ( nvifX and nvifY ) . The proportion of each virus at each condition ( wvifX and wvifY ) was then determined from the ratio of the copy number of that virus ( nvifX or nvifY ) to the total copy number ( wvifX = nvifX/[nvifX + nvifY] and wvifY = nvifY/[nvifX + nvifY] ) . The fitness difference ( WD ) for an individual experiment was calculated by determining the ratio of these proportions , always comparing the test vifX virus to the reference vifY virus ( WD = wvifX/wvifY ) . Thus , when WD > 1 , the vifX virus has won the competition , and when WD < 1 , the vifY virus has prevailed . When the WD value is based on more than one experiment , the proportions ( wvifx and wvifY ) from each experiment were averaged before calculating the WD value , to avoid giving undue weight to any single experiment . For WD values calculated from single experiments where the vifX and vifY viruses had the same env gene ( Figure 2D ) , the greatest deviation from the expected WD value of 1 was 0 . 35 . Thus , we conservatively conclude that WD values >3 or <0 . 33 indicate that two viruses in direct competition differ significantly in their replication capacities . If the minimum copy numbers per QPCR reaction produced in mono-infected cultures was ~5 × 104 ( it was frequently higher ) and the minimum copy numbers that could be detected is 50 ( it was probably lower ) , then we would be able to detect fitness differences of up to 1 , 000-fold . Hence , using these conservative estimates , we conclude that the working range for the assay is WD values of 0 . 0001 to 1 , 000 .
The GenBank ( http://www . ncbi . nlm . nih . gov/Genbank/index . html ) accession numbers for CC1/85- and CC101 . 19-derived env clones used in this study are CCcon . 20 cl . 11 ( AY357537 ) , CC1/85 cl . 6 ( AY357338 ) , CC1/85 cl . 7 ( AY357341 ) , CC1/85 cl . 8 ( AY357344 ) , CC101 . 19 cl . 3 ( AY357466 ) , CC101 . 19 cl . 7 ( AY357465 ) , and CC101 . 19 cl . 15 ( AY357468 ) . The GenBank accession number of the plasmid pNL4–3 containing the NL4–3 provirus is AF324493 . | When human immunodeficiency virus type 1 ( HIV-1 ) develops resistance in vitro or in vivo to antiretroviral drugs such as reverse transcriptase or protease inhibitors , its replicative fitness is often impaired ( i . e . , it grows at a lower rate or to a lesser extent than the parental , inhibitor-sensitive virus ) . Here , we investigated whether resistance development in vitro to a new class of antiretroviral drugs , the CCR5 inhibitors , has an associated fitness cost . These inhibitors , exemplified by the AD101 compound , are small molecules that bind CCR5 , a cell surface protein that HIV-1 uses as a co-receptor during the process of cellular entry . We performed direct-competition assays using sequence-labeled , clonal viruses derived from an AD101 escape mutant , the AD101-sensitive parental isolate , and a passage control isolate , and found that AD101 resistance was not associated with a fitness loss . Furthermore , when the escape mutant was cultured for 20 passages without AD101 , it remained resistant . Specific amino acid substitutions conferring AD101 resistance did cause a fitness loss when experimentally introduced into a sensitive clone , but in the naturally selected escape mutant they are probably compensated for by other changes . This work may help understand the development and management of resistance to CCR5 inhibitors now being evaluated clinically to treat HIV-1 infection . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods",
"Supporting",
"Information"
] | [
"viruses",
"infectious",
"diseases",
"virology",
"in",
"vitro",
"homo",
"(human)"
] | 2007 | Escape of HIV-1 from a Small Molecule CCR5 Inhibitor Is Not Associated with a Fitness Loss |
A critical problem in biology is understanding how cells choose between self-renewal and differentiation . To generate a comprehensive view of the mechanisms controlling early hematopoietic precursor self-renewal and differentiation , we used systems-based approaches and murine EML multipotential hematopoietic precursor cells as a primary model . EML cells give rise to a mixture of self-renewing Lin-SCA+CD34+ cells and partially differentiated non-renewing Lin-SCA-CD34− cells in a cell autonomous fashion . We identified and validated the HMG box protein TCF7 as a regulator in this self-renewal/differentiation switch that operates in the absence of autocrine Wnt signaling . We found that Tcf7 is the most down-regulated transcription factor when CD34+ cells switch into CD34− cells , using RNA–Seq . We subsequently identified the target genes bound by TCF7 , using ChIP–Seq . We show that TCF7 and RUNX1 ( AML1 ) bind to each other's promoter regions and that TCF7 is necessary for the production of the short isoforms , but not the long isoforms of RUNX1 , suggesting that TCF7 and the short isoforms of RUNX1 function coordinately in regulation . Tcf7 knock-down experiments and Gene Set Enrichment Analyses suggest that TCF7 plays a dual role in promoting the expression of genes characteristic of self-renewing CD34+ cells while repressing genes activated in partially differentiated CD34− state . Finally a network of up-regulated transcription factors of CD34+ cells was constructed . Factors that control hematopoietic stem cell ( HSC ) establishment and development , cell growth , and multipotency were identified . These studies in EML cells demonstrate fundamental cell-intrinsic properties of the switch between self-renewal and differentiation , and yield valuable insights for manipulating HSCs and other differentiating systems .
Stem cells are characterized by the ability to both self renew and undergo cell differentiation . Understanding the mechanisms that control the switch between renewal and differentiation is a fundamental and important problem in stem cell biology . It is likely that many key components including signaling molecules and transcription factors are involved in this process . Although a few key components that influence the switch have been found [1] , [2] , [3] , [4] , it likely that many others exist . Identification of such components and elucidation of how they function is critical for understanding this developmental switch . Blood-forming hematopoietic stem cells ( HSCs ) are one of the best-characterized stem cells , and are valuable for studying self renewal and differentiation [5] , [6] . HSCs exist in adult bone marrow , and can self-renew and differentiate into more than ten distinct mature blood cell lineages after transplantation in vivo [7] . Understanding the mechanisms that regulate differentiation of HSCs into the different cell types is expect to be important for understanding hematopoietic diseases and manipulating HSCs for therapeutic purpose . However , because HSCs are currently unable to proliferate extensively in vitro without losing their “stemness” , large cultures cannot be produced [8] . This severely limits the types of biochemical and genomic analyses that can be performed , and consequently , the mechanisms that control the decision between early-stage HSC self-renewal and differentiation remain unclear . The mouse ( Mus musculus ) EML ( Erythroid , Myeloid , and Lymphocytic ) multipotential hematopoietic precursor cell is an ideal system for studying the molecular control of early hematopoietic differentiation events . EML cells are derived from mouse bone marrow cells that have been transfected with a retrovirus expressing a dominant negative retinoic acid receptor and were subsequently cultured in the presence of stem cell factor ( SCF ) . These cells can be re-derived or repeatedly cloned and still retain their multipotentiality [9] , [10] , [11] . The ability of EML cells to propagate extensively in medium containing SCF makes them ideal for biochemical and genetic assays as well as high throughput functional screens [7] , [12] . Phenotypically , EML cells express many of the cell surface markers characteristic of hematopoietic progenitor cells , including SCA1 , CD34 , and c-KIT . Functionally , when treated with different growth factors , such as SCF , IL-3 , GM-CSF , and EPO , EML cells can differentiate into distinct cell lineages including B-lymphocyte , erythrocyte , neutrophil , macrophage , mast cell , and megakaryocyte lineages [9] . Unlike maturation of human promyelocytic cell lines , such as NB4 and HL60 , EML cell derivatives develop into mature neutrophils with segmented nuclei and azurophilic granules [9] . Thus , EML cells are biologically relevant for normal hematopoiesis . Interestingly , in culture the Lin-SCA+CD34+ subpopulation of EML cells , which can be isolated by magnetic-activated cell sorting ( MACS ) beads or Fluorescence Activated Cell Sorting ( FACS ) , gives rise in culture to a mixed population containing similar numbers of self renewing Lin-SCA+CD34+ precursor cells and partially differentiated Lin-SCA-CD34− cells ( henceforth referred to as CD34+ and CD34− cells , respectively ) [13] . Although the two populations resemble each other morphologically , only the CD34+ population propagates in SCF-containing media , while the CD34− cells do not self-renew in SCF; instead , their growth requires the cytokine IL-3 [13] . The closest normal analogs of CD34+ cells are short-term ( ST ) -HSC or multipotent progenitors ( MPP ) . Similar to short-term ( ST ) -HSC , CD34+ cells are capable of self-renewal; like MPP , when treated with cytokines such as IL-3 , CD34+ cells can give rise to CD34− cells with more restricted potential . A number of erythroid genes , such as α- and β-hemoglobin , Gata1 , Epor ( erythropoietin receptor ) , and Eraf ( erythroid associated factor ) , as well as mast cell proteases are expressed at a significantly higher level in the CD34− cell population than CD34+ cells [13] , [14] . This indicates that the CD34− cells were , at minimum , differentiated into a state with prominent erythroid potential . The ability of CD34+ cells to both differentiate and self-renew in suspension culture in the absence of any anatomical niche or other cell type suggests that CD34+ cells are regulated by a tightly controlled endogenous mechanism that guides the generation of the variety and relative abundance of the cell types in culture . Understanding the molecular events that regulate the transition between the two types of putative precursors in the EML multipotent hematopoietic cell line will give insights to the fundamental mechanisms of autonomous balanced selection of alternative cell fates available for stem cells and intermediate-stage cancer precursor cells [7] . What is the mechanism that regulates the decision between the two types of precursor cells ? One possible mechanism is by modulating the levels of key transcriptional regulators . This hypothesis is suggested by the findings that Pu . 1 or Gata1 play a determining role in downstream hematopoietic lineage decisions [15] , [16] . Higher Pu . 1 expression switches the differentiation to the myeloid lineages [16] , [17] whereas Gata1 shifts cells towards the erythroid lineage . In light of this , we examined transcription factors that were significantly up-regulated in CD34+ cells relative to CD34− cells using RNA sequencing ( RNA-Seq ) and found Tcf7 ( also referred to by the symbol Tcf1 ) to be the most strongly up-regulated transcription factor . TCF7 is a member of a family of HMG box containing factors that are known to associate with beta-catenin in the nucleus to mediate Wnt signaling [18] , [19] , [20] . Wnt signaling has been implicated in hematopoietic stem cell and precursor maintenance and affects the decision between self-renewal and differentiation [21] , [22] , [23] , [24] although its role in EML cells is not yet defined . It has been reported that TCF7 plays a role in B cell and T cell development and is a possible co-regulator in mouse embryonic stem cells , but TCF7 has not been noted for its function in earlier decisions in hematopoietic development [25] , [26] . The binding motifs of the TCF family of transcription factors are significantly enriched among genes that are expressed at a higher level in CD34+ than in CD34− cells . Therefore , we hypothesized that TCF7 is one of the key transcription factors that control a transcriptional regulatory network determining the choice between EML cell self-renewal and differentiation . We identified in vivo binding targets of TCF7 using ChIP-Seq ( chromatin immunoprecipitation in combination with high-throughput sequencing ) . We found TCF7 binds to its own promoter and the promoter of Runx1 ( Aml1 ) , a developmental determinant in hematopoietic cells that is best known for its critical role in hematological malignancies [27] , [28] . RUNX1 and TCF7 were found to bind to each other's promoters and a large number of common target genes are bound by RUNX1 and TCF7 . Analysis correlating gene expression and transcription factor binding data suggests that TCF7 is necessary to maintain cells in the undifferentiated state . We validated this hypothesis by knock-down of Tcf7 expression . Finally , through network analysis , we found that TCF7 and RUNX1 bind and regulate a network of up-regulated transcription factors in the CD34+ cells which characterize the self-renewal property of the CD34+ cells . Importantly , in EML cells TCF7 functions in the absence of autocrine Wnt signaling . Our results thus elucidate novel components and mechanisms that control stem cells renewal and differentiation .
Global identification of gene expression can provide significant insight into molecules important for the self-renewal and differentiation decisions in EML cells . Differential gene expression between CD34+ and CD34− cells was first studied using cDNA microarrays [13] . As cDNA microarrays do not cover the entire transcriptome , we decide to investigate the gene expression profiles of CD34+ and CD34− cells using the RNA-Seq technology . We generated 35 nt single end and long 75 nt paired-end reads ( sequence reads from both ends of cDNA fragments ) using Illumina technology . Although the overall patterns of mRNA levels are similar in CD34+ and CD34− cells , the expression levels of a limited number of transcription factors differ in the two cell populations ( Figure 1; Dataset S1 ) . ) Notably , the expression level of Tcf7 was found to be over 100 fold higher in the CD34+ cells relative to CD34− cells , where it is very low . In fact Tcf7 is as strongly regulated as the cell surface marker Cd34 itself . Tcf7 is a member of the T-cell factor family including Tcf3 , an essential component of the core regulatory network controlling the balance between pluripotency and differentiation in mouse embryonic stem ( mES ) cells [29] . However , the Tcf3 expression level is very low in EML cells , and neither Tcf3 nor other TCF family members displayed the remarkable differential gene expression observed for Tcf7 . Other interesting transcription factors that we found to be up-regulated in CD34+ cells include: Sox4 , Jun , Stat3 , Smad1 , Cebpa , Bcl6 , Bcl3 , Bcl9 , Fos , Runx1 and Cbfb ( Figure 1 ) . Consistent with the microarray study , we also found that transcription factors involved in erythroid differentiation , such as Gata1 , Gfi1b Klf1 and Klf3 , are up-regulated in CD34− cells ( Figure 1 ) . We next examined the functional categories of differentially expressed transcription factors in either CD34+ or CD34− cells using Ingenuity Pathway Analysis software ( IPA ) . The top three significant categories in Molecular and Cellular Functions are “Gene Expression” , “Cellular Development” , and “Cellular Growth and Proliferation” , whereas the top three significant categories in Physiological System Development and Function are: “Hematological System Development and Function” , “Hematopoiesis” , and “Tissue Morphology” ( Table S1 ) . These categories are consistent with the suggestion that the switch from CD34+ to CD34− cells represents a developmental process towards a less proliferative and more differentiated state . To further identify the potential transcription factors that control the group of genes up-regulated in CD34+ cells , we performed Distant Regulatory Elements analysis ( DiRE http://dire . dcode . org/ ) . DIRE analyzes not only the proximal promoter regions but also the full gene locus including intergenic , promoter , intronic and UTR ( upstream untranscribed regions ) . Among the DNA sequence motifs that were enriched in up-regulated genes ( >1 . 5 fold ) in CD34+ cell in comparison to CD34− cells , were binding motifs for members of the TCF family of transcription factors ( Figure S1 ) . TCF7 has previously been studied as a partner with nuclear beta-catenin , which serves as a downstream transcriptional activator in response to external Wnt signaling . TCF7 potentially acts as a transcriptional repressor in the absence of beta-catenin [18] , [19] . We therefore examined the data for expression of components of the Wnt signaling pathway . Internal components of the pathway such as Apc , Axin1 , Bcl9 , Daam1 , Gsk3a and Gsk3b were expressed at the mRNA level ( Table S2 ) in both CD34+ and CD34− cells . Activation of the canonical Wnt pathway is achieved primarily through various Wnt ligands binding to LRP5/6 and/or frizzled ( FZD ) receptors . Of the components of canonical Wnt receptors , a moderate level of Lrp5/6 , Fzd2 and Fzd7 mRNA were detected by either RNA-Seq or Illumina microarrays analysis , and even lower levels of mRNA for other Frizzled genes . Based on RNA-Seq data , the Wnt ligands were absent with the exception of Wnt9a and Wnt10a which were present in trace amounts ( Table S2 ) . The Illumina microarray also reported the presence of low levels of Wnt10a mRNA . To further test for the presence of Wnt10a mRNA , four sets of PCR primers were designed that crossed introns , and could distinguish between genomic DNA and spliced cDNA . Each set of primers gave the anticipated band from mouse embryonic fibroblast cDNA , but none showed any band of the specific product using CD34+ cell cDNA ( data not shown ) . Despite the fact Tcf7 expression is abundant; the absence or very low level of mRNA for Wnt ligands suggests that EML cells have little or no endogenous Wnt signaling . To test this possibility we utilized a Tcf/Lef GFP reporter to monitor Wnt activity in EML cells . The Tcf/Lef reporter is under the control of a minimum CMV promoter fused in tandem to Tcf/Lef transcriptional response elements . EML cells containing the positive control CMV-GFP construct showed uniform robust expression in EML cells ( Figure S2 ) . By contrast , cells containing the Tcf/Lef reporter failed to express GFP . To circumvent the possibility that the Tcf/Lef reporter is non functional in EML cells , we incubated the cells with either LiCl ( 50 mM ) , Wnt3a or Wnt5a ( 400 uM ) for 24 hours . LiCl , a GSK3beta-inhibitor , has been previously used to activate the Wnt pathway [30] , [31] . Upon stimulation with LiCl we noted a significant increase in GFP expression ( 52 fold enrichment when MOI = 5; 85 fold enrichment when MOI = 10 ) suggesting that cells are capable of receiving and activating the Tcf/Lef reporter ( Figure S2 ) . Tcf/Lef reporter activation using Wnt ligands was also observed , although it was lower than LiCl activation ( data not shown ) consistent with the findings that EML cells have low levels of frizzled receptors . These data demonstrate that while the cells are capable of response to activation of the internal Wnt response system , there is no endogenous Wnt signaling in EML cells detectable with the commonly used Wnt reporter system . In order to better understand how TCF7 may be involved in the switch from self-renewing CD34+ cells to partially differentiated CD34− cells , we identified the in vivo binding sites for TCF7 using ChIP-Seq [32] , [33] . We also performed PolII ChIP-Seq to follow genes that were activated or poised for activation . ChIP-Seq experiments identified 9696 TCF7 binding sites with a q-value ( Benjamini Hochberg corrected p-value ) < = 0 . 001 ( Dataset S2 ) . The binding sites were mapped to RefSeq genes in the UCSC mm9 database ( genome . ucsc . edu ) . The binding sites were assigned to a particular gene if the peak was present within 3 kb upstream of the transcription start site or inside of the gene ( including both exonic and intronic regions ) ( Dataset S4 ) . ∼85% of TCF7 binding peaks were assigned to known genes using this approach . These sites were mapped to 7976 TCF7 target genes . Among the interesting binding sites was the presence of TCF7 at its own promoter region in CD34+ cells ( Figure 2A ) , raising the possibility that TCF7 regulates its own transcription through an autoregulatory feedback loop . qPCR experiments verified the enrichment of TCF7 binding at the Tcf7 promoter region in CD34+ cells using TCF7 antibody compared to IgG immunoprecipitation . To explore the biological processes that are regulated by TCF7 , the functional categories associated with TCF7 target genes were examined based on annotations in the Gene Ontology ( GO ) database [34] . Enriched GO categories were identified and displayed using the BiNGO program ( http://www . psb . ugent . be/cbd/papers/BiNGO/ ) ( Figure S3 ) [35] . Genes associated with regulation of transcription were highly enriched in TCF7 targets , consistent with our hypothesis that TCF7 functions as a key transcription regulator in the decision of EML cell self-renewal and differentiation . Other significantly enriched functional categories include cell development and differentiation , metabolic processes , and signaling . In an effort to understand how the expression of Tcf7 itself is controlled to regulate hematopoietic development , we examined the promoter of Tcf7 for other potential regulators . Evolutionarily conserved RUNX1 ( AML1 ) binding sites were identified at the Tcf7 promoter region ( Figure 2B ) using the software REGULATORY VISTA ( http://ecrbrowser . dcode . org ) . In addition , the Runx1 expression pattern is consistent with Tcf7 regulation , i . e . although expressed in both CD34+ and CD34− cells , Runx1 mRNA is up-regulated 3 . 7 fold in the CD34+ cells as shown by RNA-Seq data ( Figure 1 ) . Using a previously validated anti-RUNX1 rabbit polyclonal antibody [36] , ChIP-Seq experiments identified 21932 RUNX1 binding sites with a q-value ( Benjamini Hochberg corrected p-value ) < = 0 . 001 ( Dataset S3 ) . These sites were mapped to 5393 RUNX1 target genes ( Dataset S4 ) . We performed de novo binding motif search among sequences bound in TCF7 and RUNX1 ChIP-Seq experiments , and found they overlap well with the known motifs ( from Jaspar and Transfac databases ) of RUNX1 and TCF7 ( Figure S4 ) . We found RUNX1 binds to the Tcf7 promoter ( Figure 2A ) , as well as its own promoter . Therefore , it is likely that autoregulation of RUNX1 contributes to the regulation of Tcf7 . Furthermore , TCF7 also binds to the Runx1 gene ( Figure 2C ) . Since both transcription factors bind to their own respective promoters as well as to each other , one intriguing possibility is that TCF7 and RUNX1 may co-regulate each other in a feed-back loop . We used three approaches to test whether RUNX1 functions coordinately with TCF7 in regulating target genes . In the first approach we compared the TCF7 ChIP-Seq and RUNX1 ChIP-Seq data to identify the overlapping set of target genes for the two transcription factors . Among 7976 TCF7 target genes ( Figure 3A , blue circle in the Venn diagram ) and 5393 RUNX1 target genes ( Figure 3A , red circle ) , 3915 target genes are in common ( 72% of all RUNX1 targets and 49% of all TCF7 target genes are common; see Figure 3A ) . The hypergeometric p-value of the intersection is 6 . 72294e-56 . Thereby , TCF7 and RUNX1 binding target lists showed a statistically significant overlap . In the second approach , the proximity of RUNX1 and TCF7 binding peaks were analyzed . The distance of the nearest RUNX1 peak to each of the TCF7 peak was identified . RUNX1 peaks either overlap with , or are within 500 nt upstream or downstream of 4691 of the 9696 TCF7 peaks ( 48%; Figure 3B ) . The third approach used motifs to which the factors are known to bind . Two versions of the known motifs for TCF7 and two known motifs for RUNX1 were obtained from Jaspar and Transfac databases . We assessed the occurrence of each of these motifs within the experimentally identified binding sites of TCF7 and RUNX1 . Within TCF7 peaks , at least 44% of the peaks contained one version of the TCF7 motif and 54% of the peaks contained at least one RUNX1 motif . Within the RUNX1 peaks , 46% had a TCF7 motif and 78% had a RUNX1 motif ( Figure 3C ) . Based on all above-mentioned analyses , we conclude TCF7 and RUNX1 bind to a large number of shared target genes and likely function coordinately in regulation of gene expression . To determine how the differential gene expression in CD34+ and CD34− cells is related to TCF7 regulation , Gene Set Enrichment Analysis ( GSEA ) [37] was performed to correlate transcription factor binding information with the gene expression data . The GSEA software is designed to determine whether members of a gene set , for example TCF7 binding targets , are randomly distributed throughout the gene expression data or primarily enriched among genes most highly up or down-regulated during the switch from CD34+ to CD34− cells . The expression dataset was rank-ordered by fold change such that the most up-regulated genes in CD34+ cells were at the top of the ranked list , while most up-regulated genes in CD34− cells ( down-regulated genes in CD34+ cells ) were at the bottom of the ranked list . GSEA analysis showed a statistically significant enrichment ( P near 0 ) of TCF7 targets among up-regulated genes in CD34+ cells , in comparison to the distribution expected at random ( Figure 4A ) . Overall , these observations strongly indicate that TCF7 primarily binds to genes up-regulated in CD34+ cells . A similar enrichment was observed for RUNX1 binding targets ( Figure 4B ) so that both of these factors predominantly bind to genes up-regulated in CD34+ cells . To further understand how Tcf7 affects CD34+ cell self renewal or differentiation , we used shRNA constructs targeting different regions of the Tcf7 gene . Three Tcf7 shRNA constructs caused significant reduction in levels ( 36–54% ) of gene expression as confirmed by qRT-PCR experiments and Western blots ( Figure 5A , 5B ) , and we observed obvious effects from the knockdown experiments . Consistent with the hypothesis that Runx1 and Tcf7 act coordinately , knockdown of Tcf7 caused the short isoforms of RUNX1 to disappear at the protein level , without changing the expression of the long isoform ( Figure 5B ) . Illumina bead microarray analysis of cells with reduced expression using one of the constructs revealed that 1510 genes changed >1 . 5 fold in Tcf7 knockdown lines compared to a scrambled shRNA negative control line ( 711 down-regulated and 799 up-regulated genes ) ( Microarray data is available in the Gene Expression Omnibus ( GEO ) microarray data repository; record GSE30068 ) . Gene Set Enrichment Analysis ( GSEA ) was performed to correlate these differentially expressed genes in Tcf7 knockdown lines with the differential gene expression between CD34+ and CD34− cells . We found that the down-regulated genes in Tcf7 shRNA knockdown cells are significantly enriched among up-regulated genes in CD34+ cells ( Figure 5C; P near 0 ) . On the other hand , the up-regulated genes in Tcf7 shRNA knockdown cells are significantly enriched among up-regulated genes in CD34− cells ( Figure 5D; P near 0 ) . Therefore , the gene expression profile of Tcf7 knockdown cells shifted toward a partially differentiated CD34− state . Overall these results suggest that Tcf7 is necessary for maintaining cells in the undifferentiated CD34+ state but is also necessary for switching to the partially differentiated CD34− state . To understand the intricate relationship of the transcription regulators defining the state of CD34+ cells , we employed Ingenuity Pathway Analysis software ( IPA ) to identify a network among up-regulated transcription factors in CD34+ cells ( >2 fold ) ( Figure 6 ) . TCF7 and RUNX1 and many transcription factors bound by TCF7 or RUNX1 are found in this network . Overall , three main functional groups can be identified: 1 ) . HSC establishment and development during early hematopoiesis ( marked in red in Figure 6 ) . This group includes Sox4 , Fos , Tal1 and Etv6 . Fos and Sox4 were identified as novel nuclear factors that affect hematopoietic stem cell activity . Overexpression of Fos and Sox4 induced enhanced HSC activity and resulted in an increased repopulating activity compared to the untreated cells [38] . Tal1/Scl is required during the establishment of primitive and definitive haematopoiesis , and plays a role in erythromyeloid lineage commitment [39] , [40] . Etv6/Tel is shown to regulate postembryonic HSC survival and is essential for multilineage haematopoiesis in the bone marrow ( Hock et al . 2004 ) . CBFB ( core-binding factor , beta subunit ) is the binding partner of RUNX1 . 2 ) . Cell growth control ( marked in blue in Figure 6 ) . This group includes Stats , Ppard , Erg and the Tgfβ signaling pathway components Smads etc . 3 ) . Multipotency ( marked in yellow in Figure 6 ) . The multipotency of CD34+ cells is reflected in genes involved in lineage specification . For example , Cebpa is a regulator of myelomonocytic lineage commitment . Gfi1 promotes GMP ( common myeloid progenitor ) differentiation towards the neutrophilic lineage [40] . ETS transcription factor FLI-1 interacts with RUNX1-containing multiprotein complexes through protein-protein interactions and is involved in the transcriptional regulation of megakaryocyte maturation ( Huang et al . 2009 ) . Upon differentiation into the CD34−state , there is a remarkable down-regulation of genes for other lineage specifications except for erythroid differentiation . This is consistent with the fact that CD34− cells can no longer proliferate in SCF alone; they depend on an additional cytokine , IL-3 , for growth . Therefore , TCF7 , together with RUNX1 , controls a transcriptional regulatory network determining the choice between EML cell self-renewal of multipotent cells and differentiation . The silencing of Tcf7 in the CD34− cells would contribute to the commitment to differentiation .
In this study , we identified TCF7 as a regulator in the decision of EML multipotential hematopoietic precursor cell self-renewal and differentiation . We further identified RUNX1 as a partner and effector of TCF7 function . TCF7 and RUNX1 bind to a significantly overlapping set of target genes and likely function coordinately in regulating target genes . In particular , TCF7 and RUNX1 bind to and potentially regulate a network of transcription factors which characterize the gene expression pattern of CD34+ cells . We validated our hypothesis using functional tests . Tcf7 is a member of the T-cell factor family of transcription factors that are the downstream effectors of the Wnt signal transduction pathways . Wnt signaling inhibits the degradation of beta-catenin protein by preventing its phosphorylation by GSK3 beta . In the absence of phsophorylation of N-terminal serine and threonine residues , beta-catenin accumulates and is translocated into the nucleus where it associates with members of the TCF family of transcription factors , and furnishes them with a transcriptional activation domain [20] . Wnt signaling can act in a context dependent manner to either activate or repress transcription [41] . The TCF family of transcription factors can also either activate or repress the transcription of genes responsive to Wnt signaling [18] . Wnt signaling has been implicated in self-renewal of hematopoietic stem cells and the growth of hematopoietic precursors [21] , [22] , [23] , [24] although mice with defects in the Wnt signaling pathway continue to develop normal mature cells of the hematopoietic system . Although Tcf7 knockout mice have only been reported to show a defect in thymocyte development [42] , [43] , there are multiple TCF family members in vivo which can have redundant function and compensate for the loss of Tcf7 alone in the knock-out mouse model . In the EML system the expression level of other TCF family members are low and none of them displayed as remarkable differential gene expression as observed of Tcf7 . An analysis of the repopulating activity of the subpopulations of cells in Tcf7−/− mouse bone marrow could be useful in the future to understand TCF7's roles in vivo; but such an analysis can be complicated by the stimuli from multiple signaling pathways in the bone marrow and the stem cell niche . In addition to TCF7's role in thymocyte development , T-lineage specification and differentiation [42] , [43] , [44] , the present study shows TCF7 plays a role in the EML model of hematopoietic precursor cell differentiation and function; and has a role in gene activation as well as repression , A number of components of the Wnt signaling pathway , such as Lrp5 and Bcl9 , were expressed at higher levels in CD34+ than CD34− cells . However , the data presented here suggest that regulation by Wnt molecules is not a significant factor in EML cell growth and differentiation . RNA-Seq showed only low levels of the Wnt receptors frizzled 2 , 5 and 7 , and , at most , traces of Wnt9a and 10a mRNA in the CD34+ cell population , as well as a lack of mRNA for other known Wnt molecules in these cells . Even the trace amounts of Wnt9a and 10a mRNA seen by RNA-Seq in CD34+ cells could not be detected by RT-PCR . Wnt9a and 10a mRNAs were completely absent from the CD34− cells . EML cells are grown in conditioned medium and it is possible that this medium contributed some Wnt . However , EML cells also grow well in standard medium in the presence of only purified SCF so that Wnt in the conditioned media does not seem to be a necessary factor for the growth of the cells . Finally , when a Tcf/Lef-GFP reporter was introduced into the EML cells , there was no GFP signal of Wnt induced activity in these cells , although the cells did respond to LiCl or external WNT ligands . We conclude that TCF7 may function in these cells by a pathway operative without autocrine Wnt signaling . The present study suggests that TCF7 plays a dual role in global gene network by promoting the expression of large number of genes characteristic of self-renewing CD34+ cells ( Figure 5C ) , while repressing genes activated in partially differentiated CD34− state ( Figure 5D ) . The effects of Tcf7 knock-down is a result of a combination of both direct effects of losing TCF7 binding and secondary effects of removal of the short form of RUNX1 or other TCF7 targets . RUNX1 plays multiple roles in early hematopoietic development and , unlike TCF7 or Wnt , is necessary for emergence of hematopoietic stem cells [45] . In sea-urchin embryos RUNX1 expression is linked to Wnt activity [46] . In EML cells , TCF7 and RUNX1 bind to one another's promoter and thus may regulate each other . The Runx1 gene encodes both short and long isoforms , and these have antagonistic effects . The short isoforms promote maintenance and proliferation of progenitor cells , but the long isoforms promote differentiation and inhibit progenitor cell repopulation [47] . Remarkably , knockdown of Tcf7 , caused the short isoforms of RUNX1 to disappear at the protein level , without changing the expression of the long isoforms . Therefore , TCF7 , in addition to direct effects on the transcription of individual genes , may prevent differentiation by regulating the relative abundance of RUNX1 isoforms which have opposing effects on differentiation . There may be additional unknown factors besides TCF7 and RUNX1 that are central for switching between the two cell types . Examination of the TCF7 binding targets whose expression is altered by Tcf7 inhibition showed that a STAT3 motif was one of the most frequently detectable transcription factor binding motifs ( Figure S5 ) . Interestingly STAT3 was one of the transcription factors that are up-regulated in the CD34+ cells , suggesting that increased STAT3 levels might augment TCF7 mediated transcriptional changes in the CD34+ cells . Among other transcription factors whose mRNA levels are higher in CD34+ than in CD34− cells , Bcl9 and Jun are known modulators of the TCF7-beta-catenin transcriptional response . Interestingly , analysis of the binding targets of SCL/TAL1 in a stem/progenitor cell line HPC-7 indicates that they largely overlap with TCF7 binding genes ( 163 out of 243 SCL target genes are in common with TCF7 targets ) [48] . Scl/Tal1 is one of the TCF7 target genes; however , Tcf7 is not among the SCL/TAL1 target genes . Therefore , SCL/TAL1 may be a downstream mediator of TCF7 activation . In another context Bcl3 has been reported to be increased by SCF signaling [49] , which may also contribute to the increase in Bcl3 mRNA levels seen in CD34+ as compared to CD34− cells . Overall , these results suggest that a major part of the transcriptional switch between CD34+ and CD34− cells is mediated by a small network of transcriptional mediators , with TCF7 central to the network . Finally , when Tcf7 level is knocked down by shRNA , although the transcription levels of many genes changed as mentioned earlier in this paper , CD34+ RNA is not reduced by the knockdown . Therefore , there may be additional transcription regulatory factors required for the switch and this is a topic of our future investigation . Many models of regulated stem cell differentiation , such as sperm and egg production , skin regeneration , intestinal cell regeneration , and neural differentiation , involve control of the choice of differentiation versus precursor renewal that depends on contact or signaling to the stem cell from different types of cells in anatomically circumscribed niches . The EML system provides a clear example of a mammalian precursor cell that has the intrinsic ability to produce a quantitatively balanced ratio of renewing versus differentiated progenies . We have found that TCF7 is a regulator of the self-renewal and differentiation switch and further analysis of how it is controlled will be critical for understanding how this important process is regulated .
Suspension cultures of EML cells were maintained in SCF containing medium as previously described [13] . Total EML cells were washed twice in FACS buffer ( 0 . 5% BSA , 1 mM EDTA , 1× PBS ) and resuspended in 40 µl FACS buffer per 1×107 cells . 15 µl of Mouse Lineage Depletion Cocktail biotin conjugated antibody ( Miltenyi Biotec ) was added to the cells for 20 minutes at 4°C . The labeled cells were washed twice in FACS buffer and resuspended in 80 µl FACS buffer . 20 µl of paramagnetic microbeads conjugated to anti-biotin antibody ( Miltenyi Biotec ) was added to the cells and incubated for 20 minutes at 4°C . The labeled cells were washed twice and resuspended in 2 ml FACS buffer per 1×107 cells . The cells were separated twice in an AutoMACS cell separator ( Miltenyi Biotec ) using the depletion program ( 0 . 5 mls per minute ) . The lineage negative ( Lin- ) fraction was resuspended in 100 µl ( per 1×107 cells ) of FACS buffer and CD34 biotin conjugated antibody was added ( 1 µg per 1×106 cells ) . The labeled cells were washed in FACS buffer as above and bound to anti-biotin coated beads . The cells were separated in an AutoMACS cell separator ( Miltenyi Biotec ) using a double positive separation program . The subsequent Lin-CD34− fraction was resorted with the AutoMACS separator using the depletion program . Lin-CD34+ and Lin-CD34+ cells were collected . The cell purity was checked after each separation using FACS and only cell purity >90% was used for further experimentation . Total EML cells were washed twice in FACS buffer ( 0 . 5% BSA , 1 mM EDTA , 1× PBS ) and resuspended in 100 µl FACS buffer per 1×106 cells . CD34-FITC ( 1 µg per 1×106 cells; Ebiosciences ) was added to the cells and incubated for 1 hour at 4°C . Sca1-PE ( 0 . 06 µg per 1×106 cells; Ebiosciences ) and Lineage Cocktail APC ( 5 µl per 1×106 cells; Miltenyi Biotec ) were added to the cells and incubated for an additional 30 minutes . Lin-SCA+CD34+ and Lin-SCA-CD34− cells were collected using FACS Aria ( Beckman ) . mRNA samples were prepared from 2×106 CD34+ and CD34− cells . RNA-Seq was performed as described [50] . Two biological replicas and two technical replicas were used for each cell type . The mouse genome sequence , annotation and genomic features ( genes , cDNAs , 3′ UTRs , 5′ UTRs , introns , exons , intergenic regions , ESTs ) for the mm9 database release were directly downloaded from UCSC Table Browser ( http://genome . ucsc . edu ) or obtained from Galaxy ( http://galaxy . psu . edu/ ) . Raw Illumina reads were obtained after base calling in the Solexa Pipeline version 0 . 2 . 2 . 6 . RNA-Seq reads were mapped to the mouse genome using Illumina's ELAND software . Differentially expressed gene features were identified using the ERANGE package [51] . Read coverage along the annotated transcription units was calculated using the ShortRead package [52] . Repetitive mapped reads were combined with uniquely mapped reads to produce a final RPKM ( reads per kilobase of mRNA , per million total reads ) , using the procedure defined for ERANGE , by calculating the probability that a multiread came from a particular known or candidate exon based on the distribution of counts of uniquely mapped reads in each exon . The resulting fractional counts were added to the total count for the gene locus , which was renormalized into a multi RPKM ( Gene expression values: Dataset S1 ) . Transcription factors were identified from the list provided by Luscombe et al [53] of human transcription factors . Homologous mouse genes were obtained for each from the Ensembl database using the biomaRt package . We identified genes with different expression levels in CD34+ versus CD34− cells as those with at least a two fold difference in RPKM and in which both cell types had a minimum of 2 RPKM . ChIP-Seq was performed as described [54] , [55] , [56] . 5×107 formaldehyde cross-linked Lin-CD34+ and Lin-CD34− cells were used . TCF7 goat polyclonal antibody TCF1 ( H18 ) ( Santa Cruz Biotechnology: catalog#SC8589 ) , anti-AML ( RUNX1 ) rabbit polyclonal antibody [36] ( CalBiochem: catalog#PC284 ) , and monoclonal RNA PolII antibody ( Covance: catalog#8WG16-MMS-126R ) were used . IP-western experiments were done to ensure the specificity of the antibodies . ChIP-Sequencing data were analyzed using the PeakSeq program as previously described [57] . The transcription factor binding loci were extracted with statistically significant signals ( q-value<0 . 001 ) . Subsequently , we mapped the binding sites to RefSeq genes in UCSC mm9 database ( genome . ucsc . edu ) . A gene was designated as the target gene if the peak was present within 3000 nt upstream of the transcription start site or inside of the gene ( including both exonic and intronic regions ) . ChIP-Seq data have been deposited to GEO database ( GSE31221; reviewer access link: http://www . ncbi . nlm . nih . gov/geo/query/acc . cgi ? token=lxglzumiewawcha&acc=GSE31221 . ) In TCF7 and RUNX1 coregulation analysis , known motifs for each of TCF7 and RUNX1 were obtained from Jaspar and Transfac databases . We random selected 600 sequences bound in TCF7 and RUNX1 ChIP-Seq experiments , and performed de novo binding motif search . We compared these de novo binding motifs with the known motifs ( from Jaspar and Transfac databases ) of RUNX1 and TCF7 . We used BiNGO [35] to determine the statistically over-represented ( p-value<0 . 0001 ) Gene Ontology ( GO ) categories within the target gene sets of the transcription factors , and then visualized the relationships of these GO categories with the Cytoscape software [58] . We also performed Gene Set Enrichment Analysis ( GSEA ) [37] to correlate transcription factor binding information with the gene expression data . The expression dataset was rank-ordered by fold change ( difference of classes ranking metric ) such that the most up-regulated genes in Lin-CD34+ cells were on the top of the ranked list , while the most up-regulated genes in Lin-CD34− cells ( down-regulated genes in Lin-CD34+ cells ) were at the bottom of the ranked list . GSEA analysis was used to determine whether members of a binding target list , are randomly distributed throughout the gene expression data or primarily enriched toward the top or bottom of the gene expression list using the default weighted enrichment statistic . Ingenuity Pathway Analysis ( IPA ) was performed to display the transcription factor regulatory networks ( http://www . ingenuity . com/ ) . Fresh EML cells were recovered one week prior to shRNA experiments in SCF containing growth medium ( GM ) as previously described [13] . 1 . 5×104 EML Lin-CD34+cells were double FACS sorted and infected with shRNA constructs containing Tcf7 ( Sigma Aldrich ) at an MOI = 2 in a round bottom 96 well plate . Five shRNA constructs targeting different regions of the Tcf7 gene were used ( Sigma , shRNA product numbers: TRCN0000012678 , TRCN0000012679 , TRCN0000012680 , TRCN0000012681 , TRCN0000012682 ) . SHC002VMISSION Non-Target shRNA Control Transduction Particles ( Sigma ) were used as shRNA negative control . To increase transduction efficiency the ExpressMag systems ( Sigma Aldrich ) was used according to manufacturer's instructions . 24 hours post transduction , cells were selected in EML GM containing puromycin ( 1 . 2 µg/ml ) . When indicated , selected cells were grown in expansion medium ( IMDB , 20% heat inactivated horse serum , 100 ng/ml SCF [PeproTech] ) . Cells were analyzed for knockdown efficiency using qPCR ( Tcf7 forward primer sequence ATCCTTGATGCTGGGATTCTG; Tcf7 reverse primer sequence CTTCTCTTGCCTTGGGTTCTG . CD34 forward primer sequence aggctgatgctggtgctagt; reverse primer sequence ccccagctttctcaagtcag . Two internal controls: HPRT forward primer tatgccgaggatttggaaaa; HPRT reverse primer acagagggccacaatgtgat , and/or beta Actin forward primer gatctggcaccacaccttct; reverse primer accagaggcatacagggaca ) . In addition , Western blot analysis was performed on puromycin-selected Tcf7 knockdown cell lines to examine TCF7 and RUNX1 protein expression . The polyclonal TCF7 antibody ( Sigma Aldrich , catalog#AV34782 ) and RUNX1 antibody ( Abcam , catalog# ab23980 ) were used . The anti actin ( Abcam , catalog#ab8229 ) antibody was used to indicate equal loading . For Illumina array analysis , Lin- CD34+cells ( 1×105 ) cells were transduced at an MOI = 1 with a Tcf7 targeting shRNA construct TRCN0000012679 or a shRNA negative control . After a 24-hour incubation with the shRNA-containing virus , the cells were grown in EML GM for 24 hours , then cells were selected for 24 hours in puromycin ( 1 . 2 µg/ml ) . Cells were harvested ( a total of four days after initial sort ) and total RNA extracted . Total RNA from Tcf7 shRNA knockdown cells and control cells ( transfected with scrambled shRNA ) was purified using the RNeasy Plus kit from Qiagen . Hybridization to Illumina Mouse WG-6 v2 . 0 Expression BeadChips was conducted at the Stanford Functional Genomics Facility using standard Illumina protocols . The microarray data was processed using the R version 2 . 11 Bioconductor Lumi package . The gene expression values were offset so that all values were made positive , subjected to the VST variance stabilization transformation , and were then quantile normalized . Z scores are plotted where Z = ( x−μ ) /σ , x is the log2 transformed gene expression measurement and μ and σ are the mean and standard deviations of expression of the gene . The microarray data is in compliance with MIAME guidelines . The data have been deposited in GEO database ( GSE30068 ) . To test for the presence of wnt10a mRNA , four sets of PCR primers were designed cross introns , which could distinguish between genomic DNA and spliced cDNA . Wnt10a forward primer 1:GCGCTCCTGTTCTTCCTACT , Wnt10a reverse primer 1: GATCTGGATGCCCTGGATAG; Wnt10a forward primer 2: GGCGCTCCTGTTCTTCCTAC , Wnt10a reverse primer 2: ATGCCCTGGATAGCAGAGG; Wnt10a forward primer 3: CATGAGTGCCAGCATCAGTT , Wnt10a reverse primer 3: ACCGCAAGCCTTCAGTTTAC; Wnt10a forward primer 4: CATGAGTGCCAGCATCAGTT , Wnt10a reverse primer 4: AGCCTTCAGTTTACCCAGAGC . Total EML cells ( 5×104 ) were infected with lentivirus containing either a CMV-GFP construct or a Tcf/Lef-GFP construct ( SABiosciences ) at a MOI of 1 , 5 , 10 , and 20 . After 24 hours , cells were selected in EML GM with puromycin ( 1 . 2 µg/ml ) . Selected cells were expanded in EML GM for 4–6 days . Cells were incubated for 24 hours in either LiCL ( 50 mM ) , WNT3a ( 400 uM , PeproTech ) or WNT5a ( 400 um , PeproTech ) . Cells were analyzed by FACSCalibur ( BD Biosciences ) . | The hematopoietic system has provided a leading model for stem cell studies , and there is great interest in elucidating the mechanisms that control the decision of HSC self-renewal and differentiation . This switch is important for understanding hematopoietic diseases and manipulating HSCs for therapeutic purposes . However , because HSCs are currently unable to proliferate extensively in vitro , this severely limits the types of biochemical analyses that can be performed; and , consequently , the mechanisms that control the decision between early-stage HSC self-renewal and differentiation remain unclear . Murine bone marrow derived EML multipotential hematopoietic precursor cells are ideal for studying the switch . EML cells can grow in large culture and give rise to a mixture of self-renewing Lin-SCA+CD34+ cells and partially differentiated non-renewing Lin-SCA-CD34− cells in a cell autonomous fashion . Using RNA–Sequencing and ChIP–Sequencing , we identified and validated the HMG box protein TCF7 as a regulator in this switch and find that it operates in the absence of canonical Wnt signaling . Together with RUNX1 , TCF7 regulates a network of transcription factors that characterize the CD34+ cell state . This work serves as a model for studying mechanisms of autonomous and balanced cell fate choice and is ultimately valuable for manipulating HSCs . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"medicine",
"developmental",
"biology",
"stem",
"cells",
"molecular",
"genetics",
"gene",
"expression",
"biology",
"molecular",
"biology",
"hematology",
"cell",
"biology",
"genetics",
"cellular",
"types",
"genomics",
"molecular",
"cell",
"biology",
"computational",
"biology",
"genetics",
"and",
"genomics"
] | 2012 | Tcf7 Is an Important Regulator of the Switch of Self-Renewal and Differentiation in a Multipotential Hematopoietic Cell Line |
Encoding and decoding in functional magnetic resonance imaging has recently emerged as an area of research to noninvasively characterize the relationship between stimulus features and human brain activity . To overcome the challenge of formalizing what stimulus features should modulate single voxel responses , we introduce a general approach for making directly testable predictions of single voxel responses to statistically adapted representations of ecologically valid stimuli . These representations are learned from unlabeled data without supervision . Our approach is validated using a parsimonious computational model of ( i ) how early visual cortical representations are adapted to statistical regularities in natural images and ( ii ) how populations of these representations are pooled by single voxels . This computational model is used to predict single voxel responses to natural images and identify natural images from stimulus-evoked multiple voxel responses . We show that statistically adapted low-level sparse and invariant representations of natural images better span the space of early visual cortical representations and can be more effectively exploited in stimulus identification than hand-designed Gabor wavelets . Our results demonstrate the potential of our approach to better probe unknown cortical representations .
An important goal of contemporary cognitive neuroscience is to characterize the relationship between stimulus features and human brain activity . This relationship can be studied from two distinct but complementary perspectives of encoding and decoding [1] . The encoding perspective is concerned with how certain aspects of the environment are stored in the brain and uses models that predict brain activity in response to certain stimulus features . Conversely , the decoding perspective uses models that predict specific stimulus features from stimulus-evoked brain activity and is concerned with how specific aspects of the environment are retrieved from the brain . Stimulus-response relationships have been extensively studied in computational neuroscience to understand the information contained in individual or ensemble neuronal responses , based on different coding schemes [2] . The invasive nature of the measurement techniques of these studies has restricted human subjects to particular patient populations [3] , [4] . However , with the advent of functional magnetic resonance imaging ( fMRI ) , encoding and decoding in fMRI has made it possible to noninvasively characterize the relationship between stimulus features and human brain activity via localized changes in blood-oxygen-level dependent ( BOLD ) hemodynamic responses to sensory or cognitive stimulation [5] . Encoding models that predict single voxel responses to certain stimulus features typically comprise two main components . The first component is a ( non ) linear transformation from a stimulus space to a feature space . The second component is a ( non ) linear transformation from the feature space to a voxel space . Encoding models can be used to test alternative hypotheses about what a voxel represents since any encoding model embodies a specific hypothesis about what stimulus features modulate the response of the voxel [5] . Furthermore , encoding models can be converted to decoding models that predict specific stimulus features from stimulus-evoked multiple voxel responses . In particular , decoding models can be used to determine the specific class from which the stimulus was drawn ( i . e . classification ) [6] , [7] , identify the correct stimulus from a set of novel stimuli ( i . e . identification ) [8] , [9] or create a literal picture of the stimulus ( i . e . reconstruction ) [10]–[12] . The conventional approach to encoding and decoding makes use of feature spaces that are typically hand-designed by theorists or experimentalists [8] , [9] , [11] , [13]–[16] . However , this approach is prone to the influence of subjective biases and restricted to a priori hypotheses . As a result , it severely restricts the scope of alternative hypotheses that can be formulated about what a voxel represents . This restriction is evident by a paucity of models that adequately characterize extrastriate visual cortical voxels . A recent trend in models of visual population codes has been the adoption of natural images for the characterization of voxels that respond to visual stimulation [8] , [13] . The motivation behind this trend is that natural images admit multiple feature spaces such as low-level edges , mid-level edge junctions , high-level object parts and complete objects that can modulate single voxel responses [5] . Implicit about this motivation is the assumption that the brain is adapted to the statistical regularities in the environment [17] such as those in natural images [18] , [19] . At the same time , recent developments in theoretical neuroscience and machine learning have shown that normative and predictive models of natural image statistics learn statistically adapted representations of natural images . As a result , they predict statistically adapted visual cortical representations , based on different coding principles . Some of these predictions have been shown to be similar to what is found in the primary visual cortex such as topographically organized simple and complex cell receptive fields [20] . Building on previous studies of visual population codes and natural image statistics , we introduce a general approach for making directly testable predictions of single voxel responses to statistically adapted representations of ecologically valid stimuli . To validate our approach , we use a parsimonious computational model that comprises two main components ( Figure 1 ) . The first component is a nonlinear feature model that transforms raw stimuli to stimulus features . In particular , the feature model learns the transformation from unlabeled data without supervision . The second component is a linear voxel model that transforms the stimulus features to voxel responses . We use an fMRI data set of voxel responses to natural images that were acquired from the early visual areas ( i . e . V1 , V2 and V3 ) of two subjects ( i . e . S1 and S2 ) [21] . We show that the encoding and decoding performance of this computational model is significantly better than that of a hand-designed Gabor wavelet pyramid ( GWP ) model of phase-invariant complex cells . The software that implements our approach is provided at http://www . ccnlab . net/research/ .
To learn the feature transformation , we used a two-layer sparse coding ( SC ) model of 625 simple ( i . e . first layer ) and 625 complex ( i . e . second layer ) cells [22] . Concretely , the simple cells were first arranged on a square grid graph that had circular boundary conditions . The weights between the simple and complex cells were then fixed such that each complex cell locally pooled the energies of 25 simple cells in a 55 neighborhood . There were a total of 625 partially overlapping neighborhoods that were centered around the 625 simple cells . Next , the weights between the input and the simple cells were estimated from 50000 patches of size 3232 pixels by maximizing the sparseness of the locally pooled simple cell energies . Each simple cell was fully connected to the input ( i . e . patch of size 3232 pixels ) . The patches were randomly sampled from the 1750 images of size 128128 pixels in the estimation set . To maximize the sparseness , the energy function ( i . e . square nonlinearity ) encourages the simple cell responses to be similar within the neighborhoods while the sparsity function ( i . e . convex nonlinearity ) encourages the locally pooled simple cell energies to be thinly dispersed across the neighborhoods . As a result , the simple cells that are in the same neighborhood have simultaneous activation and similar preferred parameters . Since the neighborhoods overlap , the preferred parameters of the simple and complex cells change smoothly across the grid graph . Finally , the complex cell responses of the SC model were defined as a static nonlinear function of the locally pooled simple cell energies after model estimation ( i . e . total of 625 complex cell responses per patch of size 3232 pixels and 10000 complex cell responses per image of size 128128 pixels ) . The SC model learned topographically organized , spatially localized , oriented and bandpass simple and complex cell receptive fields that were similar to those found in the primary visual cortex ( Figure 2A ) [23]– . To establish a baseline , we used a GWP model [25] , [27] , [28] of 10921 phase-invariant complex cells [8] . Variants of this model were used in a series of seminal encoding and decoding studies [8] , [13] , [14] , [16] . Note that the fMRI data set was the same as that in [8] , [13] . Concretely , the GWP model was a hand-designed population of quadrature-phase Gabor wavelets that spanned a range of locations , orientations and spatial frequencies ( Figure 2B ) . Each wavelet was fully connected to the input ( i . e . image of size 128128 pixels ) . The complex cell responses of the GWP model were defined as a static nonlinear function of the pooled energies of the quadrature-phase wavelets that had the same location , orientation and spatial frequency ( i . e . total of 10921 complex cell responses per image of size 128128 pixels ) . To learn the voxel transformation , we used regularized linear regression . The voxel models were estimated from the 1750 feature-transformed stimulus-response pairs in the estimation set by minimizing the penalized least squares loss function . The combination of a voxel model with the complex cells of the SC and GWP models resulted in two encoding models ( i . e . SC2 and GWP2 models ) . The SC2 model linearly pooled the 10000 complex cell responses of the SC model . The GWP2 model linearly pooled the 10921 complex cell responses of the GWP model . We first analyzed the receptive fields of the SC model ( i . e . simple and complex cell receptive fields ) . The preferred phase , location , orientation and spatial frequency of the simple and complex cells were quantified as the corresponding parameters of Gabor wavelets that were fit to their receptive fields . The preferred parameter maps of the simple and complex cells were constructed by arranging their preferred parameters on the grid graph ( Figure 3 ) . Most adjacent simple and complex cells had similar location , orientation and spatial frequency preference , whereas they had different phase preference . In agreement with [22] , the preferred phase , location and orientation maps reproduced some of the salient features of the columnar organization of the primary visual cortex such as lack of spatial structure [29] , retinotopy [30] and pinwheels [31] , respectively . In contrast to [22] , the preferred spatial frequency maps failed to reproduce cytochrome oxidase blobs [32] . The preferred phase map of the simple cells suggests that the complex cells are more invariant to phase and location than the simple cells since the complex cells pooled the energies of the simple cells that had different phase preference . To verify the invariance that is suggested by the preferred phase map of the simple cells , the population parameter tuning curves of the simple and complex cells were constructed by fitting Gaussian functions to the median of their responses to Gabor wavelets that had different parameters ( Figure 4 ) . Like the simple cells , most complex cells were selective to orientation ( i . e . standard deviation of 21 . 8° versus 22 . 9° ) and spatial frequency ( i . e . standard deviation of 0 . 52 versus 0 . 54 in normalized units ) . Unlike the simple cells , most complex cells were more invariant to phase ( i . e . standard deviation of 50 . 0° versus 158 . 1° ) and location ( i . e . standard deviation of 3 . 70 pixels versus 5 . 86 pixels ) . Therefore , they optimally responded to Gabor wavelets that had a specific orientation and spatial frequency , regardless of their phase and exact position . We then analyzed the receptive fields of the SC2 model ( i . e . voxel receptive fields ) . The eccentricity and size of the receptive fields were quantified as the mean and standard deviation of two-dimensional Gaussian functions that were fit to the voxel responses to point stimuli at different locations , respectively . The orientation and spatial frequency tuning of the receptive fields were taken to be the voxel responses to sine-wave gratings that spanned a range of orientations and spatial frequencies . While the eccentricity , size and orientation tuning varied across voxels , most voxels were tuned to relatively high spatial frequencies ( Figure 5A and Figure 5B ) . The mean predicted voxel responses to sine-wave gratings that had oblique orientations were higher than those that had cardinal orientations and this difference decreased with spatial frequency ( Figure 5C ) . While this result is in contrast to those of the majority of previous single-unit recording and fMRI studies [33] , [34] , it is in agreement with those of [35] . In line with [36] , [37] , the receptive field size systematically increased from V1 to V3 and from low receptive field eccentricity to high receptive field eccentricity ( Figure 6 ) . The properties of the GWP2 model were similar to those in [8] . The relationship between the receptive field parameters ( i . e . size , eccentricity , area ) of the GWP2 model were the same as those of the SC2 model . However , the GWP2 model did not have a large orientation bias . The encoding performance of the SC2 and GWP2 models was defined as the coefficient of determination ( ) between the observed and predicted voxel responses to the 120 images in the validation set across the two subjects . The performance of the SC2 model was found to be significantly higher than that of the GWP2 model ( binomial test , ) . Figures 7A and 7B compare the performance of the models across the voxels that survived an threshold of 0 . 1 . The mean of the SC2 model systematically decreased from 0 . 28 across 28% of the voxels in V1 to 0 . 21 across 11% of the voxels in V3 . In contrast , the mean of the GWP2 model systematically decreased from 0 . 24 across 24% of the voxels in V1 to 0 . 16 across 6% of the voxels in V3 . Figure 7C compares the performance of the models in each voxel . More than 71% of the voxels that did not survive the threshold in each area and more than 92% of the voxels that survived the threshold in each area were better predicted by the SC2 model than the GWP2 model . These results suggest that statistically adapted low-level sparse representations of natural images better span the space of early visual cortical representations than the Gabor wavelets . The decoding performance of the SC2 and GWP2 models was defined as the accuracy of identifying the 120 images in the validation set from a set of 9264 candidate images . The set of candidate images contained the 120 images in the validation set and the 9144 images in the Caltech 101 data set [38] . Note that the set of candidate images was ten- to hundred-fold larger than the sets in [8] but comparable to the largest set in [15] . The performance of the SC2 model was found to be significantly higher than that of the GWP2 model ( binomial test , ) . Figure 8 compares the performance of the models . The mean accuracy of the SC2 model across the subjects was 61% . In contrast , the mean accuracy of the GWP2 model across the subjects was 49% . The chance-level accuracy was 0 . 01% . These results suggest that statistically adapted low-level sparse representations of natural images can be more effectively exploited in stimulus identification than the Gabor wavelets . In principle , the SC2 and GWP2 models should have some degree of spatial invariance since they linearly pooled the responses of the complex cells that displayed insensitivity to local stimulus position . Spatial invariance is of particular importance for decoding since a reliable decoder should be able to identify a stimulus , regardless of its exact position . Furthermore , a difference between the degree of spatial invariance of the models can be a contributing factor to the difference between their performance . To analyze the spatial invariance of the models , we evaluated their encoding and decoding performance after translating the images in the validation set by ( i . e . approximately the standard deviation of the population location tuning curves of the complex cells of the SC model ) in a random dimension ( Figure 9 ) . The encoding and decoding performance of the models was found to decrease after the translations . Unlike the encoding performance of the GWP2 model , that of the SC2 model decreased less in V3 than V1 . This result suggests greater spatial invariance in V3 than V1 . The difference between the mean of the models across the voxels that survived the threshold before the translations increased from 0 . 05 to 0 . 11 . The difference between the mean accuracy of the models across the subjects increased from 12% to 24% . These results suggest that the SC2 model is more tolerant to local translations in stimulus position than the GWP2 model . Since the SC2 and GWP2 models had different nonlinearities ( i . e . pooling and static nonlinearity ) , a direct evaluation of the contribution of their components ( i . e . representations and nonlinearities ) to the difference between their encoding performance was not possible . Therefore , we estimated two control models that pooled the same static nonlinear function of the simple cell responses of the SC and GWP models . The static nonlinear function was a compressive nonlinearity ( i . e . where is a simple cell response ) . The compressive nonlinearity roughly accounts for insensitivities by increasing responses to a stimulus that is not entirely within a receptive field [39] . The simple cell responses were defined as the linear responses of the first layer of the SC model and the even-symmetric Gabor wavelets . While the performance of the compressive nonlinear SC model was significantly higher than that of the compressive nonlinear GWP model , the difference between the performance of the compressive nonlinear models was significantly lower than that of the SC2 and GWP2 models ( Figure 10 ) . This result suggests that both the representations and the nonlinearities of the SC2 model contribute to the difference between the encoding performance of the SC2 and GWP2 models . To verify the contribution of the nonlinearities to the individual encoding performance of the SC2 and GWP2 models , we estimated two more control models that pooled a linear function of the simple cell responses of the SC and GWP models . We used linear models since they retain selectivities that are discarded by nonlinearities . We found that the performance of the linear models were significantly lower than that of the compressive nonlinear , SC2 and GWP2 models ( Figure 10 ) . This result confirms the contribution of the nonlinearities that introduced the insensitivities to the individual encoding performance of the SC2 and GWP2 models .
This study addresses the question of how to model feature spaces to better predict brain activity . We introduced a general approach for making directly testable predictions of single voxel responses to statistically adapted representations of ecologically valid stimuli . Our approach relies on unsupervised learning of a feature model followed by supervised learning of a voxel model . To benchmark our approach against the conventional approach that makes use of predefined feature spaces , we compared a two-layer sparse coding model of simple and complex cells with a Gabor wavelet pyramid model of phase-invariant complex cells . While the GWP model is the fundamental building block of many state-of-the-art encoding and decoding models , the GWP2 model was found to be significantly outperformed by the SC2 model . We used control models to determine the contribution of the different components of the SC2 and GWP2 models to this performance difference . Analyses revealed that the SC2 model better accounts for both the representations and the nonlinearities of the voxels in the early visual areas than the GWP2 model . Given that the representations of the SC2 model are qualitatively similar to those of the GWP model , their contribution to this performance difference suggests that the SC model automatically learns an optimal set of spatially localized , oriented and bandpass representations that better span the space of early visual cortical representations since it adapts to the same statistical regularities in the environment as the brain is assumed to be adapted to [20] . Our approach eliminates the need for predefining feature spaces . However , the SC model does have a number of free parameters ( e . g . patch size , number of simple and complex cells , etc . ) that must either be specified by hand or using model selection methods such as cross-validation . Because of computational considerations , we used the same free parameters as those in [22] . While the choice of these free parameters can influence what the SC model can learn , the SC2 model was shown to outperform the GWP2 model even without cross-validation . Next to cross-validation , other methods that also infer these free parameters can further improve the performance of the SC2 model . One method is to first estimate voxel receptive fields using any approach and then use these estimates as free parameters ( e . g . voxel receptive field eccentricity as patch size ) of voxel-specific feature models . Another method is to use more sophisticated nonparametric Bayesian sparse factor models [40] that can simultaneously learn sparse representations while inferring their number . Furthermore , our approach included only feedforward projections such that representations and responses were solely determined by stimuli . However , taking top-down modulatory effects into account is essential to adequately characterize how sensory information is represented and processed in the brain . For example , attention has been shown to warp semantic representations across the human brain [41] , and prior expectations have been shown to bias sensory representations in visual cortex [42] . Extensions of our approach that include feedback projections can be used to address the question of how representations and responses are influenced by top-down processes . Further extensions of our approach can be used to probe mid- to high-level extrastriate visual cortical representations in a fully automated manner . In particular , the SC model can be replaced by highly nonlinear multi-layer statistical models of natural images that learn hierarchical feature spaces ( i . e . deep learning [43] ) . Some of the feature spaces that are learned by these models such as mid-level edge junctions have been shown to match well with neural response functions in area V2 [44] . Models that learn even higher-level representations such as high-level object parts [45] or complete objects [46] can be used to probe extrastriate visual cortical representations . For example , heterogenous hierarchical convolutional neural networks have been shown to predict the representational dissimilarity matrices that characterize representations in human inferior temporal gyrus [47] . Similar models have been shown to learn feature spaces that are admitted by stimulus sets other than natural images , both within the visual modality ( e . g . natural movies [48] ) as well as in other modalities ( e . g . auditory or somatosensory [49] ) . These models can be used to probe cortical representations in different sensory modalities . One approach to estimate deep models is to maximize the likelihood of all layers at the same time . However , this approach is not scalable and requires the computation of intractable partition functions that are impossible to integrate analytically and computationally expensive to integrate numerically . Nevertheless , methods such as score-matching [50] and noise-contrastive estimation [51] have been used to estimate unnormalized nonlinear multi-layer statistical models of natural images [52] , [53] . An alternative approach is to use models such as deep belief networks that comprise multiple layers of restricted Boltzmann machines . These models can be scaled by convolution [45] and estimated by maximizing the likelihood of one layer at a time , using the output of each layer as input for the subsequent layer [54] . Importantly , generative models such as deep belief networks make it possible to sample stimuli based on internal network states . Conditioning these internal network states on stimulus-evoked brain activity results in a generative approach to decoding . For example , we have previously shown that a deep belief network that comprise multiple layers of conditional restricted Boltzmann machines can reconstruct handwritten digits by sampling from the model after conditioning it on stimulus-evoked multiple voxel responses [55] . While introducing a new approach to probe cortical representations , this study complements other developments in encoding and decoding . For example , encoding models that involve computations to account for contrast saturation or heterogeneous contrast energy were shown to improve prediction of single voxel responses to visual stimuli [16] . At the same time , these modeling efforts go hand in hand with developments in fMRI such as the improvements in contrast-to-noise ratio and spatial resolution that are facilitated by increases in magnetic field strength [56] . For example , spatial features of orientation-selective columns in humans were demonstrated by using high-field fMRI [57] . Jointly , such developments can provide novel insights into how cortical representations are learned , encoded and transformed . In conclusion , we introduced a general approach that improves prediction of human brain activity in response to natural images . Our approach primarily relies on unsupervised learning of transformations of raw stimuli to representations that span the space of cortical representations . These representations can also be effectively exploited in stimulus classification , identification or reconstruction . Taken together , unsupervised feature learning heralds new ways to characterize the relationship between stimulus features and human brain activity .
We used the fMRI data set [21] that was originally published in [8] , [13] . Briefly , the data set contained 1750 and 120 stimulus-response pairs of two subjects ( i . e . S1 and S2 ) in the estimation and validation sets , respectively . The stimulus-response pairs consisted of grayscale natural images of size 128128 pixels and stimulus-evoked peak BOLD hemodynamic responses of 5512 ( S1 ) and 5275 ( S2 ) voxels in the early visual areas ( i . e . V1 , V2 and V3 ) . The details of the experimental procedures are presented in [8] . In the case of the SC model , each randomly sampled or non-overlapping patch was transformed to its principal components such that 625 components with the largest variance were retained and whitened prior to model estimation and validation . After the images were feature transformed , they were z-scored . The SC model of 625 simple and 625 complex cells was estimated from 50000 patches of size 3232 pixels that were randomly sampled from the 1750 images of size 128128 pixels in the estimation set . The details of the GWP model are presented in [8] . The SC2 and GWP2 models were estimated from the 1750 feature-transformed stimulus-response pairs in the estimation set . Voxel responses to an image of size 128128 pixels were predicted as follows . In the case of the SC model , each 16 non-overlapping patch of size 3232 pixels of the image were first transformed to the complex cell responses of the SC model ( i . e . total of 625 complex cell responses per patch and 10000 complex cell responses per image ) . The 10000 complex cell responses of the SC model were then transformed to the voxel responses of the SC2 model . In the case of the GWP model , the image was first transformed to the complex cell responses of the GWP model ( i . e . total of 10921 complex cell responses per image ) . The 10921 complex cell responses of the GWP model were then transformed to the voxel responses of the GWP2 model . The encoding performance was defined as the coefficient of determination between the observed and predicted voxel responses to the 120 images in the validation set across the two subjects . A target image was identified from a set of candidate images as follows . Prior to identification , 500 voxels were selected without using the target image . The selected voxels were those whose responses were predicted best . The target image was identified as the candidate image such that the observed voxel responses to the target image were most correlated with the predicted voxel responses to the candidate image ( i . e . highest Pearson product-moment correlation coefficient between observed and predicted voxel responses ) . The decoding performance was defined as the accuracy of identifying the 120 images in the validation set from the set of 9264 candidate images . The set of candidate images contained the 120 images in the validation set and the 9144 images in the Caltech 101 data set [38] . | An important but difficult problem in contemporary cognitive neuroscience is to find what stimulus features best drive responses in the human brain . The conventional approach to solve this problem is to use descriptive encoding models that predict responses to stimulus features that are known a priori . In this study , we introduce an alternative to this approach that is independent of a priori knowledge . Instead , we use a normative encoding model that predicts responses to stimulus features that are learned from unlabeled data . We show that this normative encoding model learns sparse , topographic and invariant stimulus features from tens of thousands of grayscale natural image patches without supervision , and reproduces the population behavior of simple and complex cells . We find that these stimulus features significantly better drive blood-oxygen-level dependent hemodynamic responses in early visual areas than Gabor wavelets–the fundamental building blocks of the conventional approach . Our approach will improve our understanding of how sensory information is represented beyond early visual areas since it can theoretically find what stimulus features best drive responses in other sensory areas . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"cognitive",
"neuroscience",
"functional",
"magnetic",
"resonance",
"imaging",
"computational",
"neuroscience",
"biology",
"and",
"life",
"sciences",
"computational",
"biology",
"neuroscience",
"neuroimaging"
] | 2014 | Unsupervised Feature Learning Improves Prediction of Human Brain Activity in Response to Natural Images |
An enormous number of alternative pre–mRNA splicing patterns in multicellular organisms are coordinately defined by a limited number of regulatory proteins and cis elements . Mutually exclusive alternative splicing should be strictly regulated and is a challenging model for elucidating regulation mechanisms . Here we provide models of the regulation of two sets of mutually exclusive exons , 4a–4c and 7a–7b , of the Caenorhabditis elegans uncoordinated ( unc ) -32 gene , encoding the a subunit of V0 complex of vacuolar-type H+-ATPases . We visualize selection patterns of exon 4 and exon 7 in vivo by utilizing a trio and a pair of symmetric fluorescence splicing reporter minigenes , respectively , to demonstrate that they are regulated in tissue-specific manners . Genetic analyses reveal that RBFOX family RNA–binding proteins ASD-1 and FOX-1 and a UGCAUG stretch in intron 7b are involved in the neuron-specific selection of exon 7a . Through further forward genetic screening , we identify UNC-75 , a neuron-specific CELF family RNA–binding protein of unknown function , as an essential regulator for the exon 7a selection . Electrophoretic mobility shift assays specify a short fragment in intron 7a as the recognition site for UNC-75 and demonstrate that UNC-75 specifically binds via its three RNA recognition motifs to the element including a UUGUUGUGUUGU stretch . The UUGUUGUGUUGU stretch in the reporter minigenes is actually required for the selection of exon 7a in the nervous system . We compare the amounts of partially spliced RNAs in the wild-type and unc-75 mutant backgrounds and raise a model for the mutually exclusive selection of unc-32 exon 7 by the RBFOX family and UNC-75 . The neuron-specific selection of unc-32 exon 4b is also regulated by UNC-75 and the unc-75 mutation suppresses the Unc phenotype of the exon-4b-specific allele of unc-32 mutants . Taken together , UNC-75 is the neuron-specific splicing factor and regulates both sets of the mutually exclusive exons of the unc-32 gene .
Alternative splicing of pre-mRNAs is a major source of proteomic complexity in metazoans . More than 90% of human multi-exon genes undergo alternative pre-mRNA processing and many alternative splicing events are controlled in tissue- and cell-type dependent manners [1] . Mis-splicing of pre-mRNAs underlie many inherited diseases [2] . A variety of auxiliary trans-acting factors and cis-acting elements regulating alternative splicing have been identified [3] , [4] , [5] , [6] . Recent genome-wide studies of protein-RNA interactions for trans-acting splicing factors led to creation of RNA splicing maps [7] . Combinations of hundreds of RNA features were used to assemble ‘splicing codes’ to predict splicing patterns in four major tissues to a significant extent [8] . However , much of our knowledge of splicing regulation relies on experiments utilizing cultured cells , and therefore complex mechanisms of the tissue-specific regulation of pre-mRNA splicing by coordination of multiple trans-factors and cis-elements in living organisms remain less understood . Mutually exclusive splicing should consist of multiple steps of strictly regulated splicing events and offers good models for elucidating regulation mechanisms for alternative pre-mRNA splicing [9] , [10] . Among them , fibroblast growth factor receptor ( FGFR ) genes have been well studied because tissue-specific and mutually exclusive selection of exons encoding a part of the extracellular domain determines the ligand specificity of the receptors [11] , [12] , [13] , [14] . The most extraordinary examples of the mutually exclusive exons are in the Drosophila Dscam gene [9] , [10] , which has four clusters of mutually exclusive exons . Selection of only one exon out of 48 candidate exons at a time for the exon 6 cluster is considered to be regulated by a complex system of competing RNA structures and a globally-acting cluster-specific splicing repressor [15] , [16] . However , the molecular mechanisms governing the selection patterns for the entire Dscam mRNA remain poorly understood [10] . A nematode Caenorhabditis elegans is intron-rich like vertebrates and is an excellent model organism for studying the regulation mechanisms of pre-mRNA processing in vivo [17] . Up to 25% of its protein-coding genes are estimated to undergo alternative pre-mRNA processing and hundreds of the events are developmentally regulated [18] . We developed a fluorescence alternative splicing reporter system and visualized spatio-temporal selection patterns of mutually exclusive exons in living worms [19] , [20] , [21] . Through genetic and biochemical analyses , we successfully identified evolutionarily-conserved and broadly-expressed RBFOX ( named after RNA binding protein , fox-1 homolog ( C . elegans ) ) family splicing regulators ASD-1 and FOX-1 and a muscle-specific RNA-binding protein SUP-12 as the co-regulators of the muscle-specific selection of exon 5B of the egl-15 gene encoding the sole homolog of the FGFRs in C . elegans [19] , [22] . The unc-32 gene of C . elegans , analyzed in this study , encodes the a subunit of V0 complex of vacuolar-type H+-ATPases considered to be proton pumps that acidify intracellular organelles [23] , [24] . The unique property of the unc-32 gene as a model for studying alternative splicing regulation is that it has two sets of mutually exclusive exons ( Figure 1A ) . Only one exon at a time is selected from three exons 4a , 4b and 4c; only one exon is selected at a time from two exons 7a and 7b . Of the six possible combinations of exons 4 and 7 , the three isoforms UNC-32A ( 4a/7b ) , UNC-32B ( 4b/7a ) and UNC-32C ( 4c/7b ) were predominantly detected [25] and appear to be developmentally regulated [18] , raising questions about the exact selection patterns and the regulation mechanisms in vivo . In the present study , we demonstrate that unc-32 exon 4 and exon 7 are selected in tissue-specific manners and that a neuron-specific RNA-binding protein UNC-75 regulates the neuron-specific selection of exons 4b and 7a .
We first confirmed mutually exclusive selection of endogenous unc-32 exon 4 and exon 7 by RT-PCR ( Figure 1B ) . Consistent with the previous report based on microarray profiling and high-throughput sequencing of mRNAs from synchronized worms [18] , the splicing patterns of both exon 4 and exon 7 appeared to be developmentally regulated; the relative amounts of the exon 4b isoform and the exon 7a isoform dramatically decreased at the L4 stage ( Figure 1B ) . Next we visualized the selection patterns of unc-32 exon 4 and exon 7 in vivo by applying our fluorescence alternative splicing reporter system [20] . A trio of symmetric reporter minigenes for exon 4 was constructed by cloning the genomic fragment spanning from exon 3 through exon 5 upstream of one of three fluorescent protein cDNA cassettes and by introducing artificial termination codons into two of the three mutually exclusive exons in each construct ( Figure 1C ) . From these minigenes , we expect expression of Venus-fusion protein ( E4a-Venus ) , monomeric red fluorescent protein ( mRFP ) -fusion protein ( E4b-mRFP ) and enhanced cyan fluorescent protein ( ECFP ) -fusion protein ( E4c-ECFP ) only when exon 4a alone , 4b alone and 4c alone are selected , respectively ( Figure 1C ) . In the same way , a pair of symmetric exon 7 reporter minigenes was constructed by cloning the genomic fragment spanning from exon 6 through exon 8 upstream of either of two fluorescent protein cDNA cassettes and by introducing an artificial termination codon into one of the two mutually exclusive exons in each construct ( Figure 1D ) . From these minigenes , we expect expression of enhanced green fluorescent protein ( EGFP ) -fusion protein ( E7a-EGFP ) and mRFP-fusion protein ( E7b-mRFP ) when exon 7a and exon 7b are selected , respectively ( Figure 1D ) . We utilized a ubiquitous promoter to drive expression of the minigenes and generated transgenic reporter worms ( Figure 1E–1G ) . Expression of the three fluorescent proteins in the exon 4 reporter worms varied among tissues; intestine , the nervous system and pharynx predominantly or exclusively expressed E4a-Venus , E4b-mRFP and E4c-ECFP , respectively ( Figure 1E ) . Expression of the two fluorescent proteins in the exon 7 reporter worms also showed tissue-specificity . Most tissues predominantly expressed E7b-mRFP and therefore the worms appear almost Red ( Figure 1F ) . Confocal microscopy revealed that neurons in head ganglia predominantly expressed E7a-EGFP ( Figure 1G ) . The expression patterns of the exon 4 and exon 7 reporters were consistent throughout development . We suspected that lack of the developmental change in the reporter expression was due to ectopic expression of the reporters in tissues that do not express the endogenous unc-32 gene . A transcriptional fusion reporter , however , revealed that the unc-32 promoter drives expression in intestine , neurons and pharynx ( Figure 1H ) , the major tissues where the exon 4 and exon 7 reporters were expressed . We therefore concluded that the mutually exclusive exons of the unc-32 exon 4 and exon 7 reporter minigenes are selected in tissue-specific and not developmentally regulated manners . To focus on the neuron-specific selection of exon 7a , we utilized the rgef-1 ( also known as F25B3 . 3 ) promoter to drive pan-neuronal expression of the exon 7 reporter . As expected , transgenic worms with an integrated reporter allele ybIs1622 [rgef-1::unc-32E7a-EGFP rgef-1::unc-32E7b-mRFP] predominantly expressed E7a-EGFP in the nervous system and appeared Green with a dual-bandpass filter ( Figure 2A ) . We therefore used the rgef-1 promoter for further analyses described below . As cis-elements regulating alternative splicing are often evolutionarily conserved in the genus Caenorhabditis [19] , [21] , [22] , [26] , we first focused on the five stretches in flanking introns of exons 7a and 7b conserved among C . elegans , C . briggsae and C . remanei ( Figure 2B , Figure S1A ) . We constructed five pairs of mutagenized exon 7 reporter minigenes M1 to M5 ( Figure 2C , Figure S1A ) and found that disruption of the UGCAUG stretch in intron 7b ( M1 ) changed the color of the exon 7 reporter from Green to Orange ( Figure 2D ) , while disruption of the other stretches had no apparent effect ( Figure S1B ) . We therefore concluded that the UGCAUG stretch in intron 7b is required for the neuron-specific selection of exon 7a . The UGCAUG stretches are known to be specifically recognized by the RBFOX family splicing regulators in metazoans including C . elegans [27] . We have previously reported that the RBFOX family proteins in C . elegans , ASD-1 and FOX-1 , redundantly repress egl-15 exon 5B by specifically binding to the UGCAUG stretch in the upstream intron [19] . The asd-1; fox-1 double mutant is defective in expression of a muscle-specific fibroblast growth factor receptor ( FGFR ) isoform EGL-15 ( 5A ) and shows the egg-laying-defective ( Egl-d ) phenotype [19] . To test whether ASD-1 and FOX-1 also regulate the neuron-specific selection of unc-32 exon 7a , we crossed the reporter allele ybIs1622 with the asd-1 and fox-1 mutants . As expected , the reporter worms turned the color from Green to Yellow in the single mutant backgrounds ( Figure 2E , top and middle ) and to Orange in the double ( Figure 2E , bottom ) , confirming that ASD-1 and FOX-1 are redundantly involved in the neuron-specific selection of exon 7a from the exon 7 reporter . To confirm direct and specific binding of ASD-1 and FOX-1 to the UGCAUG stretch in intron 7b in vitro , we performed an electrophoretic mobility shift assay ( EMSA ) using the radiolabelled RNA probes with an intact ( WT ) and a mutagenized ( M1 ) sequence as in the reporters ( Figure 2F , top ) . Recombinant full-length ASD-1 and FOX-1 proteins ( Figure 2F , bottom left ) efficiently shifted the mobility of the WT probe ( Figure 2F , bottom right , lanes 1–4 , 9–12 ) and less efficiently of the M1 probe ( lanes 5–8 , 13–16 ) in a dose-dependent manner , demonstrating direct and specific binding of ASD-1 and FOX-1 to the UGCAUG stretch . These results led to the conclusion that ASD-1 and FOX-1 regulate the selection of exon 7a from the unc-32 exon 7 reporter via the UGCAUG stretch in intron 7b in the nervous system . To identify other regulator ( s ) that confer the neuron-specificity to the exon 7 reporter , we mutagenized the ybIs1622 strain to screen for mutants exhibiting altered colors . We successfully isolated many homozygous viable strains with Yellow , Orange or Red phenotype ( Figure 3A , Figure S2 ) . In some other strains , most neurons turned red while some remained green ( Red/Green ) ( Figure 3A , Figure S2 ) . The color phenotypes were completely penetrated within the strains . Notably , all the Red and Red/Green strains also showed an uncoordinated ( Unc ) phenotype while the Orange or Yellow strains did not . By single-nucleotide polymorphism ( SNP ) -based mapping and sequencing candidate genes , we identified mutations in the unc-75 gene in the color mutants . The unc-75 gene was originally identified as the gene responsible for the Unc phenotype caused by defects in synaptic transmission [28] . The exon 7 reporter allele ybIs1622 crossed with an existing null allele unc-75 ( e950 ) , which lacks exon 1 through exon 5 and exhibits the Unc phenotype [28] , showed the RedUnc phenotype ( data not shown ) , confirming that the color phenotype is caused by loss of function of the unc-75 gene . UNC-75 belongs to the CUG-BP and ETR-3-like factor ( CELF ) family of RNA-binding proteins , which have two N-terminal RNA recognition motifs ( RRMs ) followed by a so-called divergent domain and the third RRM at the C-terminus . The CELF family can be divided into two subfamilies CELF1–2 and CELF3–6 according to sequence similarities [29] and UNC-75 is the sole member of the CELF3–6 subfamily in C . elegans [29] . Although UNC-75 has been shown to be expressed exclusively in the nervous system and localized to subnuclear speckles [28] , it is still unknown what process UNC-75 is involved in . The mutations identified in the unc-75 gene are summarized in Figure 3B . All of the five alleles with the RedUnc phenotype have nonsense mutations in exon 6 or exon 7 ( Figure 3B ) . Figure 3C shows amino acid sequence alignments of the three RRMs from the CELF family members in C . elegans and human . A missense mutation ( yb1714 ) in the conserved glycine residue in the α1β2 loop of RRM1 and four other mutations ( yb1697 , yb1705 , yb1709 and yb1718 ) in the region between exon 1 and exon 3 were associated with the Red/GreenUnc phenotype ( Figure 3B and 3C , top ) . A missense mutation ( yb1700 ) in the conserved glycine residue in the RNP1 motif of RRM2 ( Figure 3B and 3C , middle ) and a missense mutation ( yb1698 ) in the conserved arginine residue in the divergent domain ( Figure S3 ) were associated with the Yellow phenotype . A missense mutation ( yb1723 ) in the RNP2 motif and a 4-aa deletion ( yb1725 ) in the RNP1 motif in RRM3 were associated with the Orange phenotype ( Figure 3B and 3C , bottom ) . These results suggested that all the three RRMs and the divergent domain are required for UNC-75 to properly regulate the selection of exon 7a in the nervous system . During the course of cDNA cloning , we found another UNC-75 mRNA isoform lacking exon 8 corresponding to the anterior half of RRM3 ( Figure 3B and 3D , lane 1 ) . Although the skipping of exon 8 does not cause a frame-shift or nonsense-mediated mRNA decay ( NMD ) , the deletion of the half of RRM3 would more significantly affect the function of UNC-75 than the yb1723 and yb1725 mutations ( Figure 3C , bottom ) . As many splicing factors are known to regulate their own expression at the pre-mRNA splicing level , we analyzed the effect of the nonsense mutation in the unc-75 gene on its own mRNAs . The splicing patterns of the UNC-75 mRNAs were not affected in the asd-1; fox-1 mutant ( Figure 3D , lane 2 ) , while the Δexon 8 isoform was undetected in the unc-75 ( yb1701 ) mutant ( lane 3 ) , consistent with the idea that UNC-75 negatively regulates its own expression by repressing exon 8 . We noticed that the C-termini of the CELF family proteins as well as the RBFOX family proteins are evolutionarily conserved and match the consensus of the hydrophobic PY nuclear localization signal ( PY-NLS ) [30] ( Figure 4A ) . To test this idea , we analyzed the effect of substitution or deletion of the C-terminal motifs upon subcellular localization of the proteins ( Figure 4B–4G ) . The substitution of the three residues in the PY element of UNC-75 ( Figure 4A ) disrupted the nuclear localization of UNC-75 ( Figure 4B , 4C ) , confirming that the C-terminal motif of UNC-75 functions as the PY-NLS . In the same way , the deletion of the 7 and 16 residues from the C-termini of ASD-1 and FOX-1 , respectively ( Figure 4A ) , disrupted the nuclear localization of the proteins ( Figure 4D–4G ) , indicating that the C-terminal portions of ASD-1 and FOX-1 are the sole NLSs . To determine the element ( s ) in the exon 7 cluster region that UNC-75 directly and specifically recognizes in vitro , we performed EMSAs with the radiolabelled RNA probes schematically illustrated in Figure 5A ( top panel ) . Recombinant full-length UNC-75 protein shifted the mobility of Probe 2 ( Figure 5B , lanes 3 , 4 ) and Probe 2-1 ( lanes 9–12 ) and not of the other probes ( Figure 5B ) . As more than half of Probe 2-1 overlapped with Probe 1 or Probe 2-2 , we prepared a shorter probe 2-1-1 ( Figure 5A ) containing most of the sequence unique to Probe 2-1 . UNC-75 shifted the mobility of Probe 2-2-1 ( Figure 5C , lanes 1–4 , 25–28 ) , demonstrating that UNC-75 directly and specifically binds to the 2-1-1 fragment in this region . To further specify the element ( s ) necessary for the UNC-75-binding , we prepared the five mutant probes 2-1-1a to -1e , in each of which G and C residues in a short stretch were replaced with A ( Figure 5A , bottom panel ) . UNC-75 shifted the mobility of the probes 2-1-1a to -1d ( lanes 5–20 ) similarly to Probe 2-1-1 , while the mobility of Probe 2-1-1e was unaffected by UNC-75 ( lane 21–24 ) , indicating that the UUGUUGUGUUGU stretch disrupted in Probe 2-1-1e is essential for UNC-75 to specifically recognize the 2-1-1 fragment . To test whether all the three RRMs of UNC-75 are involved in the recognition of the 2-2-1 fragment , we performed EMSAs using three mutant recombinant proteins UNC-75 ( G53S ) , UNC-75 ( G165E ) and UNC-75 ( L431F ) ( Figure 6A , left ) , each of which had a single missense mutation in one of the three RRMs as found in the mutant alleles . UNC-75 ( G53S ) and UNC-75 ( G165E ) less efficiently shifted the mobility of Probe 2-1 and Probe 2-1-1 than wild-type UNC-75 ( Figure S4 , lanes 1–10; Figure 6A , right , lanes 1–13 ) . UNC-75 ( L431F ) failed to shift the mobility of these probes ( Figure S4 , lanes 11–13; Figure 6A , right , lanes 14–17 ) . These results indicated that the missense mutations affected the RNA-binding properties of UNC-75 in vitro and that all the three RRMs of UNC-75 are required for the specific recognition of unc-32 intron 7a . To specify which of the three RRMs of UNC-75 mediates the specific recognition of the elements in Probe 2-1-1 , we prepared recombinant proteins for each of the three RRMs and performed an EMSA ( Figure 6B ) . The RRM3 protein ( Figure 6B , right , lanes 12–16 ) as well as full-length UNC-75 ( lanes 17 , 18 ) shifted the mobility of Probe 2-1-1 , while the RRM1 or RRM2 protein did not ( lanes 1-11 ) , indicating that only RRM3 can bind to Probe 2-1-1 by itself . So we used only RRM3 protein for a further EMSA with the mutant 2-1-1 probes . The RRM3 protein shifted the mobility of Probe 2-2-1 ( Figure 6C , lanes 1–3 ) and the mutant probes 2-2-1a to -1d ( lanes 4–15 ) and not of Probe 2-2-1e ( lanes 16–18 ) , indicating that RRM3 specifically recognizes the UUGUUGUGUUGU stretch . To test the requirement of the UUGUUGUGUUGU stretch for the splicing regulation in vivo , we constructed another mutant pair of the exon 7 reporter minigenes M6 that has the same substitutions as in Probe 2-2-1e and generated transgenic worms . The disruption of the UUGUUGUGUUGU stretch turned the color into Orange ( Figure 6D ) , confirming that the stretch is essential for the selection of exon 7a in the nervous system in vivo . Next we analyzed the effects of the RBFOX family and UNC-75 on the endogenous unc-32 gene . In the wild-type L1 larvae , the exon 7a and exon 7b mRNA isoforms were almost equally detected ( Figure 7A , left , lane 1 ) . The relative amount of the exon 7a isoform was reduced in the asd-1; fox-1 double mutant ( lane 2 ) and unc-75 mutant ( lane 3 ) backgrounds . A double inclusion isoform or a double skipping isoform was not detected in either of the mutants . These results are consistent with their color phenotypes and the splicing patterns of the exon 7 reporter expressed in the nervous system ( Figure 7A , right ) and confirm that the RBFOX family and UNC-75 regulate the mutually exclusive splicing of exons 7a and 7b of the endogenous unc-32 gene . For mutually exclusive alternative splicing , upstream and downstream flanking introns should be sequentially excised ( Figure S5A ) . To obtain insight into the orders of intron removal for the production of the exon 7a and 7b mRNA isoforms , we analyzed the relative amounts of the four partially spliced RNA species to the unspliced RNA from the exon 7 reporter expressed in the nervous system by RT-PCR using two pairs of an intronic primer and a reporter-specific exonic primer . With one primer set , the partially spliced RNA in which exon 6 was spliced to exon 7b ( E6/E7b–E8 ) was detected but the other partially spliced RNA in which intron 6 was removed ( E6/E7a-E7b-E8 ) was almost undetectable in the wild-type , asd-1; fox-1 double mutant and unc-75 mutant worms ( Figure 7B , left ) . With the other primer set , the partially spliced RNA in which exon 7a was spliced to exon 8 ( E6–E7a/E8 ) was detected but the other partially spliced RNA in which intron 7b was removed ( E6-E7a-E7b/E8 ) was almost undetectable in these worms ( Figure 7B , right ) . The relative amounts of the four partially spliced RNAs to the unspliced RNA are summarized in Figure 7C . Of the two partially spliced RNAs that are the putative intermediates for the exon 7a isoform , E6–E7a/E8 was predominantly detected and its relative amount was decreased in the mutants . Of the two partially spliced RNAs that are the putative intermediates for the exon 7b isoform , E6/E7b–E8 was predominantly detected and its relative amount was increased in the mutants . Although these partially spliced RNAs may not necessarily be the processing intermediates but instead dead-end products , the changes in the relative amounts of the partially spliced RNAs are in good correlation with the changes in the amounts of the mature mRNA isoforms in the mutants . These results suggest that E6–E7a/E8 and E6/E7b–E8 are the major processing intermediates for the exon 7a and exon 7b isoforms , respectively . Notably , the mutations in the RBFOX family genes and unc-75 differentially affected the relative amounts of these partially spliced RNAs , suggesting their differential roles in the alternative splicing regulation of unc-32 exon 7 . We also analyzed the partially spliced RNAs from the endogenous unc-32 gene with endogenous RNA-specific pairs of primers . The result revealed consistent but weaker effects of the mutations in the RBFOX family genes and unc-75 on the partially spliced RNAs ( Figure S5B–S5C ) . Considering that the endogenous unc-32 gene is expressed not only in the nervous system but also in pharynx and intestine that select exon 7b , this result is consistent with the idea that the RBFOX family and UNC-75 regulate the selection exon 7a from the endogenous unc-32 gene in the same way as from the reporter in the nervous system . Taking the relative strength of the splice sites in this region ( Figure S5D ) into account , Figure 7D summarizes the schematic models for the mutually exclusive selection of unc-32 exon 7 , which will be discussed later ( see Discussion ) . As unc-32 exon 4b is also selected in a neuron-specific manner ( Figure 1F ) , we tested whether the RBFOX family and UNC-75 are also involved in the regulation of the exon 4 cluster . Consistent with the absence of a ( U ) GCAUG stretch in the exon 4 cluster region , the asd-1; fox-1 double mutation did not affect the splicing patterns of exon 4 of the endogenous unc-32 gene ( Figure 8A , lanes 1 , 2 ) . On the other hand , the unc-75 mutation caused marked reduction of the exon 4b isoform ( lane 3 ) . Furthermore , the neuron-specific expression of E4b-mRFP from the exon 4 reporter ybIs1891 was also abolished in the unc-75 mutant ( Figure 8B , compare with Figure 1F ) . These results indicated that UNC-75 is required for the selection of exon 4b in the nervous system . We performed an EMSA to localize the UNC-75-binding site ( s ) with four overlapping probes in the exon 4 cluster region , but none of the probes were shifted as effectively as Probe 2 in Figure 5B by full-length UNC-75 ( data not shown ) . We speculate that other cooperative factors may be required for the specific recognition of the exon 4 cluster region by UNC-75 . We next analyzed the amounts of the six theoretical partially spliced RNAs or putative processing intermediates ( Figure S6 ) from the endogenous unc-32 gene in the wild type and unc-75 mutant . Both of the two putative processing intermediate RNAs for the exon 4b isoform were detected in the wild type ( Figure 8C , left and right panels , lanes 1 , 2 ) but almost undetectable in the unc-75 mutant ( lanes 3 , 4 ) consistently with the amount of the mature exon 4b isoform . Only one ( E3–E4a/E5 ) of the two partially spliced RNAs that are the putative intermediate RNAs for the exon 4a isoform was detected and its relative amount was increased in the unc-75 mutant ( Figure 8C–8D ) . Only one ( E3/E4c–E5 ) of the two partially spliced RNAs that are the putative intermediate RNAs for the exon 4c isoform was detected and its relative amount was increased in the unc-75 mutant ( Figure 8C–8D ) . These results propose a model schematically illustrated in Figure 8E; UNC-75 represses splicing of exon 3 to exon 4c and exon 4a to exon 5 and promotes splicing of exon 4b to exons 3 and 5 . The exon 4b-specific mutation in the unc-32 ( e189 ) allele causes the uncoordinated ( Unc ) phenotype ( Figure 1A ) [25] and our results demonstrated that exon 4b is specifically selected in the nervous system in an UNC-75-dependent manner . So we speculated that the mutations in unc-75 should bypass the requirement of exon 4b in the nervous system . Consistent with this idea , the OrangeNon-Unc allele unc-75 ( yb1725 ) suppressed the Unc phenotype of the unc-32 ( e189 ) mutant ( Figure 8F ) . As neuron-specific ectopic expression of any of the three major isoforms can rescue unc-32 ( e189 ) ( Figure S7 ) , we reasoned that unc-75 ( yb1725 ) suppressed unc-32 ( e189 ) via the ectopic expression of the exon 4a or exon 4c isoform in the nervous system . Thus , UNC-75 is the critical splicing factor for the nervous system to specifically select unc-32 exon 4b in vivo .
In this study , we demonstrated that the two sets of the mutually exclusive exons of the unc-32 gene are independently regulated in tissue-specific manners by utilizing the fluorescence alternative splicing reporters . Our study revealed that intestine , neurons and pharynx express the UNC-32A ( 4a/7b ) , UNC-32B ( 4b/7a ) and UNC-32C ( 4c/7b ) isoforms , respectively . The expression patterns are consistent with the previous report that these three are the major isoforms and that the translational fusion reporter consisting of the unc-32 promoter through exon 4b is expressed in the nervous system [25] . The neuron-specific isoforms become relatively less abundant in elder stages in the RT-PCR experiments ( Figure 1B ) probably due to decrease in the relative population and/or mass of the nervous system . Our study thus demonstrated the importance of carefully analyzing alternative splicing patterns at a single cell resolution in vivo . Figure 7D illustrates the proposed models of the neuron-specific selection of exon 7a . In the non-neuronal tissues , exon 7a is skipped presumably due to its weak splice sites and exon 6 is readily spliced to exon 7b ( right panel ) . In neurons , UNC-75 specifically binds to its cis-elements in intron 7a to repress exon 7b and the RBFOX family and UNC-75 activate splicing between exon 7a and exon 8 ( left panel ) . The models may explain why the mutations in unc-75 exerted more sever effects on the selection of exon 7a in the nervous system than the disruption of the RBFOX family genes; in the absence of UNC-75 , exon 7b would be readily spliced to exon 6 , where the target exon of the RBFOX family is no longer left ( right panel ) . Figure 8E illustrates the proposed model of the mutually exclusive selection of unc-32 exon 4 . In neurons , UNC-75 activates splicing both between exon 3 and exon 4b and between exon 4b and exon 5 so that exon 4b alone is selected . In intestine and pharynx , splicing between exon 4a and exon 5 and between exon 3 and exon 4c , respectively , occurs first to determine the fate of the pre-mRNA presumably depending on other tissue-specific factor ( s ) . The proposed order of intron excision for each isoform in this model explains the fidelity of the mutually exclusive selection from the three exons of the unc-32 exon 4 cluster . The number of the mutually exclusive exons in a cluster is at most two in mammals . The fidelity of the mutually exclusive splicing relies on steric hindrance due to close proximity of the mutually exclusive exons [31] , incompatibility between U2-type and U12-type splice sites [32] , splicing regulators that repress one exon and activate the other [12] , [33] and/or mRNA surveillance system [34] . We have previously raised regulation models for two genes with mutually exclusive exons in C . elegans . In the case of egl-15 , the RBFOX family and SUP-12 cooperatively repress the splice acceptor of the upstream exon [22] . In the case of let-2 , ASD-2 activates the splice donor of the downstream exon [21] . In the present study , we demonstrate novel types of regulation; for unc-32 exons 7a and 7b , UNC-75 and the RBFOX family switch the first splicing from E6/E7b to E7a/E8; for unc-32 exons 4a , 4b and 4c , UNC-75 activates both the splice acceptor and the donor of exon 4b . It has been recently suggested that the mutually exclusive exons in the slo-1 gene are regulated in intragenic coordination with downstream alternative splicing events although the splicing patters are not analyzed at a single cell resolution [35] . Thus , the order of intron excision and the modes of regulation for the mutually exclusive exons vary from case to case even in the simple model organism . In this study , we identified the first endogenous alternative splicing events regulated by the CELF3–6 subfamily . A recent splicing-sensitive microarray analysis of the unc-75 mutant suggested only one affected gene , lec-3 [36] , but the selection patterns of the putative target exons in each tissue in vivo are not known yet and the function of UNC-75 in the splicing regulation of the lec-3 gene are to be experimentally defined . In vertebrates , the CELF1–2 subfamily proteins CELF1 ( also known as CUG-BP1 ) and CELF2 ( also known as ETR-3 and CUG-BP2 ) are broadly expressed , highest in heart , skeletal muscle and brain , and their biological functions and biochemical properties are well characterized [29] , [37] . On the other hand , CELF3 to CELF6 are predominantly expressed in the nervous system [38] , [39] , [40] , [41] , [42] and have been shown to regulate alternative splicing in heterologous minigene systems [33] , [38] , [43] , [44] , [45] , [46] . However , the in vivo functions and biochemical properties of the CELF3–6 subfamily are less characterized [29] presumably due to their functional redundancy . We identified the short fragment specifically recognized by UNC-75 in unc-32 intron 7a and provided the genetic and biochemical evidence that all the three RRMs are required for the recognition and regulation of the unc-32 pre-mRNA ( Figure 3C , Figure 6A ) . Among them , RRM3 recognizes the UUGUUGUGUUGU stretch in the target element by itself ( Figure 6C ) . On the other hand , the stretches that RRM1 and RRM2 recognize could not be determined , although our data shown in Figure 5 and Figure 6 do not preclude the possibility that RRM1 and/or RRM2 also recognize the UUGUUGUGUUGU stretch . These results suggest that recognition of target RNAs by RRM1 and RRM2 is context-dependent or cooperative , which may explain why it is difficult to determine the precise binding sites or consensus sequences for RRM1 and RRM2 . The CELF1–2 subfamily has been shown to bind to a variety of UG-rich and related sequences via the three RRMs in a context-dependent manner [47] , [48] , [49] , [50] , [51] , [52] . Considering the amino acid sequence similarities between the two subfamilies ( Figure 3C ) , it is reasonable that UNC-75 also recognizes the UG-rich sequences . Collection of the unc-75 mutant alleles revealed that the conserved stretch in the N-terminal portion of the divergent domain is also involved in the recognition and/or splicing regulation of unc-32 ( Figure S3 ) . This is consistent with the previous reports that the N-terminal portion of the divergent domain of CELF4 is involved in the RNA recognition and/or splicing regulation in minigene contexts [44] , [46] . The RedUnc mutant alleles have nonsense mutations in unc-75 exon 6 or 7 ( Figure 3A , 3B ) , while some other mutants show the Red/Green phenotype ( Figure 3A , Figure S2 ) , suggesting cell-type-dependent remaining activity of UNC-75 in such mutants . Paradoxically , most of the Red/Green alleles have nonsense mutations or splice site mutations in exon 1 , 2 or 3 ( Figure 3B ) , indicative of fatal effects on the UNC-75 expression . The remaining activity of UNC-75 in certain neurons might derive from the use of alternative promoters in the upstream region or in intron 3 to bypass exons 1–3 , although we have not experimentally identified such mRNA isoforms from the unc-75 gene . We demonstrated that the C-termini of all the CELF family and the RBFOX family proteins match the consensus of the PY-NLS and that the C-termini are indeed required for the proper nuclear localization of UNC-75 , ASD-1 and FOX-1 ( Figure 4 ) . As RRM3 of the CELF family resides at the C-terminus , the PY-NLS is overlapping with RRM3 and is highly conserved . It has been reported that deletion of a C-terminal KRP stretch affected the nuclear localization of UNC-75 in neurons [28] , consistent with our finding . In contrast to the PY-NLSs in the RBFOX and CELF families , the PY-NLS was originally identified in the internal portion of hnRNP A1 and other RNA-binding proteins including hnRNP D , TAP , HuR , hnRNP F and hnRNP M [30] . Most of the PY-NLSs predicted in many other proteins are structurally divergent and reside in the internal portion [30] . Evolutionary conservation of the sequences and positions of the PY-NLSs in the RBFOX and CELF families may suggest importance of their positions for the functions of these proteins . In this and previous studies , we demonstrated that the broadly-expressed RBFOX family proteins ASD-1 and FOX-1 regulate the neuron- and muscle-specific alternative splicing events in a target-specific manner in combination with the neuron-specific RNA-binding protein UNC-75 and the muscle-specific RNA-binding protein SUP-12 [22] , respectively . Similarly , an RBFOX family protein RBFOX2 is expressed in a variety of cell types in mammals , yet it can regulate the epithelium-specific alternative splicing of the FGFR2 gene in coordination with epithelium-specific splicing factors ESRP1 and ESRP2 [11] , [12] . The RBFOX family splicing regulators have only one RNA-binding domain that can specifically recognize the ( U ) GCAUG stretch in the target pre-mRNAs [27] , [53] . Therefore , the presence of the ( U ) GCAUG stretch in the pre-mRNAs is not the sole determinant of the tissue-specificity but can be considered to offer an opportunity for the combinatorial and context-dependent regulation of alternative splicing . Considering their broad expression , the RBFOX family may regulate alternative splicing with a variety of tissue-specificity in cooperation with other tissue-specific factors in both mammals and C . elegans .
To construct the unc-32 exon 7 reporter cassettes , the unc-32 genomic fragment was cloned upstream of either mRFP1 [54] or EGFP ( Clontech ) cDNA in the Entry vectors by using In-Fusion system ( BD Biosciences ) and the artificial termination codons were introduced with QuickChange ( Stratagene ) . The reporter minigenes for unc-32 exon 4 and the unc-32 transcriptional fusion were constructed as described previously [20] . The sequences of the primers used in the plasmid construction are available in Table S1 . The worms were cultured following standard methods . Generation of transgenic worms , mutant screening and mapping were performed as described previously [20] . The images of the fluorescence reporter and mutant worms were captured using fluorescence stereoscopes ( MZ16FA and M205FA , Leica ) equipped with color , cooled CCD cameras ( DP71 , Olympus and DFC310FX , Leica ) or a confocal microscope ( FV500 , Olympus ) and processed with Metamorph ( Molecular Devices ) or Photoshop ( Adobe ) . The RT-PCRs were performed essentially as described previously for amplifying the mature mRNAs [20] and the partially spliced RNAs [55] . The RT-PCR products were analyzed by standard agarose gel electrophoresis or by using BioAnalyzer ( Agilent ) and the sequences of the RT-PCR products were confirmed by direct sequencing or cloning and sequencing . The sequences of the primers used in the RT-PCR assays are available in Table S2 . The amino acid sequences of the proteins used in the alignments were retrieved from the protein sequence databases derived from GenBank and RefSeq . The accession numbers are as follows: human CELF1 , NP_006551; CELF2 , NP_001020247; CELF3 , AAK07474; CELF4 , NP_064565; CELF5 , NP_068757; CELF6 , NP_443072; hnRNP A1 , AAH02355; hnRNP D , BAA09525; TAP , AAD20016; hnRNP F , NP_004957; hnRNP H1 , NP_005511; hnRNP H2 , NP_062543; RBFOX1 , Q9NWB1; RBFOX2 , NP_001026865; mouse RBFOX3 , NP_001034256; Drosophila A2bp1 , AAQ22527; C . elegans UNC-75 , AAQ19851; ETR-1 , NP_493673; EXC-7 , CAA85327; HRPF-1 , AAK21490; ASD-1 , NP_497841; FOX-1 , NP_508446 . The amino acid sequences were aligned by Clustal W using Lasergene ( DNASTAR ) . The rabbit polyclonal anti-UNC-75 antiserum ( 9493R2R ) was generated by using denatured His-tagged full-length UNC-75 protein as described previously [55] . The rabbit polyclonal anti-ASD-1 ( RbD8211 ) and -FOX-1 ( RbD8209 ) antisera were generated with the mixtures of synthetic peptides TVEKLNDFDYKVAL+C and C+RGVPQPGRIPTSTA for anti-ASD-1 and C+GKVKDDPNSDYDLQ and C+LPSYQMNPALRTLN for anti-FOX-1 by Operon Biotechnologies ( Tokyo , Japan ) . The expression vectors for untagged UNC-75 and HA-tagged ASD-1 and FOX-1 were constructed by using Destination vectors pDEST-cDNA3 and pDEST-ME18S-3HA ( H . K . ) , respectively . HeLa cells were transfected with the expression vectors by utilizing GeneJuice ( Novagen ) . For UNC-75 , the cells were stained with anti-UNC-75 ( 9493R2R ) , Alexa488-conjugated goat anti-rabbit IgG ( Molecular Probes ) and DAPI ( Vector Laboratories ) and fluorescence images were captured with a compound microscope ( Eclipse E600 , Nikon ) and a CCD camera ( DP71 , Olympus ) . For ASD-1 and FOX-1 , the cells were stained with anti-ASD-1 ( RbD8211 ) or -FOX-1 ( RbD8209 ) , Cy3-conjugated goat anti-rabbit IgG ( Jackson ) and TO-PRO3 ( Molecular Probes ) and the confocal images were acquired with FV1000 ( Olympus ) . The expression vectors for FLAG-tagged ASD-1 , FOX-1 and UNC-75 proteins were constructed using the primers listed in Table S3 and the recombinant proteins were prepared as previously described [55] . The 32P-labelled RNA probes were prepared as described previously [55] using the template oligo DNAs listed in Table S4 and the PCR products amplified with the primes in Table S5 . The EMSAs were performed as described previously [55] . | Tissue-specific and mutually exclusive alternative pre–mRNA splicing is a challenging model for elucidating regulation mechanisms . We previously demonstrated that evolutionarily conserved RBFOX family RNA–binding proteins ASD-1 and FOX-1 and a muscle-specific RNA–binding protein SUP-12 cooperatively direct muscle-specific selection of exon 5B of the C . elegans egl-15 gene . Here we demonstrate that two sets of mutually exclusive exons , 4a–4c and 7a–7b , of the unc-32 gene are regulated in tissue-specific manners and that ASD-1 and FOX-1 , expressed in a variety of tissues , can regulate the neuron-specific selection of unc-32 exon 7a in combination with the neuron-specific CELF family RNA–binding protein UNC-75 . We determine the cis-elements for the RBFOX family and UNC-75 , which separately reside in intron 7b and intron 7a , respectively . By analyzing the partially spliced RNA species , we propose the orders of intron removal and the sites of action for the RBFOX family and UNC-75 in the mutually exclusive selection of exon 7a and exon 7b . We also demonstrate that UNC-75 regulates the neuron-specific selection of exon 4b and propose the models of the mutually exclusive selection of exons 4a , 4b , and 4c . These studies thus provide novel modes of regulation for tissue-specific and mutually exclusive alternative splicing in vivo . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"genetic",
"mutation",
"gene",
"regulation",
"neuroscience",
"animal",
"models",
"caenorhabditis",
"elegans",
"model",
"organisms",
"mutation",
"types",
"molecular",
"genetics",
"gene",
"expression",
"gene",
"splicing",
"biology",
"molecular",
"biology",
"mutagenesis",
"rna",
"processing",
"synapses",
"genetic",
"screens",
"gene",
"identification",
"and",
"analysis",
"genetics",
"molecular",
"cell",
"biology",
"neurophysiology",
"genetics",
"and",
"genomics"
] | 2013 | CELF Family RNA–Binding Protein UNC-75 Regulates Two Sets of Mutually Exclusive Exons of the unc-32 Gene in Neuron-Specific Manners in Caenorhabditis elegans |
Hippo signaling acts as a master regulatory pathway controlling growth , proliferation , and apoptosis and also ensures that variations in proliferation do not alter organ size . How the pathway coordinates restricting proliferation with organ size control remains a major unanswered question . Here we identify Rae1 as a highly-conserved target of the Hippo Pathway integrating proliferation and organ size . Genetic and biochemical studies in Drosophila cells and tissues and in mammalian cells indicate that Hippo signaling promotes Rae1 degradation downstream of Warts/Lats . In proliferating cells , Rae1 loss restricts cyclin B levels and organ size while Rae1 over-expression increases cyclin B levels and organ size , similar to Hippo Pathway over-activation or loss-of-function , respectively . Importantly , Rae1 regulation by the Hippo Pathway is crucial for its regulation of cyclin B and organ size; reducing Rae1 blocks cyclin B accumulation and suppresses overgrowth caused by Hippo Pathway loss . Surprisingly , in addition to suppressing overgrowth , reducing Rae1 also compromises survival of epithelial tissue overgrowing due to loss of Hippo signaling leading to a tissue “synthetic lethality” phenotype . Excitingly , Rae1 plays a highly conserved role to reduce the levels and activity of the Yki/YAP oncogene . Rae1 increases activation of the core kinases Hippo and Warts and plays a post-transcriptional role to increase the protein levels of the Merlin , Hippo , and Warts components of the pathway; therefore , in addition to Rae1 coordinating organ size regulation with proliferative control , we propose that Rae1 also acts in a feedback circuit to regulate pathway homeostasis .
The Hippo Pathway ( also called the Salvador-Warts-Hippo Pathway ) plays a well-appreciated and strongly conserved developmental role in establishing and maintaining organ size . Aberrations in signaling pathways can increase rates of cellular growth or proliferation , but once appropriate organ size is reached , what is coming to be called an “organ size checkpoint” blocks further growth and proliferation; organs do not overgrow unless these aberrations also bypass the “organ size checkpoint” [1] . The precise nature of the signal that restricts cell division in response to organ size checkpoint activation remains unknown . Given that loss of Hippo signaling ( 1 ) results in both tissue and organ overgrowth in Drosophila and vertebrates and ( 2 ) is implicated in a range of cancers including colorectal cancer , liver cancer , melanoma , lung cancer , leukemia , and ovarian cancer [2–11; for review see 12–19] , elucidating this link between proliferation control and organ size control within the Hippo Pathway has important implications for development and disease . The Hippo Pathway consists of a core cassette: Hippo ( Hpo ) , Warts ( Wts ) , Salvador ( Sav ) and Mats [19–25] . Hpo ( homologous to mammalian Mst1 and Mst2 ) , the upstream serine/threonine kinase in the cassette , phosphorylates the scaffold protein Sav ( hWW45 or SAV1 in mammals ) , the downstream kinase Wts ( Lats1 and Lats2 in mammals ) , and Wts co-activator Mats ( Mob1 in mammals ) . Activated Wts then phosphorylates transcriptional co-activator Yorkie ( Yki ) ( YAP and TAZ in humans ) [26] promoting its cytoplasmic retention where it cannot regulate transcription of cell death , cell division , and cell growth regulators such as DIAP1 , cyclin E , and others [27–28] . The pathway is also subject to feedback through Yki/YAP-dependent transcription of upstream regulators such as Merlin ( Mer ) and expanded ( ex ) in Drosophila tissues [26 , 29] , and Lats2 and NF2 in mammalian cultured cells [30] . The core components and Yki/YAP thus play a crucial role in the Hippo Pathway’s global regulation of organ homeostasis . Early characterization of Hippo Pathway mutants uncovered a role for the pathway in regulating mitotic progression , consistent with a role for yeast homologs in the mitotic exit/septation initiation networks . Hpo depletion in Drosophila S2 cells causes mitotic and central spindle defects [31] . Similarly , mats mutant embryos show chromosome segregation defects [32] and Mats over-expression has been shown to regulate cytokinesis [33] , suggesting a role for mats in mitotic exit in Drosophila . Importantly , mats mutant imaginal discs show increased cyclin A ( cycA ) and cyclin B ( cycB ) levels [32] and wts mutant discs show increased cycA levels [34] . The restriction of cycA is functionally important in restricting organ size downstream of Wts [32] . Mutations in the mammalian tumor suppressor components of the pathway have also been extensively characterized for their regulation of centrosomal dynamics [35–36] , mitotic exit/cytokinesis [37–39] , and tetraploidy checkpoint [40] . Together , the data suggest that Hippo Pathway components control appropriate mitotic cyclin levels in Drosophila cells and also have more specific roles regulating the mitotic spindle and genome integrity . YAP and TAZ have not been characterized as regulators of mitotic exit and cytokinesis , so it remains unclear how the pathway regulates cyclin levels and mitotic progression . Understanding this process will shed light on the complicated mechanism by which Hippo signaling coordinates proliferation and organ homeostasis . Here we identify Rae1 as an important and highly conserved regulator of proliferation and organ size . Rae1 is a WD repeat protein first identified for a role in RNA export from the nucleus in yeast [41] and now with reported roles in spindle assembly [42] , regulation of the Anaphase Promoting Complex/Cyclosome ( APCC ) [43–45] , regulation of the E3 Highwire [46] , and spermatogenesis [47] . In this work , we present genetic and biochemical studies showing that Hippo signaling promotes Rae1 degradation downstream of Wts/Lats , and parallel to the pathway’s regulation of Yki/YAP . Importantly , Rae1 is epistatic to Wts in its regulation of cycB , and Hippo signaling regulation of Rae1 is functionally relevant to its organ size functions . Instead of an “on/off” switch for organ growth , our data implicates Rae1 as a molecular rheostat for organ size control . Complementing Yki/YAP’s role to transcriptionally regulate upstream Hippo Pathway components , we also define a role for Rae1 to regulate the levels and activity of Hippo Pathway components post-transcriptionally in a proposed feedback circuit to ensure Hippo Pathway homeostasis .
We identified Rae1 in a Drosophila in vitro expression cloning ( DIVEC ) screen [48–50] for in vitro translated ( IVT ) proteins whose stability or migration on a gel was affected by supplementing IVT reticulocyte lysates with recombinant Mst1 and Mst2 proteins ( S1A–S1C Fig , [50] , experimental detail is provided in the Materials and Methods section ) . To evaluate if Rae1 played a role in Hippo signaling , we first characterized Hippo signaling regulation of Rae1 stability in vitro in tissue culture cells and in vivo in Drosophila . In S2 cells , co-transfected Hippo Pathway tumor suppressor components Mer , hpo , or wts each promoted a reduction in Rae1 levels ( Fig 1A ) . Longer exposures showed a slower migrating band ( Fig 1A ) that decreased when incubated with phosphatase ( S2A Fig ) , and experiments in S2 extracts which preserve proteasomal activity ( see Materials and Methods for extract protocol ) showed accumulation of this band in the presence of MG132 and phosphatase inhibitors ( Fig 1B ) , suggesting that Hippo signaling promotes a phosphorylation-dependent mobility shift and Rae1 degradation by the proteasome . Consistent with this , reducing the gene dosage of hpo or wts in vivo in Drosophila or impairing proteasome function by heterozygosity in E1 ( Uba1 , the most upstream enzyme in the Ubiquitin Pathway ) , increased Rae1 protein levels as monitored by the levels of a GFP-tagged Rae1 transgene , Rae1GFP [46] ( Fig 1C and S2B–S2D Fig ) . RNAi knockdown of hpo or wts stabilized co-transfected Rae1 , and RNAi to wts prevented Hpo-induced degradation of Rae1 in S2 cells ( Fig 1D and S2E Fig ) . Consistent with this , Rae1 protein levels were negatively regulated by Hippo and Warts kinase activity in vivo in Drosophila imaginal discs and salivary glands ( Fig 1E and 1F and S2F Fig ) . The ability of co-transfected wts to destabilize Rae1 ( Fig 1A ) and of wts inhibition ( through RNAi in vitro , Fig 1D and S2E Fig or over-expression of a kinase-dead transgene in vivo , Fig 1F ) to stabilize Rae1 in the presence of over-expressed Hpo indicates that Wts activity is required downstream of Hpo for regulating Rae1 protein levels in Drosophila cells and in tissues . Importantly , regulation of Rae1 by Hippo signaling is highly conserved . Activating Hippo signaling by over-expressing Mst1 and/or Lats1 promoted loss of co-transfected Rae1 in immortalized HEK-293T cells ( Fig 1G ) and loss of endogenous Rae1 in tumorigenic U87-MG ( Fig 1H ) or HeLa cells ( S2G Fig ) . Rae1 loss was dose-responsive to Hippo signaling ( S2H Fig ) and not due to cell death ( S2I Fig ) . To address if Rae1 is a direct target of the Warts/Lats kinase , we tested if immunoprecipitated Rae1 was recognized by a phospho-RXXS antibody ( the consensus Lats1 site [53] ) ( S3A Fig ) . The percentage of immunoprecipitated Rae1 phosphorylated at the RXXS site increased in a dose-responsive manner to increased pathway activation ( S3A Fig ) . Like many WD repeat proteins , recombinant Drosophila or human Rae1 purified from bacteria was insoluble and refolding attempts resulted in largely aggregated protein unsuitable for direct kinase assays . Therefore , we utilized small Rae1 peptides containing the putative Rae1 phosphorylation site ( S3B Fig ) . Despite recognition of endogenous Rae1 by the phospho-RXXS antibody , purified system kinase assays using recombinant Lats2 and a Rae1 peptide of this site failed to show phosphorylation even when showing robust phosphorylation of a control YAP peptide ( S3B Fig ) . Kinase assays with full-length baculovirus-produced Rae1 ( a gift from Y . Ren and the Blobel lab , [54] ) showed insignificant phosphorylation by Lats2 compared to a positive control ( S3C Fig ) . Recognition by phosho-RXXS antibodies but failure of Lats2 to recognize Rae1 peptides or baculovirus-expressed Rae1 may reflect that Warts/Lats kinase directly phosphorylates full length Rae1 when in specific complexes with other proteins or requires a priming phosphorylation . Alternatively , the Warts-dependent Rae1 targeting observed ( Fig 1 and S2 Fig ) may occur further downstream . To exclude that changes in Rae1 are in part due to a transcriptional effect of Yki/YAP , we conducted qRT-PCR of adult heads expressing a yki transgene ( GMR>Yki , Fig 1I ) and of mammalian cells over-expressing activated YAP ( YAPS127A and YAPS5A , Fig 1J ) . Both cases showed increased levels of well-characterized transcriptional targets ( expanded , a Yki target in flies [26] and CTGF , a transcriptional target of YAP in mammalian systems [52] ) , confirming increased Yki/YAP transcriptional activity , but did not show increased Rae1 transcripts . Our in vitro extract experiments in the presence of cycloheximide ( Fig 1B ) showed accumulation of a slower-migrating form of Rae1 from the initial time point to the 30 minute time point; this does not rule out that Hippo signaling affected Rae1 via transcriptional means in cells before extract creation but does suggest a means by which Hippo signaling regulates Rae1 post-translationally . If this occurred by a non-transcriptional role of Yki to regulate Rae1 protein levels , modulating the levels of Yki should modulate Rae1 levels . Reducing yki levels by RNAi had no substantial effect on Rae1 protein levels or localization in S2 cells or in larval tissues ( Fig 1K and 1L and S3D–S3G Fig ) . Consistent with this , over-expressing activated YAP in 293T cells did not increase Rae1 protein levels ( S3H Fig ) . These findings suggest that the Hippo Pathway does not downregulate Rae1 levels through Yki/YAP via transcriptional or post-translational mechanisms . To investigate if the negative regulation of Rae1 by the Hippo Pathway is functionally relevant in restricting proliferation , organ size , and promoting apoptosis , we first characterized the phenotypes of reducing or over-expressing Rae1 in vivo in the fly . We used previously characterized Rae1 deletion allele Rae1ex28 [46] and four inducible RNAi lines corresponding to three independent inverted repeat alleles: P{GD14705}v29303 from the VDRC referred to here as Rae1IRV; 9862R-2 and 9862R-3 from the NIG collection , referred to here as Rae1IRN2 and Rae1IRN3 , and P{TRIP . HMS00670} from the Transgenic RNAi Project referred to here as Rae1IRT . The inverted repeat in Rae1IRN2 and Rae1IRN3 is partially overlapping with Rae1IRV; Rae1IRT is entirely non-overlapping with Rae1IRV , Rae1IRN2 , and Rae1IRN3 . To increase Rae1 gene dosage , we created inducible transgenic alleles Rae102 and Rae103 , and used previously characterized Rae1 transgenic allele Rae1GFP [46] ( relative mRNA levels for a subset of these is shown in S4A Fig ) . Larvae homozygous for deletion of Rae1 or undergoing strong , constitutive Rae1 RNAi died as small wandering third-instars ( Fig 2A and 2B , [46] ) . Their imaginal discs were smaller than control heterozygous animals . Reducing Rae1 levels by low level RNAi resulted in viable adults of reduced weight and size ( Fig 2C ) . Rae1 RNAi in the developing wing disc using nubgal4 resulted in adult flies with smaller wings ( Fig 2D–2F ) . Similar phenotypes were observed using different wing drivers and additional RNAi lines ( S4B–S4J Fig ) or by RNAi in a stripe in the wing ( S4K Fig ) . Rae1 RNAi in the proliferating cells of the developing eye disc using eygal4 , resulted in adult flies with smaller eyes ( Fig 2G–2J , quantified in S2L Fig ) . Eyes containing primarily homozygous Rae1ex28 mutant tissue were also small ( Fig 2K and 2K’ , quantified in S4M Fig ) . Furthermore , organ size reduction was seen with non-overlapping RNAi lines , and Rae1 over-expression rescued RNAi phenotypes in the eye and wing ( S4N–S4P Fig ) indicating that the reduced organ size phenotypes resulted specifically from Rae1 reduction and not off-target effects . In contrast , Rae1 RNAi in differentiating eye cells ( GMR>Rae1IRV ) resulted in no obvious phenotype ( Fig 1L and 1L’ and S4Q Fig ) . The Rae1 loss-of-function phenotypes could result from an essential cell function or from a normal role of Rae1 to promote organ size . If Rae1 positively regulates organ size , its over-expression should increase organ size . Constitutive Rae1 over-expression increased overall organism size in terms of weight and body length ( Fig 2M and S4R Fig ) , increased wing size ( Fig 2O and 2P , quantified in 2Q ) , and increased eye size ( S4S-S4S’ Fig ) . Over-expressing Rae1 in the whole wing or specific compartments also increased wing size ( Fig 2N ) . Over-expressing Rae1 in the proliferating cells ( Fig 2R and 2R’ ) ( but not the differentiating cells only , Fig 2S and 2S’ , quantified in S4Q Fig ) of the eye increased eye size . Despite larger overall size , Rae1 over-expressing eyes appeared normally-patterned and showed no change in ELAV ( a neuronal marker ) expression in third instar eye discs ( S4T Fig ) . This is in stark contrast to Yki over-expression in the early eye . Wild-type and constitutively active Yki expression in proliferating cells of the eye with eygal4 or constitutively with Actgal4 resulted in loss of eye structures ( S4U-S4V’ Fig ) reminiscent of ex loss [55–56] and as seen with Yki expression limited to the dorsal-ventral margins with bigal4 [57] . As with loss of ex , the block in differentiation from Yki over-expression with eygal4 in our study or bigal4 [57] was suppressed by loss of wingless ( wg ) so likely resulted from effects of increased wg blocking progression of the Morphogenetic Furrow ( MF ) . All together , the organ size phenotypes of Rae1 loss-of-function or over-expression are consistent with a role for Rae1 to promote organ size and are consistent with Rae1 inhibition by the Hippo Pathway ( Fig 1 and S2 Fig ) to restrict organ size . Effects on organ size can result from changes in cell size . Forward scatter of cells dissociated from dissected mosaic wing discs containing clones of Rae1 RNAi using Rae1IRV and Rae1IRN2 showed no difference in size of cells undergoing RNAi to Rae1 ( GFP-positive cells ) compared to control cells ( GFP-negative cells ) ( S5A Fig ) . Similarly , forward scatter showed no difference in size for Rae1 over-expressing cells ( GFP-positive cells ) compared to control cells ( GFP-negative cells ) ( S5B Fig ) . This suggests that the smaller organ size of Rae1 RNAi and the larger organ size of Rae1 over-expression did not result from changes in cell size . Smaller organs could also result from increased cell death or from differentiation into other structures . We saw no obvious increase in anti-activated caspase 3 staining or in TUNEL assays upon Rae1 RNAi ( S5C and S5D Fig for TUNEL ) . Moreover , co-expressing caspase inhibitor p35 did not suppress the eye size phenotype of Rae1 RNAi ( S5E Fig ) . Consistent with previous studies that Rae1 RNAi in S2 cells did not promote apoptosis [58] , these findings suggest that decreased organ size did not result from increased apoptosis . We observed no effects on ELAV staining upon Rae1 RNAi in actively dividing cells in the early eye ( using eygal4 ) ( S5F Fig ) suggesting that reducing Rae1 does not cause premature differentiation to reduce organ size . Smaller organs could result from decreased proliferation . In the Drosophila larval eye , a wave of differentiation , the MF , passes from posterior to anterior . A subset of cells undergo an additional round of coordinated division called the Second Mitotic Wave ( SMW ) which appears as a synchronized stripe of dividing cells just posterior to the MF . Cells posterior to the SMW in the eye disc are not dividing at this stage and so do not normally stain for pHH3 and do not undergo BrdU incorporation . When cells in this region are induced to undergo ectopic division ( for example , due to over-expression of an oncogene ) , individual cells will incorporate BrdU or stain for cell cycle markers cycA , cycB , or pHH3 ( appropriate for their cell cycle phase ) . If cells enter S-phase but are endoreplicating or stall before mitosis ( so are not actively cycling ) , they do not stain for pHH3 . When cells that have entered the cell cycle arrest in mitosis , cells in the tissue do not incorporate BrdU but stain for pHH3 . Effects of Rae1 on the cell cycle can thus be assessed with cell cycle markers in the third instar larval eye . Tissue undergoing Rae1 RNAi ( GFP-positive ) prior to differentiation showed reduced BrdU incorporation most obviously in the SMW compared to adjacent wild-type tissue ( GFP-negative ) , consistent with decreased entry into S-phase ( Fig 3A and 3A’ and S5G-S5H’ Fig ) . This is consistent with a previously reported role for Rae1 in the G1-S transition in cell culture [46] . Despite decreased BrdU incorporation , Rae1 RNAi clones in the eye disc did not show decreased phospho-histone H3 ( pHH3 ) staining ( Figs 3B , 3B’ , S5I and S5I’ ) . Reduced BrdU incorporation but no obvious reduction in pHH3 staining is perplexing but suggests that cells undergoing Rae1 RNAi that do enter the cell cycle endure a prolonged stay or arrest in mitosis . To further investigate this mitotic phenotype , in addition to constitutive Rae1 RNAi in small clones , we examined eye discs undergoing Rae1 RNAi across all cells anterior to the MF using eygal4 . These discs were smaller than controls but , surprisingly , showed increased pHH3 staining in the antenna , anterior to the MF , and , strikingly , posterior to the MF compared to controls ( Fig 3C–3E; because the increased pHH3 staining made it difficult to distinguish the posterior border of the SMW , pHH3 staining was quantified for the regions anterior to the MF versus posterior to the MF including the SMW ) . Consistent with findings in the eye , we also observed an increase in pHH3-positive cells in larval wing discs ( Fig 3F and S5J–S5M Fig ) and in S2 cells ( S5N Fig ) upon Rae1 loss . As noted , cells posterior to the SMW do not divide during the third instar larval stage and therefore should not stain positive for pHH3 . We saw no significant BrdU incorporation or inappropriate cycA and cycB staining posterior to the SMW in these discs ( S5O–S5R Fig ) indicating that ( 1 ) these pHH3-positive cells were not actively cycling and ( 2 ) presumably they had completed cyclin degradation potentially placing them in anaphase or telophase . Visual examination of nuclei in S2 cells undergoing Rae1 RNAi showed significant abnormalities including multipolar spindles , inappropriately localized tubulin , and lagging chromosomes ( S6 Fig ) , consistent with reports that Rae1 depletion causes disorganized or multipolar spindles as well as chromosome alignment and segregation defects in cultured human and plant cells [42 , 59–60] and in Drosophila neuroblasts and spermatocytes [47] . The requirement for Rae1 in cellular proliferation is conserved in mammalian cells; transient Rae1 knockdown in various mammalian transformed and tumorigenic cell lines reproducibly restricted proliferation ( Fig 3G–3I and S7A–S7D Fig ) . These cells have intact p53 signalling and did not show elevated p21 transcription ( S7E–S7G Fig ) , suggesting the proliferative arrest is likely independent of p53 . A decrease in proliferation upon Rae1 loss might indicate a role for Rae1 to promote proliferation . Over-expressing Rae1 in the early eye resulted in increased BrdU incorporation ( Fig 3J and S7H and S7I Fig ) and increased pHH3 staining anterior to MF ( Fig 3K and 3K’ ) suggesting that Rae1 plays a role to promote proliferation . Analogously , Rae1 over-expression promoted proliferation in both 293T and HeLa cells ( Fig 3L and 3M ) . Taken together , these findings reflect a highly conserved role for Rae1 in proliferation . Previous studies reported genetic interactions between wts and cycA to regulate organ size [32] , and wts loss affected cycE , cycA , and cycB levels ( S8A and S8B Fig ) . Although accumulation of cycA and cycB has been reported in a variety of Hippo Pathway mutants , the functional mechanism underlying their regulation by Hippo signaling has remained unresolved . A prior report linked decreased S-phase entry of Rae1 loss to cycE [58] , and regulation of cycE by the Hippo Pathway has already been established . Therefore , to further investigate the Rae1 loss-of-function mitotic phenotypes and establish if they underlie Hippo signaling regulation of mitosis , we proceeded by examining cycA and cycB levels upon Rae1 knockdown . Clones undergoing Rae1 RNAi anterior to the MF and in the SMW showed subtly decreased cycA and cycB staining compared to adjacent control tissue ( Fig 4A–4D and S8C and S8C’ Fig ) . To establish if decreased cyclin levels were functionally relevant to reduced organ size caused by Rae1 knockdown , we reduced cycA and cycB gene dosage . Heterozygous mutation in cycA and cycB across the fly each dominantly enhanced the reduced organ size phenotypes of Rae1 RNAi in the eye and wing ( Fig 4E–4H and S8D–S8G Fig ) but did not reduce the size of control wings ( S8H Fig ) . Conversely , individually over-expressing cycE , cycA , or cycB3 partially suppressed reduced eye size caused by reduction in Rae1 ( S8I–S8M Fig ) despite producing no overgrowth of control eyes ( S8N–S8R Fig ) . The reduced cyclin levels upon reduction of Rae1 together with the genetic interaction studies suggest a normal role for Rae1 to positively regulate cycA and cycB levels to promote proliferation and organ size . Consistent with this , cycA and cycB staining increased in Rae1-over-expressing clones and discs ( Fig 4I and 4I’ and S8S–S8U Fig ) . Furthermore , increasing Rae1 in the larval wing disc increased cycB protein levels in a dose-dependent way ( Fig 4J ) . How does Rae1 regulate cycA and cycB ? Rae1 was identified for a role in RNA export [41] and was later reported to inhibit the Anaphase Promoting Complex/Cyclosome ( APCC ) activator Cdh1/Fizzy-related ( fzr , also called rap; referred to here as Cdh1/fzr ) in mammalian cells [43–45] . The APCC is a ubiquitin ligase that targets the mitotic cyclins and has an essential role in mitosis . When coupled to substrate-specific activators Cdc20/Fizzy ( fzy ) and Cdh1/fzr , the APCC ubiquitinates substrates to direct them for proteasomal degradation . Excess Cdh1/Fzr activity upon Rae1 loss could explain reduction of cycA and cycB . To establish if Cdh1/Fzr misregulation played a role in Rae1 organ size phenotypes , we tested interactions with APCC components and regulators . We saw no obvious change in reduced organ size of Rae1 RNAi in the eye by removing one copy of APCC subunit Cdc27 or Cdc20/fzy . However , removing one copy of Cdh1/fzr across the fly dominantly suppressed the reduced organ size of Rae1 RNAi in the eye and wing ( Fig 4K–4L” ) but did not increase the size of control organs ( S8V Fig ) . Although we cannot rule out parallel regulation of cycB by Cdh1/Fzr-APCC , these findings are consistent with a model that the reduced organ size resulted from specific effects of Rae1 on Cdh1/Fzr-APCC , not Cdc20-APCC . Proliferation is necessary to achieve appropriate organ size , but changes in cell cycle regulation are not sufficient to increase organ size . For example , loss of cell cycle regulators such as cyclin E ( cycE ) can be dramatic enough that they cannot be compensated for sufficiently to achieve normal organ size ( S8X and S8Y Fig ) , but over-expression of cycE , cycA , or cycB3 on their own do not increase organ size [32] ( S8N–S8R Fig ) . Over-expressing Hpo , Sav and Wts , or Wts alone in differentiating cells of the Drosophila eye reduces eye size . Eyes become smaller and rougher and black tissue appears with increased expression of Hpo ( S9A–S9D Fig ) ; these phenotypes are suppressed by reducing wts gene dosage ( S9E–S9G Fig ) . If these phenotypes result in part by promoting Rae1 degradation , they would be enhanced by further reducing Rae1 and suppressed by restoring Rae1 levels . Removing one copy of Rae1 on its own ( with Rae1ex28 ) or with a deficiency that uncovers it ( Df ( 2R ) ED3923 or by Rae1 RNAi in differentiating eye cells ( GMR>Rae1IRV ) resulted in no obvious phenotype ( Fig 2L and 2L’ ) but enhanced the phenotype of over-expressing Hpo ( GMR Hpo , Fig 5A–5A’ and S9H–S9J’ Fig ) , Sav and Wts ( GMR Sav , Wts , Fig 5B and 5B’ and S9K–S9K’ Fig ) , and Wts alone ( GMR Wts , Fig 5C–5C’ ) in differentiating eye cells in terms of both eye size and the appearance of black tissue . In contrast , co-over-expressing Rae1 in differentiating eye cells ( GMR>Rae102 ) or constitutively ( Act>Rae1GFP ) strongly suppressed the small eye caused by GMR Hpo ( Fig 5D and 5D’ and S9L and S9L’ Fig ) but resulted in no obvious phenotype on its own ( Fig 2S and 2S’ ) . Similarly , the small wing phenotype caused by Hpo over-expression in the wing was suppressed by reducing the gene dosage of wts or by Rae1 over-expression ( Fig 5E–5G and S9M–S9N Fig ) . These in vivo findings are consistent with tissue culture findings that Hippo signaling negatively regulates Rae1 downstream of Wts ( Fig 1 and S2 Fig ) and provide evidence that this regulation , regardless of direct targeting by Wts or targeting further downstream , plays a role in the Hippo-mediated restriction of organ size . Generally , RNAi reduces but does not eliminate gene expression; Rae1 RNAi should lead to Rae1 protein at lower levels subject to its endogenous post-translational regulation . Therefore , reducing the gene dosage of negative regulators of Rae1 protein should suppress Rae1 RNAi organ size phenotypes . Indeed , mutations in Mer , ex , hpo , or wts dominantly restored eye and wing size in organs undergoing Rae1 RNAi ( S10A and S10B Fig and S1 Table for hpo , Mer , and additional effectors on the eye , and S10C–S10E Fig for hpo and wts effects on the wing ) . Together with our in vitro findings ( Fig 1 and S2 Fig ) , these genetic interactions support a role for Hippo signaling to negatively regulate Rae1 in vivo to restrict organ size . Hippo Pathway downregulation of Rae1 ( Fig 1 and S2 Fig ) , Rae1 inhibition of Cdh1/Fzr [43–45] , and Rae1 interaction with Cdh1/Fzr in vivo ( Fig 4K–4L” ) would further suggest that Hpo promotes Cdh1/Fzr activation by relieving Cdh1/Fzr inhibition by Rae1 . Therefore , Hpo over-expression phenotypes may in part result from excess Cdh1/Fzr activity . Indeed , removing one copy of two distinct alleles of Cdh1/fzr partially restored GMR Hpo eye size ( Fig 5H–5H” for fzrG0326 , S10F Fig ) suggesting a possible functional link between the Hippo Pathway and the essential cell cycle ubiquitin ligase , APCC . Previous studies reported genetic interactions between wts and cycA to regulate organ size [32] . We observed that wts loss affected cycE , cycA , and cycB levels ( S8A and S8B Fig ) . If the cyclin decreases in the context of Hippo signaling result from Rae1 depletion , then restoring Rae1 levels should prevent Wts-mediated cyclin decrease . In the context of Wts over-expression , expressing a low level of Rae1 restored cycB protein to control levels in S2 cells ( Fig 5I ) . By the logic above , if accumulation of the cyclins in the absence of Hippo signaling resulted from Rae1 accumulation , then reducing Rae1 levels in those contexts should suppress cyclin accumulation phenotypes . As explained shortly , it is difficult to perform epistatic analysis with double mutant tissue , so we were limited in the contexts in which to perform epistasis experiments . Normal cyclin levels were restored in homozygous mutant wts or sav tissue in eye discs ( generated using MARCM tools ) upon reducing Rae1 ( by low level Rae1 RNAi or removing one copy of Rae1 ) ( Fig 6A and 6B ) . Together with the Wts-over-expression studies in Fig 5I , these findings indicate that the Wts regulation of cycB occurs through the downregulation of Rae1 . Knocking down Mer , ex , hpo , or wts in the posterior or whole wing or inhibiting Hippo signaling with a kinase dead version of Hpo resulted in dramatic wing overgrowth ( Fig 6D , 6G , 6J , 6M , 6O and 6R and S11C , S11G , S11N , S11R , S12A , S12C , S12E and S12H Figs ) compared to controls ( Fig 6C , 6I and 6Q and S11A , S11E , S11L , S11P and S12G Figs ) . Reducing the levels of critical target Yki suppresses this overgrowth ( [26]; Fig 6S and S12F and S12J Fig ) . Similarly , in contexts of little overgrowth , low-level Rae1 RNAi causing mild or no change in wing size on its own or removing one copy of Rae1 significantly reduced overgrowth due to loss hpo or over-expression of kinase dead Hpo ( Fig 6E and 6H and S11D , S11I and S11J Fig ) . This suggests that the accumulation of Rae1 upon loss of Hippo signaling ( Fig 1C and 1D and S2B , S2D and S2E Fig ) is important for the resulting tissue overgrowth . Surprisingly , reducing Rae1 gene dosage in many of these overgrowth contexts also gave rise to shriveled and blackened wings , a “tissue synthetic lethality” phenotype . As noted above , in contexts of little overgrowth , Rae1 knockdown suppressed overgrowth ( Fig 6E , 6H and 6J and S11D , S11I and S11J Fig ) or led to blistering ( S12B Fig ) . Upon much lower levels of Hippo Pathway activity that causes more severe overgrowth , Rae1 knockdown caused catastrophic tissue loss ( Fig 6F , 6H” , 6K–6L , 6N , 6P , 6U and 6U”; S11H , S11I’ and S11J’ Fig ) . Enhanced overgrowth can cause tissue collapse; however , in these cases overall wing size decreased coincident to tissue collapse , reflecting suppressed not enhanced overgrowth . This is also highlighted when we reduce hpo , ex , or Mer in only the posterior compartment using engal4 ( Fig 6K–6L , 6N and 6P and S12B and S12D Fig ) ; anterior tissue remains intact indicating that the wings inflated and only posterior tissue was collapsing . We saw tissue loss throughout the body including in the thorax , the legs , and the eye when we created random clones in proliferating tissue ( S12K Fig for the eye ) and compelling growth suppression and tissue loss using multiple gal4 drivers including nubgal4 ( Fig 6C–6F ) , ms1096gal4 ( Fig 6G–6H”‘ ) , engal4 ( Fig 6I–6P and S12A–S12D Fig ) , and c765gal4 ( referred to as c5gal4 Fig 6Q–6V ) . We did not observe catastrophic tissue loss when using GMRgal4 , suggesting this phenomenon may reflect sensitivity of proliferating , not differentiated , tissue . Importantly , this synthetic tissue lethality upon reducing Rae1 in the context of limited Hippo Pathway activity is characteristically distinct from Yki , reduction of which suppressed overgrowth upon loss of Hippo signaling but did not compromise the survival of overgrowing tissue ( Fig 6S and S12F and S12J Fig ) . Tissue lethality instead of straight-forward suppression was unanticipated because ( 1 ) we reduced Rae1 to a level with minimal or no phenotypes on its own ( removing one copy or low-level RNAi ) , ( 2 ) even significant knockdown or knockout of Rae1 does not cause tissue lethality , and ( 3 ) Rae1 reduction would be expected only to decrease proliferation and decrease organ size . When lethality results in genetic interactions upon modulating two genes each of whose individual modulation is not lethal , the term”synthetic lethality” is applied and usually results from perturbing genes with parallel , redundant roles or between genes of the same pathway . The “tissue synthetic lethality” was specific for Hippo signaling and not a global response of overgrowing tissue . We did not see tissue lethality upon Rae1 RNAi or removing one copy of Rae1 with expression of oncogenic Ras , myc , or caspase inhibitor p35 ( S12L–S12O Fig for Myc and p35 ) . The “synthetic lethality” in this instance , therefore , might suggest that tissue with impaired Hippo signaling requires Rae1 for survival and further supports a role for Rae1 in Hippo signaling . To determine whether Rae1 could promote “synthetic lethality” of Yki-over-expressing tissue , we reduced Rae1 levels in a range of Yki over-expression contexts including contexts that matched or exceeded the extent of overgrowth seen for reducing Mer , ex , hpo , and wts . When Yki caused moderate overgrowth matching those shown for loss of Mer , ex , hpo , or wts , Rae1 reduction in some cases suppressed the wing overgrowth but did not cause tissue ablation ( Fig 7A–7D and S13A–S13R Fig ) . The lack of tissue collapse of Rae1 knockdown upon Yki over-expression emphasizes that there is a fundamental difference between loss of Hippo signaling and Yki over-activation and that the “tissue synthetic lethality” phenomenon is restricted to specific components of the pathway including Mer , ex , hpo , and wts . In cases of Yki over-expression where there was even more dramatic overgrowth , Rae1 reduction did not suppress the overgrowth but unexpectedly enhanced Yki over-expression adult phenotypes ( S13D and S13S–S13T’ Fig ) . We used a moderately activated transgene ( YkiS168A ) to determine whether the enhancement caused by Rae1 loss was more robust with a higher threshold of Yki activity than that caused by wild-type Yki over-expression . Importantly , this moderately activated form of Yki , YkiS168A , is still responsive to Hippo Pathway regulation since co-expression of Hpo and Wts reduces the size of YkiS168A clones and suppresses Yki-mediated eye overgrowth [61] . Because this transgene causes adult lethality with promoters that drive expression in proliferative tissues , we analyzed interactions between Rae1 and YkiS168A in carefully-staged larval organs . Removing a genomic copy of Rae1 dramatically enhanced the Yki-mediated overgrowth phenotypes in imaginal discs ( shown for wing and eye discs in Fig 7E–7G and 7I–7K but also seen in leg imaginal discs ) . Rae1 knockdown using RNAi further enhanced these phenotypes ( Fig 7H and 7L ) . In the salivary glands , YkiS168A mis-expression restricted salivary gland size ( Fig 7M and 7N as reported previously , [62] ) which was enhanced with Rae1 loss ( Fig 7O and 7P ) . In addition to their changes in size , these tissues showed stronger YkiS168A fluorescence upon Rae1 reduction ( for example , the disc in Fig 7H compared to 7F ) , suggesting that Rae1 may negatively affect Yki protein levels . Rae1 reduction reproducibly increased the levels of both V5 and FLAG-tagged wild-type Yki in wing disc lysates ( Fig 8A and 8B ) . In salivary glands , mis-expressed Yki migrated as a doublet , presumably because of phosphorylation mediated by Hippo Pathway activity ( reported to be high in the salivary glands [62] ) . Rae1 loss caused an increase in Yki levels and reduced the proportion of the slower-migrating form ( Fig 8 and S13U Fig ) . Effects on Yki were conserved in mammalian cells; HeLa cells knocked down for Rae1 showed increased YAP levels compared to control-treated cells ( Fig 8C ) . Consistent with studies of Yki localization in the wing [28] , immunofluorescence of Yki mis-expressing salivary glands showed that both wild-type and YkiS168A are predominantly cytoplasmic ( Fig 8D–8D” for YkiS168A ) . Rae1 loss promoted nuclear localization of YkiS168A ( Fig 8E–8E” ) . This restriction of Yki localization is conserved in mammalian cells; Rae1 knockdown promoted accumulation and nuclear localization of YAP in transformed , non-tumorigenic mammary epithelial cells ( Fig 8G–8G” ) compared to control cells ( Fig 8F–8F” ) . Furthermore , this YAP accumulation and relocalization promotes YAP transcriptional activity ( Fig 8H ) . The increase in Yki/YAP protein levels upon loss of Rae1 suggests that Rae1 plays a role to limit Yki/YAP levels . Consistent with this , Rae1 over-expression reduced mis-expressed Yki protein levels ( Fig 8I ) . Given that a genome-wide mass-spec study reported direct binding between Rae1 and Yki [63] , we cannot exclude that some effects of Rae1 on Yki/YAP might be mediated by direct interactions . However , in experiments exploring the role of Rae1 in Wts-mediated cyclin regulation , we had observed that increased Rae1 levels stabilized myc-Wts ( Fig 5I ) . Because post-translational effects on Yki/YAP levels , localization , and activity are known to result from targeting by the Hippo Pathway [26–28 , 61] , the Rae1 effects on Wts protein in Fig 5I could explain the regulation of Yki/YAP . This raises the possibility that in addition to serving as a target of the pathway , Rae1 could act in a feedback circuit to promote Hippo Pathway activity at a step upstream of Yki/YAP . To investigate the potential for Rae1 to regulate upstream components , we examined their levels and activity in vitro and in vivo . Rae1 over-expression increased Wts protein levels in S2 cells and Drosophila tissue ( Fig 9A and 9B and Fig 5I ) . Conversely , Rae1 loss reduced Wts protein levels in Drosophila tissue ( Fig 9C ) and reduced Lats1 activation in mammalian cells ( Fig 9D ) , indicating that Rae1 regulation of Wts/Lats is conserved . Rae1 over-expression increased Hpo and Mer protein levels in S2 cells and Hpo activation in Drosophila tissue ( Fig 9E–9G ) , suggesting that Rae1 might act at or upstream of Mer , or on multiple components when in complex together . Because Rae1 did not promote accumulation of other proteins tested including GFP ( Fig 1E ) and promoted reduction in Yorkie/YAP levels ( Fig 8I ) , effects on Mer , Hpo , and Wts protein levels are unlikely due to a non-specific effect to stabilize all proteins . Activation of Hippo signaling requires proper recruitment of the Hpo and Wts kinases to specific regions in the apical membrane from distinct domains by upstream components of the pathway [64–65] . Mer and Ex are membrane-associated proteins that facilitate this activation of Hippo signaling by recruiting Wts/Lats to the membrane where it receives its activating phosphorylation from Hippo/MST [64–65] . To assess how Rae1 , a primarily nuclear protein , could affect the protein levels and activation of these components , we looked more closely at Rae1 localization . Consistent with previous reports about Rae1 localization in other systems [42 , 59–60 , 66] , a Rae1-GFP fusion protein was strongly enriched in the nucleus and nuclear periphery in various Drosophila tissues ( Fig 9H and S14A–S14A” Fig ) . Rae1 was also found to be associated with a mesh-like network in the cytoplasm . Importantly , at higher levels of Rae1 ( resulting from reduced Hpo or Uba1 or from increased Rae1 expression ) a pool of Rae1 localized to the membrane ( Fig 9I and 9J and S14C–S14C” Fig ) whereas at lower levels of Rae1 ( such as upon co-expression of Hpo , S14B–S14B” Fig and S14D–S14D” Fig ) , this pool disappeared . Curiously , in addition to promoting accumulation of Wts protein levels ( Figs 5I and 9A and 9B ) over-expressing Rae1 increased the membrane association of Wts protein , including some areas of co-localization ( Fig 9L and 9L’ compared to Fig 9K and 9K’ ) . Increased Wts at the membrane could reflect the increase in overall Wts levels ( Figs 5I and 9A and 9B ) . Because Mer acts to increase Wts recruitment to the membrane , increased Wts at the membrane could also reflect the increased Mer levels ( Fig 9G ) . Mer and Rae1 both bind microtubules in purified systems [42 , 67] . Mer’s association with microtubules is increased upon microtubule acetylation , and interaction with acetylated microtubules is critical in regulating YAP [67–69] . Therefore another potential explanation for the increased Wts recruitment to the membrane could be via Rae1 effects on microtubules or microtubule acetylation which could affect Mer . Rae1 over-expression in Drosophila tissues dramatically increased the proportion of acetylated tubulin while tissue undergoing Rae1 RNAi showed reduced acetylated tubulin ( Fig 9M and 9N ) . All together , these findings would be consistent with higher levels of Rae1 activating the Hippo Pathway at the membrane my mutiple mechanisms: ( 1 ) Rae1 could act at an upstream step to promote tubulin acetylation to regulate Mer , and/or ( 2 ) Rae1 could promote accumulation of core components Mer , Hpo , and Wts which would then promote downregulation of both Rae1 and Yki/YAP ( Fig 10 ) .
The Hippo Tumor Suppressor Pathway plays a highly conserved role from Drosophila to mammals in organ homeostasis , in restricting growth and proliferation , and in promoting apoptosis . By searching for post-translational targets , we identified Rae1 as a novel target of Hippo signaling downstream of Wts with a role to regulate proliferation , cycB , and organ size . Our studies showed that Rae1 is regulated by Wts in vitro and Wts kinase activity in vivo . Whether Rae1 is a direct target of Wts remains an open question . Rae1 binds both Wts and Lats in Drosophila and mammalian cultured cells and increasing Hippo activity shows increased recognition of Rae1 by a pRXXS antibody . However , an in vitro kinase assay using a peptide encompassing the Wts consensus motif and full-length purified Rae1 showed no phosphorylation . These data raise the possibilities that Rae1 phosphorylation by Wts would require an additional co-factor or a priming phosphorylation , or that Rae1 is targeted downstream of Wts by another kinase . Ongoing studies are centered on resolving how the Hippo Pathway targets Rae1 for degradation . Unlike DIAP1 , which is regulated both transcriptionally by Yki [26] and post-translationally by phosphorylation by the pathway [19 , 21] , we found no evidence that Rae1 transcript or protein levels are regulated by Yki in Drosophila tissues . This is consistent with previous RNA-seq and microarray studies that showed no major changes in Rae1 mRNA in Yki over-expressing or wts mutant contexts [75–77] . Our studies in mammalian cell lines also show no evidence for Rae1 transcriptional regulation by YAP activity . Consistent with this , microarray studies of activated YAP [9 , 52 , 78–81] as well as ChIP-on-chip , ChIP-seq and RNA-seq studies [52 , 81–83] did not show statistically significant regulation of Rae1 mRNA in a variety of mouse tissues and human cell lines . Rae1 regulation thus may represent another example of Yki-independent functions of the Hippo pathway including a variety of developmental processes such as dendritic tiling [84] , planar polarity [85] , salivary gland and neuronal autophagy [62 , 86] as well as in growth control contexts including that elicited by F-actin accumulation [87] and alcohol [88] , the mechanisms of which remain unresolved . Recent work also identified novel Yki-independent pathway effectors such as Enabled , Mud , and Canoe in processes such as collective migration [89] , and spindle orientation [90–91] . Phenotypic characterization revealed that Rae1 acts as a regulator of organ and organism size and as a critical regulator of mitosis . While both loss-of-function and gain-of-function studies showed increased mitotic index , cells in the former context were not actively cycling and each context produced distinct outcomes in terms of cyclin levels suggesting that Rae1 normally acts to promote mitotic progression and its loss results in a prolonged stay or arrest in mitosis . Consistent with our findings , synchronization experiments in BY-2 plant cells showed that Rae1 depleted cells entered mitosis normally but showed delayed progression [59] . This mitotic phenotype may result from mitotic cyclin dysregulation . We observed that the mitotic cells knocked down for Rae1 did not stain for cycA and cycB , consistent with arrest in anaphase or telophase . Rae1 loss of function results in decreased cycA and cycB levels , while Rae1 over-expression promotes high cycA and cycB protein levels . Our genetic interaction studies show that the restriction of cycA and cycB levels is critical for Rae1 loss-of-function reduced organ size phenotypes . Furthermore , Rae1 acts epistatically to wts and sav loss of function in regulating cyclin protein levels . In this work we thus established a Hippo Pathway-dependent but Yki-independent role for Rae1 in mitotic cyclin regulation . A summary of the similarities and differences between the roles and phenotypes of Rae1 and Yki is shown in S2 Table . How could Rae1 regulate the mitotic cyclins ? Rae1 has been reported to regulate APCC activator Cdh1/fzr [43–45] . Indeed , we showed that Cdh1/fzr genetically interacts with both Rae1 and Hippo to regulate organ size . Not necessarily mutually exclusive , Rae1 is an RNA binding protein and may directly regulate cycB transcripts , suggested by studies in plants and yeast [62 , 92] . Thirdly , mass spec screens in yeast identified binding between Rae1 homolog Gle2 and B-type cyclins CLB2 and CLB3 [93] so may affect cycB by direct protein-protein interaction . As noted , “organ size checkpoint” mechanisms normally compensate for changes in proliferation to ensure that final organ size is not altered . Rae1 over-expression increased both proliferation and organ size suggesting an exciting role for Rae1 in the Hippo signaling network to integrate regulation of proliferation and overall organ size . We propose that Rae1 acts as a “rheostat” for organ size rather than an “on/off” switch for tissue growth: decreasing Rae1 levels tunes the dial down to a lower organ size while increasing Rae levels tunes the dial up to a larger organ size . However , increased cyclin levels are insufficient to increase organ size . It will be interesting for future studies to investigate which Rae1-dependent processes mediate increased organ size . Our studies show that Rae1 restricts Yki levels and localization in Drosophila tissues and YAP localization , levels , and activity in mammalian cells , potentially as a result of Rae1-mediated functional regulation of upstream components of the Hippo Pathway . Increasing Rae1 levels leads to Rae1 deposition outside the nucleus including at the plasma membrane where activation of the pathway occurs . Rae1 is a conserved regulator of actin and microtubule networks in vitro [41–42 , 66] , and we showed that Rae1 promotes microtubule acetylation in vivo . Rae1 effects on microtubules could reflect a role in mitosis and spindle assembly [41–45 , 47 , 59–60 , 66] . Alternatively , tubulin acetylation is implicated in activation of Hippo signaling in some contexts [94] . Mer interacts with acetylated microtubules; mutants disrupting this association promote YAP nuclear localization [67–69 , 94] . Mer is crucial in recruiting Wts to the membrane to signal [64–65] . Collectively , these observations are consistent with a model that Rae1 could regulate tubulin acetylation and/or stability to correctly localize Mer to allow for Wts recruitment for proper Hippo signaling ( Fig 10 ) . We observed that Rae1 increases the protein levels of Mer , Hpo and Wts . Rae1 regulation of Wts is conserved; Rae1 loss reduces Wts levels in Drosophila tissues and reduces pLats levels in mammalian cells , and other reports indicate Rae1 interactions with Mst2 in HEK-293T cells [95] . Previous reports showed that upstream component Fat promotes accumulation of Wts protein in Drosophila [96–97] . Future work will address if stabilization of Hpo and Wts occurred via upstream stabilization of Mer , akin to Fat regulation of Wts protein levels . Previous work addressing the instability of Sav protein showed Hpo/Mst association with Sav is stabilizing in both flies and mammalian systems [98–99] . Our observations are consistent with an alternate but not mutually exclusive model that Rae1 could promote stabilization of a complex of core pathway components by promoting their assembly following proper recruitment of Wts to the membrane or by direct association . Disrupting the tight link between proliferation and organ size can have serious consequences in normal development and in diseases such as cancer . Hippo Pathway dysregulation is associated with a broad spectrum of cancers , and mutations in upstream component Merlin are associated with the familial tumor syndrome Neurofibromatosis Type 2 ( NF2 ) [2–19] . Our findings in both Drosophila and mammalian cells demonstrate that ( 1 ) high levels of Rae1 promote proliferation , ( 2 ) that Rae1 levels are controlled by Hippo signaling , and ( 3 ) that this increased proliferation due to high levels of Rae1 allows cells to evade the organ size checkpoint . We showed that Rae1 over-expression could promote proliferation of human cancer cells in culture , and our data suggests that Rae1 protein may accumulate upon loss of Hippo signaling in cancer cells . TCGA data indicates Rae1 amplification in a range of cancers [100–101] and Rae1 protein levels accumulate in gliomas [102] , a tumor type where loss of Mer and Hippo Pathway function are frequently implicated [11] . Importantly , we showed that decreasing Rae1 dramatically compromised the survival of tissue with abrogated Hippo signaling . This means that maintaining sufficient Rae1 was crucial in the context of Hippo Pathway loss; when Rae1 levels did not reach a critical threshold , the tissue underwent massive catastrophe . Elucidating this phenomenon could have tremendous impact for cancer therapeutics . In recent years , evidence has emerged that cancer cells rely heavily on individual genes for survival ( oncogene and non-oncogene “addiction” ) [103–106] It has also been proposed that “second site mutations” that do not impair viability of wild-type tissue can disadvantage cancer cells with specific primary lesions , and that this “synthetic lethality” can be exploited therapeutically [107] . Given the role of Rae1 to promote cell proliferation and increase organ size , its conserved regulation by the Hippo pathway in both cultured insect and mammalian cells , and the synthetic tissue lethality phenomenon observed in Hippo-compromised tissue , we propose that Rae1 may represent a novel therapeutic target in cancers arising due to loss of Hippo Pathway tumor suppression ( Fig 10 ) .
Flies were raised on standard media at 25°C unless otherwise stated . Genotypes are detailed are detailed separately for larval and adult tissues appearing in image panels and tissue analyzed in Western blots . The coding region of Rae1 was cloned into pUAST . Genetic Services , Inc . performed vector injection of pUAST-Rae1 and isolated independent transgenic lines . Adult wings of progeny were photographed , all at the same magnification . For quantitation , between 4 and 20 wings per genotype were traced using Adobe Photoshop CS5 or ImageJ , and wing areas were normalized to the average area in control . For engal4 wings , we measured area posterior to vein L4 . For c5gal4 , total wing area is shown . Because of the effect on eye shape and size with both Hippo and Yki over-expression and Rae1 loss of function phenotypes , we found side-by-sides the best way to represent genetic interactions with Rae1 with respect to eye size . To rule out unintentional observer bias , experiments were scored blind with lab members evaluating eyes without knowledge of genotypes . We also indicate quantification of eye outlines traced and measured using Image J . The data shown in the figures are representative experiments that have been performed independently at least 3 times . Larvae were dissected and stained using standard protocols and imaged on a Leica TSC-SP confocal . S2 cells were stained using standard protocols and imaged on Zeiss Axio Imager . Z1 . Antibodies , anti-FlagM2 ( 1:500 , Sigma ) , anti-pHH3 ( 1:1000 , Upstate ) , anti-BrdU ( 1:500 , BD Biosciences ) , anti-cyclin B ( 1:25 , DSHB ) anti-cyclin A ( 1:25 , DSHB ) , Alexa-Fluor 488 and 555 goat anti-mouse ( 1:4000 ) , Alexa-Fluor 555 goat anti-rabbit ( 1:4000 ) , Molecular Probes/Invitrogen . Antibodies , anti-FlagM2 ( 1:2000 , Sigma ) , anti-myc 9E10 ( 1:1000 , mouse , Santa Cruz Biotechnology , SCB ) , anti-myc A14 ( 1:1000 , rabbit , SCB ) , anti-cyclin B d-300 ( 1:500 , SCB ) , anti-phosphoMst1/2 ( 1:1000 , Cell Signaling ) , anti-Mst1/2 ( 1:1000 , Cell Signaling ) , anti-Cdc2 ( PSTAIR ) ( 1:1000 , SCB ) , anti-Rae1 ( 1:8000 , Sigma ) , anti-HA ( 1:1000 , Roche ) anti-Tubulin ( 1:8000 , Sigma ) anti-Lats1 ( 1:500 , Cell Signaling ) , anti-Yap ( 1:500 , SCB ) , anti-pRXXS ( 1:1000 , Cell Signaling ) , Alexa-Fluor goat anti-mouse 680 ( 1:20 , 000 ) , Alexa-Fluor goat anti-rabbit 800 ( 1:20 , 000 , Molecular Probes/Invitrogen ) , anti-rabbit IgG-conjugated HRP ( 1:4000 , GE Healthcare ) or anti-mouse IgG-conjugated HRP ( 1:4000 dilution; GE Healthcare ) . Westerns of cultured cell extracts were visualized with the Li-Cor Odyssey . Westerns of Drosophila tissue extracts were developed using Clarity Western ECL Substrate ( Bio-Rad ) . Results from immunohistochemical staining and Western Analysis were reproduced in at least three independent trials . w; Rae1ex28/+ ( left larva in Fig 2A ) w; Rae1ex28 ( right larva in Fig 2A ) w; actgal4/+ ( 2O; left eye in S4S Fig; black bars in Fig 2B , 2C; 2M and 2Q and S4A , S4R Fig; black tracings in Fig 2P and S4S’ Fig ) w; UAS Rae1IRV/+; actgal4/+ ( pink bars in Fig 2B and 2C , white bar in S4A Fig ) w; UAS Rae1IRN2/+; actgal4/+ ( hashed pink bars in Fig 2B and 2C ) w; actgal4/UAS Rae1IRN3 ( striped pink bars in Fig 2B and 2C ) UAS dcr2; nubgal4/+ ( Figs 2D and 6C and S11L , S11P and S13I Figs; black bars in Fig 2F ) UAS dcr2; nubgal4/UAS Rae1IRV ( Fig 2E; pink bars in Fig 2F ) w; nubgal4/+ ( S4H Fig ) w; nubgal4/+; Rae1IRT/+ ( S4I Fig ) w; nubgal4/+; Rae1IRT ( S4J Fig ) w; eygal4/+ ( Fig 2G and Figs 3C and 7I , left eyes in 2R and S4V , S5C , S5O , S5Q , S7H , S8N and S8X Figs; solid bars in Fig 3E and 3K’ , black bars in Fig 2J and S4L Fig , pink bars in S5E and S8R Figs , black tracing in Fig 2V’ ) w; eygal4/UAS Rae1IRV ( Figs 2H and 3D and S5P Fig; pink bar in S4L Fig , hashed bars in Fig 3E ) UAS dcr2/+; eygal4/UAS Rae1IRV ( Fig 2I and 2J and S8D and S8I Fig; left eyes in S4N , S10A and S10B Figs; hashed and striped bars in S4L Fig , pink bars in S8G and S8M Fig ) yweyFLP/+; FRT42D/FRT42D l ( 2 ) pW+ ( left eye in Fig 2K; black tracing in Fig 2K’; black bar in S4M Fig ) yweyFLP/+; FRT42D Rae1ex28/FRT42D l ( 2 ) pW+ ( left eye in Fig 1K; pink tracing in Fig 1K’; pink bar in S4M Fig ) w; GMRgal4/+ ( left eye in Fig 2L and 2S; black tracing in Fig 2L’ and 2S’; black bar in Fig 1I and S4Q Fig ) w; UAS Rae1IRV; GMRgal4/+ ( right eye in Fig 2L; pink tracing in Fig 2L’; pink bar in S4Q Fig ) UAS dcr2/+; eygal4/UAS Rae1IRV; UAS-Rae102/+ ( Right eye in Fig 2N ) UAS dcr2/+; engal4/+ ( Figs 6I and 7A and S4B and S13M Figs ) UAS dcr2/+; engal4/UAS Rae1IRV ( Figs 4E and 7B and S4C , S4D , S4O and S10C Figs; pink tracing in Fig 4F–4G and 4L–4L’ and S4P , S10D and S10E Figs; pink bar in Fig 4L” ) UAS dcr2/fzrG0326; engal4/UAS Rae1IRV ( Fig 4L; black tracing in Fig 4L’; black bar in Fig 4L” ) UAS dcr2/+; engal4/UAS Rae1IRV; UAS-Rae102/+ ( S4P Fig ) w; c5gal4/+ ( Figs 6Q and 7E and S4E , S5J , S11A , S11E , S12G , S13A and S13P Figs; black bar in Fig 2J; black tracing in Fig 6T and 6V ) w; Rae1ex28/+; c5gal4/+ ( S5K Fig; pink bar in Fig 3F ) w; UAS Rae1IRV/+; c5gal4/+ ( S5L and S6F Figs; light pink bar in Fig 3F ) w; Rae1ex28/UAS Rae1IRV; c5gal4/+ ( S5M Fig; lightest pink bar in Fig 3F ) w; dppgal4 , UAS GFP/+ ( S4K’ Fig; black bars in S2N and S4K Figs ) w; UAS Rae1IRV/+; dppgal4 , UAS GFP/+ ( Pink bars in S2K Fig ) w1118 ( Pink bars in S8H Fig ) w; eygal4/UAS Rae1IRN2 ( S5D , S5F and S5R Fig ) w; eygal4/UAS Rae1IRV; UAS p35/+ ( Hashed pink bar in S5E Fig ) yw hsFLP UAS GFP; Act>y+>gal4/+ ( Pink line in S5A and S5B Fig ) yw hsFLP UAS GFP/+; UAS Rae1IRV/Act>y+>gal4 ( Fig 3A and 3A’ and S5H-S5I’ Fig , green line in S5A Fig ) yw hsFLP UAS GFP/+; UAS Rae1IRN2/Act>y+>gal4 ( Fig 3B and 3B’ and S5G and S5G’ , S8C and S8C’ Figs ) UAS dcr/fzrG0326; eygal4/UAS Rae1IRV ( Right eye in Fig 4K; black tracing in Fig 4K’ ) UAS dcr2/+; engal4/UAS Rae1IRV; cycAc05304/+ ( Fig 4F , yellow bar in Fig 4H ) UAS dcr2/+; engal4/UAS Rae1IRV; cycB2/+ ( Fig 4G , orange bar in Fig 4H ) UAS dcr2/+; eygal4/UAS Rae1IRV; cycAc05304/+ ( S8E Fig , yellow bars in S8G Fig ) UAS dcr2/+; eygal4/UAS Rae1IRV; cycB2/+ ( S8F Fig , orange bars in S8G Fig ) w; cycAc05304/+ ( yellow bars in S8H Fig ) w; cycB2/+ ( orange bars in S8H Fig ) UAS dcr2/+; eygal4/UAS Rae1IRV; UAS cycE/+ ( S8J Fig , red bars in S8M Fig ) UAS dcr2/+; eygal4/UAS Rae1IRV/UAS cycA ( S8K Fig , yellow bars in S8M Fig ) UAS dcr2/+; eygal4/UAS Rae1IRV/ UAS cycB3 ( S8L Fig , orange bars in S8M Fig ) UAS dcr2/fzrG0326; engal4 , UAS Rae1IRV/+ ( Fig 4L; black tracing in Fig 4L’; black bar in Fig 4L” ) w; eygal4/+; UAS Rae102/+ ( right eye in Fig 2R and S8U Fig; hashed bars in Fig 3K’; pink tracing in Fig 2R’ ) w; eygal4/+; UAS Rae103/+ ( S5A and S7I Figs; pink bar in Fig 3J ) yw hsFLP UAS GFP; Act>y+>gal4/+; UAS Rae102/+ ( green line in S5B Fig ) yw hsFLP UAS GFP; Act>y+>gal4/+; UAS Rae103/+ ( S8S and S8S’ Fig ) w; GMRgal4/ UAS Rae102 ( right eye in Fig 2S; pink hashed bar in S4Q Fig; pink tracing in Fig 2S’ ) w; GMRgal4/ UAS Rae103 ( pink striped bar in S4Q Fig ) w; actgal4/UAS Rae102 ( Fig 2P; pink bar in Fig 2M and 2Q and S4A and S4R Fig ) w; actgal4/UAS Rae103 ( Pink hashed bar S4A and S4R Fig ) w; actgal4/UAS Rae1GFP ( Fig 9H , right eye in S4S Fig; pink tracing in S4S’ Fig ) w; actgal4/UAS ykiV5 ( eye in S4U Fig ) w; eygal4/+; UAS ykiV5/+ ( right eye in S4V Fig; yellow tracing in S4V’ Fig ) w; eygal4/+; UAS cycE/+ ( S8X Fig ) w; eygal4/+; UAS cycEIRT/+ ( S8Y Fig ) w; GMR Hpo/+ ( S9A–S9D Fig and S10B Fig; left eyes in Fig 5A and 5H and S9E , S9F , S9H and S9J Fig; black tracing in Fig 5A’ and S9H’ and S9J’ Fig , blue tracing in Fig 5H , blue bars in Fig 5H” , black bar in S10F Fig ) w; GMR Hpo/Rae1ex28 ( right eye in Fig 5A and S9H and S10B Figs; pink tracing in Fig 5A’ and S9H’ Fig ) w; GMR Sav , Wts/+ ( left eye in Fig 5B; black tracing in Fig 5B’ ) w; Rae1ex28/+; GMR Sav , Wts/+ ( right eye in Fig 5B; pink tracing in Fig 5B’ ) w; GMR Wts/+ ( left eye in Fig 5C; black tracing in Fig 5C’ ) w; Rae1ex28/+; GMR Wts/+ ( right eye in Fig 5C; pink tracing in Fig 5C’ ) w; GMR Hpo/GMRgal4 ( left eye in Fig 5D; black tracing in Fig 5D’ ) w; GMR Hpo/GMRgal4; UAS Rae102/+ ( right eye in Fig 5D; pink tracing in Fig 5D’ ) w; UAS hpo/+; c5gal4/+ ( Figs 5E and S9M; blue tracing in Fig 5F and S9N Fig; blue bar in Fig 5G ) w; UAS hpo/+; c5gal4/UAS Rae1GFP ( Fig 5F; pink bar in Fig 5G ) fzrG0326/+; GMR Hpo/+ ( right eye in Fig 5H; black tracing in Fig 5H’; black bar in Fig 5H” ) fzrG0418/+ ( blue bar in S8V Fig ) fzrG0418/+; GMR Hpo/+ ( blue bar in S10F Fig ) w; GMR Hpo/+; wts3-17/+ ( S9G Fig; right eyes in S9E and S9F Fig ) w; UAS hpo/+; c5gal4/wtsX1 ( S9N Fig ) w; GMR Hpo/Df ( 2R ) ED3923 ( right eye in S9J Fig; pink tracing in S9J’ Fig ) w; GMR Sav , Wts/GMRgal4 ( left eye S9K Fig; black tracing in S9K’ Fig ) w; UAS Rae1IRV/+; GMR Sav , Wts/GMRgal4 ( right eye S9K Fig; pink tracing in S9K’ Fig ) w; GMR Hpo/+; actgal4/+ ( left eye in S9L Fig; black tracing in S9L’ Fig ) w; GMR Hpo/+; actgal4/Rae1GFP ( right eye in S9L Fig; pink tracing in S9L’ Fig ) UAS dcr2/+; eygal4 , UAS Rae1IRV/hpoMGH1 ( right eye in S10A Fig ) UAS dcr2/mer4; eygal4 , UAS Rae1IRV/+ ( right eye in S10B Fig ) UAS dcr2/+; engal4 , UAS Rae1IRV/hpoKS240 ( S10F Fig ) UAS dcr2/+; engal4 , UAS Rae1IRV/+; wts3-17/+ ( S10E Fig ) UAS dcr2/+; engal4/+; UAS hpoIRT/+ ( Fig 6J and S12C Fig; blue tracing in Fig 6K and 6L ) UAS dcr2/+; engal4 , UAS Rae1IRV; UAS hpoIRT/+ ( Fig 6K and S12D Fig ) UAS dcr2/+; engal4/Rae1ex28; UAS hpoIRT/+ ( Fig 6L ) UAS dcr2/+; nubgal4/+; UAS hpoIRT/+ ( Fig 6D; blue tracing in Fig 6E and 6F ) UAS dcr2/+; nubgal4/Rae1ex28; UAS hpoIRT/+ ( Fig 6E ) UAS dcr2/+; nubgal4/UAS Rae1IRV; UAS hpoIRT/+ ( Fig 6F ) ms1096gal4/+; UAS hpoIRT/+ ( Fig 6G ) ms1096gal4/+; Rae1ex28/+; UAS hpoIRT/+ ( Fig 6H’ and 6H” ) UAS dcr2/+; engal4/UAS MerIRN ( Fig 6M and S12A Fig; green tracing in Fig 6N and S12B Fig ) UAS dcr2/+; engal4 , UAS Rae1IRV/UAS MerIRN ( Fig 6N and S12B Fig ) UAS dcr2/+; engal4/+; UAS exIRT/+ ( Fig 6O; purple tracing in Fig 6P ) UAS dcr2/+; engal4 , UAS Rae1IRV/+; UAS exIRT/+ ( Fig 6P ) w; c5gal4/UAS wtsIRT ( Fig 6R and S12H Fig; purple tracing in Fig 6T and 6V , S12I and S12J Fig ) w; UAS ykiIRN/+; c5gal4/UAS wtsIRT ( Fig 6S and S12J Fig; yellow tracing in Fig 6T ) w; Rae1ex28/+; c5gal4/UAS wtsIRT ( Fig 6U–6U” and S12I Fig; pink tracing in Fig 6V ) UAS dcr2/+; engal4/+; UAS ykiV5/+ ( Fig 7 and S13N Fig; yellow tracing in S13O Fig ) UAS dcr2/+; engal4/UAS Rae1IRV; UAS ykiV5 /+ ( S13O Fig ) w; c5gal4/UAS ykiFLAG ( S13Q Fig; yellow tracing in S13R Fig ) w; Rae1ex28/+; c5gal4/UAS ykiFLAG ( S13R Fig ) w; GMRgal4/UAS ykiV5 ( Left eye in S13S Fig; black tracing in S13S’ Fig; orange bars in Fig 1I ) w; UAS Rae1IRV/+; GMRgal4/UAS ykiV5 ( Right eye in S13S Fig; pink tracing in S13S’ Fig ) w; GMRgal4/UAS ykiS168A . GFP ( Left eye in S13T Fig; black tracing in S13T’ Fig ) w; UAS Rae1IRV/+; GMRgal4/UAS ykiS168A . V5 ( Right eye in S13T Fig; pink tracing in S13T’ Fig ) w; UAS ykiS168A . GFP/+; c5gal4/+ ( Fig 7F ) w; UAS ykiS168A . GFP/Rae1ex28; c5gal4/+ ( Fig 7G ) w; UAS ykiS168A . GFP/UAS Rae1IRV; c5gal4/+ ( Fig 7H ) w; UAS ykiS168A . GFP/eygal4 ( Fig 7J ) w; UAS ykiS168A . GFP/eygal4 , Rae1ex28 ( Fig 7K ) w; UAS ykiS168A . GFP/eygal4 , UAS Rae1IRV ( Fig 7L ) w; dppgal4/+ ( Fig 7M ) w; UAS ykiS168A . GFP/+; dppgal4/+ ( Figs 7N and 8D–8D” ) w; UAS ykiS168A . GFP/Rae1ex28; dppgal4/+ ( Figs 7O and 8E–8E” ) w; UAS ykiS168A . GFP/UAS Rae1IRV; dppgal4/+ ( Fig 7P ) w; ptcgal4/+; UAS Rae1GFP/+ ( S14C and S14C’ Fig ) w; ptcgal4/UAS Hpo; UAS Rae1GFP/+ ( S14D and S14D’ Fig ) w; actgal4/hpoMGH1; UAS Rae1GFP/+ ( Fig 9I ) w; actgal4/Uba1B2; UAS Rae1GFP/+ ( Fig 9J ) w; c5gal4/Rae1IRT ( S11B and S11F Fig ) w; UAS HpoKD/+; c5gal4/+ ( S11C and S11G Fig ) w; UAS HpoKD/+; c5gal4/UAS Rae1IRT ( S11D and S11H Fig ) w; UAS HpoKD/Rae1ex28; c5gal4/+ ( S11I and S11I’ Fig ) w; UAS HpoKD/UAS Rae1IRV+; c5gal4/+ ( S11J and S11J’ Fig ) UAS dcr2/+; nubgal4/+; UAS Rae1IRT/+ ( S11M and S11Q Fig ) UAS dcr2/+; nubgal4/UAS HpoKD ( S11N and S11R Fig ) UAS dcr2/+; nubgal4/UAS HpoKD; UAS Rae1IRT/+ ( S11O and S11S Fig ) w; engal4/+; UAS hpoIRT/+ ( S12E Fig; blue tracing in S12F Fig ) w; engal4/UAS YkiIRN1; UAS hpoIRT/+ ( S12E Fig; blue tracing in S12F Fig ) UAS dcr2/+; engal4/+; UAS mycWT/+ ( S12L Fig; black tracing in S12M Fig ) UAS dcr2/+; engal4/UAS Rae1IRV; UAS mycWT/+ ( S12M Fig ) UAS dcr2/+; engal4/+; UAS p35/+ ( S12N Fig; black tracing in S12O Fig ) UAS dcr2/+; engal4/UAS Rae1IRV; UAS p35/+ ( S12O Fig ) yw hsFLP UAS GFP; Act>y+>gal4/UAS Rae1IRV; UAS hpoIRT/+ ( S12K Fig ) w; c5gal4/UAS YkiV5 ( S13B Fig; yellow tracing in S13C and S13D Fig ) w; Rae1ex28/+; c5gal4/UAS YkiV5 ( S13C Fig ) w; Rae1IRV/+; c5gal4/UAS YkiV5 ( S13D Fig ) w; engal4/+ ( S13E Fig ) w; engal4/+; UAS YkiV5 ( S13F Fig ) w; engal4/Rae1ex28; UAS YkiV5 ( S13G Fig ) w; engal4/Rae1IRV; UAS YkiV5 ( S13H Fig ) UAS dcr2/+; nubgal4/+; UAS YkiV5 ( S13J Fig ) UAS dcr2/+; nubgal4/ Rae1ex28; UAS YkiV5 ( S13K Fig ) UAS dcr2/+; nubgal4/Rae1IRV; UAS YkiV5 ( S13L Fig ) w; ptcgal4/+; UAS wtsmyc ( Fig 9K–9K” ) w; ptcgal4/+; UAS wtsmyc/Rae1GFP ( Fig 9L–9L” ) w; UAS Rae1GFP/c5gal4 ( Lane 1 in Fig 1C ) w; hpoMGH1/+; UAS Rae1GFP/c5gal4 ( Lane 2 in Fig 1C ) w; Uba1B1/+; UAS Rae1GFP/c5gal4 ( Lane 3 in Fig 1C ) w; dppgal4 , UAS GFP /+ ( Lane 1 in Fig 1E , lane 1 in S2C Fig ) w; dppgal4 , UAS GFP /UAS Rae1GFP ( Lane 2 in Fig 1E , lane 2 in S2C Fig ) w; UAS hpo/+; dppgal4 , UAS GFP/UAS Rae1GFP ( Lane 3 in Fig 1E ) w; UAS hpoKD/+; dppgal4 , UAS GFP/UAS Rae1GFP ( Lane 4 in Fig 1E ) w; ptcgal4/+; UAS Rae1GFP/+ ( Lane 1 in Fig 1F , lane 1 in Fig 1K , lane 2 in Fig 8I , lane 2 in Fig 9M , lane 1 in S2B Fig , lane 1 in S2D Fig , lane 1 in S2F Fig ) w; ptcgal4/hpoMGH1; UAS Rae1GFP/+ ( Lane 2 in S2B Fig , lane 2 in S2D Fig ) w; ptcgal4/+; UAS Rae1GFP/wtsX1 ( Lane 3 in S2B Fig ) w; ptcgal4/Uba1B1; UAS Rae1GFP/+ ( Lane 3 in S2D Fig ) w; Uba1B1/+; dppgal4 , UAS GFP/UAS Rae1GFP ( Lane 3 in S2C Fig ) w; ptcgal4/UAS hpo; UAS Rae1GFP/+ ( Lane 2 in Fig 1F , lane 2 in S2F Fig ) w; ptcgal4/UAS hpoKD; UAS Rae1GFP/+ ( Lane 3 in S2F Fig ) w; ptcgal4/UAS hpo; UAS Rae1GFP/UAS wtsKD ( Lane 3 in Fig 1F ) w; ptcgal4/UAS YkiIRV; UAS Rae1GFP/+ ( Lane 2 in Fig 1K ) w; engal4/+ ( Lane 1 in Fig 4J ) w; engal4/+; UAS Rae1GFP/+ ( Lane 2 in Fig 4J ) w; engal4/+; UAS Rae1GFP ( Lane 3 in Fig 4J ) ey ( 3 . 5 ) -FLP; Act>y+>gal4/+ , UAS GFP/+; FRT82B ( Lane 1 in Fig 6A , lane 1 in Fig 6B , lane 1 in S8A Fig , lane 1 in S8B Fig ) ey ( 3 . 5 ) -FLP; Act>y+>gal4/+ , UAS GFP/+; FRT82B wtsX1 ( Lane 2 in Fig 6A , lane 2 in S8A Fig , lane 2 in S8B Fig ) ey ( 3 . 5 ) -FLP; Act>y+>gal4/+ , UAS GFP/+; FRT82B wtsX1 ( Lane 2 in Fig 6A ) ey ( 3 . 5 ) -FLP; Act>y+>gal4/+ , UAS GFP/UAS Rae1IRV; FRT82B wtsX1 ( Lane 3 in Fig 6A ) ey ( 3 . 5 ) -FLP; Act>y+>gal4/+ , UAS GFP/Rae1ex28; FRT82B wtsX1 ( Lane 4 in Fig 6A ) ey ( 3 . 5 ) -FLP; Act>y+>gal4/+ , UAS GFP/+; FRT82B sav4 ( Lane 2 in Fig 6B ) ey ( 3 . 5 ) -FLP; Act>y+>gal4/+ , UAS GFP/UAS Rae1IRV; FRT82B sav4 ( Lane 3 in Fig 6B ) ey ( 3 . 5 ) -FLP; Act>y+>gal4/+ , UAS GFP/Rae1ex28; FRT82B sav4 ( Lane 4 in Fig 6B ) w; c5gal4/+ ( Lane 1 in Fig 8A and 8B ) w; c5gal4/UAS YkiV5 ( Lane 2 in Fig 8A ) w; Rae1ex28/+; c5gal4/UAS YkiV5 ( Lane 3 in Fig 8A ) w; UAS Rae1IRV/+; c5gal4/UAS YkiV5 ( Lane 4 in Fig 8A ) w; c5gal4/UAS YkiFLAG ( Lane 2 in Fig 8B ) w; Rae1ex28/+; c5gal4/UAS YkiFLAG ( Lane 3 in Fig 8B ) w; UAS Rae1IRV/+; c5gal4/UAS YkiFLAG ( Lane 4 in Fig 8B ) w; ptcgal4/+ ( Lane 1 in S13U Fig , Lane 1 in Fig 8I , Lane 1 in Fig 9F , Lane 1 in Fig 9M ) w; ptcgal4/+; UAS YkiV5/+ ( Lane 2 in S13U Fig ) w; ptcgal4/ Rae1ex28; UAS YkiV5/+ ( Lane 3 in S13U Fig ) w; ptcgal4/UAS Rae1IRV; UAS YkiV5/+ ( Lane 4 in S13U Fig ) w; UAS ykiS168A . GFP/ptcgal4; UAS Rae1GFP/+ ( Lane 3 in Fig 8I ) w; UAS ykiS168A . GFP/ptcgal4 ( Lane 4 in Fig 8I ) w; ptcgal4/+; UAS myc wts/+ ( Lane 1 in Fig 9B ) w; ptcgal4/+; UAS myc wts/UAS Rae1GFP ( Lane 2 in Fig 9B ) w; UAS myc wts/c5gal4+ ( Lane 1 in Fig 9C ) w; Rae1ex28/+; UAS myc wts/c5gal4 ( Lane 2 in Fig 9C ) w; ptcgal4/UAS hpo ( Lane 2 in Fig 9F ) w; ptcgal4/UAS hpo; UAS Rae1GFP/+ ( Lane 3 in Fig 9F ) w; nubgal4/+ ( Lane 1 in Fig 9N ) w; nubgal4/+; UAS Rae1IRT/+ ( Lane 2 in Fig 9N ) w; nubgal4/+; UAS Rae1IRT ( Lane 3 in Fig 9N ) S2 cells cultured at 25°C were transfected with Actin-Gal4 , pIE1-4-myc-Hippo , pIE1-4-myc-Warts , pIE1-4-HA-Merlin and UAS-FLAG-His6- Rae1 , pAc5 . 1-His6-FLAGx3-Rae1 , pAc5 . 1-Rae1-V5-His using Cellfectin II ( Invitrogen ) or Effectene ( Qiagen ) . HEK-293T , U87MG and HeLa cells were cultured in DMEM ( Invitrogen ) containing 10% FBS ( Gemini ) and 50 μg/mL penicillin/streptomycin ( Gemini ) . Transfection with pCMV5-FLAG-Mst1 , pCMV2-FLAG2 Lats1 , pCMV-FLAG Yap2 S127A , pCMV-FLAG-Yap2 5SA , using Effectene ( Qiagen ) was performed according to the manufacturer's instructions . The percentage of anti-phosphorylated histone H3 ( Cell Signalling Technology ) -positive S2 cells per total cells ( mitotic index ) was determined by scoring a total of at least 400 cells in each of four independent experiments . Bacterial stocks containing plasmids of Drosophila Gene Collection Releases 1 and 2 ( representing more than 11 , 000 genes in the Drosophila genome ) were grown as individual 1 ml cultures then pooled for isolation of plasmid DNA in pools of 12 and 16 . Pools containing 12–16 plasmids were in vitro translated ( IVT ) using the Promega TNT combined in vitro transcription/translation kits and labeled with 35S-methione . Pool IVTs were incubated in the presence or absence of two unrelated kinases or a combination of recombinant Mst1 and Mst2 ( Invitrogen ) and run on a gel ( Drosophila In vitro Expression Cloning ) [48–50] . Bands representing individual clones in the pools were considered positive hits if showing a gel shift , smear , or other change after incubation with Hpo compared to the control incubation or incubation with two unrelated kinases . S2 cells were harvested by centrifugation at 1000 g for 3 minutes . The cell pellet was washed once and resuspended in either 8M urea , 150 mM NaCl , 25 mM Tris , 1% NP40 and 1 mM EDTA or in 8M urea dissolved in PhosphoSafe Extraction Reagent ( Novagen ) . In both cases , the lysis buffers were supplemented with 1 mM PMSF and protease inhibitor cocktail ( Roche ) . S2 cells were harvested by centrifugation at 400 g for 5 minutes . The cell pellet was resuspended in 300 μL of extract buffer ( 20 mM HEPES KOH pH 7 . 4 , 50 mM KCl , 1 . 5 mM MgCl2 , 1 mM EDTA , 1 mM EGTA , 1 mM DTT , 250 mM sucrose , 100 μg/mL cycloheximide , 1 mM PMSF , protease inhibitor cocktail ( Roche ) , 1 mM NaF , 20 mM NaOPO7 , 40 mM b-glycerophosphate and para-Nitrophenylphosphate , supplemented with either 20 μM MG132 or DMSO ) and homogenized using 30 strokes of a Dounce homogenizer . Cell extracts were incubated at room temperature and stopped by the addition of 6x Laemmli sample buffer and boiling for 10 minutes 100 μg of protein was incubated with 200 units of calf-intestinal phosphatase ( New England Biolabs ) with NEB Buffer 3 supplemented with protease inhibitor cocktail ( Roche ) and 1 mM PMSF at room temperature for 10 minutes . Reactions were stopped by the addition of 6x Laemmli sample buffer and boiling for 10 minutes . The DNA template was generated using the following primer sets ( Table 1 ) : dsRNA was generated using the T7 RiboMAX Express Large Scale RNA Production System ( Promega ) , followed by DNAse digestion using RQ1 RNAse-free DNAse ( Promega ) . 1x106 cells were treated with 15 μg dsRNA for 48–72 hours and then transfected with appropriate plasmids ( Fig 4C and 4M ) or co-transfected with plasmids and dsRNA ( S8F Fig ) . RNA from approximately 15 larvae of each genotype or 30 adult heads was extracted using TRIzol reagent ( Invitrogen ) . For mammalian cells , RNA was extracted using the GeneJet RNA Purification kit ( Thermo ) . In all cases , RNA was treated with RQ1 RNase-free DNase ( Promega ) . 1 μg of RNA was reverse-transcribed using iScript cDNA synthesis kit ( BioRad ) and diluted 1:50 for each quantitative PCR reaction ( QPCR SYBR Green ROX Mix ( Fisher Scientific ) ) . The probes used were ( Table 2 ) : Statistical analysis was performed using Excel . For all quantitative changes , T-test ( with one-sided using equivalent variance ) were conducted . For changes in categorical data ( for example , incidence of black tissue or pHH3 cells ) , chi tests were conducted . All data presented represent typical findings from experiments performed a minimum of three times with appropriate controls . Approximately 20 third instar larval wing discs of each genotype were dissected in PBS and transferred to 0 . 5 mL of 10x Trypsin-EDTA solution ( Sigma-Aldrich ) and incubated for 3 hours at room temperature on a nutator . The cells were analyzed using FACScalibur and BD CellQuest Pro . 24 hours after transfection , 2 . 5x104 ( for 293T , Rae1 over-expression ) and 3 . 5x104 ( for HeLa , Rae1 over-expression ) cells were added to each well of multiple12-well cell culture dishes . For the Rae1 knockdown experiment , 1 . 0x105 cells ( for 293T , Rae1 knockdown ) were added to each well of multiple 6-well cell culture dishes . Cells were incubated at 37°C until harvest . At approximately 24 , 48 , and 72 hours post-seeding , one dish was retrieved from incubation and the contents of each well were aspirated . Each well was washed twice in 1x PBS . The adherent cells were dissociated with 0 . 05% Trypsin/EDTA and subsequently counted using a Coulter Counter ( Beckman ) or using Cell Countess ( Invitrogen ) . Third instar larval eye discs were dissected in serum-free Schneider’s media and then incubated with 8 μg/mL BrdU for 30 minutes at room temperature . Discs were washed first in serum-free media then in PBS and fixed in 4% paraformaldehyde ( diluted in PBS ) for 30 minutes . They were permeabilized in PBS/0 . 1%Triton 100 ( PBT ) and incubated in 2N HCl solution ( diluted in PBT ) for 30 minutes . Cells were washed in PBT and incubated overnight in anti-BrdU ( 1:500 , BD Biosciences ) and standard protocols were followed for secondary staining . BrdU incorporation was imaged on a Zeiss AxioImager Z1 and AxioVision Release 4 . 8 and/or also on a Leica TSC-SP confocal . Rae1 and YAP Peptides were synthesized by Genscript and incubated individually with commercially available Lats2 ( Invitrogen ) , spotted onto P81 phosphocellulose cation exchange paper ( Whatman ) . The P81 paper was washed at least five times in 0 . 5% orthophosphoric acid until counts were no longer detectable in the washes , rinsed with ethanol , and air-dried . The dried P81 papers were mixed with Ready Safe scintillation mix ( Beckman ) and counted in a Beckman liquid scintillation counter . Purified Rae1 ( a kind gift from Y Ren and the Blobel lab ) or MBP ( Sigma ) were diluted in kinase assay buffer ( 50 mM Tris pH8 , 10 mM MgCl2 , 1 mM DTT ) incubated in the presence of 100 μM cold ATP , 10 μCi [γ-32P]ATP and recombinant Lats2 ( Invitrogen ) . The reaction mixtures were incubated for 20 minutes at 30°C , terminated with SDS sample buffer , and subjected to SDS-PAGE and autoradiography . | Exquisite control of organ size is critical during animal development and its loss results in pathological conditions . The Hippo Tumor Suppressor Pathway coordinates regulation of proliferation , growth , apoptosis , and autophagy to determine and maintain precise control of organ size . However , the genes responsible for Hippo-mediated regulation of mitosis or coordination of proliferation within organ size control have evaded characterization . Here , we describe Rae1 , an essential WD-repeat containing protein , as a new organ size regulator . By genetic analysis , we show that Rae1 acts downstream of the Hippo Pathway to regulate mitotic cyclins and organ size . In contexts where organ size control is lost by compromised Hippo signaling , we show that there is a requirement for Rae1 that is distinct from the requriement for Yki: reducing Yki levels causes suppression of overgrowth , while reducing Rae1 levels dramatically compromises the survival of Hippo-deficient tissue . Lastly , our studies of Rae1 uncovered a potential post-transcriptional feedback loop that reinforces Yorkie-mediated transcriptional feedback for the Hippo Pathway . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"invertebrates",
"medicine",
"and",
"health",
"sciences",
"rna",
"interference",
"cell",
"cycle",
"and",
"cell",
"division",
"cell",
"processes",
"cloning",
"animals",
"cell",
"differentiation",
"animal",
"models",
"developmental",
"biology",
"drosophila",
"melanogaster",
"model",
"organisms",
"epigenetics",
"molecular",
"biology",
"techniques",
"eyes",
"drosophila",
"digestive",
"system",
"research",
"and",
"analysis",
"methods",
"specimen",
"preparation",
"and",
"treatment",
"staining",
"genetic",
"interference",
"gene",
"expression",
"exocrine",
"glands",
"head",
"molecular",
"biology",
"insects",
"arthropoda",
"biochemistry",
"rna",
"cell",
"staining",
"anatomy",
"nucleic",
"acids",
"cell",
"biology",
"genetics",
"salivary",
"glands",
"biology",
"and",
"life",
"sciences",
"ocular",
"system",
"organisms",
"cyclins"
] | 2016 | The Hippo Pathway Targets Rae1 to Regulate Mitosis and Organ Size and to Feed Back to Regulate Upstream Components Merlin, Hippo, and Warts |
Blinding trachoma , caused by ocular infection with Chlamydia trachomatis , is targeted for global elimination by 2020 . Knowledge of risk factors can help target control interventions . As part of a cluster randomised controlled trial , we assessed the baseline prevalence of , and risk factors for , active trachoma and ocular C . trachomatis infection in randomly selected children aged 0–5 years from 48 Gambian and 36 Tanzanian communities . Both children's eyes were examined according to the World Health Organization ( WHO ) simplified grading system , and an ocular swab was taken from each child's right eye and processed by Amplicor polymerase chain reaction to test for the presence of C . trachomatis DNA . Prevalence of active trachoma was 6 . 7% ( 335/5033 ) in The Gambia and 32 . 3% ( 1008/3122 ) in Tanzania . The countries' corresponding Amplicor positive prevalences were 0 . 8% and 21 . 9% . After adjustment , risk factors for follicular trachoma ( TF ) in both countries were ocular or nasal discharge , a low level of household head education , and being aged ≥1 year . Additional risk factors in Tanzania were flies on the child's face , being Amplicor positive , and crowding ( the number of children per household ) . The risk factors for being Amplicor positive in Tanzania were similar to those for TF , with the exclusion of flies and crowding . In The Gambia , only ocular discharge was associated with being Amplicor positive . These results indicate that although the prevalence of active trachoma and Amplicor positives were very different between the two countries , the risk factors for active trachoma were similar but those for being Amplicor positive were different . The lack of an association between being Amplicor positive and TF in The Gambia highlights the poor correlation between the presence of trachoma clinical signs and evidence of C . trachomatis infection in this setting . Only ocular discharge was associated with evidence of C . trachomatis DNA in The Gambia , suggesting that at this low endemicity , this may be the most important risk factor . ClinicalTrials . gov NCT00792922
Trachoma is caused by ocular infection with serovars A , B , Ba or C of the bacterium Chlamydia trachomatis . It is the leading infectious cause of blindness worldwide [1] with an estimated 40 . 6 million people suffering from active trachoma ( trachomatous inflammation , follicular ( TF ) and/or intense ( TI ) ) and 8 . 2 million having trichiasis [2] . As part of the “SAFE” ( Surgery , Antibiotics , Facial cleanliness , Environmental improvement ) trachoma control strategy , the World Health Organization ( WHO ) recommends mass antibiotic treatment annually for at least three years of all individuals in any district or community where the prevalence of TF in children aged 1–9 years is at least 10% . After three or more years of A , F and E interventions , the prevalence is reassessed and a decision is made regarding the need to continue or cease treatment [3] . Mass antibiotic treatment aims to clear infection from the community , most of which is found in children [4] . Trachoma is endemic in both The Gambia and Tanzania , with estimated active trachoma prevalences in children aged 1–9 years of 10 . 4% and 27% , respectively [5] , [6] . Accordingly , they have both recently qualified for a donation of the antibiotic azithromycin for mass treatment by Pfizer via the International Trachoma Initiative . Given the different endemicities of these two countries , one in which trachoma is almost disappearing and one in which trachoma shows only modest signs of being reduced , the question of whether the same risk factors are predictive of trachoma is of interest . In addition , since the presence of trachoma clinical signs is often poorly correlated with that of ocular C . trachomatis infection [5] , [7] , [8] , [9] , [10] , [11] , the risk factors for these markers of trachoma may also differ . Studies have shown that although young age is a common risk factor for active trachoma , other risk factors may be setting-specific . Furthermore , few studies have simultaneously reported risk factors for active trachoma and ocular C . trachomatis infection within the same setting [12] , [13] . Information on risk factors can contribute to our understanding of trachoma transmission within the study area , and the targeting of trachoma control interventions can be aided through knowledge of risk factors . We aimed to assess the prevalence of , and risk factors for , both active trachoma and ocular C . trachomatis infection pre-treatment in The Gambia and Tanzania , as part of the Partnership for the Rapid Elimination of Trachoma ( PRET ) cluster randomised controlled trial . The aims of PRET are to test the impact on the prevalence of active trachoma and ocular C . trachomatis infection , as detected by Amplicor PCR , after three years in communities mesoendemic for trachoma ( between 20% and 50% TF ) or hypoendemic ( between 10% and 20% TF ) , when communities are randomised to different mass treatment population coverage levels and a different number of rounds of treatment , with a graduation rule if the prevalence of TF or detected ocular C . trachomatis infection falls below 5% ( Stare et al . submitted ) . The data presented here are from the baseline surveys of PRET , where data on the prevalence of TF and evidence of ocular C . trachomatis infection were collected , and risk factors for these outcomes were obtained in a standardised fashion .
The research was done in accordance with the declaration of Helsinki . Ethical approval was obtained from the London School of Hygiene & Tropical Medicine ( LSHTM ) , UK , Ethics Committee; The Gambia government/Medical Research Council ( MRC ) Joint Ethics Committee , The Gambia; the Johns Hopkins Institutional Review Board; and the Tanzanian National Institute for Medical Research . Oral consent was obtained from the village leaders , and written ( thumbprint or signature ) consent from the child's guardian at the time of examination , which was signed by an independent witness . In The Gambia , 48 census Enumeration Areas ( EAs ) , designed to have similar population sizes of between 600–800 people , were randomly selected from within 4 strata consisting of the following districts: Foni Bintang and Foni Kansala in Western Region , and Central Baddibu and Lower Baddibu in North Bank Region ( 12 EAs per district ) ( Figure 1 ) . In Tanzania , 32 communities ( geographically distinct areas within a village with an average population of approximately 1500 people ) were selected in Kongwa district , Dodoma region ( Figure 2 ) . Tanzanian communities were selected based on having an active trachoma prevalence above 20% in preliminary surveys and were therefore not randomly selected as they were in The Gambia . A week-long workshop was conducted in February 2008 to standardise all fieldwork methods , including trachoma grading , photography , sample collection , form filling , facial cleanliness status grading , and data entry . For trachoma grading , graders were standardised against a senior grader ( RB ) every day by examining participants in the field . A kappa of >0 . 6 for TF grading was required between the senior grader and the graders in the final grading exam . All other procedures had to be performed correctly five times in the field under observation by senior investigators before certification was given . Fieldwork in The Gambia took place between 19th May 2008 and 29th July 2008 . In Tanzania , data collection was between 15th May and 1st November 2008 . Data were entered into a customised database ( MS Access v2007 ) developed at the Dana Center , Johns Hopkins University . Key fields were double-entered by different entry clerks . Reports of discrepant , missing or query entries were generated in the database and resolved by reference to the forms , or in some cases by return field visits . Further queries of data inconsistencies were produced in statistical packages ( Stata v10 , STATA Corp . , College Station , TX , USA for the Gambian data; SAS v9 . 2 , SAS Institute Inc . , Cary , NC , USA for the Tanzanian data ) prior to analysis . All queries were verified against the original paper forms . The analyses presented here were conducted using Stata , v10 . Baseline characteristics of household attributes and population size were summarised for both countries . Evidence of variability between communities ( clusters ) and households was assessed using random effects logistic regression models assuming a 3-level hierarchy to the data structure ( community , household and individual ) in null regression models . Univariate associations with TF and ocular C . trachomatis infection in children aged 0–5 years were tested using random effects logistic regression , accounting for between-cluster and between-household variation ( variance ) , comparing models with and without covariates using the likelihood ratio test ( LRT ) . Multivariate model building for TF and C . trachomatis infection in both countries employed the same stepwise strategy; age and sex were considered a priori risk factors and included in all models . Covariates associated with TF or evidence of C . trachomatis infection at the 10% significance level in univariate analyses were added in turn ( a forward stepwise approach ) and covariates retained in the model if the LRT p-value was ≤0 . 1 . In The Gambia , the final multivariate model also adjusted for district to account for sampling stratified by district .
In The Gambia , 5033 children aged 0–5 years were examined . In Tanzania , 3198 children were examined but ocular C . trachomatis data were missing from 76 of these children . In The Gambia , there were 9 households with missing data for awareness of a village face-washing education programme . In Tanzania , the number of missing values was 5 for household head education , 6 for time to water , 20 for latrine access , and 693 ( of which 677 were recorded as “unknown” ) for knowledge of a face-washing health education programme . Community randomisation units were larger in Tanzania than in The Gambia , containing more , smaller households , as seen from the total population size and average household sizes ( Table 1 ) , although similar proportions of the total population were children aged under 10 years . Household heads in The Gambia had less formal education than in Tanzania , whereas latrines and water were less easily accessible in Tanzania . Around one third of households in both countries reported awareness of receiving community face-washing health education programmes . The low prevalence of Amplicor positives in The Gambia provided little power for formal risk factor analyses ( Table 5 ) . Chi-squared tests of association suggested that ocular discharge was a possible risk factor for an Amplicor positive result ( p = 0 . 044 ) and that prevalence varied by district ( p<0 . 001 ) . In Tanzania , Amplicor positivity was associated in univariate analyses with being aged ≥1 year , having ocular or nasal discharge , flies on the child's face , lack of household head education , and poor access to water or a latrine ( Table 5 ) . In multivariate models , being Amplicor positive was only significantly related to being aged 2–5 years , having discharge , and a head of household educational level of less than 7 years , and possibly poor access to water ( Table 5 ) . Other factors were not related to evidence of C . trachomatis infection .
In summary , this study showed that despite different prevalences of active trachoma and evidence of infection between the Tanzanian and Gambian study sites , the risk factors for TF were similar . The risk factors for being Amplicor positive in Tanzania were similar to those for TF , whereas in The Gambia , only ocular discharge was associated with evidence of C . trachomatis DNA , suggesting that at this low endemicity , this may be the most important risk factor . The lack of an association between being Amplicor positive and having TF in The Gambia highlights the poor correlation between the presence of trachoma clinical signs and evidence of C . trachomatis infection in this setting . | Trachoma is caused by Chlamydia trachomatis and is the leading infectious cause of blindness . The World Health Organization's ( WHO ) control strategy includes antibiotic treatment of all community members , facial cleanliness , and environmental improvements . By determining how prevalent trachoma is , decisions can be made whether control activities need to be put in place . Knowing what factors make people more at risk of having trachoma can help target trachoma control efforts to those most at risk . We looked at the prevalence of active trachoma and C . trachomatis infection in the eyes of children aged 0–5 years in The Gambia and Tanzania . We also measured risk factors associated with having active trachoma or infection . The prevalence of both active trachoma and infection was lower in The Gambia ( 6 . 7% and 0 . 8% , respectively ) than in Tanzania ( 32 . 3% and 21 . 9% , respectively ) . Risk factors for active trachoma were similar in the two countries . For infection , the risk factors in Tanzania were similar to those for TF , whereas in The Gambia , only ocular discharge was associated with infection . These results show that although the prevalence of active trachoma and infection is very different between the two countries , the risk factors for active trachoma are similar but those for infection are different . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"evidence-based",
"healthcare/clinical",
"decision-making",
"ophthalmology/eye",
"infections",
"infectious",
"diseases/neglected",
"tropical",
"diseases",
"public",
"health",
"and",
"epidemiology/epidemiology",
"public",
"health",
"and",
"epidemiology/infectious",
"diseases",
"infectious",
"diseases/bacterial",
"infections",
"infectious",
"diseases/epidemiology",
"and",
"control",
"of",
"infectious",
"diseases"
] | 2010 | Trachoma Prevalence and Associated Risk Factors in The Gambia and Tanzania: Baseline Results of a Cluster Randomised Controlled Trial |
Achieving a theoretical foundation for malaria elimination will require a detailed understanding of the quantitative relationships between patient treatment-seeking behavior , treatment coverage , and the effects of curative therapies that also block Plasmodium parasite transmission to mosquito vectors . Here , we report a mechanistic , within-host mathematical model that uses pharmacokinetic ( PK ) and pharmacodynamic ( PD ) data to simulate the effects of artemisinin-based combination therapies ( ACTs ) on Plasmodium falciparum transmission . To contextualize this model , we created a set of global maps of the fold reductions that would be necessary to reduce the malaria RC ( i . e . its basic reproductive number under control ) to below 1 and thus interrupt transmission . This modeling was applied to low-transmission settings , defined as having a R0<10 based on 2010 data . Our modeling predicts that treating 93–98% of symptomatic infections with an ACT within five days of fever onset would interrupt malaria transmission for ∼91% of the at-risk population of Southeast Asia and ∼74% of the global at-risk population , and lead these populations towards malaria elimination . This level of treatment coverage corresponds to an estimated 81–85% of all infected individuals in these settings . At this coverage level with ACTs , the addition of the gametocytocidal agent primaquine affords no major gains in transmission reduction . Indeed , we estimate that it would require switching ∼180 people from ACTs to ACTs plus primaquine to achieve the same transmission reduction as switching a single individual from untreated to treated with ACTs . Our model thus predicts that the addition of gametocytocidal drugs to treatment regimens provides very small population-wide benefits and that the focus of control efforts in Southeast Asia should be on increasing prompt ACT coverage . Prospects for elimination in much of Sub-Saharan Africa appear far less favorable currently , due to high rates of infection and less frequent and less rapid treatment .
Plasmodium falciparum , the most virulent of the Plasmodium species that cause malaria in humans , is responsible for hundreds of millions of cases per year [1] . The number of fatal outcomes is a matter of considerable debate , with estimates for 2010 ranging from 655 , 000 to 1 , 238 , 000 [2] , [3] . Studies nonetheless agree that overall levels of morbidity and mortality have declined over the past decade , due at least in part to the worldwide scaling up of insecticide-treated bed nets and the use of artemisinin-based combination therapies ( ACTs ) . ACTs , which pair fast-acting short-lived artemisinin derivatives with longer-lasting partner drugs , are now the first-line antimalarial drugs in almost the entire malaria-endemic world [4] , [5] . Public health and malaria infection experts are increasingly promoting the goal of malaria elimination in areas of low transmission [6] , [7] while planning ways to achieve significant reductions in higher-transmission areas [8] , [9] . Major obstacles , however , stand in the way . These include insecticide and drug resistance [10] , [11] , under-developed health care systems , shifting public funding priorities , donor fatigue [6] , malaria importation [12] and economic constraints [13] . The complex life cycle of P . falciparum also presents unique challenges [14] . P . falciparum stages differ markedly in their levels of metabolic activity , within-host locations , and susceptibilities to antimalarials . Mathematical modeling can help guide elimination efforts by providing quantitative predictions to assess the feasibility of different intervention and control strategies [15]–[18] . ACTs and other antimalarial drugs reduce transmission in three ways: by killing the disease-causing asexual blood stages and thus preventing continued production of the intra-erythrocytic sexual gametocyte forms; by killing existing gametocytes and reducing or preventing onward transmission to the mosquito ( thereby reducing parasite oocyst numbers in mosquito midguts ) ; and by post-treatment drug prophylaxis wherein residual drug levels can protect against new infections [19] . Here we utilize mathematical modeling to quantify how ACTs reduce malaria transmission , with or without late-stage gametocytocidal agents such as primaquine ( PQ ) or methylene blue . These agents are receiving considerable interest within the malaria community as to how their action might be leveraged to help interrupt parasite transmission [20]–[22] . Here , we report outputs from our within-host model of P . falciparum infection and transmission , and overlay these findings onto geospatial maps of malaria endemicity in order to predict the benefits of extended coverage of infected individuals and incorporation of transmission-blocking agents into current ACT regimens . In low-transmission settings we predict that if at least 93–98% of all symptomatic infections , corresponding to an estimated 81–85% of all infected individuals , were treated within five days of first fever , then the RC could be reduced to below one and malaria would progress towards elimination in regions harboring over ∼91% of at-risk populations in Southeast Asia and ∼74% of the global at-risk population . Our findings suggest that increasing treatment coverage with ACTs would be more effective than adding additional transmission-blocking agents in driving towards the goal of malaria elimination in Southeast Asia .
We used our recently developed within-host model of the progression of P . falciparum infection [23] to simulate the densities of asexual blood stage parasites and gametocytes in a population of individuals with no acquired immunity to malaria . The variability in densities among individuals was matched to the variability observed in malaria therapy studies , in which syphilitic individuals with no history of malaria infection were infected with P . falciparum to induce a fever and clear the syphilis infection [23] . Of note , our model incorporates three different types of antimalarial immunity: an innate response that establishes an upper limit for parasite density; a PfEMP1 variant-specific response that regulates short-term periodic oscillations in density; and a variant-transcending response that causes a steady log-linear decrease in density over time , clearing the infection [23] . In our simulations , these responses were calibrated such that the infection dynamics matched those of experimental challenge volunteers who had no previous malarial infections , i . e . , we assumed that individuals either had no prior episode of malaria or had acquired malaria so long ago that their responses were equivalent to that of individuals without prior infections . Further , we simulated only single infections , i . e . we did not simulate infections that overlapped in time . Figure 1 illustrates six runs from our within-host infection model; untreated individuals are denoted by ‘Untreated’ or ‘U’ . Figure 1A shows the log10 parasitized red blood cells ( PRBC ) per µL , while Figure 1B depicts the daily gametocytemias over time , which are typically two logs lower for untreated individuals . The Figure 1A inset illustrates the asexual densities for the first 50 days post emergence of parasites into the bloodstream; colored triangles illustrate the onset of first fever [23] . Once the daily gametocytemias were simulated , a gametocyte density-to-infectivity relationship was utilized to translate these values into predicted infectiousness to mosquitoes over time . Figure 1C illustrates the predicted human-to-mosquito infectivity for each of the three untreated individuals ( U ) using the Jeffery-Eyles ( JE ) relationship between gametocyte densities and infectivity [23] . As described in detail below , we then incorporated the effects of drug treatment into our within-host model to illustrate how this modeled treatment affects malaria transmission . Figure 1 includes the results of a hypothetical drug treatment on three individuals infected with P . falciparum; treated individuals are denoted by ‘Treated’ or ‘T’ . As with the untreated cases , Figure 1A shows the asexual parasite densities in log10 PRBC/µL , while Figs . 1B and 1C depict the daily gametocytemias and human-to-mosquito infectivities . Treatment was assumed to begin 4 , 9 and 14 days after the onset of fever . The effects of drug treatment can be seen immediately on the asexual blood stage population , which showed a steep drop in numbers after dosing , as well as the gametocytemias that were lower among the treated individuals following treatment . To calculate the net infectivity of individuals to mosquitoes , gametocytes densities were transformed into infectivity probabilities , and the area under the infectivity curve was derived ( AUIC ) . This approach was previously used with field data [24] to estimate onward infectivity following treatment . In the case where drugs reduced oocyst numbers independently of their effects against gametocytes , the gametocyte density-to-infectivity relationships were adjusted in a drug-dependent manner . Infectivities of treated patients ( T ) were of relatively brief duration and had largely disappeared within 30 days ( Figure 1C ) . This hypothetical example illustrates many of the processes involved in modeling the effects of drugs on transmission . Below we describe how we have used in vitro and field data to parameterize the various components of drug activity and predict the effects of various antimalarial therapies in real-world settings . To model the effects of different drugs on asexual parasite densities , we first modeled the within-host concentrations of the partner drugs of two ACTs , artesunate+mefloquine ( AM ) and artemether+lumefantrine ( AL ) . AM is a frequently used first-line therapy in parts of Southeast Asia [24] , while AL has recently become the most widely-used ACT worldwide [25] . Because we calculated our asexual parasite densities daily , we did not model the explicit concentrations of the artemisinin derivatives , as these drugs have half-lives of 1 to 3 hr [4] . However , we did incorporate their fast-acting PD potency on asexual blood stage parasite densities [4] . As a point of reference , we also modeled the drug concentrations of chloroquine ( CQ ) , the former first line therapy that is highly active against asexual blood stages and has some activity against very early stage gametocytes but is inactive against mature gametocytes [21] . For lumefantrine ( LMF ) uptake and clearance we used a two-compartment pharmacokinetic model parameterized from field data; for mefloquine ( MFQ ) and CQ we used two-compartment non-parametric models , also parameterized from field data . Model equations , model parameters , and a description of the model fitting to average concentrations as well as population variation are detailed in Text S1 . Notably , our study assumed that individuals were fully compliant with treatment and that parasites were sensitive to the various drug combinations . In future work we hope to examine how differences in patient compliance and parasite drug susceptibility impact transmission . Figure 2 ( panels A , C , E ) illustrates the results of our PK simulations for LMF , MFQ , and CQ , respectively . The black lines indicate the median ( LMF ) or mean ( MFQ , CQ ) of the population concentrations , while the blue lines indicate simulated individual concentration profiles . Model outputs revealed wide variations in concentrations within a population , due to differences in rates of drug uptake and clearance . The number of concentrations depicted in each panel corresponds to the number of patients in each of the studies that provided data for model fitting . Of note , LMF plasma concentrations achieved considerably higher levels than the other two agents . To simulate the effects of these drug concentrations against asexual parasites , we first calculated hourly plasma and/or blood concentration levels from our PK modeling . We then translated the hourly drug concentrations into asexual activities , assuming that the dose-response relationships could be modeled as Hill functions . These Hill functions were parameterized from in vitro and field data ( see the SI ) . The asexual activities of drugs were quantified as 48-hour parasite reduction ratios ( PRRs ) , i . e . , the fold decreases in parasite numbers every 48 hr , corresponding to one cycle of intra-erythrocytic development and reinvasion . Figure 2 ( panels B , D , F ) illustrates the drug concentrations translated into these PRRs over time . To calculate the asexual parasite densities in our model under the effects of drugs for a generic day t , we took the densities from day t-1 , applied our within-host model to calculate densities on day t including both the effects of parasite growth and host immune responses , then multiplied densities by the square root of the mean 48-hour PRR for drug concentrations during day t . The resulting density was then used to calculate densities at t+1 , and so on until the end of the simulation . Regardless of the drug regimens simulated here , asexual parasite densities fell rapidly when an individual was treated . The PRRs nevertheless showed substantial differences between drugs in later time periods following treatment . For example , the LMF PPRs rapidly declined within 5–15 days of treatment , whereas CQ took longer to decline while showing more heterogeneity . MFQ was also heterogeneous but always showed lower PPRs even at peak plasma concentrations . The maximal PRRs , and the differential rates of absorption , clearance , and volumes of drug distribution are described more fully in the SI . While effective drug treatment rapidly clears asexual blood stage parasites , even successful regimens differ markedly in their effects on gametocytes . As gametocytes develop , they become metabolically more inert , thus reducing the number of drug targets available . Specifically , gametocytes mature over the course of ∼10–15 days through Stages I–V , with each stage differing in metabolic activity and drug susceptibility [26] . Stages I–IV are sequestered in the microvasculature , spleen or bone marrow and are not present in the blood circulation . Only Stage V gametocytes have matured to the point of releasing from sequestered sites , thus becoming visible by blood smear . Here , we simulated differential effects of drugs against each developmental stage . This feature is novel to the literature so far as we know , although a recent article did allow for a differential effect on non-circulating vs . circulating gametocytes [27] . To parameterize the modeled effects of drugs on gametocytes , we first conducted a literature review to examine the types of datasets that could inform the model . There are few in vitro studies that have measured the stage-specific effects of drugs on gametocytes [21]; however , many field studies have examined the clearance of gametocytes within a population after drug treatment . These studies typically measured the proportion of individuals who were gametocytemic by microscopy after treatment ( threshold for detection: ∼5–10 gametocytes per µL [28] , [29] ) . Figure S1 illustrates the prevalence of gametocytes from field studies of patients treated with a variety of antimalarials [28] , [30]–[38] . The field studies are disaggregated by drug type: treatment with SP [28] , [30] , [33]; CQ or amodiaquine ( AQ ) ( sometimes in combination with SP ) [28] , [30]–[33]; various ACTs [28] , [30]–[32] , [34]–[36] , [38]; or ACTs plus PQ [35] , [36] , [38] . The inter-study variability observed in Figure S1 was likely due to a variety of factors , including the levels of acquired immunity in the population , exact timing of treatment , age of treated individuals , differences in parasite biology , and drug treatment regimens . These data were used to calibrate the gametocyte component of our drug effects model . To generate a set of model outputs to compare against field data , we simulated treating individuals with a three-day course of AM and varied the assumed killing properties of the two component drugs . We assumed that treatment started relatively early in the infection , i . e . , 5 days after first fever , in agreement with field studies from Thailand and Indonesia [34] , [38] . Treatment timing will vary from place to place given the treatment-seeking behaviors of the local population; if treatment is delayed significantly beyond this point , the ability of drugs to reduce transmission is likely to be diminished . We began our simulations assuming that the treatment had no effect on gametocytes at all; we called this therapy purely schizonticidal . We then increased the simulated killing properties of the combination , assuming that the components only killed early stages of gametocytes ( e . g . as for CQ ) . We gradually increased this presumed killing power against early gametocytes ( from mild to varying levels of moderate to strong , as in Figure 3A ) : as the killing power increased , gametocyte prevalence post-treatment decreased . We then assumed that the short-lived component killed both early and late stage gametocytes , with a larger effect on the former ( e . g . as for an ACT ) . Finally , we simulated adding a single dose of a third drug that killed both early and late stage gametocytes , with varying levels of activity ( e . g . as for ACT+PQ ) . Additionally , we varied the timing of this third component as this has been a topic of debate [39] , by assuming that treatment occurred either on the first day of ACT treatment , denoted by ‘<day 0>’ , or last day , denoted by ‘<day 2>’ . Our studies used the simulated prevalence of post-treatment gametocyte positive individuals as the output to be compared against field data , since this metric was most often tracked in the field . Once the model outputs had been generated , these were compared to field studies of drug treatments with similar activity ( see Figures S1 , S2 ) : the modeled schizonticidal treatment data were compared to field trials with SP; mild to moderate gametocytocidal outputs were compared to data from field trials of CQ , CQ+SP or AQ+SP; strong gametocytocidal outputs were compared to ACT clinical trials data , and the modeled triple-combination data were compared to field data for ACTs+PQ . For each drug activity type we then chose the sets of simulations that most closely resembled the mean , maximum , and minimum of observed responses to represent the effects of each class of drugs against gametocytes . We also included some intermediate sets of simulations for the sake of comparison . Figure S2 illustrates the model outputs that best approximated the mean and observed variation in the field data; all model means were from 1 , 000 runs for each parameterization . We modeled the entire range of observed variation in post-treatment gametocytemias to allow for sensitivity and robustness analyses in our results . This ‘ensemble modeling’ approach has been used previously to model the effects of vaccines on malaria transmission [40] as well as within-host P . falciparum dynamics [41] . Figure 3A illustrates the results of the gametocyte activity model fitting . The untreated model gametocyte prevalence 5 days after first fever is shown in black . The modeled post-treatment gametocyte prevalence assuming a pure schizonticidal combination is shown in green . The four model parameterizations that best correspond to the observed field patterns after CQ treatment are labeled ‘CQ mild’ , ‘CQ moderate 1’ , ‘CQ moderate 2’ , and ‘CQ strong’ , respectively ( indicating increasing levels of activity against very early stage gametocytes ) . The three model parameterizations that best correspond to the ACT field patterns are labeled ‘ACT mild’ , ‘ACT moderate’ , and ‘ACT strong’ , respectively . The four model parameterizations that correspond to the ACT+PQ field studies are labeled ‘ACT+PQ moderate <day 0>’ , ‘ACT+PQ moderate <day 2>’ , ‘ACT+PQ strong <day 0>’ , and ‘ACT+PQ strong <day 2>’; the bracketed number indicates the day on which the simulated PQ component was administered , relative to the other two drug components . Once the gametocyte parameterizations were fitted for each type of drug combination , we then transformed the daily gametocytemias before and after treatment into predicted infectivities to mosquitoes . These transformations utilized gametocyte density-to-infectivity relationships derived from mosquito feeding studies , as described in [23] . We chose two transformations , one derived from studies of mosquito feeding on malaria therapy patients with no prior history of malaria infection ( ‘Jeffery-Eyles’ or JE ) and the other derived from feeding studies conducted in field trials in Africa ( ‘Carter & Graves’ or CG ) . In short , the JE transformation assumes that 1 ) gametocytes appearing in the first few days of infection are non-infectious ( immature ) ; 2 ) low density gametocytemias are relatively non-infectious; 3 ) high density infections are highly infectious . The CG relationship assumes that 1 ) gametocytes are immediately infectious; 2 ) low density gametocytemias are relatively infectious; 3 ) high density infections are not as infectious as for JE [23] . These two functions are substantially different and represent some of the possible types of density-to-infectivity relationships . Calculation using both relationships allows us to highlight where differences in density-to-infectivity assumptions play an important role in interpreting model outputs [23] . It was assumed for both parameterizations that modeled gametocytemias were infectious at densities below the level of detection by microscopy ( ∼5–10 gametocytes per µL ) . Our model is thus able to capture the effects of ‘submicroscopic’ infections [23] , [35] , [42] . However , as densities decrease , infectivity decreases asymptotically toward 0 ( see Figure 3 of [23] ) . As a simplification we assumed that gametocyte densities below 2 gametocytes per 3 µL were non-infectious ( given the need for 2 gametocytes per mosquito bite that typically collects ∼3 µL of blood ) [23] . Figure S3 illustrates the modeled gametocytemias of Figure 3A transformed into probabilities of mosquito infection , along with data from feeding studies in the field [28] , [30]–[33] , treatment with various ACTs [28] , [30]–[32] , [34]–[36] , [38] , and treatment with ACTs plus PQ [35] , [36] , [38] . Both the JE and CG transformed probabilities are shown; the data transformed using the JE assumptions are in bold . The coloring of modeled infectivity data corresponds to that of Figure 3A . The field feeding study data were disaggregated according to the same criteria as the gametocyte clearance data . Once the model gametocyte and infectivity parameters were fitted to data , we then calculated the AUIC for each drug parameterization [23] . Table 1 provides the unadjusted net human-to-mosquito infectivity for each of the drug parameterizations in Figure 3A . All data are from the mean of 1 , 000 model runs . To determine the transmission reduction achieved with each treatment , we divided the untreated AUIC by the treated AUIC . For example , to calculate the transmission reduction post-treatment , we first took untreated individuals and calculated the mean AUIC for 5 days after first fever until the end of simulation; the mean untreated AUIC values were 31 . 8 for the JE parameterization and 29 . 3 for the CG parameterization ( Table 1 ) . The post-treatment net infectivities of treated individuals , i . e . the treated AUIC values , were 0 . 70 for JE and 0 . 94 days for CG ( Table 1 ) . Mean fold reductions in transmission , post ACT treatment , were then 47 . 1 and 32 . 4 for JE and CG transformations , respectively ( Table 1 ) . The quantity most relevant for control efforts is the total effect size , i . e . , the reduction in transmission that includes the period of transmissibility before treatment . The longer that individuals wait to be treated , the less the maximum effect size achievable , because these individuals could transmit the parasite prior to treatment . After adding pretreatment infectivity ( 0 . 03 or 3 . 06 infectious days for JE and CG transformations respectively ) , the total mean fold-reductions for ACTs were 45 . 3 and 8 . 1 , respectively . For ACT+PQ , total mean effect sizes were predicted to be 92 . 2 ( JE ) and 8 . 9 ( CG ) , respectively . The reason for the large differences between these two transformations is how they incorporate pre-treatment infectivity . Thus , pretreatment infectivity plays a major role in the total effect size . For JE , gametocytes were assumed to be non-infectious early in the course of an infection , thus pretreatment infectivity was almost nonexistent and the effect size was determined by post-treatment infectivity . In contrast , the CG model assumed that gametocytes were infectious upon emergence , thus pretreatment infectivity was relatively large compared to post-treatment infectivity . Purely schizonticidal treatments ( with zero gametocytocidal activity ) were predicted to reduce post-treatment transmission 6 . 2 to 5 . 7 fold ( JE and CG , respectively ) . Including pretreatment infectivity , the mean effect size of a pure schizonticide was 6 . 2 and 3 . 9 ( JE and CG , respectively ) . For CQ , which is moderately gametocytocidal against stage I–II gametocytes , total effect sizes were estimated to be 15 . 6 and 6 . 0 for the JE and CG parameterizations , respectively . These findings highlight the substantial benefit of drugs with more potent gametocytocidal activity in reducing transmission . The fold reductions in Table 1 illustrate how antimalarials reduce transmission assuming 100% treatment coverage . However , these calculations do not incorporate the oocidal effects of some antimalarials . Treatments that are also oocidal ( such as SP , LMF , and MFQ [42] ) will have larger effect sizes than predicted in Table 1 because of greater reductions in overall human-to-mosquito transmission . To calibrate oocidal drug effects , we compared unadjusted model-predicted infectivity and observed field infectivity post-treatment ( see Text S1 and Figure S3 ) . For our ACT model infectivity , at day 7 the feeding studies indicated an infectiousness of approximately 2–3 . 5% , whereas the model predicted 5–8% under the JE parameterization ( see Figure S3C ) . This small difference between model and field studies nearly disappeared by day 14 ( possibly because of a dose-response effect of LMF on mosquito stage development ) . To incorporate oocidal activity , we thus assumed that ACTs ( AL or AM ) reduced onward infectivity by 50% compared to the mean values in Table 1 . Mean post-treatment infectivity values became 0 . 35 ( JE ) and 0 . 47 ( CG ) net infectious days while the total effect sizes became 83 . 7 ( JE ) and 9 . 2 ( CG ) . These data are shown in Table 2 . Again , the assumed importance of pretreatment infectivity played a crucial role in determining the effect sizes of treatment when the oocidal effects of treatment were included . When examining the effects of ACTs plus PQ , our adjustment for the oocidal effects of PQ was different than that for LMF or MFQ , because PQ is active against mosquito stages for only a few days after treatment , but reduces infectivity almost completely during its period of activity [43] , [44] . If we assumed that infectivity in the first three days post ACT+PQ treatment was zero , then the net infectivity post-treatment was 0 . 19 ( JE ) and 0 . 30 ( CG ) and the total mean effect sizes were 162 . 1 ( JE ) and 9 . 6 ( CG ) , respectively . Figure 3B illustrates the total effect sizes for each of the modeled drug parameterizations as a function of treatment coverage , including the oocidal effects of drugs from Table 2 , assuming that the mean period of infectivity in untreated individuals is 30 . 5 days ( mean untreated infectivity from Table 1 ) and assuming the JE density-to-infectivity parameterization for treated individuals . Each drug class is depicted in a different color with individual lines showing simulation outputs assuming varying levels of gametocytocidal activity ( CQ: mild , moderate 1 , moderate 2 , strong ) ; ( ACT: mild , moderate , strong ) ; ( ACT+PQ: moderate 0 days delay , moderate 2 days delay , strong 0 days delay , strong 2 days delay ) . The black lines clustered within each drug class indicate the mean effect size across all simulations . The horizontal line illustrates a six-fold reduction in transmission , a threshold discussed below . The dotted vertical lines indicate the levels of treatment coverage needed to reach the six-fold reduction in total human-to-mosquito transmission for each drug class . The y-axis is in log-scale . This figure graphically illustrates the importance of treatment coverage in determining the effect size of a control program . At 100% coverage ( i . e . all infected individuals ) , the effect size of ACTs is 87 . 3 ( JE parameterization , Table 2 ) , assuming untreated infectivity is 30 . 5 days . However , this value drops quickly , yielding 16 . 3 at 95% coverage , 9 . 0 at 90% coverage , 4 . 8 at 80% coverage , and 3 . 2 at 70% coverage . For ACT+PQ , the effect sizes are 162 . 1 , 17 . 7 , 9 . 4 , 4 . 9 , and 3 . 3 at 100% , 95% , 90% , 80% , and 70% coverage , respectively . Treatment with ACTs exceeds a six-fold reduction threshold at ∼84 . 3% coverage of the total population , whereas the ACT+PQ regimen exceeds a six-fold reduction at ∼83 . 9% coverage of the total population . The above effect sizes ( and those in Figure 3 ) were calculated assuming the untreated infectivity is 30 . 5 days ( the mean of the JE and CG parameterizations ) and using the JE parameterization for treated individuals . If we use the average of the JE and CG parameterizations for treated individuals , and still assume that the untreated net infectivity is 30 . 5 days , the pretreatment net infectivity is 1 . 55 ( average from Table 1 ) , the post-treatment net infectivity for ACTs is 0 . 41 ( average from Table 2 ) , and the computed effect sizes of ACTs become 15 . 6 at 100% coverage , 9 . 0 at 95% treatment coverage , 6 . 3 at 90% treatment coverage , 4 . 0 at 80% coverage , and 2 . 9 at 70% coverage . Assuming a post-treatment net infectivity of 0 . 245 for ACT+PQ ( average from Table 2 ) , these values become 17 . 0 , 9 . 4 , 6 . 5 , 4 . 0 , and 2 . 9 at the coverage levels listed above . Treatment with ACTs reaches a five-fold reduction in transmission at 85 . 5% coverage of the total population; with ACTs+PQ , the coverage level required is 85 . 0% . Treatment with ACTs exceeds a six-fold reduction threshold at 89 . 1% coverage , whereas the ACT+PQ regimen exceeds a six-fold reduction at 88 . 5% coverage . These outputs suggest a barely detectable impact of adding PQ to ACTs in the context of reducing transmission levels with these model assumptions . To contextualize the fold-reductions in transmission theoretically achievable with various treatments , we developed a set of maps of the fold-reductions in malaria transmission necessary to achieve elimination in low-transmission settings . These maps were derived from the worldwide maps of the basic reproductive number of malaria , R0 [45] , assuming the malaria control coverage of 2010 as the baseline . We can also consider these maps as calculating the RC , i . e . the reproductive number under control efforts , as of 2010 , though here we use the terms R0 and RC interchangeably . R0 is a threshold criterion for transmission: if R0>1 over a given region , the disease will spread within this region ( unless there is significant migration ) , if R0<1 , the disease will disappear within this region ( unless there is significant importation ) . In brief , the R0 values described in [45] were developed by regressing various malariometric data ( such as elevation and rainfall ) on tens of thousands of parasite rate surveys and modeling the spatio-temporal autocorrelation structure of the residual variation . The regressions used were Bayesian and geostatistical , producing a full Bayesian posterior distribution for the age-standardized parasite rate at each pixel ( with a per-pixel size of 5×5 km , i . e . 5 km2 ) . These worldwide maps of predicted parasite rates varied in intensity from pixel to pixel , given different magnitudes of various malaria covariates . By utilizing empirical and theoretical relationships , the maps were combined with aspects of malaria that remain constant over time and space to calculate R0 . One such malaria invariant is the net infectivity of infected humans to mosquitoes , assuming no acquired immunity developed over the course of repeated infections [23] . To calculate this invariant , we used our within-host model of malaria transmission to simulate the progression of infectivity in thousands of simulated individuals , and then calculated the mean area under the human-to-mosquito infectivity curves [23] . To achieve elimination in a given area , the fold-reduction in transmission under control must be greater than or equal to the R0 [46] . Thus , we took the worldwide maps of R0 in [45] and calculated the fold-reductions necessary over each pixel to reduce the estimated RC to below 1 . Our maps of the transmission reductions include estimates of uncertainty inherited from the RC posterior densities at each pixel . Table 3 summarizes these maps by providing the number of people living in areas requiring <2 fold reductions in transmission to interrupt transmission , as well as 2–5 , 5–10 , 10–20 , 20–50 , 50–100 , and >100 fold reductions to interrupt transmission . Table 3 calculates these necessary fold-reductions with 75% confidence , i . e . , our posterior estimates of RC fall within the given regions with 75% confidence . Because our within-host model simulates the progression of infections for individuals with no prior history of malaria infection , our modeling conclusions are most relevant for areas of low transmission where individuals have accumulated little acquired immunity from prior infections . We thus restricted our analyses to areas with R0<10 , which excludes the more endemic areas of Africa and aligns with the transmission levels prevalent in Southeast Asia . In terms of biting intensity , an R0 of 10 translates to a yearly entomological inoculation rate of approximately 3 infectious bites per person per year , which would result in approximately 1 . 5 infections every year [47] . This upper limit of analysis can be compared with the intensity reported in an area of ‘low and seasonal’ transmission in Thailand , where individuals had one infection every other year [34] . Figure 4 provides a worldwide map of the probabilities that areas can interrupt malaria transmission ( RC<1 ) assuming a five-fold reduction in transmission . Areas with R0>10 are masked , as these regions have such high transmission that our modeling predictions are less relevant . Fold reductions were organized into 6 bins for clarity . Figure 5 shows these probabilities of interruption for Southeast Asia , where transmission is generally much lower than in Africa . Figures S4 and S5 present maps of the probabilities of interrupting transmission assuming two- or ten-fold reductions in transmission , respectively . These maps are discussed below in the context of the reductions achievable with antimalarial drugs .
This study calculates the effects of antimalarial therapies on P . falciparum transmission , using a within-host model of malaria infection [23] and a PK/PD model parameterized from field studies . The effects of drugs are modeled on both asexual and sexual stages of parasite development under different assumptions of gametocyte-to-infectivity relationship . We also generate global maps of the fold reductions in malaria transmission , i . e . the effect sizes , necessary to achieve elimination in regions of low endemicity ( defined as having local R0 values less than 10 ) . From our model outputs , we can generate three major conclusions . First , the infectivity of individuals before treatment plays a crucial role in determining effect size . If treatment is delayed more than only a few days after the onset of fever , and gametocytes are infectious during this period , then the effect sizes achievable even with first-line ACT therapies plus the gametocytocidal agent PQ are limited . Second , if we account for the effects that the partner drugs LMF and MFQ exert upon mosquito stages of the parasite life-cycle , then there is little difference in the benefits of ACTs versus ACTs+PQ in terms of transmission reductions . Both regimens are extremely effective at stopping onward transmission , with many fold greater benefits versus purely schizonticidal treatments that act only upon asexual blood stage parasites . Third , the proportion of individuals receiving treatment has a major impact on reductions in transmission ( Figure 3B ) . In Tables 1 and 2 our effect size calculations assumed 100% coverage . Because untreated individuals are so much more infectious than treated individuals , leaving even a few individuals untreated drastically reduces the effectiveness of a control program . We can put these fold-reductions in context using our maps of transmission reductions necessary for elimination . Figure 4 illustrates a worldwide map of the probabilities that a five-fold reduction in transmission would interrupt the spread of malaria . In this map the pixel size is 5 km2 . Because our model is most applicable in regions of relatively low transmission , we masked out the regions where R0 , the basic reproductive number , is predicted to be greater than 10 ( with a probability exceeding 50% ) . Higher transmission regions are more difficult to model , given the complex interactions of immunity , superinfection , and control . As can be seen from the map , many areas of Africa have such intense transmission that R0 exceeds 10 , and we cannot say how transmission might be affected by the use of drugs in such areas . However , examining the map , one can visualize many regions of Africa , including the Sahel , most of East Africa , and parts of Southern Africa , where elimination would appear possible with a five-fold reduction in transmission . Further , much of India and Southeast Asia have low enough transmission that elimination would be possible at this level of control . Prospects for elimination in Myanmar and southern Thailand , however , do not appear to be favorable . Figure 5 provides a zoomed-in view of the probabilities of malaria elimination with an effect size of five , focusing on Southeast Asia . Table 3 provides a quantification of the populations at risk both worldwide and in Southeast Asia , as well as the populations where elimination is possible at different levels of control . Worldwide , regions where malaria can be interrupted with five-fold reductions contain 74% of the population at risk; in Southeast Asia regions that can interrupt transmission with five-fold reductions harbor 91% of the population at risk . Given these maps and quantifications of populations at risk , we can apply our modeling results to determine the percentage of the population that needs to be treated promptly with antimalarials to interrupt transmission in various areas . Figure 3B illustrates the relationship between treatment coverage with different antimalarials and the resulting fold reductions in transmission , assuming the Jeffery-Eyles gametocyte density-to-infectivity relationship ( used to calculate the needed treatment coverage levels below ) . To achieve a five-fold reduction in transmission , approximately 81% of the total infected population would need to be treated with ACTs or ACTs+PQ . As a comparison , achieving this fold reduction with CQ ( with weak activity against early-stage gametocytes ) or a schizonticide ( with no gametocytocidal activity ) would require treatment coverage of ∼86% and ∼96% respectively . To achieve a six-fold reduction in transmission , approximately 84% of the total infected population would need to be treated with ACTs or ACTs+PQ ( illustrated in Figure 3B ) . Our modeling thus suggests that the addition of PQ to an ACT would provide almost negligible benefits at these levels of coverage , reducing the fraction of the population needing to be treated by less than 1% versus treatment with ACTs alone that already provide quite potent gametocytocidal activity ( Figure 3B ) . Combining the maps and the within-host modeling results based on our JE parameterization , we thus estimate that promptly treating ∼81% of the total infected population with ACTs and/or ACTs+PQ would interrupt transmission in areas covering 91% of the population in Southeast Asia . These coverage rates are for the infected population as a whole , regardless of whether individuals are symptomatic or not . In a study conducted in a region of western Thailand with low and seasonal transmission , most infections ( 87% ) were found to be symptomatic [48] . In experimental challenge studies among human volunteers , all subjects displayed some degree of symptoms [49] . If we take the former proportion ( 87% ) as the percentage of the infected population that is symptomatic , then 93% of the symptomatic population would need to be treated with ACTs and/or ACTs+PQ to achieve interruption in the areas of Figure 5 based on the predicted five-fold reduction; the percentage rises to 97% to achieve six-fold reductions . Thus , it is possible that treating only symptomatic individuals may be sufficient to eliminate transmission throughout most of Southeast Asia . Almost all of these individuals , however , would need to be reached with treatment ( either through a public campaign or private sector provisioning or a combination of both ) in order to interrupt transmission , using drugs alone . Our result that prospects for malaria elimination are favorable for most of Southeast Asia is supported by two other studies that also find that elimination efforts are feasible using antimalarials in this region [50] , [51] . Figures S4 and S5 illustrate the probabilities assuming control interventions with two-fold and ten-fold reductions , respectively . At the two-fold level , much of central Thailand can interrupt transmission , but there are significant portions of Myanmar , Cambodia , Southern Laos , and Southern Thailand where elimination is not likely . At the ten-fold level , there are small pockets in Southern Thailand as well as large areas in Myanmar where interruption is still not likely . These estimates may be somewhat optimistic because we are using only the JE density-to-infectivity relationship when calculating the infectivity of treated individuals . As we are focusing on low transmission areas , this assumption seems reasonable ( the JE relationship was derived from individuals with no prior infections ) . However , if we take the mean of the JE and CG relationships for treated individuals , then we predict that it would require treating 85 . 5% of the total population , or 98 . 3% of symptomatic patients with ACTs to achieve a five-fold reduction ( 85 . 0% of the total or 97 . 7% of symptomatic patients with ACTs+PQ ) . Further , we find that it is not possible to achieve a six-fold reduction only treating symptomatic individuals with either the ACTs or the ACTs+PQ modeled here , assuming the mean of the JE and CG relationships for treated individuals . Thus , the assumed density-to-infectivity relationship has a large effect on the calculated effectiveness of control programs . We note that our maps predict the levels of control necessary to interrupt transmission at the per-pixel level ( 5 km2 ) , incorporating uncertainty analysis . These average reductions needed to interrupt malaria transmission are not at the per-village or per-household level . Hotspots of transmission will need to be identified and treated in order to achieve elimination in a given region . The uncertainty for each pixel takes into account this heterogeneity to some degree , but nevertheless caution is advised before using these maps for local-scale planning . We would suggest our maps be used to guide elimination planning at a regional or national level; for elimination planning at a district or city level more intensive surveillance will likely be needed . If we consider the timelines to elimination , the narrower the margin by which the effect size exceeds the threshold for elimination , the longer elimination will take , as population-wide transmission will decay more slowly [46] . Conversely , the higher the proportion of individuals above the needed threshold , the faster elimination will be achieved [46] . We do not compute the quantitative benefits of mass drug administration here , and instead focus on individual-level treatment . However , we would qualitatively expect that mass drug administration may provide a benefit to elimination efforts by speeding an area toward faster elimination , assuming that the critical level of coverage can be reached ( i . e . the RC of the region drops to below 1 ) . Once an area has eliminated malaria , the costs of maintaining elimination may be less than those needed to achieve elimination in the first place , though more research is needed on strategies to maintain elimination in previously endemic areas [7] . In areas where antimalarials are predicted to be insufficient to achieve elimination , other interventions may be included in control efforts to increase the effect size of the combined control effort . The combined effect size is simply the product of both component interventions . For example , if the coverage level with antimalarials reduces transmission by three-fold and distribution of bed nets reduces transmission another three-fold , the combined effects are a nine-fold reduction in transmission ( as long as there are no antagonistic interactions between the two efforts ) . Thus , high fold-reductions can be achieved by bundling interventions . While we do not compute the effect sizes of other interventions here , the results in this paper can be combined with other modeling efforts for the purposes of an integrated elimination effort . Given these conclusions , serious efforts to eliminate malaria will require extensive planning and sustained support [6] . We note the encouraging prediction that high coverage ( at least 81–85% of total infections , corresponding to an estimated 93–98% of symptomatic infections ) with ACTs that act against P . falciparum asexual , sexual , and mosquito stages might suffice to interrupt transmission throughout most of Southeast Asia , especially if complimented by insecticide-treated bed net distribution to reduce population infectivity . We also note that the addition of a single dose of a purely gametocytocidal drug such as PQ to ACTs can reduce onward transmission slightly . However , the focus of control efforts should be on maintaining a high level of treatment coverage . Based on our modeling , PQ and similarly gametocytocidal therapies added to ACTs do not appear to be a magic bullet ensuring elimination and add only nominally to the transmission reductions achievable with ACTs that act against the various parasite stages at feasible levels of coverage . Efforts are ongoing to utilize our model to predict the effects of possible emerging artemisinin resistance , which threatens existing ACT control strategies [10] , [52] , [53] . Additional modeling is also required to delineate better what measures beyond expanded ACT and insecticide-treated bed net coverage would help reduce the RC in Africa to levels that might make elimination an achievable goal .
Our recently reported within-host mathematical model [23] was utilized to simulate asexual and sexual blood stage parasite densities over time in untreated and treated individuals . This model reproduces the range of observed parasite densities among individuals undergoing malaria therapy , wherein adult males with tertiary syphilis ( and no acquired immunity to malaria ) were infected with various strains of P . falciparum to induce a fever in order to clear the syphilis [29] , [54]–[56] . Our model uses a combination of parasite antigenic variation and host immune responses to reproduce the observed range of responses in these patients . The model calculates the density of asexual parasites every two days and uses log-linear interpolation to generate daily counts . The model also calculates the daily human-to-mosquito infectivity using gametocyte density-to-infectivity relationships derived from mosquito feeding studies on human volunteers [23] . The source code for our model is provided in Dataset S1 ( see SI ) . We note that our modeling uses discrete-time difference equations rather than a continuous time model , to calculate both asexual densities and gametocyte densities over time . We chose the former , as the calculation of gametocyte densities from asexual densities is difficult with a continuous-time model because gametocyte densities are a function of weighted cumulative asexual densities and are highly stochastic . For a thorough description , we refer to [57] . We also note that an insightful report by Kay and Hastings [58] , building on earlier work from this group [59] , simulated the concentrations of both artemisinin derivatives and ACT partner drugs using compartmental PK/PD modeling . These authors also simulated the killing effects of these drugs against asexual parasites assuming Michaelis-Menten dose-response functions ( which are similar to the Hill functions used here ) . These authors focused on the effects of emerging artemisinin resistance and provided substantial data on artemisinin PK/PD properties . Our two studies differ in that we simulated parasite densities daily and so we did not explicitly model artemisinin PK values because of the short half-lives of the artemisinins . Further , our stochastic difference equation model of within-host parasite growth and immune response was calibrated to match the range of variation in the malaria therapy data [23] . We also modeled the effects of drugs against asexual , sexual , and mosquito stages of development . Thus , The Kay and Hastings study [58] is a representation of the effects of drugs against asexual stages incorporating resistance , whereas our study focused on the effects of drugs on total human malarial transmission . Importantly , their simulations predict that the spread of ART resistance would result in a potentially rapid decline in ACT effectiveness . In future studies , we plan to investigate the effect of ART resistance on effect sizes and how this would impact the required treatment coverage to drive the RC to below 1 in low-transmission settings . PK modeling was used to simulate the concentrations of antimalarial drugs after uptake . The concentrations of CQ and its active metabolite , monodesethyl-chloroquine ( mdCQ ) , were simulated using a non-compartmental model parameterized with data from Papua New Guinean children [60] . For LMF , a two-compartmental model was developed from plasma concentration data from the treatment of uncomplicated individuals in western Thailand ( Mae La ) [61] . For MFQ , a non-compartmental model was developed that incorporated data from two studies , one in Thailand [62] and the other in Peru [63] ( the Peruvian study used whole blood concentrations , rather than plasma ) . The plasma concentrations of the artemisinins were not modeled , although such data exist [62] . This is because the half-lives of the artemisinins are so short ( ∼1–3 hr ) that effective concentrations are gone within one day after uptake [62] . A full description of the PK modeling is provided in the SI . The dose-response effect of antimalarials against asexual parasites was assumed to follow the commonly used ‘Hill function’ [64] , a four-parameter dose-response function: , where a is set to 0 , b is set to 1 , c is what we term the EC50 , d is the Hill slope , and x is the plasma concentration of the drug . The Hill function dose-response curve type was chosen because this function type was utilized by both of the references that provided in vivo EC50 values for MFQ [65] , [66] , and because many of the in vitro studies using CQ and LMF used Hill dose-response relationships to model the effects of drugs against asexual blood stage parasites . Because each drug has a characteristic maximum inhibitory effect , this dose-response function was scaled by the maximum parasite reduction ratio ( PRR ) for each drug . To determine the effect on asexual parasites of a drug concentration on a given day t of modeling , the asexual parasite densities from day t-1 were used as inputs into the within-host model . The predicted densities on day t were then calculated , incorporating the effects of host immunity and parasite growth . The mean of over day t was then subtracted from the within-host simulations after appropriate log transformation to calculate the end of day asexual parasite densities , incorporating the effects of host immune responses , parasite growth , and drug concentrations . These densities were then used to calculate the asexual parasite densities on day t+1 , and so on until the end of the simulation time . A full description of the PD modeling against asexual blood stages is provided in the SI . The dose-response effect of antimalarials against gametocytes was assumed to follow a binary model , where antimalarials act against gametocytes only if drug concentrations are above a certain drug-specific threshold . This binary model was adopted because of the current paucity of dose-response data against gametocytes , as compared to asexual blood stage parasites where the many data sets permit the use of Hill slopes to define dose-response relationships . The threshold for gametocyte activity was chosen to be the in vitro IC50 against asexual blood stage parasites scaled by a factor of five [21] . For CQ , LMF , and MFQ , this value is 40 , 174 , and 322 ng/ml , respectively ( SI ) . To determine the stage-specific gametocytocidal effects of drugs , the within-host malaria model was first run assuming treatment with a purely schizonticidal combination therapy . The post-treatment gametocyte clearance curves from these simulations were then compared to clearance curves from field studies using SP [28] , [30]–[38] to validate the model outputs . In separate simulations , we assumed that individuals were treated with a combination that was weakly gametocytocidal , and these results were compared to field data from CQ trials [28] , [30]–[33] to choose a drug parameterization . We also performed simulations assuming treatment with a stronger gametocytocidal combination , and modeled our parameters by comparing these results with ACT field trial data [28] , [30]–[32] , [34]–[36] , [38] . Finally , we simulated treatment with a stronger gametocytocidal combination paired with a third highly gametocytocidal drug , and compared these results to ACT+PQ field data [35] , [36] , [38] to parameterize this combination . Outputs from modeled drug parameterizations that were consistent with observed trends were considered representative of that type of treatment . A full description of the process of model parameterization of antimalarial effects against gametocytes is provided in the SI . Once the ensembles of gametocyte densities after treatment had been generated for various drug combinations , we used two different gametocyte density-to-infectivity relationships ( ‘Jeffery-Eyles’ and ‘Carter & Graves’ ) to translate the daily gametocyte densities into predicted human-to-mosquito infectivities [23] . These modeled daily infectivities were then compared to field studies in which mosquitoes were fed on human volunteers after treatment . For the effects of a single dose of PQ on drug transmission , we assumed that the first three days post-treatment were non-infectious ( including the day of treatment ) . For the effects of partner drugs with longer half-lives that are active against sexual-stage parasites for longer periods ( LMF , MFQ ) , we scaled the area under the infectivity curve ( AUIC ) . To quantify the uncertainty associated with our predictions , we utilized an ensemble modeling approach [40] , [41] . In ensemble modeling , various scenarios are simulated to illustrate the effects of changing assumptions of model outputs . Ensemble modeling is especially appropriate when insufficient or conflicting data exist to determine the relative likelihoods of the possible scenarios . For our ensembles , we used different sets of assumptions about the stage-specific effects of drugs against gametocytes and the type of gametocyte density-to-infectivity relationship to map out the uncertainties associated with our best-estimate predictions of the effects of drugs on transmission . | We utilize a within-host mathematical model of malaria transmission to predict the effects of antimalarial treatment across the globe . We predict that areas containing 91% of the at-risk population of Southeast Asia can achieve elimination if at least 93–98% of symptomatic individuals are promptly treated with effective artemisinin-based combination therapies ( ACTs ) , based on assessments of treatment and transmission levels as of 2010 . The benefit of attaining this level of coverage far outperforms that of adding additional gametocyte-specific transmission-blocking drugs to current ACTs . We advocate for elimination programs in Southeast Asia to focus on maximizing ACT coverage . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"mathematical",
"computing",
"mathematics",
"biology",
"microbiology",
"parasitology"
] | 2014 | Modeling Within-Host Effects of Drugs on Plasmodium falciparum Transmission and Prospects for Malaria Elimination |
Centrosomes are critical sites for orchestrating microtubule dynamics , and exhibit dynamic changes in size during the cell cycle . As cells progress to mitosis , centrosomes recruit more microtubules ( MT ) to form mitotic bipolar spindles that ensure proper chromosome segregation . We report a new role for ATX-2 , a C . elegans ortholog of Human Ataxin-2 , in regulating centrosome size and MT dynamics . ATX-2 , an RNA-binding protein , forms a complex with SZY-20 in an RNA-independent fashion . Depleting ATX-2 results in embryonic lethality and cytokinesis failure , and restores centrosome duplication to zyg-1 mutants . In this pathway , SZY-20 promotes ATX-2 abundance , which inversely correlates with centrosome size . Centrosomes depleted of ATX-2 exhibit elevated levels of centrosome factors ( ZYG-1 , SPD-5 , γ-Tubulin ) , increasing MT nucleating activity but impeding MT growth . We show that ATX-2 influences MT behavior through γ-Tubulin at the centrosome . Our data suggest that RNA-binding proteins play an active role in controlling MT dynamics and provide insight into the control of proper centrosome size and MT dynamics .
As the primary microtubule-organizing centers , centrosomes are vital for the maintenance of genomic integrity in animal cells [1] . The centrosome consists of two barrel-shaped centrioles surrounded by a network of proteins termed pericentriolar materials ( PCM ) . To maintain the fidelity of cell division , each cell must duplicate a pair of centrioles precisely once per cell cycle , one daughter per mother centriole . Mishaps in centrosome assembly result in chromosome missegregation and other cell cycle defects . Thus , stringent regulation of centrosome assembly is imperative for proper cell division and survival . Studies in C . elegans have discovered five evolutionarily conserved proteins ( ZYG-1 , SPD-2 , SAS-4 , SAS-5 and SAS-6 ) that are required for centrosome assembly [2–5] . Many other factors , including protein phosphatase 2A , also regulate the production , activity , or turnover of core regulators , and are equally important in regulating centrosome assembly [6 , 7] . Like other biological processes , centrosome assembly is regulated by a combined action among negative and positive regulators [8] . While the kinase ZYG-1 promotes centriole duplication in C . elegans , szy-20 acts as a genetic suppressor of zyg-1 [9 , 10] . The szy-20 gene encodes a centrosome-associated RNA-binding protein that negatively regulates centrosome assembly by opposing ZYG-1 . Centrosomes in szy-20 mutants exhibit elevated levels of centrosomal proteins , resulting in defective microtubule ( MT ) behavior and embryonic lethality . SZY-20 contains putative RNA-binding domains ( SUZ , SUZ-C ) . Mutating these domains has been shown to perturb in vitro RNA-binding of SZY-20 and its capacity to regulate centrosome size in vivo [10] . Other studies have shown that a number of RNAs and RNA-binding proteins are associated with centrosomes and MTs , and influence proper mitotic spindles and other aspects of cell division . In mammalian cells , several RNA-binding proteins ( e . g . , RBM8A , Hu antigen R , and the Ewing sarcoma protein ) associate with centrosomes and play a role in regulating centrosome assembly during cell division [11–15] . In yeast , spindle pole body duplication is linked to translational control via the action of RNA-binding proteins [16] . In Xenopus , MT-guided localization of transcripts followed by spatially enriched translation is important for proper MT behavior and cell division , suggesting the importance of local translational control [17–19] . Despite the finding that SZY-20 negatively regulates ZYG-1 , no direct interaction between the two proteins has been found . Thus , identifying additional factors that function between SZY-20 and ZYG-1 should provide further insights into the molecular mechanism by which the putative RNA-binding protein , SZY-20 , influences centrosome assembly . Toward this end , we report here our identification of an RNA-binding protein ATX-2 that physically associates with SZY-20 . ATX-2 is the C . elegans ortholog of human Ataxin-2 that is implicated in human neurodegenerative disease [20] . Specifically , human spinocerebellar ataxia type 2 is shown to be associated with an extended poly-glutamine ( Q ) tract in Ataxin-2 [20–22] . Ataxin-2 is an evolutionarily conserved protein that contains an RNA-binding motif ( LSm: Sm-like domain ) and a PAM domain for binding the poly- ( A ) binding protein ( PABP1 ) [23–25] . It has been shown that Ataxin-2 binds directly to the 3’UTR of mRNAs and stabilize target transcripts , and that poly-Q expansion blocks the RNA-binding by Ataxin-2 in vitro [26] . Ataxin-2 homologs have been implicated in a wide range of RNA metabolism-dependent processes including translational control of circadian rhythm [27 , 28] . While ATX-2 in C . elegans is known to be responsible for embryonic development and translational control in germline development [29–32] , the action of this RNA-binding protein in centrosome assembly and cell division has not been fully explored . In this study , we investigate the role of C . elegans ATX-2 in early cell cycles and how ATX-2 acts together with SZY-20 and ZYG-1 in controlling centrosome size and MT behavior . We show that ATX-2 negatively regulates the key centriole factor ZYG-1 . In the centrosome assembly pathway , SZY-20 acts upstream of ATX-2 and positively regulates embryonic levels of ATX-2; proper levels of ATX-2 contributes in turn to normal centrosome size and subsequent MT dynamics .
To further elucidate the role of SZY-20 in regulating centrosome assembly , we looked for additional factors interacting with SZY-20 . By immunoprecipitating ( IP ) endogenous SZY-20 from worm protein lysates with anti-SZY-20 , followed by mass spectrometry , we generated a list of proteins associated with SZY-20 in vivo . As for a putative RNA-binding protein SZY-20 , we found many known RNA-binding proteins co-precipitated with SZY-20 , including ATX-2 ( 11 peptides , 19% coverage = 181/959aa ) and PAB-1 ( 15 peptides , 32% coverage = 206/646aa ) ( S1A Fig ) . In C . elegans , ATX-2 is shown to form a cytoplasmic complex with PAB-1 [29] , and the ATX-2-PAB-1 interaction is conserved from yeast to mammals [25 , 26 , 29 , 33] . pab-1 encodes Poly-A-binding protein , a C . elegans homolog of a human PABP1 that plays a role in RNA stability and protein translation [29 , 34] . To confirm the physical interaction between SZY-20 and ATX-2 , we used anti-SZY-20 to pull down SZY-20 and its associated proteins from embryonic lysates and examined co-precipitates by western blot ( Fig 1A ) . Consistent with our mass spectrometry data , we detected ATX-2 and SZY-20 in the SZY-20-immunoprecipitates from wild-type embryonic extracts . Given that this protein complex consists of RNA-binding proteins , we asked if the physical association is mediated through RNA . To test RNA dependence , we repeated IP to pull down SZY-20 interacting proteins in the presence of RNaseA or RNase inhibitor and found that they co-precipitated in either condition ( Fig 1A , S1B Fig ) , suggesting ATX-2 and SZY-20 physically interact in an RNA-independent manner although we cannot exclude the possibility that RNA bound by the ATX-2-SZY-20 complex could have been protected from RNase treatment . The szy-20 ( bs52 ) mutation results in a truncated protein , deleting the C-terminal 197 aa residues including the SUZ-C domain , one of the putative RNA-binding domains in SZY-20 [10] . We utilized szy-20 ( bs52 ) embryos for IP analysis to determine if the C-terminal truncation of SZY-20 affected physical association of SZY-20 with ATX-2 ( S1B Fig ) . Whereas ATX-2 co-precipitated with SZY-20 in wild-type extracts , ATX-2 was undetectable in co-precipitates from szy-20 ( bs52 ) extracts , suggesting that the C-terminus of SZY-20 influences physical interaction with ATX-2 . IP assay using embryos expressing SZY-20-GFP-3xFLAG yielded a similar result: ATX-2 is undetectable in co-precipitates of SZY-20 tagged with GFP-3xFLAG at the C-terminus ( S1E Fig ) , supporting that proper folding of the C-terminal domain is critical for SZY-20 to interact with ATX-2 . Further , the C-terminal deletion in either ATX-2 or SZY-20 appears to alleviate its interaction with SZY-20 ( S1C and S1D Fig ) . By additional IP assays with anti-GFP using embryos expressing various GFP-tagged proteins , we further confirmed that both ATX-2 and PAB-1 physically interact with SZY-20 or with each other ( S1E Fig ) . Together , our data suggest SZY-20 forms a complex with known RNA-binding proteins ATX-2 and PAB-1 in vivo , via direct or indirect interaction . We next asked if these SZY-20 interacting proteins play a similar role to SZY-20 during early cell division . In C . elegans , ATX-2 is required for translational control in gonadogenesis and for normal cytokinesis during embryogenesis [29–32] . To date , PAB-1 is only known to be required for germline development [29 , 35] . Consistent with prior studies , we observed strong embryonic lethality in both atx-2 ( RNAi ) and atx-2 ( ne4297 ) but a very low embryonic lethality in pab-1 ( RNAi ) , and sterility by loss of pab-1 or atx-2 ( Fig 1B ) . Embryonic lethality by loss of atx-2 might result from defective cell division . To examine what role ATX-2 plays in cell division , we immunostained embryos for microtubules ( MTs ) , centrosomes and DNA ( Fig 1D–1H ) . Confocal microscopy of immunofluorescence ( IF ) revealed that knocking down atx-2 by RNAi results in multiple cell division defects including polar body extrusion failure ( 22%; S3 Movie ) , abnormal spindle positioning ( 3% ) , chromosome missegregation ( 10% ) and cytokinesis failure ( 36%; n = 114 ) . We observed similar cell division phenotypes , but with higher penetrance in temperature sensitive ( ts ) atx-2 ( ne4297 ) mutants . By 4D time-lapse confocal microscopy , we observed that incomplete cytokinesis following successful centrosome duplication results in tetrapolar spindles in one-cell embryo ( S1 and S2 Movies ) . In these embryos , the cytokinetic furrow initiates but cytokinesis fails to complete , resulting in a multi-nucleated cell with four centrosomes after the second mitosis . All of these cell division phenotypes resemble cell cycle defects observed previously in szy-20 ( bs52 ) embryos [10] , suggesting that ATX-2 functions closely with SZY-20 in cell division . In contrast , pab-1 ( RNAi ) produced only minor cell cycle defects such as atypical spindle positioning ( S2A Fig ) . Since ATX-2 appears to play a similar role to SZY-20 , we then asked if atx-2 genetically interacts with szy-20 . Genetic analysis by combining atx-2 ( RNAi ) or pab-1 ( RNAi ) with a hypomorphic mutant allele szy-20 ( bs52 ) suggest a positive genetic interaction among these factors ( Fig 1B , Table 1 ) . At 24°C , the semi-restrictive temperature , szy-20 ( bs52 ) animals fed with atx-2 ( RNAi ) produced higher embryonic lethality than single knockdown of either szy-20 ( bs52 ) or atx-2 ( RNAi ) . Similarly , pab-1 ( RNAi ) in szy-20 ( bs52 ) led to a synergistic increase in embryonic lethality compared to szy-20 ( bs52 ) or pab-1 ( RNAi ) alone . Cytological analyses further confirmed the functional interaction among these factors . Double-knockdown significantly enhanced the penetrance of cell division phenotypes , compared to mock-treated szy-20 ( bs52 ) or RNAi alone: 55% ( n = 128 ) of szy-20 ( bs52 ) ; atx-2 ( RNAi ) embryos exhibit a failure of polar body extrusion , compared to 22% ( n = 114 ) of atx-2 ( RNAi ) and 18% ( n = 145 ) of szy-20 ( bs52 ) . This synergistic effect is even stronger for cytokinesis: 36% of atx-2 ( RNAi ) and 6% of szy-20 ( bs52 ) , but 68% of szy-20 ( bs52 ) ; atx-2 ( RNAi ) embryos failed to complete cytokinesis at first division ( Fig 1C and 1H ) . Consistent with embryonic lethality , pab-1 ( RNAi ) ; szy-20 ( bs52 ) produced highly penetrant cytokinesis defect , in contrast to the weak phenotype by pab-1 ( RNAi ) alone ( Fig 1C ) . Combining pab-1 ( RNAi ) and szy-20 ( bs52 ) mutation produced 82% ( n = 68 ) of cytokinesis failure and 24% ( n = 68 ) of polar body extrusion failure , while pab-1 ( RNAi ) ( n = 68 ) alone showed a minor cell cycle defect ( S2A Fig ) . In addition , atx-2 ( RNAi ) or pab-1 ( RNAi ) enhances sterility in szy-20 ( bs52 ) animals , consistent with the sterility associated with all three genes . Together , our data suggest that RNA-binding proteins ATX-2 and PAB-1 function in close association with SZY-20 during embryogenesis . As szy-20 is known as a genetic suppressor of zyg-1 [10] , we asked if ATX-2/PAB-1 functions in centrosome assembly as well . To address this , we first examined centrosome duplication in zyg-1 ( it25 ) embryos fed with atx-2 or pab-1 ( RNAi ) at 24°C ( Fig 2 ) . When grown at 24°C the restrictive temperature , zyg-1 ( it25 ) embryos fail to duplicate centrosomes during the first cell cycle , producing monopolar spindles at the second cell division [9] . By recording live imaging of zyg-1 ( it25 ) embryos expressing GFP-α-Tubulin , mCherry-γ-Tubulin and mCherry-Histone , we scored centrosome duplication at the second mitosis ( Fig 2A–2C ) . Over 60% of atx-2 or pab-1 ( RNAi ) treated zyg-1 ( it25 ) embryos produced bipolar spindles , indicating successful centrosome duplication during the first cell cycle , while only 3% of mock treated zyg-1 ( it25 ) embryos formed bipolar spindles . We noticed , however , that a great majority of atx-2 ( RNAi ) treated zyg-1 ( it25 ) embryos exhibit four centrosomes in one-cell embryo after the second cell cycle , suggestive of cytokinesis failure due to loss of atx-2 ( Fig 2C , S1 and S2 Movies ) . Consistent with RNAi-mediated knockdown , we made a similar observation in zyg-1 ( it25 ) ; atx-2 ( ne4297 ) double mutants ( S3A and S3B Fig ) . Given the positive genetic interactions among szy-20 , atx-2 and pab-1 , we further asked if co-depleting these factors could enhance the suppression of zyg-1 ( Fig 2D and 2E ) . At 24°C , co-depleting either atx-2 or pab-1 with szy-20 restored nearly 100% of centrosome duplication to zyg-1 ( it25 ) embryos , while single depletion produced ~60% duplication in zyg-1 ( it25 ) . Despite the restoration in centrosome duplication at 24°C , none of these embryos hatched , owing to other cell cycle defects such as cytokinesis failure described above . However , we were able to show partial restoration of embryonic viability in zyg-1 ( it25 ) by co-depleting either atx-2 or pab-1 with szy-20 at semi-restrictive temperature 23°C , whereas single depletion of atx-2 or pab-1 showed no effect on embryonic viability of zyg-1 ( it25 ) . Our data indicate that like szy-20 , atx-2 and pab-1 act as genetic suppressors of zyg-1 . Thus , these RNA-binding proteins in a complex function together to negatively regulate centrosome assembly . Finally , we asked if any RNA-binding protein could suppress zyg-1 through global translational control . We chose the RNA-binding protein CAR-1 that is required for cytokinesis [36 , 37] to test if car-1 suppresses zyg-1 ( S3C Fig ) . We found no sign of centrosome duplication in car-1 ( RNAi ) ; zyg-1 ( it25 ) embryos , while atx-2 ( RNAi ) restored centrosome duplication to zyg-1 ( it25 ) in a parallel experiment . This result suggests that RNA-binding proteins are not general suppressors of zyg-1 , but that actions of RNA-binding proteins ( SZY-20/ATX-2/PAB-1 ) are specific to the ZYG-1 dependent centrosome assembly pathway . To understand how these RNA-binding proteins coordinate their actions within the cell , we examined the subcellular localization in early embryos . Immunostaining embryos with α-ATX-2 shows a diffuse pattern throughout the cytoplasm with occasional small foci; as expected , this signal is significantly diminished in atx-2 ( RNAi ) or atx-2 ( ne4297 ) embryos ( Fig 3A and 3B ) . We also generated a transgenic strain expressing atx-2-gfp-3x flag ( S1 Table ) and observed the dynamics of ATX-2-GFP using time-lapse movies , which shows consistent patterns to endogenous ATX-2 ( S4 Movie ) . C . elegans PAB-1 is also found in the cytoplasm and is often associated with P-bodies or stress granules [38] and SZY-20 localizes to the nuclei , centrosome and cytoplasm in early embryos [10] . Co-staining embryos show that ATX-2 partially coincides with cytoplasmic SZY-20 ( Fig 3C , S2B Fig ) or PAB-1 ( Fig 3D , S2C Fig ) . A similar co-localization is also observed for GFP-PAB-1 and SZY-20 ( S2D Fig ) . Thus , these RNA-binding proteins appear to function together in the cytoplasm to regulate early cell division . To gain insights into the hierarchical relationship among these factors , we used mutant strains to determine how the embryonic expression is affected by the other factors ( Fig 3E–3I ) . Quantitative immunofluorescence revealed that ATX-2 levels in the cytoplasm are significantly reduced ( ~50% ) in the hypomorphic szy-20 ( bs52 ) embryo relative to wild-type control ( Fig 3E and 3H ) , while the zyg-1 ( it25 ) mutation had no significant effect on the cytoplasmic ATX-2 ( Fig 3E and 3I ) . SZY-20 localization was , however , unaffected by either atx-2 or pab-1 depletion ( S2D Fig ) . Consistently , prior study showed that szy-20 acts upstream of zyg-1 [10] . Thus , it seems likely that szy-20 function upstream of atx-2 , with both acting upstream of zyg-1 . Quantitative western blots further support this hierarchical relationship ( Fig 3J–3L ) . szy-20 ( bs52 ) embryos exhibit significantly reduced ATX-2 levels ( ~40%; Fig 3K ) relative to wild-type control , while atx-2 ( ne4297 ) embryos show normal levels of SZY-20 ( ~98%; Fig 3L ) and zyg-1 mutants contain normal levels of both ATX-2 and SZY-20 ( Fig 3J–3L ) . However , no significant changes are found in the level of a centriole factor , SAS-6 in either mutant embryo ( Fig 3J ) . Taken together , our data suggest that SZY-20 acts upstream of ATX-2 , and positively regulates ATX-2 abundance in early embryos . Because atx-2 acts as a genetic suppressor of zyg-1 , we reasoned that inhibiting ATX-2 might enhance ZYG-1 activity , thereby restoring centrosome duplication and embryonic viability to zyg-1 ( it25 ) embryos . By staining embryos for ZYG-1 and microtubules ( Fig 4A ) , we quantified the fluorescence intensity of ZYG-1 at first metaphase centrosomes , finding that atx-2 mutant centrosomes possess twice as much ZYG-1 levels as those in control embryos ( p <0 . 001 , Fig 4B ) . Using the CRISPR-Cas9 method [39 , 40] , we also generated a strain expressing HA-tagged ZYG-1 at endogenous levels from the native genomic locus ( S4A–S4C Fig , S2 Table ) . By labeling endogenous ZYG-1 with anti-HA ( S4A and S4B Fig ) , we observed a similar pattern of ZYG-1 localization at centrosomes to that seen by anti-ZYG-1 ( S4E Fig ) . The levels of centrosomal HA-ZYG-1 are also increased by two-fold in atx-2 ( RNAi ) embryos ( S4C Fig ) . Given that ZYG-1 localizes to centrosomes in a cell cycle-dependent manner , the observed increase in centrosomal ZYG-1 at first metaphase in atx-2 mutants could result from a shift in the cell cycle due to loss of atx-2 . To examine cell cycle dependence of ZYG-1 localization to centrosomes , we utilized a strain expressing GFP-ZYG-1-C-term that contains a C-terminal portion ( 217–706 aa ) lacking most of the kinase domain , but including the Cryptic Polo Box ( CPB ) that is sufficient for centrosomal targeting [5 , 41] . To observe dynamics of centrosome-associated ZYG-1 over time , we acquired 4D time-lapse movies of early embryos starting from pronuclear meeting up to separation of the centriole pair at first anaphase ( Fig 4C and 4D , S5 and S6 Movies ) . Using these recordings , we first quantified the fluorescence intensity of the intensely labeled sub-centrosomal GFP signal , which presumably reflects the centriolar structure . Then , we measured pericentriolar GFP signal that likely represents PCM . Throughout the cell cycle , we observe a nearly two-fold increase in both centriolar and PCM-associated GFP signal in atx-2 ( RNAi ) embryos compared to controls . It could be that the elevated levels of centrosomal ZYG-1 in atx-2 mutants reflect a global increase in ZYG-1 throughout the cells . To test this possibility we compared overall ZYG-1 levels by measuring cytoplasmic GFP signals , but found no increase in the cytoplasmic levels of atx-2 ( RNAi ) embryos . In fact , we noticed a small decrease ( p = 0 . 2 ) in the cytoplasmic GFP signal in atx-2 ( RNAi ) embryos compared to controls , suggesting that increased centrosomal ZYG-1 levels are unlikely due to an increase in overall ZYG-1 expression . Similar results were also observed in atx-2 ( ne4297 ) mutants ( S6 Movie ) . Together , our data show that inhibiting ATX-2 results in elevated levels of ZYG-1 at centrosomes without affecting overall ZYG-1 levels . Thus , we speculate that elevated levels of centrosomal ZYG-1 in atx-2 depleted embryos might partially compensate for the reduced activity of mutant ZYG-1 ( P442L ) in zyg-1 ( it25 ) [8] , restoring centrosome duplication in zyg-1 mutants . Prior work has shown that SZY-20 negatively regulates centrosome size in a ZYG-1 dependent manner [10] . Here , our data indicate that ATX-2 , acting downstream of SZY-20 , affects centrosomal ZYG-1 levels . We thus tested if ATX-2 influences centrosome size as well . Quantitative IF using α-SPD-5 revealed that at first metaphase , atx-2 mutant centrosomes possess twice as much SPD-5 as wild-type centrosomes ( Fig 5A and 5B ) . Consistent with the results in szy-20 mutants [10] , centrosomal SPD-2 levels are also significantly increased without affecting overall levels in atx-2 mutants ( S5A and S5B Fig ) . The centrosome factors SPD-5 and SPD-2 , acting as a scaffold , are required for localization of other PCM factors including γ-Tubulin [42–44] . γ-Tubulin is a PCM factor that plays a key role in MT nucleation and anchoring [45–47] . By 4D-time lapse confocal microscopy of embryos expressing mCherry-γ-Tubulin [48] , we measured the fluorescent intensity of centrosomal mCherry signal at first metaphase ( Fig 5C and 5D ) . Consistent with SPD-5 levels , atx-2 ( RNAi ) embryos display a nearly two-fold increase in centrosomal γ-Tubulin , equivalent to the increase seen in szy-20 ( bs52 ) embryos [10] . A similar observation is made in a strain overexpressing GFP-γ-Tubulin ( S5C Fig ) . Note that szy-20 ( bs52 ) embryos possess ~50% of normal ATX-2 levels , which is comparable with the 50% average knockdown of ATX-2 by feeding atx-2 ( RNAi ) ( Fig 3E and 3K ) . Remarkably , atx-2 ( ne4297 ) mutant centrosomes show even greater increase ( 4-fold ) than szy-20 ( bs52 ) or atx-2 ( RNAi ) embryos , suggesting a dose-dependent correlation between the amount of ATX-2 and centrosomal γ-Tubulin . Compared to single knockdown by szy-20 ( bs52 ) or atx-2 ( RNAi ) , co-depletion by szy-20 ( bs52 ) ; atx-2 ( RNAi ) further exacerbated centrosomal enlargement , in a manner nearly equivalent to those in the strong loss-of-function mutation atx-2 ( ne4297 ) embryo . Thus , centrosomal γ-Tubulin levels correlate inversely with embryonic ATX-2 levels . Cytoplasmic mCherry-γ-Tubulin levels , however , are found to be similar in all embryos examined ( Fig 5D ) . Furthermore , quantitative immunoblot confirms no significant effect on endogenous γ-Tubulin levels in atx-2 ( ne4297 ) compared to wild-type embryos ( Fig 5E ) . Therefore , our data show that centrosomal PCM levels ( SPD-2 , SPD-5 , γ-Tubulin ) correlate inversely with ATX-2 abundance , suggesting that ATX-2 negatively regulates centrosome size in proportion to its abundance , while not affecting overall cellular levels of centrosome factors . The PCM factors , SPD-2 , SPD-5 and γ-Tubulin , play a critical role in positively regulating the MT nucleating capacity of the centrosome [42 , 45–47] . As atx-2 mutant embryos exhibit enlarged centrosomes with increased centrosomal SPD-5 and γ-Tubulin , we examined if atx-2 mutant centrosomes affected MT nucleating capacity . To investigate MT nucleation , we used a strain expressing EBP-2-GFP to mark the plus-ends of growing MTs [47] and acquired a series of 500 msec-interval snap shots at the center plane of first metaphase centrosomes ( Fig 6 , S7 Movie ) . First , we found that atx-2 mutant embryos exhibit a three-fold increase ( p<0 . 0001 ) in centrosomal EBP-2-GFP signal compared to wild-type controls ( Fig 6A and 6B ) . To assess the level of MT nucleation by the centrosome , we measured the fluorescent intensity of EBP-2-GFP in line regions proximal ( 25 pixels ) to the centrosome , finding that atx-2 mutant centrosomes exhibit a two-fold increase ( p<0 . 001 ) in MT nucleation ( Fig 6C ) . Consistent with increased MT nucleation , kymograph analysis at the proximal region to the centrosome revealed a drastic increase ( p = 0 . 002 ) in the number of astral MTs emanating from atx-2 mutant centrosomes over a 5-sec period , compared to controls ( Fig 6F ) . Further , line scans of the kymograph in a 355° arc around the centrosome show that mutant centrosomes emanate increased numbers of spindle fibers as well as astral MTs ( Fig 6D ) . Thus , atx-2 mutant centrosomes possessing elevated levels of PCM factors nucleate the increased number of astral and spindle MTs . Next we examined how these MTs nucleated by mutant centrosomes continue to grow out toward the cortex ( Fig 6E , 6G , 6H , S7 Movie ) . To measure the number of MTs reaching the cortex , we used kymograph analysis over a 5-sec period and counted the number of EBP-2-GFP dots crossing an arc drawn 1 . 5 μm ( 10 pixels ) inside the cortex . In mutant embryos , fewer MTs grew out to reach the cortex than in controls ( p = 0 . 001; Fig 6H ) . Consistently , only a small portion of MTs nucleated by mutant centrosomes grew beyond the midpoint between the centrosome and cortex compared to controls ( p = 0 . 029; Fig 6G ) . In fact , astral MT growth rates are significantly slower in mutant embryos ( 0 . 70 μm/sec ± 0 . 007 ) than in controls ( 0 . 88 μm/sec ± 0 . 013 , p = 0 . 0014 ) ( S5E Fig ) . Thus , MT growth appears to be impeded in the mutant embryo , presumably due to an excess of MT nucleation . In C . elegans embryos , it has been shown that fast MT growth is subject to the MT stabilizing complex ( ZYG-9/TAC-1 ) and the amount of free tubulin [47] . atx-2 mutant embryos , however , exhibit a significant increase in both centrosomal and overall TAC-1 levels ( S5F and S5G Fig ) , indicating that defective MT growth in atx-2 mutants is unlikely due to insufficient MT stabilization . Studies in other systems also showed that the amount of free tubulin influences the rate of MT polymerization in vitro [49 , 50] and such mechanism appears to be pertinent in C . elegans embryos [10 , 47 , 51] . We thus hypothesized that atx-2 mutant centrosomes possessing increased PCM nucleate more MTs , reducing the supply of free tubulin available for MT polymerization . Subsequently MT growth is interfered , resulting in shorter than normal MTs and cytokinesis failure . To test this , we generated atx-2 mutants overexpressing GFP-Tubulin to see if increasing Tubulin levels in the atx-2 mutant could partially rescue cytokinesis defects . As predicted , the incidence of cytokinesis failure is significantly ( p<0 . 001 ) reduced in GFP-Tubulin overexpressing atx-2 embryos compared to control mutant embryos ( Fig 7A ) . This partial rescue of cytokinesis failure further led to a significant decrease ( p<0 . 001 ) in embryonic lethality ( Fig 7B ) , suggesting that MT growth in atx-2 embryos is partly affected by limited supply of free tubulins , likely resulting from excessive MT nucleation . Next , we examined atx-2 mutant embryos overexpressing GFP-γ-Tubulin , a PCM factor , to see if further enhancing MT nucleation by increasing γ-Tubulin levels could exacerbate cytokinesis defects through worsened MT growth . Consistent with our hypothesis , cytokinesis defects in mutant embryos were significantly increased by GFP-γ-Tubulin overexpression compared to control mutants , leading to an increase in embryonic lethality ( Fig 7A and 7B ) . Together , our data suggest that excessive MT nucleation and subsequent MT growth defects result in cytokinesis defects in atx-2 mutants . If defective cytokinesis results from elevated PCM levels at mutant centrosomes through excessive MT nucleation , reducing PCM levels should reconcile proper centrosomal PCM levels and MT nucleation , thereby restoring proper cytokinesis in atx-2 mutants . Whereas tbg-1 ( RNAi ) produces cytokinesis failure in wild-type embryos , reducing γ-Tubulin in atx-2 embryos partially restores normal cytokinesis ( p<0 . 001 , Fig 7C ) , and embryonic viability ( p<0 . 01 , Fig 7D ) to mutant embryos . Interestingly , not only did tbg-1 ( RNAi ) rescue cytokinesis defects in atx-2 mutants , but atx-2 mutation also reduced cytokinesis failure in tbg-1 ( RNAi ) embryos , suggesting a mutual suppression between atx-2 and tbg-1 ( Fig 7C and 7D ) : At 20°C , the semi-permissive temperature for atx-2 mutants , single knockdown of atx-2 ( ne4297 ) and tbg-1 ( RNAi ) produces 24% ( n = 91 ) and 66 . 5% ( n = 73 ) , respectively , compared to 10 . 8% ( n = 57 ) of cytokinesis failure by double knockdown in atx-2 ( ne4297 ) ; tbg-1 ( RNAi ) embryos . Consistently , atx-2 ( ne4297 ) ; tbg-1 ( RNAi ) embryos exhibited centrosome-associated γ-Tubulin nearly to the normal level , although tbg-1 ( RNAi ) almost abolished centrosomal γ-Tubulin in wild-type embryos ( Fig 7E ) . It appears that ATX-2 influences MT nucleation and MT dynamics through γ-Tubulin at the centrosome , suggesting an epistatic relationship between atx-2 and tbg-1 . A prior study reported a mutual suppression between szy-20 and zyg-1 [10] . Given the genetic and functional relationship between atx-2 and szy-20 , we speculated that loss of zyg-1 might have a similar effect on atx-2 . At semi-restrictive conditions for the atx-2 mutant , double homozygous mutants zyg-1 ( it25 ) ; atx-2 ( ne4297 ) produce a small but significant increase ( p = 0 . 02 ) in embryonic viability compared to the atx-2 ( ne4297 ) single mutant , accompanied by the restoration of normal cytokinesis ( Table 1 ) . Restoring normal cytokinesis appears to correlate with zyg-1 ( it25 ) -mediated restoration of centrosome size in atx-2 embryos ( Fig 7F ) . In fact , zyg-1 ( it25 ) embryos exhibit decreased levels of centrosomal γ-Tubulin at the permissive temperature 20°C where centrosome duplication occurs normally . Moreover , inhibiting ZYG-1 reduces centrosomal γ-Tubulin levels in atx-2 ( RNAi ) or mutant embryos , which likely leads to the partial restoration of normal cytokinesis and embryonic viability in ATX-2 depleted embryos ( Fig 7F and 7G , Table 1 ) . Together , our results suggest a model in which loss of ATX-2 leads to elevated PCM levels at centrosomes and excessive MT nucleation , resulting in MT growth defects and subsequent cytokinesis failure ( Fig 7H ) . Consistently , embryos depleted of ATX-2 exhibit a prominent cytokinesis failure phenotype , accompanied by enlarged centrosomes that cause excessive MT nucleation and aberrant MT growth , which results in cytokinesis failure and embryonic lethality . Thus , the proper level of centrosomal PCM factors is critical for normal MT nucleating activity to support normal MT dynamics . In this pathway , ATX-2 acts upstream to establish the proper centrosome size and MT nucleating activity .
In this study , we identified that ATX-2 , together with PAB-1 , physically associates with SZY-20 in vivo . While RNA-binding proteins ATX-2 and SZY-20 , each containing unique RNA-binding motifs ( LSm , SUZ and SUZ-C , respectively ) , form a complex , these two proteins appear to physically interact independently of RNA . While it remains unknown what RNA molecules associate with these RNA-binding proteins , each RNA-binding protein in a ribonucleoprotein ( RNP ) complex might recruit a specific group of RNA through each own RNA-binding motif . In fact , Ataxin-2 has been shown to bind directly to mRNAs through its LSm domain and promote the stability of transcripts independently of its binding partner , poly ( A ) -binding protein ( PABP ) in flies and humans [25 , 26] . Our prior study in C . elegans embryos showed that SUZ and SUZ-C RNA-binding motifs in SZY-20 exhibit RNA-binding capacity in vitro and that mutating these domains perturbs in vitro RNA-binding of SZY-20 and its capacity to regulate centrosome size in vivo [10] . However , there is no evidence that RNA-binding role of ATX-2 is directly involved in centrosome regulation , while C . elegans ATX-2 is shown to function in translational regulation during germline development [29] . In a multi-protein complex , ATX-2 and SZY-20 function closely to regulate cell division and centrosome assembly . While knocking down ATX-2 phenocopies a loss of function szy-20 ( bs52 ) mutation , the different degree of phenotypic penetrance in atx-2 ( RNAi ) , a strong loss of function atx-2 ( ne4297 ) mutation , and a hypomorphic szy-20 ( bs52 ) mutation suggests a dose-dependent regulation of ATX-2 . Double knockdown by combining atx-2 ( RNAi ) and szy-20 ( bs52 ) mutation further enhances embryonic lethality , cytokinesis failure , the restoration of centrosome duplication to zyg-1 ( it25 ) embryos and the levels of centrosome-associated factors ( ZYG-1 , SPD-5 , γ-Tubulin ) . Furthermore , the atx-2 ( ne4297 ) mutation produces a similar effect to that of atx-2 ( RNAi ) combined with the szy-20 ( bs52 ) mutation . Our data thus suggest that atx-2 exhibits a positive genetic interaction with szy-20 in regulating cell cycle and centrosome assembly . In this pathway , SZY-20 acts upstream of ATX-2 to promote ATX-2 levels , with both acting upstream of zyg-1 . We propose that SZY-20 influences centrosome size and MT dynamics indirectly through ATX-2 [10] . While it remains unclear whether C . elegans ATX-2 directly acts on ZYG-1 , it has been shown that Plx4 ( a Xenopus homolog of Plk4/ZYG-1 ) forms a complex with Atxn-2 ( Xenopus homolog of ATX-2 ) [52] . Although a direct role for Atxn-2 in centrosome assembly has not been demonstrated , a physical connection between Xenopus Atxn-2 and Plx4 suggests a possible role of Xenopus Atxn-2 in centrosome assembly , via a mechanism that seems likely to be conserved between nematodes and vertebrates . Our data indicate that ATX-2 acts as a negative regulator of centrosome size . A priori , increased PCM levels at atx-2 mutant centrosomes could be achieved by several mechanisms . First , centrosome factors might be overexpressed in atx-2 mutant cytoplasm via translational control , leading to increased recruitment of these factors to the centrosome by equilibrium , as shown by Decker et al . , [53] . Second , atx-2 mutants might enhance the recruitment of factors to centrosomes post-translationally , without affecting overall levels of these factors . Third , loss of ATX-2 might promote local translation near centrosomes , leading to locally enriched centrosome factors . The first scenario is unlikely because our quantitative analyses reveal no significant changes in overall levels of centrosome factors ( SPD-2 , SAS-6 , γ-Tubulin ) or cytoplasmic levels of ZYG-1 by loss of ATX-2 . Our current data do not differentiate between the second and third scenarios , but certain observations in C . elegans and other systems are consistent with the latter . It has been shown that Ataxin-2 assembles with polyribosomes and that ribosomes are associated with the MT cytoskeleton [25 , 54] . Furthermore , neuronal RNA granules are shown to contain translational machinery , allowing local translation upon arrival of transcripts at the right location [55] . We also observed ribosomal protein S6 spatially associated with C . elegans centrosomes ( S3D Fig ) , suggesting the possibility of local translation around the centrosome . Indeed mRNA localization linked to the cytoskeleton and the following local translation is shown to be an efficient means of concentrating proteins at the functional site [19 , 56 , 57] . For example , spatial control of β-actin translation is executed by localizing its transcripts to actin-rich protrusions , which facilitates neuronal outgrowth [56] . Also , RNA-binding protein ( CPEB/maskin ) mediated localization of cyclin B1 transcripts to the mitotic apparatus leads to locally enriched Cyclin B proteins in the vicinity of spindles and centrosomes , supporting cell cycle progression [19] . Together such mechanism has been demonstrated to provide a tight control for temporal and spatial translation . How then , might ATX-2 regulate translation ? For negative translational control , it has been proposed that Ataxin-2 binding to PABP inhibits translation by blocking the interaction between PABP and translational machinery or by directly blocking translation of mRNA targets at the initiation stage [58 , 59] . Alternatively , ATX-2 may facilitate the interaction between a microRNA and its mRNA target , leading to translational inhibition [60] . ATX-2 might also mediate translational control via its RNA-binding role as a component of the RNP complex comprising SZY-20/ATX-2/PAB-1 . In fact , C . elegans PAB-1 is shown to associate with stress granules and processing bodies ( P-bodies ) that have been implicated in translational repression [38] . It has been observed that RNP complexes including P-bodies are involved in subcellular targeting of RNAs and precise timing of local translation at the final subcellular destination [58 , 61] . Further identification of specific RNAs that bind ATX-2 and/or SZY-20 will help to understand how RNA-binding role of an ATX-2 associated RNP complex plays a role in defining proper centrosome size . We have shown here that ATX-2 plays an essential role in cell division . atx-2 mutant embryos with enlarged centrosomes exhibit multiple cell division defects including cytokinesis failure and aberrant spindle positioning . In animal cells , contact between astral MT and the cortex is critical for the initiation of the cleavage furrow that is required for proper cytokinesis [62 , 63] . Thus cell cycle defects in atx-2 ( ne4297 ) embryos might associate with aberrant MT behavior , likely due to enlarged centrosomes . Recent work showed an additional role of ZYG-1 in regulating centrosome size , independently of its role in centriole duplication [10] . PCM factors SPD-2 , SPD-5 and γ-Tubulin are known to positively regulate the MT nucleating activity of the centrosome [42 , 45 , 46 , 64] . atx-2 mutant centrosomes possessing increased ( 2–5 folds ) levels of centrosome factors nucleate the drastically increased number of MTs . Such increase in MT nucleation by mutant centrosomes may lead to a substantial reduction in cytoplasmic free tubulins available for timely MT polymerization . Given the rapid cell divisions ( ~20 min/cell cycle ) and high MT growth rate ( 0 . 88 μm/sec ) in early C . elegans embryos , timely and sufficient supply of free tubulins in the cytoplasm must be immensely critical for proper cell cycle progression . In support of this , either overexpressing Tubulin or reducing positive regulators ( ZYG-1 or γ-Tubulin ) of MT nucleation partially restores normal cytokinesis , MT growth and embryonic viability to atx-2 mutants . Partial restoration of normal MT- dependent processes correlates with proper levels of centrosomal γ-Tubulin . Therefore , enlarged centrosomes are likely to be a primary cause of embryonic lethality in atx-2 mutants . Mutant centrosomes possess significantly increased levels of γ-Tubulin , and reducing its levels at mutant centrosomes partially reinstalls free tubulins available for MT growth , which in turn restores normal cell divisions and embryonic viability in atx-2 mutants . Our data suggest that ATX-2 contributes to MT nucleation and MT dynamics , in part , through γ-Tubulin at centrosomes , although it remains unknown how RNA-binding protein , ATX-2 , regulates centrosomal levels of γ-Tubulin . Given our observation that overall levels of γ-Tubulin are not altered in atx-2 mutant embryos , it is curious how γ-Tubulin levels are locally enriched at the centrosome , perhaps via the regulation of recruitment to the centrosome or locally enriched translational control . Thus , it will be interesting to see if tbg-1 transcripts are elevated at the proximity of centrosomes by in situ hybridization . A recent study in Xenopus egg extracts reported that RNase treatment leads to defective spindle assembly due to hyperactive MT destabilizer MCAK ( the ortholog of C . elegans KLP-7 ) by depleting RNA , suggesting a direct involvement of RNA in regulating MT organization [65] . Interestingly , we also found that atx-2 mutants show increased levels of KLP-7 at centrosomes and cytoplasm ( S5H Fig ) , and that atx-2 and klp-7 exhibit a synergistic effect on MT dynamics and embryonic lethality ( S6 Fig ) . Another centriole factor , SAS-4 , positively regulates centrosome size in C . elegans [66] . Interestingly , while we identify ATX-2 as a negative regulator of centrosome size , depleting ATX-2 had no effect on centriolar SAS-4 levels ( S4G Fig ) , suggesting that ATX-2 regulates centrosome size independently of SAS-4 , perhaps through a separate pathway . Thus , a balance between positive ( e . g . , SAS-4 ) and negative ( e . g . , ATX-2 ) regulators may contribute to establish proper centrosome size and MT nucleating activity , which in turn influences MT dynamics during the cell cycle . Improper levels of ATX-2 disrupt this balance , resulting in deregulated MT dynamics and subsequently abnormal cell divisions . In summary , our work uncovers a role for ATX-2 , the C . elegans ortholog of human Ataxin-2 in regulating centrosome size and MT cytoskeleton . In this pathway , SZY-20 positively regulates levels of ATX-2 , which contributes to defining the proper centrosome size , leading to proper levels of MT nucleation and subsequent MT growth . While human Ataxin-2 and its poly Q stretch have been implicated in spinocerebellar ataxia [20] , it remains largely unknown how this RNA-binding protein is linked to human diseases . Many RNA-binding proteins in neuronal RNA granules have been implicated in neuronal functions and growth via controlled local translation [55 , 58] , suggesting a link between RNA-binding role and neuronal activity . In this regard , our work provides insights into a mechanistic link between the RNA-binding role of Ataxin-2 and the MT cytoskeleton .
All C . elegans strains were grown on MYOB plates seeded with E . coli OP50 , and were derived from the wild-type Bristol N2 strain [67] . All strains were maintained at 18 or 20°C unless otherwise indicated . Transgenic strains were generated by standard particle bombardment transformation [68] and the CRISPR/Cas-9 method ( S2 Table ) [39 , 40] . The list of worm strains used in this study is shown in S1 Table . Standard genetics were used for strain construction and genetic manipulations [69] . RNAi feeding was performed as described and L4440 clone with the empty dsRNA-feeding vector served as a negative control [70] . For immunostaining , we used the following antibodies at a 1:400–3000 dilution: α-ATX-2 [29] , DM1A ( Sigma ) , α-GFP ( Roche ) , α-SAS-4 [10] , α-SPD-2 [43] , α-SPD-5 [42] , α-SZY-20 ( α-S20N ) [10] , and secondary antibodies Alexa Fluor 488 and 568 ( Invitrogen ) . Affinity-purified rabbit polyclonal antibodies were generated ( YenZym ) against the following peptides: for γ-Tubulin ( S4D Fig , aa31-44 ) : Ac-HGINERGQTTHEDD-amide , for ZYG-1 ( S4E Fig , aa671-689 ) : Ac-PAIRPDDQRFMRTDRVPDR-amide . Immunofluorescence and confocal microscopy were performed as described in [10] . For confocal microscopy , MetaMorph software ( Molecular Devices ) was used to acquire images from a Nikon Eclipse Ti-U microscope equipped with a Plan Apo 60 X 1 . 4 NA lens , a Spinning Disk Confocal ( CSU X1 ) and a Photometrics Evolve 512 camera . Fluorescence intensity measurements were made using the MetaMorph software , and image processing with Photoshop CS6 . For quantification of centrosomal signals , the average intensity within a 25-pixel ( pixel = . 151 μm ) diameter region was measured throughout a region centered on the centrosome . For each region , a focal plane with the highest average intensity was recorded . The same sized regions were drawn in the cytoplasm for cytoplasmic signal , and outside the embryo for background subtraction unless otherwise indicated . For centriolar signals , analysis was done the same way , but a 5-pixel diameter region was used . For EBP-2-GFP movies , pre-mitotic embryos were selected and monitored under DIC until before metaphase I . Upon entrance into metaphase , 61 images were captured every 500 msec for a 30 sec period . For spindle MTs , a 355°circular region was drawn around each centrosome , and kymographs were generated for the first 5 seconds of recordings . Linescan measurements of the kymographs were taken to obtain average EBP-2 signal along every point of the circular region . Values for each linescan were averaged and plotted on a graph . To measure MT nucleation , the fluorescence intensity was quantified at a line region drawn in a semi-circle ( 25 pixel radius ) around the centrosome . Linescan ( 5 pixel wide ) measurements were taken in 5-sec time projections from recordings . Pixel intensities were generated for each point along the line and averaged for comparison . For background subtraction , cytoplasmic intensity in the same size region was used . To quantify the number of growing astral microtubules proximal to the centrosome , line regions were drawn in a semi-circle ( 25 pixel radius ) around the centrosome . For midpoint measurements , line regions were drawn at a radius of 60 pixels from the centrosome . Kymographs were generated for the first 5 seconds of each movie . Individual pixel measurements from the kymographs were obtained , and pixels with GFP intensity over 40 were counted . Cortex measurements were obtained by generating 5 sec kymographs of a region drawn at 1 . 5 μm inside of the cell cortex . The number of EBP-2 signal was counted manually . For MT polymerization rate , 4 . 5 μm-long lines were drawn along the individual EBP-2-GFP tracks outward to the cortex . Kymographs were generated , and rate was calculated by distance over time ( μm/sec ) . 20–25 μl of Dynabeads Sheep-anti-Rabbit , Protein A magnetic beads ( Invitrogen ) or mouse-anti- GFP magnetic beads ( MBL ) were used per IP reaction . Beads were washed 2x 15 min in 1 ml PBS with 0 . 1% Triton-X ( PBST ) . For SZY-20 IP , beads were incubated overnight at 4°C in a ratio of 1mg/ml ( α-SZY-20 or α-HA/beads volume ) and 3 mg/ml ( α-IgG/beads volume ) . Embryos were collected by bleaching worms grown in liquid culture , snap-frozen in liquid nitrogen , and stored at -80°C until use . Embryo pellets were ground up in microcentrifuge tubes in equal volumes of worm lysis buffer ( 50 mM HEPES; pH 7 . 4 , 1 mM EDTA , 1 mM MgCl2 , 200 mM KCl , 10% glycerol ) [71] containing complete protease inhibitor cocktail ( Roche ) and MG132 ( Tocris ) . For RNase A or RNase Inhibitor treatment , RNase ( 10 μg/ml , Roche ) or RNase Inhibitor ( 200U/ml , Roche ) was added to lysis buffer prior to grinding . Embryos were ground for 5 min , and sonicated for 3 min . The lysate was then centrifuged at 4°C for 2x 20 min at 15K rpm in a desktop centrifuge , and the supernatant were collected . For enzyme reaction , lysates were incubated for 15 min at room temperature ( RT ) . Protein concentration was determined and adjusted before IP . Beads were washed for 2x 15 min with PBST + 0 . 5% BSA at RT , followed by 1x 15 min wash with 1X worm lysis buffer . Beads were resuspended in 1X lysis buffer , mixed with embryonic lysates and incubated at 4°C for 1 hour . Following IP , beads were washed 2x 5 min in PBST . Samples were dissolved in 2X Laemmli Sample Buffer ( Sigma ) with 10% β-mercaptoethanol , fractionated on a 4–12% NuPAGE Bis-Tris Gel ( Invitrogen ) and blotted to nitrocellulose membrane using the iBlot Gel Transfer System ( Invitrogen ) . α-ATX-2 [29] , α-GFP ( Roche ) , mouse α-HA ( Sigma ) , α-SAS-6 [7] , α-SPD-2 [10] , α-SZY-20 [10] , DM1A ( Sigma ) , α-TAC-1 [72] and α-γ-Tubulin at 1:400–2500 dilution , and IRDye secondary antibodies ( LI-COR Biosciences ) were used at 1:10 , 000 dilution . Blots were imaged on a Licor Odyssey IR scanner , and analyzed using the Odyssey Infrared Imaging System ( LI-COR Biosciences ) . Mass spectrometry analysis was performed as described [7] . | The microtubule ( MT ) cytoskeleton undergoes dynamic rearrangements during the cell cycle . As the primary microtubule-organizing center , centrosomes orchestrate MT dynamics and play a key role in establishing bipolar spindles in mitosis . Errors in centrosome assembly lead to missegregation of genomic content and aneuploidy . Thus , stringent regulation of centrosome assembly is of vital importance for the fidelity of cell division and survival . Using the nematode Caenorhabditis elegans ( C . elegans ) as a model , we study the role of the RNA-binding protein , ATX-2 , a C . elegans homolog of Human Ataxin-2 in early cell division . A number of RNAs and RNA-binding proteins are shown to be associated with centrosomes and MTs , and influence the assembly of mitotic spindles . In C . elegans , the RNA-binding role of SZY-20 is implicated in regulating centrosome size . We show that ATX-2 functions together with SZY-20 in centrosome size and MT behavior . SZY-20 promotes ATX-2 protein levels , and the amount of ATX-2 influences centrosome size and subsequent MT dynamics . Our work provides evidence that RNA-binding proteins have an active role in controlling MT dynamics . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"invertebrates",
"rna-binding",
"proteins",
"centrosomes",
"caenorhabditis",
"metaphase",
"cell",
"cycle",
"and",
"cell",
"division",
"cell",
"processes",
"condensed",
"matter",
"physics",
"animals",
"animal",
"models",
"developmental",
"biology",
"caenorhabditis",
"elegans",
"model",
"organisms",
"cytokinesis",
"embryos",
"cellular",
"structures",
"and",
"organelles",
"research",
"and",
"analysis",
"methods",
"embryology",
"proteins",
"physics",
"biochemistry",
"cell",
"biology",
"nematoda",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"nucleation",
"organisms"
] | 2016 | ATX-2, the C. elegans Ortholog of Human Ataxin-2, Regulates Centrosome Size and Microtubule Dynamics |
Mucor circinelloides is a zygomycete fungus and an emerging opportunistic pathogen in immunocompromised patients , especially transplant recipients and in some cases otherwise healthy individuals . We have discovered a novel example of size dimorphism linked to virulence . M . circinelloides is a heterothallic fungus: ( + ) sex allele encodes SexP and ( − ) sex allele SexM , both of which are HMG domain protein sex determinants . M . circinelloides f . lusitanicus ( Mcl ) ( − ) mating type isolates produce larger asexual sporangiospores that are more virulent in the wax moth host compared to ( + ) isolates that produce smaller less virulent sporangiospores . The larger sporangiospores germinate inside and lyse macrophages , whereas the smaller sporangiospores do not . sexMΔ mutants are sterile and still produce larger virulent sporangiospores , suggesting that either the sex locus is not involved in virulence/spore size or the sexP allele plays an inhibitory role . Phylogenetic analysis supports that at least three extant subspecies populate the M . circinelloides complex in nature: Mcl , M . circinelloides f . griseocyanus , and M . circinelloides f . circinelloides ( Mcc ) . Mcc was found to be more prevalent among clinical Mucor isolates , and more virulent than Mcl in a diabetic murine model in contrast to the wax moth host . The M . circinelloides sex locus encodes an HMG domain protein ( SexP for plus and SexM for minus mating types ) flanked by genes encoding triose phosphate transporter ( TPT ) and RNA helicase homologs . The borders of the sex locus between the three subspecies differ: the Mcg sex locus includes the promoters of both the TPT and the RNA helicase genes , whereas the Mcl and Mcc sex locus includes only the TPT gene promoter . Mating between subspecies was restricted compared to mating within subspecies . These findings demonstrate that spore size dimorphism is linked to virulence of M . circinelloides species and that plasticity of the sex locus and adaptations in pathogenicity have occurred during speciation of the M . circinelloides complex .
Zygomycetes and chytridiomycetes are basal lineages of the fungal kingdom and both are paraphyletic and encompass several phyla [1] , [2] . Within the Zygomycota , the order Mucorales is a monophyletic group that has been relatively well studied compared to other basal fungal groups . However , molecular data and our knowledge of sex and virulence in this fungal lineage is still limited . M . circinelloides belongs to the order Mucorales and is a dimorphic fungus that grows as a budding yeast anaerobically and as a filamentous fungus aerobically [3] , [4] . M . circinelloides is a causal agent for the rare but lethal fungal infection mucormycosis ( also known as zygomycosis ) . Mucormycosis is an emerging infectious disease [5] , [6] and is recognized as a prevalent fungal infection in patients with impaired immunity [7] . Recent data indicate a significant increase in mucormycosis due to an increasing population of immunocompromised patients with , for example , diabetes or AIDS , hematologic malignancies , hematopoietic stem cell/solid organ transplantation , or trauma [7]–[11] . High serum iron levels are also a risk factor that increases susceptibility to mucormycosis [7] , [8] , [12] , and the high affinity iron permease , Ftr1 , is known to be a virulence factor in the zygomycete Rhizopus oryzae in a murine host model [13] . Recently the host endothelial cell receptor GRP78 was shown to be overexpressed during R . oryzae infection in human umbilical vein endothelial cells , resulting in increased susceptibility to mucormycosis [14] . A major concern with mucormycosis is the high mortality rate , which is ∼50% in general and >90% in disseminated infections [8]–[10] , [12] , [15] . Causal agents are zygomycetes including Rhizopus spp . , Mucor spp . , Rhizomucor , Absidia spp . , Cunninghamella spp . , among others [6] , [16] . A recent study found that in 50 cases in solid organ transplant recipients Mucor spp . caused 37% of mucormycosis cases followed by Rhizopus spp ( 35% ) and Mycocladus ( 13% ) [17] . In a European survey of 230 mucormycosis cases , Rhizopus spp . caused 24% of the cases , followed by Mucor spp . ( 22% ) and Absidia spp . ( 14% ) [18] . The predicted economic burden in the US health care system caused by mucormycosis is ∼$100 , 000 per case [19] . However , surprisingly little is known about the genetics of pathogenesis for zygomycetes compared to other fungal pathogens [20] . M . circinelloides is a heterothallic [ ( + ) and ( − ) strains] zygomycete and propagates through both asexual and sexual life cycles . In the asexual life cycle , spores germinate and undergo hyphal growth , and complex mycelia are formed , from which aerial hyphae form culminating at their apices in sporangia harboring multinucleate asexual spores ( sporangiospores ) . In the sexual life cycle , hyphae from the two different mating types recognize each other and then fuse to form zygospores in which meiosis occurs . The zygospores later send a hypha to produce a sporangium containing meiospores at the apex . Sexual development is mediated by a zygomycete specific pheromone , trisporic acid . Minus ( − ) and plus ( + ) mating types secrete and exchange trisporic acid precursors that are converted in the opposite mating type to mature trisporic acid [21] . Trisporic acid triggers the formation of specialized hyphae ( zygophores ) supporting the zygospores followed by hyphal fusion of the opposite mating types to form a zygote . Meiosis then occurs . The sex locus of zygomycetes , including M . circinelloides , Phycomyces blakesleeanus , and R . oryzae , governs and orchestrates sexual reproduction and consists of a high mobility group ( HMG ) transcription factor gene flanked by genes encoding a triose phosphate transporter homolog ( TPT ) and an RNA helicase [22]–[24] . The HMG domain proteins are designated SexP for the ( + ) and SexM for the ( − ) mating types . The sequences of the genes encoding SexP and SexM are divergent but allelic in the ( + ) and ( − ) mating types , in contrast to the idiomorphic nature of MAT in many ascomycetes and basidiomycetes encoding entirely divergent proteins [25] . The evolutionary trajectory of sex in fungi is an intriguing subject , and provides a forum to elucidate the basis of sexual development and the evolution of sex [26] . For example , complete genome sequences of several pathogenic and non-pathogenic Candida species revealed a dramatic divergence of MAT loci and sexuality in the Candida clade [26]–[28] . The studies reveal that sexual development and its specification are differentially adapted in each species . Additionally , Cryptococcus species were also found to be divergent in MAT locus structure and sexuality [26] , [29] , [30] . In contrast to the bipolar species C . neoformans and C . gattii , Cryptococcus heveanensis retains a sexual cycle involving a tetrapolar system with unfused gene clusters , one containing the homeodomain genes and the other pheromone/pheromone receptor genes [29] . Within the Cryptococcus lineage , C . heveanensis represents an evolutionary intermediate in the trajectory from a tetrapolar to a bipolar mating system . The M . circinelloides complex has been characterized based on physiological characteristics and includes M . circinelloides f . lusitanicus ( Mcl ) , M . circinelloides f . griseocyanus ( Mcg ) , and M . circinelloides f . circinelloides ( Mcc ) [31] . In this study , we examined the mating , virulence , and sex locus of the three subspecies of M . circinelloides . Multi-locus sequence typing ( MLST ) was applied to construct a phylogeny of the M . circinelloides subspecies complex . We also tested the virulence of each subspecies in larvae of the wax moth , Galleria mellonella , a heterologous host model; a significant difference in virulence in the M . circinelloides subspecies was observed . Spore size was found to be correlated with virulence in that larger spores of ( − ) mating type were found to be more virulent than smaller spores of ( + ) mating type . We disrupted the sexM gene in the ( − ) mating type , and found that sexMΔ mutants are sterile in genetic crosses , functionally verifying a key role in sex determination and sexual development for the first time in this basal fungal phylum . The virulence analysis was extended to a diabetic murine host model including analysis of clinical M . circinelloides isolates , revealing the Mcc subspecies is highly virulent in mice which is well correlated with its more frequent occurrence in human clinical isolates . A comparison of the sex loci between the subspecies is presented here and the evolutionary trajectory of the sex locus in the M . circinelloides complex is posited . Expansion of the sex locus in one subspecies of M . circinelloides into the RNA helicase promoter region contrasts with the sex locus of two related zygomycetes , P . blakesleeanus and R . oryzae , and reveals the evolutionary plasticity of this dynamic region of the genome involving either expansion or contraction .
M . circinelloides is recognized as an agent responsible for mucormycosis , a rare fungal infection associated with a high mortality rate . The Mucor circinelloides complex is known to consist of three extant subspecies: M . circinelloides f . lusitanicus ( Mcl ) , M . circinelloides f . circinelloides ( Mcc ) , and M . circinelloides f . griseocyanus ( Mcg ) [31] . Among them , the genome of one of Mcl isolate , CBS277 . 49 ( Table 1 ) , has been sequenced ( US Department of Energy Joint Genome Institute M . circinelloides genome project ) . Asexual sporangiospores are involved in dissemination , whereas sexual zygospores are considered to be dormant . Therefore , sporangiospores were tested in this study , in which spores indicate sporangiospores unless otherwise stated . We observed that in Mcl spore size and shape differ between ( − ) and ( + ) mating type isolates of M . circinelloides ( Figure 1A ) . Minus ( − ) strains produce larger , irregularly shaped spores that are on average 12 . 3±2 . 7 µm , while ( + ) isolates are smaller in spore size ( 4 . 7±0 . 9 µm ) and more homogenous in shape ( Table 2 ) . We further analyzed all currently available Mcl isolates ( 14 total ) and found that asexual spores of ( + ) mating type isolates are homogenously smaller compared to those of ( − ) mating type isolates , in which three ( − ) isolates produce larger spores and the other 7 ( − ) mating type isolates produce spores of an intermediate size ( Figure 1 and Figure S1 ) . When visualized inside of the sporangia , the spores already exhibit a difference in size and shape ( Figure S2 ) . The nuclei of ( − ) and ( + ) spores were stained with DAPI and observed by confocal microscopy . Larger spores contain more nuclei than smaller spores , where larger ( − ) spores contain multiple nuclei ( from 1 to 16 ) whereas smaller ( + ) spores are uniformly uninucleate ( Figure 1B ) . The ( − ) spore population is also mixed with a minority of smaller spores with sizes comparable to those of the ( + ) isolates . Interestingly , these smaller ( − ) spores contain fewer nuclei ( 1 or 2 ) compared to larger ( − ) spores ( Figure 1B ) . When we tested the virulence of Mcl strains in a heterologous host , Galleria mellonella , which has been used as a host for several human fungal pathogens and pathogenic bacteria ( [32] , [33] and references therein ) , a correlation between larger spore size and enhanced virulence was apparent ( Figure 1C and Figure S1 ) . Five hundred sporangiospores of each strain were suspended in PBS and injected through the pseudopod of the wax moth larvae . We monitored the viability of infected larvae at one day intervals . Interestingly , strains with larger spores were more virulent than ones with smaller spores; for example R7B ( − ) was significantly more virulent than NRRL3631 ( + ) ( p<0 . 0001 ) ; however , NRRL3631 ( + ) was not significantly virulent compared to the PBS control ( p = 0 . 3173 ) . Intermediate sized spores of NRRL1443 ( − ) showed no significant virulence in comparison with PBS or ( + ) strains [p = 0 . 3173 for NRRL1443 ( − ) vs PBS , p = 1 . 0000 for NRRL1443 ( − ) vs NRRL3631 ( + ) ] . When nine additional Mcl isolates were tested for virulence in the wax moth host , the larger spore producing isolates were all more virulent , further substantiating the conclusion that larger spore isolates are more virulent than smaller spore producing isolates ( Figure S1 ) . These results provide evidence that spore size dimorphism is an important virulence factor in the invertebrate model . Based on these findings , we propose two possible hypotheses: 1 ) fungal biomass could affect virulence , wherein the ( − ) spores challenge the host with more fungal material compared to the ( + ) spores , or 2 ) the host may respond differently to larger spores . To test these hypotheses , we examined pathogenesis in the wax moth model with increased numbers of ( + ) spores ( Figure 1D ) . In this experiment , 50 , 000 ( + ) spores displayed similar virulence to 500 larger ( − ) spores ( p = 0 . 8826 ) . Furthermore significantly more ( + ) spores ( 10 , 000 ) were less virulent than 500 ( − ) spores ( p = 0 . 0421 ) . With the assumption that the spores are spherical , the larger spores are ∼18 times greater in volume than the smaller spores . Our virulence test indicates that a 20-fold numerical excess of smaller spores is less virulent than the larger spores ( despite the roughly equivalent biomass of the two inocula ) , and a 100-fold numerical excess of smaller spores is required for equivalent virulence to the larger spores , suggesting that the first hypothesis about the possible effect of biomass is not sufficient to explain the marked difference in virulence of ( − ) vs . ( + ) spores . We observed and analyzed the germination of large and small spores ( Figure 2A and B , Videos S1 and S2 ) . Interestingly , the larger spores display a shorter isotropic growth phase or bypass the isotropic growth stage resulting in a rapid and immediate germ tube emergence after exiting dormancy . In constrast , the smaller ( + ) spores grow isotropically for a longer time until their size is comparable to that of the larger ( − ) spores , and they then start sending germ tubes . This difference in germination kinetics between the larger and smaller spores may contribute to the differences in their virulence . To address this , smaller spores were grown isotropically and then used to infect wax moth larvae to test the effect on virulence . Smaller spores were grown in liquid YPD media for 3 . 5 hours until they attained a size comparable to the larger spores . We found that the isotropic growth of small spores yields large multinucleate spores similar to ( − ) larger spores ( data not shown ) . These spores were collected for inoculation and 1 , 000 each of the larger ( − ) spores ( LS ) , smaller ( + ) spores ( SS ) , and isotropically grown ( + ) spores ( IS ) were injected into ten wax moth larvae and survival was monitored . Interestingly IS are as virulent as LS ( Figure 2C ) . These findings further support the conclusion that spore size is a virulence factor . The host response to different sized spores is of interest to consider . We observed that cultured murine macrophage cells ( J774 ) respond differently to LS , SS , and IS ( Figure 3 and Videos S3 , S4 , and S5 ) . When co-cultured , spores of all sizes were phagocytosed by macrophages . A characteristic difference is that the larger spores germinated inside of the macrophages , whereas smaller spores remained dormant inside macrophages without isotropic growth or germination , and grew significantly more slowly compared to the small spores outside of macrophages . Interestingly , IS also germinated inside macrophages similar to the LS . Thus , LS and IS are both likely to undergo invasive hyphal growth in hosts and may therefore exhibit higher virulence . We also observed that , with longer co-incubation , macrophages harboring LS or SS or IS all underwent lysis ( data not shown ) , which warrants further study to elucidate how macrophage cell lysis is triggered by the encounter with fungal spores . The difference in virulence and early germination prompted us to examine the detailed microscopic structure to assess differences between the larger and smaller spores . Interestingly , SEM analyses revealed that the larger ( − ) spores are decorated with ‘bumps’ on the surface , whereas the surface of the smaller ( + ) spores is smooth ( Figure 4A and B ) . As described above , the ( − ) isolates producing larger spores also produce a subpopulation of smaller spores ( Figure 1B ) . We also observed that small , uninucleate ( − ) spores have a smooth surface unlike the bumpy larger ( − ) spores ( Figure 4C ) . Based on TEM , the spore surface bumps may result from trafficking processes from cytosol to the cell surface involved in cell wall construction ( Figure 4D ) . Two sexM mutants [MU423 ( sexMΔ1 ) and MU424 ( sexMΔ2 ) ] were obtained by transformation and homologous recombination with the pyrG cassette flanked with sequences 5′ and 3′ end of the sexM gene ORF . To obtain the transformants , 50 µg of the pyrG cassette DNA was co-incubated with protoplasts of the MU402 strain ( See Materials and Methods ) [34] . A total of 25 pyrG+ transformants were obtained , which were grown on selective medium for several vegetative cycles to obtain homokaryotic transformants . PCR analysis designed to distinguish homologous from ectopic integration indicated that two of these transformants amplified the expected fragments from the 5′ and 3′ flanking regions ( Figure S3 ) . The homologous replacement was confirmed through Southern blot hybridization with a probe homologous to the sexM gene that can discriminate between wild-type and mutant alleles ( Figure 5A ) . The absence of the 4 . 39 kb SalI wild-type fragment in the sexMΔ mutants confirmed that the wild-type allele had been successfully replaced in all of the nuclei of these mutant strains . We tested mating ability by co-inoculating spores of either mating type with the sexMΔ mutants on YPD or YXT media with incubation for two weeks in the dark at room temperature . The sexMΔ mutants failed to form zygospores in any combination of crosses with wild-type ( + ) , wild-type ( − ) , or the other sexMΔ mutant ( Figure 5B and Figure S3 ) . Moreover , no self-fertile development was observed , excluding models in which SexM represses sexual development . That two sexMΔ mutants are both sterile provides evidence that sexM is essential for mating . The sexMΔ mutants have no apparent difference in spore size compared to ( − ) wild-type isolates , and in virulence tests in wax moth larvae the sexMΔ mutants were as virulent as the ( − ) wild-type ( p = 0 . 8235 and 0 . 8619 for MU423 and MU424 , respectively ) ( Figure 5C ) . Thus , SexM does not appear to be involved in spore size determination or virulence in the wax moth model . Although the sexM gene is not involved in virulence , the successful disruption of the sexM gene and functional verification of a role for the sex locus in sexual reproduction in this basal fungal lineage is a major advance in our understanding of sex in the Zygomycota basal fungal lineage . Previous characterization of the Mucor circinelloides subspecies complex was based on morphological and physiological characteristics [31] . To obtain a high-resolution phylogeny for the M . circinelloides subspecies ( Table 1 ) , a phylogenetic analysis based on multi-locus sequence typing ( MLST ) was performed with three of the genes analyzed in the fungal tree of life project [1] . These include an RNA polymerase subunit gene ( RBP1 ) , a large ribosomal RNA subunit gene ( rDNA2 ) , and one intragenic spacer region ( ITS ) . All DNA sequences obtained were aligned and maximum likelihood trees were constructed for each of the three genes ( Figure 6 ) . Trees constructed with RPB1 , ITS , and rDNA2 sequences all revealed similar patterns , where three notable clusters are formed that correspond to the M . circinelloides f . lusitanicus ( Mcl ) ( ATCC1216a , ATCC1216b , CBS277 . 49 , NRRL3631 ) , M . circinelloides f . griseocyanus ( Mcg ) ( ATCC1207a , ATCC1207b ) , and M . circinelloides f . circinelloides ( Mcc ) ( NRRL3614 , NRRL3615 , ATCC11010 ) subspecies . There was no phylogenetic incongruence observed demonstrating that the M . circinelloides subspecies are sufficiently diverged to be designated as at least subspecies . ATCC1209b was found to be distinct from all three subspecies based on this MLST analysis and may prove to be an intermediary , hybrid , or distinct subspecies within the M . circinelloides complex . The clinical Mucor circinelloides isolates ( Table 1 ) were classified at the subspecies level based on molecular analysis . Based on the previously analyzed MLST alleles ( Figure 6 ) , we found that most clinical strains ( 5 out of 8 ) grouped within the Mcc subspecies with the exception of AS71 and UIC-1 ( Figure S4 ) . These findings were consistent at all three MLST loci . In the phylogenetic analyses , we additionally found evidence for a fourth group containing ATCC1209b and UIC-1 ( Table 1 ) . These isolates clustered together in all three trees . ATCC1209b was found to be sterile with all other tested isolates ( Table 2 ) and these two lines of evidence may indicate that ATCC1209b and UIC-1 could be isolates of a different subspecies . To elucidate possible differences in the pathogenicity between the M . circinelloides subspecies , we tested virulence using wax moth larvae . M . c . f . lusitanicus strain R7B was the most pathogenic ( Figure 7A ) ; for example , R7B caused 100% lethality within 3 days compared to 20% lethality with M . c . f . griseocyanus ATCC1207B over the course of 10 days ( p value<0 . 001 ) . All other strains tested were less virulent and a correlation between spore size and pathogenicity was observed ( Table 2 and Figure 7 ) . Interestingly , although we observe greater virulence of Mcl in the wax moth , the Mcc subspecies is more common in human clinical cases . To assess whether the virulence of a given M . circinelloides subspecies in a mammalian host system differs from that observed in the wax moth heterologous host , we employed a diabetic mouse model ( [14] and references therein ) for zygomycosis . One million ( 1×106 ) spores of M . circinelloides subspecies ( Table 1 ) were intravenously inoculated into 5 mice for each strain . In this experiment , one Mcc species , NRRL3615 , displayed the highest virulence ( p = 0 . 0091 ) ( Figure 7B ) compared to Mcl and Mcg . The Mcc isolate NRRL3615 showed 100% mortality and NRRL3614 displayed 40% mortality by 4 days post infection . However the other Mcc isolate , ATCC11010 , and Mcl species were avirulent . Thus , only Mcc isolates ( but not all ) show virulence in the murine host , which may explain the prevalence of Mcc species in clinical isolates . MU423 ( sexMΔ mutant of Mcl ) , ATCC1207b ( Mcg ) , and ATCC1209b were also tested and did not display mortality for the duration of the experiment ( data not shown ) . More clinical isolates ( Table 1 ) were tested in the murine host confirming that only Mcc isolates based on our phylogenetic analysis display virulence but less virulent Mcc strains were also found ( Figures S4 and S5 ) . Notably , we observed that the Mcc isolates exhibited better growth at 37°C compared to the other subspecies ( Figure 7C ) indicating that temperature sensitivity/resistance might contribute to the differences between Mc subspecies in virulence in the murine host . The mating of three M . circinelloides subspecies was examined in response to different conditions and media . Isolates from each M . circinelloides subspecies were co-cultured with each other ( Table 3 ) . Spores of each strain were inoculated onto PDA medium . After a 24-hour incubation , 5 mm×5 mm agar blocks containing mycelia of each strain were placed in abutting pairs on YXT media . The plates were wrapped with tin foil and incubated at room temperature or , in some cases , the mating plate was placed at a lower incubation temperature ( see Materials and Methods ) . When co-cultured without light , a dark zygospore line formed in the middle between two opposing strains of opposite mating type ( for example , ATCC1216a and ATCC1216b ) ( Figure 8A ) ; however , two strains of the same mating type did not form zygospores based on morphological analysis by light microscopy . Zygospore formation was largely restricted to matings within a subspecies , with two notable exceptions in which zygospores were sporadically observed , involving the mating crosses of ATCC1207b ( Mcg ) ×NRRL3615 ( Mcc ) and R7B ( Mcl ) and ATCC1207a ( Mcg ) ( Table 3 ) . Interestingly , the ATCC11010 Mcc isolate did not mate with any of the Mcc strains tested . The unclassified ATCC1209b isolate did not mate with any strain tested . Mating assays performed with the clinical isolates found evidence of mating in two isolates . Both CNRMA03 . 371 and CNRMA04 . 805 were found to mate with NRRL3614 , indicating they are of the ( + ) mating type ( data not shown ) . Crosses performed with other isolates did not reveal conclusive zygospore formation . Zygospores of zygomycetes remain dormant for a long period from months to a year before germination occurs [21] , [35] . P . blakesleeanus zygospores germinate after a 3 to 4 month dormancy period , enabling the analysis of progeny [23] , [36] , [37] . Although zygospore germination of other Mucor spp . under laboratory conditions has been reported [38]–[40] , our extensive trials for M . circinelloides zygospore germination were not successful with the conditions and isolates tested . Mating ultrastructures were observed by scanning electron microscopy analysis . Mating between strains ATCC1216a ( − ) and ATCC1216b ( + ) , and also vegetatively grown ATCC1216b cells , were examined to investigate sexual and asexual morphologies . Zygospores are morphologically distinct from asexual spore-harboring structures , sporangia , which develop at the apex of aerial hyphae ( Figure 8B ) . The zygospores were thick-walled and enveloped by repeated asterisk-like structures . Zygospores are the dormant , stress-tolerant stage , and thus these structures may contribute to the increased rigidity of the sexual spores . M . circinelloides formed coiled hyphae , possibly during the process of conjugation of two mating type hyphae . These early stages of sexual development resemble mating structures in some dimorphic ascomycetes including Histoplasma capsulatum [41] and the dermatophyte Microsporum gypseum [42] ( Figure 8C ) . A related zygomycete , P . blakesleeanus , forms a twisted rope-like structure prior to zygophore formation [23] , whereas in M . circinelloides it is speculated that the formation of coiled hyphae is followed by hyphal fusion between the two mating types , and then by zygophore and mature zygospore formation . In previous studies , the sex locus of P . blakesleeanus was defined and found to contain one of two divergent HMG domain genes , sexM or sexP [23] . The P . blakesleeanus sex locus was compared with that of M . c . f . lusitanicus ( Mcl ) [22] . The synteny of the TPT-HMG-RNA helicase genes was found to be conserved in the two other subspecies of M . circinelloides , including Mcg and Mcc ( Figure 9 ) . Two Mcg strains , ATCC1207a ( + ) and ATCC1207b ( − ) , were analyzed using primers ( Table S2 ) designed from the genes for the triose phosphate transporter ( tptA ) and RNA helicase genes ( rnhA ) of the sequenced Mcl strain . The architecture of the Mcg sex locus , including gene order and orientation , was identical to that of Mcl ( Figure S6 ) , whereas the orientation of the sexP gene in P . blakesleeanus is opposite to that in the three M . circinelloides subspecies . DNA dot plot analyses were performed to establish the boundaries of the sex locus of Mcg and Mcc ( Figure 9A ) . A comparison with the R . oryzae ( + ) sex locus ( ∼9 kb ) revealed that the sex locus of the three M . circinelloides species is substantially shorter [1552 bp for Mcl ( + ) , 1579 bp for Mcl ( − ) ; 1573 bp for Mcg ( + ) , 1811 bp for Mcg ( − ) ; 1495 for Mcc ( + ) , 1673 bp for Mcc ( − ) ] and does not contain a BTB/Ankyrin/RCC1 domain protein gene spanning approximately 4 , 000 bp in R . oryzae . Additionally , the tptA gene of R . oryzae is in the opposite orientation from the tptA gene of the M . circinelloides subspecies and P . blakesleeanus ( Figure S6 ) . The R . oryzae ( − ) sex locus allele also lacks the additional ORF that is found in the ( + ) sex locus [24] , [26] . Sequence comparisons of the sex loci of the ( + ) and ( − ) mating types of the M . circinelloides subspecies are detailed in Tables 4 and 5 . Although the overall architecture was similar , there was an interesting difference in the sex locus of the Mcg species , where the border of the sex locus includes the promoters of both the tptA and rnhA genes whereas the sex locus of Mcl includes only the tptA promoter ( Figure 9C ) [22] . In P . blakesleeanus , neither the tptA nor the rnhA gene promoters are part of the sex locus [23] . This is a strong indication of the plasticity of the sex locus involving expansion/contraction in the M . circinelloides subspecies complex ( Figure S7 ) .
Mcl , especially ( − ) mating type , was found to be highly virulent in the wax moth host . The difference in virulence between closely related species is of interest . One important difference between the mating types is spore size , in which only the ( − ) mating type of Mcl is highly virulent and produces larger spores . In the pathogenic basidiomycete Cryptococcus neoformans , the MAT locus is linked to virulence [43]; α mating type isolates are more prevalent in clinical isolates , the α MAT locus genes are highly expressed during infection in macrophages [44] , and α isolates are more pathogenic in certain strain backgrounds [45] , [46] or during co-infection [44] , [47] , [48] . The sex locus might therefore be similarly involved in the pathogenesis of this zygomycete species . We found evidence that in M . circinelloides , the sex locus may be involved in virulence via a role in the asexual spore size of different mating types . These results prompted us to consider three models for the relationship between the sex locus and spore size . First , the sex locus could control spore size . In this model , SexM could have been required for larger spore size; however , the large spore size of the sexMΔ mutants isolated and characterized here exclude this model ( Figure 5 ) . Alternatively , SexP may promote smaller spore size , and this can be addressed by constructing isogenic mating type strains in which sexM is replaced with sexP or in which SexP has been deleted . Second , the sex locus and other unlinked genomic loci may together control spore size . In this model , spore size is a quantitative trait , and the sex locus may be one of several genes that contribute to control spore size . The ( − ) mating type isolate NRRL1443 has an intermediate spore size , possibly lending support to this hypothesis . In this model , deletion of sexM or sexP could lead to an intermediate spore size , possibly dependent on strain backgrounds , rather than strictly large or small spores . Third , the sex locus could play no role in controlling spore size . In this model , the apparent linkage observed between mating type and spore size could be the result of analysis of a small sample size . And it may not be the case that the sex locus contributes to virulence in ways other than spore size because the sexMΔ mutants are as virulent as wild-type ( Figure 5 ) . For example , the larger spore isolates could represent naturally occurring mutants that bypass a hypothetical cell cycle inhibition stage during spore dormancy and the multinucleate sporangiospores of larger size may reflect uncontrolled cell cycle: inside sporangia the spores would therefore break dormancy and undergo rounds of nuclear division . In this case , activation or overexpression of cell cycle inhibitors may reduce spore size . Why are larger spores more virulent ? The short or absent isotropic growth period for larger spores , compared to the long phase observed prior to germ tube emergence for smaller spores , could be involved in the differences in virulence ( Figure 2 , Videos S1 and S2 ) . The larger spores are likely poised to undergo rapid invasive hyphal growth compared to the smaller spores . The extended isotropic growth phase of the smaller spores would result in slow germ tube formation or a block or delay in germ tube emergence , and could reduce virulence in the host . Our studies on the response of macrophages to spores further supported this hypothesis , where larger spores engulfed by macrophages are still able to send germ tubes ( Figure 3 and Videos S3 , S4 , and S5 ) . This observation could reflect a recent study in zygomycosis that shows the germ tubes of R . oryzae cause more damage compared to spores in in vitro experiments with an umbilical vein cell line [14] , [49] . Although we observed that both the larger and smaller M . circinelloides spores trigger cell death of cultured macrophages with prolonged incubation ( data not shown ) , the invasive hyphal growth contributes to render larger spores more virulent than smaller spores . The macrophage cell death elicited by Mucor is also of considerable interest . A previous study showed that an aqueous R . oryzae extract can trigger apoptosis in cultured human leukemia cells [50] . Thus , Mucor spores may produce unknown components that trigger host immune cell death responses resulting in susceptibility . The relationship of spore size to virulence is also a central question to emerge from our study . Although large spores are more virulent in Galleria , it is also possible that we would observe an opposite result in pulmonary infections in a murine inhalation model , because larger spores may be less likely to penetrate lung alveolar spaces compared to smaller spores . Comparing results between intranasal and intravenous routes of infection in the murine model will provide insight into whether the route of infection influences the relative virulence of larger and smaller spores following inhalation . Recent findings on size dimorphism involving Cryptococcus are interesting to consider in light of our findings on Mucor spore size dimorphism . C . neoformans often forms significantly enlarged cells called giant or titan cells , which are known to be more virulent and less susceptible to the host immune system [51] , [52] . Aspects of fungal cell gigantism differ between the two different pathogenic fungi: C . neoformans giant/titan cells are mononucleate and polyploid , but Mucor large spores are multinucleate . However , in both cases it is clear that enlarged fungal cells confer benefits to the fungal pathogens during host infection . The human pathogenic ascomycete Coccidioides immitis is also known to exhibit cell giantism during host infection , where ‘smaller’ athroconidia undergo multiple cell cycles resulting in the formation of enlarged multinucleate cells , spherules , which escape from host immune systems [53] , [54] . Given these precedents , other examples of fungal size dimorphism linked to virulence likely remain to be discovered . In the diabetic murine host system , Mcc displays higher virulence compared to Mcl and Mcg tested in this study ( Figure 7 and Figure S5 ) . This is an intriguing observation , which is in accord with the increased prevalence of the Mcc species in Mucor clinical isolates ( Table 1 and Figure S4 ) . Mcl isolates , especially larger spore isolates , are highly virulent in the wax moth host , however , they are less virulent in the murine host indicating that virulence traits may have been differentially adapted during speciation . For example , the Mcl isolates tested exhibited limited growth at 37°C but Mcc species are relatively resistant to the human/mouse body temperature ( Figure 7 and data not shown ) . Interestingly not all Mcc strains show the same level of virulence ( Figure S5 ) , and further investigation is necessary to examine differences between virulent and less-virulent Mcc strains . The M . circinelloides complex comprises three subspecies , M . c . f . lusitanicus ( Mcl ) , M . c . f . griseocyanus ( Mcg ) , and M . c . f . circinelloides ( Mcc ) , and has been historically characterized based on physiological and morphological characteristics [31] , or by comparing enzymes of certain M . circinelloides species [55] . Based on our MLST analysis , the M . circinelloides subspecies are closely related when compared to the P . blakesleeanus and R . oryzae outgroups , but there is enough sequence divergence in the ITS , rDNA2 , and RPB1 regions to support at least the current subspecies classifications ( Figure 6 ) . We found no examples of MLST marker exchange between the 3 subspecies and if more comprehensive analyses of the population substantiate this observation , they may represent cryptic species . Our mating assays showed mating occurs predominantly within the subspecies with only a few exceptional intersubspecies fertile combinations , further suggesting that they could represent cryptic species with genetic isolation limiting or preventing introgression . Phylogenic analysis with SexP and SexM indicates that allelic sex determinant genes may have evolved before speciation within zygomycetes , especially in the Mucorales ( Figure S8 ) . Allele compatibility tests support evidence for recombination in the clinical Mcc population ( Figure S9 ) . Several criteria have been used to define the sex locus in the heterothallic zygomycetes [22] , [23] , [26] . First , the ( + ) and ( − ) mating types are defined by the presence of the sexP or sexM genes , respectively . Mating is only observed between opposite mating types . Additionally , rare Phycomyces disomic strains containing both sexP and sexM are self-fertile , producing spiral , zygospore-like structures [23] . Furthermore , the sex locus region has been genetically mapped with crosses and RFLP analysis , linking the Phycomyces sexP gene to the ( + ) mating type and likewise , the Phycomyces sexM gene to the ( − ) mating type within a 38 kb interval linked to mating type . Most importantly , the Mucor sexMΔ mutants isolated in this study are sterile ( Figure 6 ) . Finally , the sex-determining region has been corroborated across R . oryzae , M . circinelloides , and P . blakesleeanus , representing three species within the Mucorales . Our SEM analyses of the early development of zygospores revealed a coiled hyphae structure before the rest of the zygophore is produced ( Figure 4 ) . Formation of coiled hypha was previously described in H . capsulatum , in which the hyphae of one mating type extends and enwraps the hyphae of the opposite mating type [56] . Branches form around this knob-like structure and anastomoses between the hyphae result in a larger hyphal mass that eventually becomes the ascocarp . It is possible that zygospore formation in M . circinelloides could follow a similar process of hyphal mass aggregation , followed by anastomoses culminating in formation of the zygospore . However , the factors that contribute to the remarkable rigidity of the zygospore have yet to be discovered . Interestingly , the promoter of the TPT gene is included within the sex locus for Mcl while the promoter lies outside the sex locus of P . blakesleeanus [26] . In this study , we found that in addition to the TPT gene , the promoter for the RNA helicase gene is also included in the sex locus for Mcg . This is indicative of a possible expansion , or contraction , of the sex locus in the different subspecies by changing the recombination block that punctuates the evolutionary trajectory of this dynamic region of the genome ( Figure S6 ) . Evolutionarily , the TPT and RNA helicase region may have been included or excluded from the sex locus over time [26] . Thus , our observations may imply an expansion or contraction of the sex locus in zygomycetes , especially in the M . circinelloides complex ( Figure S7 ) . MAT locus expansion/contraction has been observed in ascomycetes . The MAT locus of ascomycetes is generally characterized as a syntenic region with APN1-MAT1-1 ( alpha box ) -SLA2 or APN1-MAT1-2 ( HMG ) -SLA2 gene clusters [57] , [58] . In two evolutionarily related ascomycetous fungal groups , the dermatophytes and dimorphic fungi , the APN1 , MAT1-1 or MAT1-2 , and SLA2 genes span ∼3 kb in Microsporum gypseum compared to ∼9 kb in Coccidioides immitis/posadasii in which flanking genes have been recruited into the MAT locus [42] . Comparison of sex and sex-related loci in zygomycetes and microsporidia also revealed additional ORFs in the sex/sex-related locus [59] . This might suggest gene eviction from or capture into the sex locus . Examples can be found in pathogenic basidiomycete tetrapolar MAT loci . The basidiomycete tetrapolar MAT locus involves a homeodomain locus and a pheromone/receptor locus . In Ustilago maydis , a plant pathogenic basidiomycete , the MAT locus harbors unlinked homeodomain and pheromone/receptor loci and the MAT locus spans ∼4 kb for the homeodomain and ∼4 . 4–8 kb for the pheromone/receptor locus . In C . neoformans the two domains are linked and novel genes have been captured into a bipolar MAT locus that now spans >100 kb [26] . The bipolar MAT loci of U . hordei and Malassezia globosa are even larger and span more ∼500 kb or ∼170 kb , respectively [60]–[62] . The MAT loci of C . neoformans , U . hordei , and M . globosa serve as dramatic examples of MAT locus expansion and share features with sex chromosome evolution in plants and animals . Additionally , a comparison of the MAT locus within the Cryptococcus complex suggests an evolutionary trajectory involving gene evictions [63] , [64] . In C . neoformans var . grubii and C . gattii , the MAT locus includes three genes ( IKS1 , NCM1 , and BSP3 ) , although in C . neoformans var . neoformans these three genes have been evicted from the MAT locus by gene conversion leading to a relative contraction of the MAT locus . Similar molecular events may have punctuated the evolution of the zygomycete sex locus .
The animal studies at Duke University Medical Center were in full compliance with all of the guidelines of the Duke University Medical Center Institutional Animal Care and Use Committee ( IACUC ) and in full compliance with the United States Animal Welfare Act ( Public Law 98–198 ) . The Duke University Medical Center IACUC approved all of the vertebrate studies . The studies were conducted in the Division of Laboratory Animal Resources ( DLAR ) facilities that are accredited by the Association for Assessment and Accreditation of Laboratory Animal Care ( AAALAC ) . The strains used in this study are listed in Table 1 and in Table S1 . M . circinelloides strains were grown on yeast and dextrose agar ( YPD ) or potato dextrose agar ( PDA ) media for spore production at room temperature . For mating , YXT ( 4 . 0 g yeast extract; 10 g malt extract; 4 g glucose; 15 g agar; 1000 mL water , with the pH adjusted to 6 . 5 ) [65] , YPD ( 10 g yeast extract; 20 g peptone; 20 g dextrose; 20 g agar; 1000 mL water ) , and V8 media ( 50 mL V8 juice; 0 . 5 g KH2PO4; 950 mL dH2O; 40 g bactoagar at pH between 7 . 0–7 . 2 adjusted with 5 M KOH ) were used . Plasmids in this study were maintained in Escherichia coli One Shot MAX Efficiency DH5α-T1R competent cells ( Invitrogen Co . , Carlsbad , CA ) and manipulated as previously described [66] . Microbial strains were grown under appropriate Biosafety Level 2 conditions ( BSL2 ) . All chemicals for media , buffer , and supplements were from Difco Laboratories ( Detroit , MI ) unless otherwise indicated . According to the Centraalbureau voor Schimmelcultures ( CBS ) , American Type Culture Collection ( ATCC ) , and ARS Culture Collection ( NRRL ) databases , the strains CBS277 . 49 , ATCC1216b , and NRRL3631 are the same isolate with different designations . However , previous studies indicate that these isolates differ in karyotype [67] , [68] . Our further analysis demonstrated the three isolates including another ATCC isolate , ATCC1216a , have different distinguishing SNPs in the RPB1 gene . The database mating type designation for these isolates are inconsistent with our data; we find that CBS277 . 49 and ATCC1216a are ( − ) ; NRRL3631 and ATCC1216b are ( + ) . CBS277 . 49 , NRRL3631 , and ATCC1216a are different based on analyses by Random Amplification of Polymorphic DNA ( RAPD ) . These results document that while those isolates are all clearly isolates of Mcl they are genetically distinct , despite the records of the stock culture collections ( See Supplementary Text S1 ) . Spores were resuspended in a phosphate buffered saline ( PBS ) . PBS containing 500 or 1 , 000 spores or 5 µl of PBS alone were injected into wax moth ( Galleria mellonella ) larvae ( 10 or 20 larvae per strain ) . For the murine host model , groups of BALB/c mice were rendered diabetic with 190 mg per body kg streptozocin ( in citric acid buffer pH = 4 . 5 ) through intraperitoneal injection 10 days prior to fungal challenge [14] . A cohort of injected mice ( 10 ) was randomly chosen and confirmed to exhibit glycosuria with Keto-Diastix reagent strips ( Bayer Co . Elkhart , IN ) . After 10 days , the mice were infected with 106 spores in 200 µl PBS through tail vein injection . Survival rate of the host was monitored twice a day and body weight was measured daily . Animals that appeared moribund or in pain were sacrificed appropriately . Significance of mortality rate data was evaluated by using Kaplan–Meier survival curves with the PRISM statistics software ( GraphPad Software , Inc . , La Jolla , CA ) . Spores were observed with a Zeiss Axioskop 2 Plus with an AxioCam MRm camera ( Carl Zeiss Inc . , Thornwood , NY ) . To analyze nuclei , spores were fixed with 3 . 7% formaldehyde in 50 mM potassium phosphate buffer , pH 7 . 0 , containing 0 . 2% Triton X-100 . Then the spores were mounted on a coverslip with ProLong Gold antifade reagent with DAPI ( Invitrogen , Carlsbad , CA . ) . For confocal microscopic analyses , a Zeiss LSM 510 inverted confocal microscope was used ( Carl Zeiss Inc . , Thornwood , NY ) . Time lapse analyses for large and small spore germination , and interactions between macrophages and spores were performed by using the Zeiss Axio Observer Z1 microscope system ( Carl Zeiss Inc . , Thornwood , NY ) equipped with an Opto-electonically motorized XY stage , Pecon XL S1 incubator , and Coolsnap ES2 high resolution CCD camera ( Photometrics Inc . , Huntington Beach , CA ) . A 6-well plate was layered with YPD media and inoculated with spores of R7B and NRRL3631 . For macrophages , J774 murine macrophage cell lines were layered on a 6-well plate at 105 cells/ml prior to the fungal spore challenge . The same number of spores ( 105 spores/ml ) was inoculated into 6-well plates and the plates were immediately observed by microscopy . The images were obtained every 30 seconds and reconstructed as a movie by using MetaMorph 7 . 6 . 5 ( Molecular Devices Inc . , Sunnyvale , CA ) . For scanning electron microscopy ( SEM ) , the mating/culture plates were washed with 0 . 1 M Na cacodylate buffer ( pH = 6 . 8 ) , and 1 mm3 blocks of mating areas were excised , and incubated in fixation buffer at 4°C . Samples were then rinsed in cold 0 . 1 M Na cacodylate buffer three times , post-fixed in 2% osmium tetroxide in 0 . 1 M Na cacodylate buffer for 2 . 5 h at 4°C , critical point dried , and sputter coated before being viewed by SEM . Additional SEM of M . circinelloides spores was accomplished as follows . The R7B and NRRL3631 strains were inoculated on PDA and spores were collected after 4 days . Spores were suspended in 0 . 1 M sodium cacodylate and immobilized on Millipore Nitrocellulose filters ( Millipore HAWP 0 . 46 µm ) . The spores and membrane were immediately fixed in 2% glutaraldehyde ( Electron Microscopy Sciences , EMS , Hatfield , PA ) , 0 . 05% malachite green oxalate ( EMS ) in 0 . 1 M sodium cacodylate buffer , and incubated at 4°C until further processing . The fixation buffer was then removed and the membrane was washed in 0 . 1 M sodium cacodylate prior to being subjected to an ethanol dehydration series ( 2 times for 10–15 min in 25% , 50% , 75% , 95% , and 3 times in 100% ethanol ) . Samples were then critical point dried ( Pelco CPD2 , Ted Pella , Inc . , Redding , CA ) , sputter coated , and viewed and imaged with the FEI XL30 SEM-FEG ( FEI Company , Hillsboro , OR ) at the Shared Materials Instrument Facility ( SMIF ) at Duke University . Transmission electron microscopy ( TEM ) of M . circinelloides spores was accomplished as follows . Spores were collected as described above and washed in 0 . 1 M sodium cacodylate buffer ( pH = 6 . 8 ) , collected by centrifugation ( ∼4 , 000 rpm , 3 min in a table top centrifuge ) , resuspended in 2% glutaraldehyde plus 0 . 05% malachite green oxalate in 0 . 1 M sodium cacodylate buffer , and incubated at 4°C for 2 days . Fixed spores were collected by centrifugation , resuspended , washed once with 0 . 1 M sodium cacodylate buffer , centrifuged , the supernatant was removed , and 100–200 µl 1 . 6% agarose was added to the tube on ice to immobilize the cells in a 0 . 8% agarose pellet . The agarose pellet containing spores was then dehydrated by an ethanol series ( 2 times for 10–15 min in 25% , 50% , 75% , 95% , and 3 times in 100% ethanol ) , and then stained with 0 . 8% K3Fe ( CN ) 6 , 1% OsO4 , 0 . 1 M sodium cacodylate for 1 hr at room temperature . The agar pellet was then washed two times with 0 . 1 M sodium cacodylate buffer and stained with 1% tannic acid for 1 hr at room temperature . The pellet was then washed with 0 . 1% sodium cacodylate buffer for 5 min followed by two washes in ddH20 for 5–10 min each , and then stained with 1% uranyl acetate in water overnight at 4°C . Sample were then prepared for embedding in Embed812 ( EMS ) as follows , one 5 min incubation in 50/50 ethanol-propylene oxide , three 10 min incubations at room temperature in 100% propylene oxide , 50/50 Embed812-propylene oxide overnight at room temperature with gentle rotation , 10 min in 100% uncatalyzed Embed812 , and 1 hr in 100% catalyzed Embed812 . Catalyzed Embed812 was then drained off , agar pellets are immersed in 100% catalyzed Embed812 beam capsule , and cured at 65°C for 72–96 hrs . The cured peg was then trimmed , sectioned , and mounted on copper grids . Grids were post-stained prior to viewing . Sections were viewed and imaged with a Philips/FEI CM 12 Transmission EM instrument ( FEI Company , Hillsboro , OR ) with Advanced Microscopy Techniques , Corp . ( AMT ) 2k×2k digital camera ( Danvers ) , Duke University Department of Pathology or by the FEI Tecnai G2 Twin instrument at the Shared Materials Instrument Facility ( SMIF ) at Duke University . To disrupt the sexM gene in the sex locus , we constructed a disruption allele containing the pyrG gene flanked by 1 kb each of the 5′ and 3′ regions of the sexM gene by using overlap PCR . The 5′ end was amplified with primers JOHE20368 and JOHE20369 and the 3′ end was amplified with primers JOHE20372 and JOHE20373 ( Table S2 ) . The pyrG fragment was amplified with primers JOHE20370 and JOHE20371 from the genome of wild type Mcl strain CBS277 . 49 . The three fragments were then subjected to an overlap PCR to isolate a disruption allele as described [69] . The cassette was purified and strain MU402 ( leuA− , pyrG− ) was transformed to disrupt the target gene and transformation was carried out essentially as described previously [70] . In brief , protoplasts were obtained from 2 . 5×108 germinated spores of strain MU402 ( pyrG− , leuA− ) [34] by incubation with 0 . 03 unit/ml chitosanase RD ( US Biologicals ) and 1 mg/ml Lysing Enzymes ( L-1412; Sigma ) at 30°C for ∼90 min . Protoplasts were incubated with 50 µg DNA and pyrG+ transformants were selected in MMC medium ( 1% casamino acids , 0 . 05% yeast nitrogen base without amino acids and ammonium sulfate , 2% glucose ) pH 3 . 2 , supplemented with 0 . 5 M sorbitol [34] . Because the initial transformants are usually heterokaryons due to the presence of several nuclei in the protoplasts , transformants were grown in MMC selective medium for several vegetative cycles to increase the proportion of transformed nuclei . Two pyrG-positive transformants were finally selected and homologous replacement was confirmed by PCR and Southern blotting . In brief , primers; P1 and P2 , were used to identify the integration of the pyrG gene at the sexM locus . The genomic DNA of the disruption candidates and wild-type were digested with SalI and Southern blotting was performed with the 3′ fragment as a probe as described [66] . M . circinelloides hypha is coenocytic , therefore a vegetative passage is required to select progeny only with transformed nuclei . During this step , the two initial transformants contained different proportions of transformed nuclei and we interpret these to be the result of independent transformation events though they were isolated from the same transformation experiments . Primers used in this study are listed in Table S2 . Primers JOHE19916 and JOHE19917 were used to amplify the entire MAT locus for all three subspecies of M . circinelloides . All other primers were used for subsequent PCR analysis and sequencing . Primers JOHE19868 and JOHE19869 were used to amplify the TPT gene region and JOHE22785 , JOHE22786 , JOHE22787 , JOHE22788 , JOHE22789 , and JOHE22790 were used to amplify the RNA helicase gene region . For sequencing of the sex locus and MLST analysis of M . circinelloides strains , PCR products were cloned into plasmid pCR2 . 1-TOPO following the manufacturer's instructions ( Invitrogen , Carlsbad , CA ) or subjected to direct sequencing of the PCR product . Sequencing reactions were performed with an Eppendorf epgradient thermal cycler using standard BigDye Terminator chemistry ( Applied Biosystems , Foster City , CA ) and sequencing was carried out at the IGSP sequencing facility at Duke University ( http://www . genome . duke . edu/cores/sequencing ) using an Applied Biosystems 3730xl DNA Analyzer . DNA sequences were analyzed with Sequencher version 4 . 10 and BLAST [71] . Genes were annotated with FGENESH or ORF finder ( NCBI ) . Obtained sequences were deposited in GenBank ( Table S3 ) . Sequences were analyzed with CLUSTALW and phylogenies were constructed using a PhyML 3 . 0 software [72] , which allowed phylogenies to be inferred and levels of support ascertained , and PAUP 4 . 0 ( Sunderland , MA ) . A species tree was constructed by concatenating two conserved loci , rDNA2 and RPB1 , from the fungal tree of life project [2] . Phylogenetic trees were drawn with the Dendroscope program [73] with aligned sequences . All mating assays were performed on YPD , YXT , or V8 media . Fungal strains were grown on PDA for 2 days , and two agar blocks ( 5 mm×5 mm ) containing mycelia of each strain were then placed approximately 1 cm apart at the edge of a mating plate . All possible combinations of strains listed in Table 2 were tested . Plates were incubated for 3 days at 15°C and then for another 4 days at room temperature . All mating plates were incubated in the dark . Zygospore and mating specific structure formation were monitored with a Nikon Eclipse E400 microscope , equipped with a Nikon DXM1200F digital camera ( Nikon Instrument Inc . , Melville , NY ) . | Mucormycosis is recognized as an emerging infectious disease . Compared to other fungal infections , mucormycosis results in high mortality: ∼50% of overall infections and >90% in disseminated infections . There is therefore an ongoing need to study these fungal pathogens . However , surprisingly little is known about the pathogenesis of mucormycosis . Our findings reveal a correlation between sporangiospore size and virulence: larger sporangiospores are more virulent than small spores . Larger spores start invasive hyphal growth immediately upon phagocytosis by host immune cells , whereas smaller spores have a long period of isotropic growth . Differential host immune response explains their difference in virulence , whereby larger spores escape host immune cells by germinating inside of macrophages . Our findings revealed an example of adaptation through fungal cell gigantism , which enable pathogenic fungi to survive within and establish infection in the host . Knowledge of the mechanisms of pathogenicity and the molecular basis of the sexual cycle in M . circinelloides will contribute to advance our understanding of pathogenic zygomycetes . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"medicine",
"infectious",
"diseases",
"zygomycosis",
"fungal",
"diseases"
] | 2011 | Sporangiospore Size Dimorphism Is Linked to Virulence of Mucor circinelloides |
A microsphere immunoassay ( MIA ) utilising Luminex xMap technology that is capable of determining leptospirosis IgG and IgM independently was developed . The MIA was validated using 200 human samples submitted for routine leptospirosis serology testing . The traditional microscopic agglutination ( MAT ) method ( now 100 years old ) suffers from a significant range of technical problems including a dependence on antisera which is difficult to source and produce , false positive reactions due to auto-agglutination and an inability to differentiate between IgG and IgM antibodies . A comparative validation method of the MIA against the MAT was performed and used to determine the ability of the MIA to detect leptospiral antibodies when compared with the MAT . The assay was able to determine samples in the reactive , equivocal and non-reactive ranges when compared to the MAT and was able to differentiate leptospiral IgG antibodies from leptospiral IgM antibodies . The MIA is more sensitive than the MAT and in true infections was able to detect low levels of antibody in the later stages of the acute phase as well as detect higher levels of IgM antibody earlier in the immune phase of the infection . The relatively low cost , high throughput platform and significantly reduced dependency on large volumes of rabbit antisera make this assay worthy of consideration for any microbiological assay that currently uses agglutination assays .
Leptospirosis is considered to be the most widespread zoonotic disease in the world [1] with clinical diagnosis proving challenging due to the non-specific nature of symptoms associated with the disease . There are some 300 leptospiral serovars belonging to a number of different serogroups . Currently there are 24 sero-groups of pathogenic leptospires based on their antigenic relatedness [2] . Leptospirosis was first reported in Australia in 1933 in the state of Queensland and has since been isolated Australia wide [3] with Queensland reporting the majority of these cases ( 57 . 6% ) [4] . In 2011 the reported incidence of leptospirosis in Queensland was 3 . 4 cases per 100 , 000 people and overall in Australia the incidence was 0 . 84 cases per 100 , 000 people [5] . At present , 24 serovars of Leptospira spp are recognised in Australia and in recent years a dramatic increase in the incidence of leptospirosis cases in Australia ( particularly Queensland ) has been noted with environmental factors believed to be the main influence on this increase [6] . Diagnosis of leptospirosis occurs at two stages—during the acute phase the live organism can be detected by two methods . Polymerase chain reaction ( PCR ) testing is a useful molecular detection tool for rapid qualitative diagnosis of leptospirosis in its earliest stage [7] . Serum or blood samples provided for PCR testing must be collected within a precise timeframe ( 0–8 days post onset ) to enable diagnosis . Blood culture isolation can also be utilised in the early stages of leptospiral infection ( 0–10 days post onset ) , however this method is time consuming , requires specialised media and equipment and can take months for a serovar specific result [8] . The immune phase of a leptospiral infection is characterised by the presence of leptospiral antibodies and diagnosis is based on serological methods with the microscopic agglutination test ( MAT ) considered the current gold standard [9] . If the stage of the disease is unknown , both acute and immune phase tests are performed . Other serological test methods have previously been developed including flow cytometry [10] , complement fixation testing [11] , indirect hemagglutination assay [12] an IgM dipstick assay [13] and an IgM enzyme-linked immunosorbent assay ( ELISA ) in a number of formats [14 , 15] . Each of these assays has its advantages and disadvantages [16] and the type of assay used for diagnosis is generally dependant on the facilities available . Serological diagnosis of leptospirosis in humans in Queensland , Australia is currently performed by screening with a commercially available leptospirosis IgM ELISA followed by the MAT as a reference and confirmatory test . The MAT method has many disadvantages as it requires specialist expertise , fresh leptospirosis cultures , is labour intensive , costly and is capable of determining total antibody only . The current endemic routine panel for MAT testing in Queensland , Australia consists of 16 serovars , with representatives from a number of different serogroups . Each sample submitted for MAT is screened against this panel and any reactive samples are then serially diluted and retested to determine an end point . Results are reported as a titre with the end point being the final dilution of serum at which 50% or more of the leptospires are agglutinated . This assay permits the testing of up to 20 samples per day on a routine basis . The MIA has the ability to simultaneously test large numbers of samples against large numbers of serovars as well as determine individual IgG and IgM titres . These factors alone would be enormously beneficial in the laboratory diagnostics and epidemiological studies of leptospirosis . Bead based suspension array technology ( xMap , Luminex ) has the capacity to multiplex up to 500 individual analytes in a single well and has been shown to be a successful diagnostic tool for serology in many applications [17 , 18 , 19] . This assay platform is based on magnetic coated polystyrene beads filled with two coloured fluorescent dyes in differing ratios resulting in 500 distinct bead sets . Each bead set can be coated with a different antigen and mixed to allow the simultaneous measurement of antibody response to up to 500 different antigens . This high-throughput screening system allows processing of high numbers of patient samples per day . Its speed , sensitivity , and accuracy of multiple binding events measured in the same small volume have the potential to replace many clinical diagnostic and research methods and deliver data on hundreds of analytes simultaneously [20] . The microsphere immunoassay ( MIA ) that has been validated in this study was adapted from the method described by Luminex Corp ( 2000 ) and can be utilised as a routine serology testing protocol for leptospirosis . The development and validation of a high quality , reliable serological assay is pertinent to the ability of a laboratory to sero-diagnose diseases in humans . Assay development begins with the identification of a need for improved diagnostic capabilities and the benefits that can be obtained from such an assay . A Luminex microsphere immunoassay ( MIA ) for leptospirosis antibody detection has the potential to function both as a high sensitivity , high throughput screening assay as well as a high specificity assay for determination of serovar level antibodies . This paper assesses the leptospirosis MIA in human samples as a screening assay to determine reactive , equivocal and non-reactive samples . Validation is performed by comparison to the leptospirosis IgM ELISA and the current gold standard , the microscopic agglutination test ( MAT ) as the basis for defining the performance characteristics of the MIA .
The study protocol was approved by the Public and Environmental Health Research Committee and the Humans Ethics Committee , Queensland Health Forensic and Scientific Services . All human samples utilised in this study were de-identified and allocated a generic number . Sixteen Australian endemic pure leptospiral cultures , Table 1 , were grown for 5–7 days in 3mL EMJH broth at 30°C . These antigens were then quantitated using a Petroff-Hausser grid and centrifuged at 4°C for 25 mins . The supernatant was removed and the pellet resuspended in 500μL phosphate buffered saline ( pH 7 . 5 ) . All cultures were then diluted to obtain a concentration of 1 . 8 x 109 per mL . These diluted antigens were used to coat 16 individual Bio-Plex Pro Magnetic COOH Bead-sets . Coupled beads were then checked for sensitivity and specificity using rabbit anti-sera of known serovar and titre , obtained from MAT results ( See method below ) . This study utilised 200 serum samples which were selected from human serum samples submitted for routine leptospirosis serology to the WHO/FAO/OIE Collaborating Centre for Leptospirosis Reference and Research during 2012 and 2013 . These samples were submitted from Queensland hospitals and private laboratories . One hundred and eighty of these samples had leptospirosis IgM ELISA reactive serology , 12 had non-reactive leptospirosis IgM ELISA serology and the remaining 8 samples were not tested previously using leptospirosis IgM ELISA . All leptospirosis IgM ELISA testing was performed at a Queensland hospital or private laboratory prior to the samples being received at the WHO/FAO/OIE Collaborating Centre for Leptospirosis Reference and Research . Routine diagnostic MAT was performed on all samples at the WHO/FAO/OIE Collaborating Centre for Leptospirosis Reference and Research and results recorded against 16 routinely used , endemic serovars . Forty-eight additional samples with reactive serology for Dengue Virus ( 24 ) , Barmah Forest Virus ( 8 ) , Ross River Virus ( 8 ) or Rabies Virus ( 8 ) antibodies were obtained from the Queensland Health Public and Environmental Health Virology Laboratory . These samples had previously been tested by ELISA IgM ( Dengue virus ) , ELISA IgG ( Rabies virus ) or Alphavirus Hemagglutination Inhibition total antibody ( HAI ) ( Ross River virus and Barmah Forest virus ) and were used to assess whether cross reactions exist in the leptospirosis MIA . In addition to the 200 samples used for the validation , 20 sets of paired samples with a non-reactive leptospirosis acute sample and reactive leptospirosis convalescent sample on the MAT were also obtained and analysed using the MIA to determine a timeline for the detection of leptospiral antibody . The results for these twenty additional samples are shown separately . Leptospiral antigens were covalently coupled to individual Bio-Plex Pro Magnetic COOH bead-sets ( Table 2 ) using the Bio-Rad Amine Coupling kit and methods from Luminex Corp . Coupling is achieved via carbodiimide reactions involving the primary amino groups on the protein and the carboxyl functional groups on the bead surface . The bead yield per coupling reaction is approximately 2 , 500 beads per well ( in a 96-well microtitre plate ) . For optimum results in the MIA , the coupled beads were diluted 1:4 in Triton-X detergent and 100 beads in 100μL buffer were used for the immunoassay . Each individual coupled bead-set was diluted in phosphate buffered saline ( PBS ) to give a reading of approximately 100 beads per bead-set per well . The working dilution and specificity of each bead-set was validated prior to use in a diagnostic capacity by utilising serovar-specific rabbit antisera and the IgG method as described below , substituting the secondary antibody with an anti-rabbit IgG ( RPE ) . Bead-sets were considered to be valid for use if the targeted serovar produced an antibody response to that specific bead-set . Two microsphere immunoassays ( IgG and IgM ) were performed on 200 serum samples taken from the routine MAT submissions which included samples with MAT titres ( serial dilutions ) ranging from < 1:50 ( non-reactive ) to 1:6400 . Samples with an MAT titre between 1:50 and 1:200 were considered equivocal and samples with a titre 1:400 or above were considered reactive . Pooled convalescent serum from patients with recent leptospirosis infections , confirmed by PCR ( on acute sample ) and MAT , was used as the positive control serum in each microsphere immunoassay . Negative patient serum , confirmed by negative PCR and serology ) was pooled and used as negative control serum . These controls were monitored each run to ensure the assay was consistent . A 96-well filter plate was pre-wetted with 150μL PBS per well and vacuum applied . One-hundred μL of the diluted coupled beads were then added to each required well of the pre-wetted 96-well microtitre filter plate and vacuum applied . Serum samples for the IgG immunoassay were diluted 1:400 in PBS in 1mL micronic tubes . One-hundred μL of the diluted samples were added to the plate which was then incubated for 45 minutes on a shaker ( 750rpm ) at room temperature . The plate was then vacuum-washed three times with 150μL PBS per well . 100μL of a diluted secondary antibody ( anti-human IgG ) with a fluorescent tag ( RPE ) was added to each well followed by a second 45 minute incubation and vacuum wash as per previous step . Finally , 150μL PBS was added to each well and the plate placed back on a shaker at room temperature for at least 10 minutes prior to analysis . Serum samples for the IgM immunoassay were treated with Siemens Rheumatoid factor ( RF ) absorbent ( at a dilution of 1:2 ) and diluted to a final concentration of 1:800 in PBS . The plate was prepared as per the IgG immunoassay . The secondary antibody—anti-human IgM with a fluorescent RPE tag—was used in this assay for conjugation . All plate wells were then analysed using Luminex xMap technology on a BioPlex 200 Platform . The MIA results were reported as mean fluorescent intensity ( MFI ) and were deemed congruent or incongruent relative to the standard of comparison ( MAT ) . Cut-off values for reactive samples were determined using five reactive sera for each MAT titre ranging from 1:100 to 1:6400 ( Table 3 ) , and developing a standard curve ( R-Biopharm , 2012 ) using the titres obtained from MAT testing and comparing them with the mean fluorescent intensities from the MIA titrations . Fig . 1 shows the reactive sera MAT titres plotted against the MFI’s and the standard curve that resulted . From this curve , cut-off points were determined ( Table 4 ) . Positive/negative ratios were used to determine the cut-off point for non-reactive samples . During the validation and determination of cut-off points the results reactive high and reactive low were used to ensure that the MAT and the MIA results were comparable . All patient results were reported as reactive , non-reactive or equivocal . Sensitivity ( the ability of the MIA to correctly determine the presence of leptospiral antibody ) was determined by running known reactive ( true positive ) samples on the MIA and calculating the proportion of reactive samples detected . True positive samples are samples known to be reactive by paired sample testing with the Gold standard , the microscopic agglutination test . Assay specificity was assessed using two methods . The first involves running known non-reactive ( true negative ) samples and calculating the proportion of non-reactive samples detected by the MIA . True negatives are defined as non-reactive samples known to be non-reactive by paired sample testing with the Gold standard assay , microscopic agglutination test . False positives are reactive samples determined by the test assay ( MIA ) that are non-reactive by the gold standard . The second test of specificity ensured that samples that have been shown to have reactive serology for other pathogens are not cross reacting with the leptospirosis MIA . Within-run repeatability was determined by running four samples 20 consecutive times on one assay run for both IgG and IgM assays . Two of these samples had an equivocal result for at least one serovar on both assays , one sample had a reactive result for at least one serovar on both assays and the remaining sample was non-reactive for all 16 serovars for both IgG and IgM assays . Repeatability is also monitored continuously as a quality control measure by monitoring positive ( reactive ) and negative ( non-reactive ) controls with expected and accepted MFI ranges for each control serum in every assay . If the control serum results were outside of these ranges , the run was deemed to have failed and was repeated .
Of the 200 samples tested , 180 samples were reactive for leptospirosis IgM by ELISA ( Table 5 ) . Twelve samples were IgM ELISA non-reactive and eight samples did not have previous IgM ELISA results; comparisons could only be made with the MAT and MIA for these eight samples . The MAT confirmed 27 of the leptospirosis IgM ELISA reactive samples had evidence of leptospiral total antibody and suggested that the remaining 153 IgM ELISA reactive samples were non-reactive ( titre of < 1:50 ) . These results suggest a substantial gap in the diagnostic performance of the ELISA and the MAT . The MIA results ( in mean fluorescent intensity—MFI ) for the 27 MAT reactive samples also indicated reactive serology ( MFI > 1200 ) . Of the 173 non-reactive MAT samples , 126 were non-reactive on the MIA and the remaining 47 had low reactivity on the MIA , suggesting better sensitivity in the MIA . The results for five of these 47 samples , which have been confirmed as true leptospiral infections by PCR or blood culture , are shown in Table 6 . The MIA detected leptospiral antibody in 74 ( 41% ) of the 180 ELISA IgM reactive samples . The remaining 106 ELISA IgM reactive samples were non-reactive on the MIA and the non-reactive IgM ELISA samples were also non-reactive on the MIA . The 8 samples that were not previously tested by ELISA were non-reactive on the MIA also ( Table 5 ) . Of the 20 sets of additional paired samples with an MAT non-reactive acute sample and MAT reactive convalescent sample , 12 of these pairs demonstrated equivocal or reactive IgM MFI results for the acute samples with a significant rise in MFI in the convalescent samples on the IgM MIA . The results for the remaining eight pairs of samples were consistent between the MAT and the MIA . Table 7 shows the results for the paired samples comparing the MAT titre and the MIA IgM and IgG results . These samples were included in this study to show that IgM can be detected earlier or , at least at the same time , by the MIA when compared with the MAT in true leptospiral infections , as determined by a four-fold rise in serology . Of the 48 reactive viral serology samples only one showed reactive IgG and IgM serology for leptospirosis ( this sample was previously reactive for Dengue virus serology ) and the remaining 47 samples were non-reactive for both leptospirosis IgG and IgM . The four samples used to test within-run repeatability showed comparable results in each well across each of the 16 serovars . Table 8 shows the mean fluorescent intensity and standard deviation values for each of the four samples used in the repeatability testing for one of the serovars in the IgM immunoassay . Samples 1 and 2 were non-reactive . Sample 3 was reactive and sample 4 was in the equivocal range . The expected values were derived from comparison of the MIA mean fluorescent intensity with MAT titres . Repeatability was assessed across one run with one operator as , at the time of testing , only one operator was available to perform this testing .
The aim of diagnostic serology is to determine reactive and non-reactive samples for a particular infectious agent . By definition , a validated assay consistently provides test results that identify samples as being reactive or non-reactive for a selected analyte , and , by inference , accurately predicts the disease status of patients with a predetermined degree of statistical certainty [21] . The aim of this study was to validate a microsphere immunoassay ( MIA ) using Luminex xMap technology for diagnostic leptospirosis serology screening . The validation process was performed using a comparative method—that is comparing the new assay with the current gold standard assay . Sixteen leptospiral antigens have been coupled to 16 individual magnetic bead-sets and validated as a panel for routine diagnostic leptospirosis serology . This assay gives a qualitative result—Reactive , Equivocal or Non-Reactive and has the ability to determine recent from past infection by differentiating between IgM and IgG antibodies—something that is more difficult to achieve with microscopic agglutination testing ( MAT ) as this test can only determine total antibody . The class of antibody detected by the MIA can be used to determine the stage of the infection which is valuable for clinicians as it can determine treatment regimens for patients or in the case of a past infection , can suggest that something other than leptospirosis is causing symptoms . Information regarding new infections is also vital from a public health perspective as it can provide information on what serovars of leptospirosis are currently circulating and indicate the areas where these infections are occurring . All leptospirosis serology reactive samples by MAT were detected by MIA suggesting that congruence is 100% when compared to the MAT . Results from the non-reactive samples , as well as the paired samples suggest , however , that the MIA is more sensitive than the MAT . In true infections ( as demonstrated by paired sample serology testing with a minimum four fold rise in titre ) the MIA was able to detect low level antibody in the later stages of the acute phase as well as pick up higher levels of IgM antibody earlier in the immune phase of the infection . The MAT results indicated that these samples were non-reactive in the acute/early immune phase . The MAT generally becomes positive between day 8 and day 10 of infection [22] however , results from this validation suggest that the MIA could detect antibody in the earlier stages of infection development and increase the likelihood of the clinician submitting a convalescent sample for confirmation of infection status . The leptospirosis IgM ELISA has previously been shown to have poor specificity , as low as 41% , when used according to the manufacturer’s instructions [22] . All leptospirosis IgM ELISA reactive samples tested in Queensland pathology laboratories are sent to the WHO/FAO/OIE Collaborating Centre for Leptospirosis Reference and Research for confirmation testing . In this study it was found that of 180 leptospirosis IgM ELISA reactive samples only 15% ( 27/180 ) of these showed reactive results on the MAT . This could be due to a lower level of antibody which is not detected by the MAT at a dilution of 1 in 50 or a non-specific antibody reaction . In this study , 41% ( 74/180 ) of the leptospirosis IgM ELISA reactive samples had reactive IgG and/or IgM serology on the MIA , again suggesting the level of antibody in these particular samples may be too low for the MAT to detect . Also , this again shows that there may be some non-specific reactions occurring in the IgM ELISA , which are not seen on the MIA . The MIA is therefore advantageous as a screening test as it reduces the large numbers of samples that are unnecessarily sent for confirmation testing by MAT . It has also been suggested that false positivity can also occur in the leptospirosis IgM ELISA due to the presence of persistent IgM from past infections [23] . The MIA screening test eliminates these results by looking at the levels of the individual IgG and IgM antibodies across paired specimens . A low level or non-reactive IgM result and a plateaued reactive IgG would be suggestive of a past infection—something not currently visible on the leptospirosis IgM ELISA or the MAT . The MIA results suggest that the beads coated with leptospiral antigen are specific for leptospiral antibodies and show no cross-reactivity with other viral agents . The one case in this study where a Dengue Virus reactive serology sample also showed leptospiral antibodies is likely to be a true leptospirosis infection occurring simultaneously with a Dengue Virus infection . Leptospirosis and Dengue Virus infections are both common in northern parts of Queensland ( where this sample was from ) as they are both associated with tropical and sub-tropical regions where extreme weather events occur [24] . In many cases samples are submitted for both arbovirus testing ( including Dengue virus ) and leptospirosis testing at the same time . The results from the MIA show that reproducibility is possible and accurate when compared to the MAT . A major disadvantage of the MAT is attenuation of the live leptospiral cultures . It has been shown that over time , leptospiral cultures lose their antigenicity and therefore become less effective [25] . Also , day to day , the cultures can be different—more or less dense or contaminated—which makes reproducing results accurately a difficult task on the MAT . This issue is overcome with the MIA as the antigens ( leptospiral cultures ) are wild type cultures with a known passage number and are all diluted to a known concentration ( 1 . 8 x 109 ) prior to the bead coupling process . This ensures that there are equal amounts of each antigen available in every test . Another major advantage of the MIA over the MAT is that there is no need to maintain stocks of live leptospiral cultures for daily use . Pure cultures are only used in the MIA as antigens for bead coupling and these antigens can be centrifuged , diluted and frozen at -20°C for up to six months [26] . Currently , performing the MAT on a routine basis requires sub-culturing more than 200 tubes per week , maintaining four stocks of cultures . When comparing the MAT and the MIA the advantages of the latter are obvious . Firstly , the MIA is less time consuming—a full plate of 88 samples can be run in around three hours . To run the same number of samples on the MAT , it would take twice the time for a full panel of 16 serovars excluding analysis . The MIA is also less labour intensive as it does not require adding 16 individual cultures to each well on a 96 well plate for each individual patient . These savings combined as well as the reagent costs suggest that the MIA is also less costly than the MAT . An analysis of laboratory and assay costs shows that the current diagnostic serology method ( MAT ) is performed at a cost of $AUD6 . 95 ( excluding labour ) to the leptospirosis reference laboratory per sample per 16 serovars [27] . In comparison , the MIA costs $AUD4 . 95 per sample ( excluding labour ) per 16 serovars . Secondly , the MIA uses a total of 7μL of serum ( 2μL for the IgG assay dilutions and 5μL for the IgM assay dilutions ) compared with 50μL of serum used in the MAT . Thirdly , the MIA has the ability to detect and differentiate both IgG and IgM antibodies whereas the MAT can only detect total antibody and cannot give an accurate indication of the stage of infection in a single sample . The MIA can potentially include up to 500 analytes in the one assay , therefore , there is potential to be able to include all known leptospirosis serovars ( ~250 ) in one test at one time . Given the number of bead-sets available for microsphere immunoassays other applications could potentially involve the inclusion of a number of different viral and bacterial agents in one assay . For example , leptospirosis antibody detection and Dengue Virus antibody detection could be combined into one routine diagnostic test . In conclusion , the results from this validation suggest that the leptospirosis MIA is a beneficial diagnostic screening tool for leptospirosis serology testing . This assay is able to determine reactive , equivocal and non-reactive samples when compared to the MAT . It is able to differentiate leptospiral IgG antibodies from leptospiral IgM antibodies which will provide vital diagnostic information as well as provide a better epidemiological picture . Further investigations will include validation of each individual serovar to enable serovar specific results to be reported and validation of a microsphere immunoassay for detection of leptospiral antibodies in animal samples will also be looked at in the future . | Leptospirosis is a zoonotic disease caused by spirochaetes of the genus Leptospira and affects millions of people , worldwide , each year . Laboratory diagnosis of leptospirosis currently relies on methods that are flawed in many areas . Current methods are outdated , time consuming and expensive . They rely on a continuous supply of animal products ( rabbit anti-sera ) and require specialist expertise and equipment . The current gold standard diagnostic assay for leptospirosis ( MAT ) cannot determine IgG from IgM antibodies and relies on live cultures , which presents problems in the way of maintenance and attenuation . Development of a new diagnostic assay for serological diagnosis of leptospirosis that is specific , sensitive and able to discriminate between IgG and IgM classes of antibodies—as well as being more cost effective—will significantly improve the capabilities for detecting leptospirosis infections . It will provide medical professionals with more valuable diagnostic information and public health professionals with improved epidemiological information . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [] | 2015 | Validation of a Microsphere Immunoassay for Serological Leptospirosis Diagnosis in Human Serum by Comparison to the Current Gold Standard |
Neurocysticercosis ( NCC ) is the most common cause of acquired epilepsy in Taenia solium endemic areas , primarily situated in low-income countries . Diagnosis is largely based upon the “Del Brutto diagnostic criteria” using the definitive/probable/no NCC diagnosis approach . Neuroimaging and specific T . solium cysticercosis antibody detection results are at the mainstay of this diagnosis , while antigen detection in serum has never been included . This study aimed at evaluating the addition of antigen detection as a major diagnostic criterion , especially in areas where neuroimaging is absent . The B158/B60 monoclonal antibody-based enzyme-linked immunosorbent assay ( ELISA ) for the detection of circulating cysticercus antigen was carried out retrospectively on serum samples collected during a hospital-based study from 83 people with epilepsy ( PWE ) in an endemic area . The addition of antigen results as a major criterion allowed the correct diagnosis of definitive NCC in 10 out of 17 patients as opposed to 0/17 without antigen results in the absence of neuroimaging . A sensitivity of 100% and a specificity of 84% were determined for the diagnosis of active NCC using antigen ELISA . While the use of a higher cutoff improves the specificity of the test to 96% , it decreases its sensitivity to 83% . In areas where neuroimaging is absent , NCC diagnosis according to the existing criteria is problematic . Taking into account its limitations for diagnosis of inactive NCC , antigen detection can be of added value for diagnosing NCC in PWE by supporting diagnostic and treatment decisions . Therefore , we recommend a revision of the “Del Brutto diagnostic criteria” for use in resource poor areas and suggest the inclusion of serum antigen detection as a major criterion .
More than 80% of people with epilepsy ( PWE ) live in low-income countries [1] , where the prevalence of active epilepsy is approximately twice that of high-income countries [2] . Moreover , in many of those countries over 75% of PWE have no access to treatment with anti-epileptic medication [3] . Infectious diseases play a major role in the etiology of epileptic seizures and epilepsy in developing countries [1] . A recent review reported that 29% of PWE also had neurocysticercosis ( NCC ) [4] , caused by the larval stage of Taenia solium , a zoonotic parasite . The treatment of NCC depends on the stage of the disease and the number and localization of lesions . The determination of an optimal treatment is still a developing field of research , it may have to be tailored to individual cases and relies largely on results of neuroimaging techniques . However , there is frequently no or very limited access to/availability of these neuroimaging tools in low-income endemic countries . To assist in diagnosis , a number of immunodiagnostic tests have been developed , among which is the enzyme-linked immunoelectrotransfer blot ( EITB ) that detects specific antibodies against T . solium cysticerci in serum and was reported to have a high specificity ( 100% ) and sensitivity ( 98% ) [5] , [6] . This test is widely recognized; unfortunately it is expensive and in a format ( Western Blot ) not very applicable in most resource-poor laboratories in endemic areas . More field applicable enzyme-linked immunosorbent assay ( ELISA ) formats have been developed to detect specific antibodies and antigens in the serum , although they have until now failed to produce consistently good results of high specificity and high sensitivity [6] . However , research is ongoing into the development/identification of new markers for diagnostic tools [7]–[9] . The current antigen detecting ELISA's are based on monoclonal antibodies that detect excretory/secretory proteins produced by viable cysts [10] , [11] . As such , these tests detect viable cysts only , which has several epidemiological and clinical implications . In epidemiological studies , the presence of antigens indicates presence of infection , whereas presence of antibodies indicates exposure to the parasite , but not necessarily establishment of infection [12] . For the B158/B60 monoclonal antibody-based antigen ELISA a sensitivity of 90% ( 95% CI: 80%–99% ) and a specificity of 98% ( 95% CI: 97%–99% ) were determined for the detection of infected individuals , based on Bayesian analyses [12] . Currently , the only published diagnostic criteria are the “Del Brutto diagnostic criteria” [13] . However , these criteria have not been systematically validated [14] . Neuroimaging and EITB results provide the basis for most absolute and major criteria , while antigen detection in serum has never been included in the criteria . The aim of this study was to determine the added value of specific antigen detection in the diagnosis of NCC related epilepsy . Detection of circulating T . solium cysticercosis antigen was performed retrospectively on samples from PWE obtained from a hospital-based study carried out in northern Tanzania , in which clinical examinations , CT scanning and antibody detection had been carried out [15]–[17] .
The study and the use of human subjects for the study were approved by the National Institute for Medical Research ( NIMR ) , Tanzania . The samples were anonymized and transported in accordance to a material transfer agreement between HLH , NIMR , the Medical University Innsbruck , CDC , and ITM . Written informed consent was obtained from all participants or legal custodians in case of minors . PWE received free treatment for epilepsy and , in case of active NCC , anthelmintic treatment according to national guidelines . The study took place at Haydom Lutheran Hospital ( HLH ) situated in a remote area in the North of Tanzania . The serum and CSF samples were collected during a hospital-based study that has been described elsewhere [15]–[17] . Briefly , 212 PWE , all above 10 years of age , diagnosed by a neurologist ( ASW ) were followed up and computer tomography ( CT ) scans were performed . Epilepsy was defined as two or more unprovoked epileptic seizures and categorized according to an International League Against Epilepsy adjusted classification for resource-poor countries [18] . Venous blood samples were taken from 83 of the 212 examined PWE , including 28 of 29 PWE with highly suggestive or definite NCC lesions on CT scan , 7 of 9 PWE with lesions compatible with NCC and 48 of 126 without NCC lesions on CT scan . Due to ethical reasons it was not possible to take blood from all PWE . In addition 11 CSF samples from PWE with multiple cysts or calcifications on CT scan were collected . The diagnosis of NCC was based on the diagnostic criteria proposed by Del Brutto et al . ( 2001 ) [13] . The combination of absolute , major , minor and epidemiological criteria lead to the diagnoses of definitive , probable or no NCC ( Annex S1 ) . All participants had an epidemiological criterion , because Tanzania is endemic for cysticercosis [19] . Epilepsy is a clinical manifestation suggestive of NCC; hence all PWE had one minor criterion . In the present study , CT scans with contrast were available from all PWE . The CT scanner was a Toshiba Auklet Slice Spiral CT . The thickness of slices was 5 mm at skull base and 10 mm above the skull . Cystic lesions showing the scolex were regarded as an absolute criterion , multiple parenchymal calcifications and ring enhancing lesions , which are highly suggestive of NCC , as a major criterion and lesions that are compatible with NCC including equivocal single parenchymal calcifications as a minor criterion ( Annex S1 ) . Cystic and ring enhancing lesions were regarded as active NCC and calcifications only as inactive NCC . All samples were collected between May and August 2006 at HLH . After clotting , blood samples were centrifuged at 1000×/min for 5 minutes and serum was separated . All samples were initially kept at 4–8°C and after transport in September 2006 at −20°C . The Centers for Disease Control and Prevention , Atlanta , USA ( CDC ) performed the CDC-developed EITB [5] . Results of the latter analysis were published previously [16] . Samples were analyzed using the B158/B60 antigen ELISA ( Ag-ELISA ) at the Department of Biomedical Sciences of the Institute of Tropical Medicine , Antwerp , Belgium ( ITM ) [20] . Eight negative and 2 positive control serum samples were run on each plate . The plates were read using an automated spectrophotometer at 490 nm with a reference of 655 nm . The optical density ( OD ) of each serum sample was compared with a sample of negative serum samples ( n = 8 ) at a probability level of p = 0 . 001 ( cut-off calculation ) . A ratio for each sample was calculated by dividing the mean OD of the sample ( samples were tested in duplicate ) by the cut-off [21] . These ratios were used in the Receiver Operating Characteristic analysis ( see data analyses ) . CSF samples were run using the same methodology and cut-off calculation , albeit the pre-treatment of samples with trichloroacetic acid was not carried out . CSF samples were diluted ½ in phosphate buffered saline . The evaluation of the accuracy of Ag-ELISA was performed using a Receiver Operating Characteristic ( ROC ) analysis . This was done using “probable and definitive NCC” versus “no NCC” and “active” versus “inactive and no NCC” as reference tests ( as determined by the “Del Brutto diagnostic criteria” ) each in turn . The optimal cut-off ratio value was selected as the point on the ROC curve ( which displays estimated percentages of sensitivity and specificity at a selected cut-off value ) with the minimum distance to the ( 0 , 1 ) coordinate . The ROC curves were generated using the R software package [22] . The Fisher's exact test was used to compare circulating antigen levels in different diagnostic groups . Statistical significance was arbitrated at the 5% level . The Mann-Whitney U test was used to compare two groups in a non parametric variable .
Of the 83 PWE in whom antigen testing was performed , 34 were diagnosed with NCC ( following the “Del Brutto diagnostic criteria” ) , of which 17 were cases of definitive NCC and 17 of probable NCC . Six out of the 17 definitive NCC cases had active NCC lesions on CT scan . About twenty seven percent ( 22/83 ) of PWE were positive on serum Ag-ELISA . In the group of PWE with NCC ( according to the “Standard Del Brutto Diagnostic criteria” ) , circulating antigens were detected in 44 . 1% ( 15/34 ) ; 58 . 8% ( 10/17 ) in people with definitive NCC and 29 . 4% ( 5/17 ) in people with probable NCC . In the group of PWE without NCC , 14 . 3% ( 7/49 ) were positive on serum Ag-ELISA . The difference in proportion between PWE with NCC and PWE without NCC appeared to be statistically significant ( p = 0 . 005 , Fisher's exact test ) . In the group of people with NCC , 100% of active NCC cases ( 6/6 ) had a positive Ag-ELISA result in serum , which was significantly higher compared to people with inactive lesions , in whom 33 . 3% ( 8/24 ) were positive on antigen detection ( p = 0 . 005 , Fisher's exact test ) ( Figure 1 ) . The number of NCC lesions ( active or inactive ) was significantly associated with a positive Ag-result ( p = 0 . 009 , Mann-Whitney U ) . Four out of eleven ( 36 . 4% ) CSF samples were positive on Ag-ELISA . All patients with a positive result based on CSF were also positive based on serum . Regarding diagnosis of NCC , all people with probable NCC were negative on CSF Ag-ELISA ( 0/5 ) and 66 . 7% ( 4/6 ) with definitive NCC were positive . Whereas all samples from people with active NCC ( 4/4 ) were positive , all samples from people with inactive NCC were negative ( 0/7 ) . In Table 1 , a comparison is made between different types of NCC lesions and circulating antibody and antigen results in serum and CSF . In Table 2 , the “Standard Del Brutto diagnosis” ( = using neuroimaging , EITB , clinical , serological and epidemiological data; annex S1 ) is compared with the Del Brutto diagnosis without neuroimaging and 1 ) only EITB is used as a major criterion; 2 ) only Ag-ELISA is used as a major criterion; 3 ) EITB and Ag-ELISA are used as major criteria . In the last row , we have determined the change of diagnosis using the “Standard Del Brutto diagnosis” with Ag-ELISA as an added major criterion . The 83 PWE from our study are according to the “Standard Del Brutto diagnosis” , divided into 17 cases of definitive NCC , 17 of probable and 49 of no NCC . In the absence of neuroimaging , and if only EITB was considered as a major criterion , the 17 cases of definitive NCC could only be diagnosed as probable NCC . If Ag-ELISA was the only major criterion ( in absence of EITB results and neuroimaging ) , 10 cases would be diagnosed as probable NCC and the others as no NCC . If considering both EITB and Ag-ELISA as major criteria , 10 cases would be diagnosed as definitive NCC and 7 as probable NCC . Of the 17 cases with probable NCC only 6 and 5 cases can be diagnosed with EITB and Ag-ELISA as only majors , respectively . When combining both tests , 3 diagnoses of definitive NCC are made . When all criteria are considered , with inclusion of Ag-ELISA ( last row in the Table ) , 5 cases are allocated a diagnosis of definitive NCC . In the group of 49 cases with no NCC diagnosis , the addition of Ag-ELISA as a criterion yields 7 probable NCC diagnoses . In Figure 2 , the ROC curve shows the relationship between sensitivity and the complement of specificity for Ag-ELISA for the detection of NCC in PWE . The dot on the curve indicates the optimal cut-off value corresponding to the maximum sensitivity and specificity ( shortest distance to the point ( 0 , 1 ) in the diagram ) . The optimal cut-off ratio value of 0 . 81 corresponded to an estimated sensitivity of 53% ( 95% CI: 37–69% ) , specificity of 71% ( 95% CI: 58–82% ) . The area under the curve was 0 . 63 ( 95% CI: 0 . 52–0 . 75 ) . If a maximum specificity is requested , an optimal cut-off ratio value of 1 . 08 can be used , with a sensitivity of 44% ( 95% CI: 29–61% ) and a specificity of 90% ( 95% CI: 78–96% ) . In Figure 3 , the ROC curve shows the relationship between sensitivity and the complement of specificity for Ag-ELISA for the detection of active NCC in PWE . The optimal cut-off ratio value of 1 . 17 corresponded to an estimated sensitivity of 100% ( 95% CI: 61–100% ) , specificity of 84% ( 95% CI: 75–91% ) and the area under the curve was 0 . 95 ( 95% CI: 0 . 90–1 ) . Considering a prevalence of active NCC of 7 . 2% ( 6/83 ) in our group of PWE , a positive predictive value of 33% and negative predictive value of 100% can be calculated . If a maximum specificity is requested , an optimal cut-off ratio value of 39 . 97 can be used , with a sensitivity of 83% ( 95% CI: 44–97% ) and a specificity of 96% ( 95% CI: 90–99% ) . At this cut-off ratio , a positive predictive value of 63% and negative predictive value of 99% can be calculated .
A less frequent , but nevertheless important manifestation of NCC is extraparenchymal NCC . Due to the design of our study , which included only patients with epilepsy , most of our patients had parenchymal cysts and our results cannot be extrapolated to people with extraparenchymal NCC . However , using the “Del Brutto diagnostic criteria” might lead to an underestimation of Cysticercus racemosus forms , because a scolex is not visible [28] . Here also , serum circulating antigen detection could possibly be of great value , as racemose cysts are easily detectable by Ag-ELISA [29] . Presence or absence of live cysts can influence treatment decisions as it has consequences on the use of anthelminthics . In the absence ( availability and/or access ) of neuroimaging , antigen detection could be important to guide further diagnostic and treatment decisions , although treatment with anthelminthic medication based on positive serology results alone clearly is not advisable in resource-poor settings . Furthermore , antigen detection is considered helpful to follow up patients after treatment , where a relatively fast decrease in antigen levels is expected in contrast to antibody levels that can remain positive up to one year after treatment [30] . Clearly the use of the Ag-ELISA to identify NCC cases in a population of PWE ( NCC versus no NCC ) is less efficient as for the identification of active NCC cases . This can be explained by the number of inactive NCC cases , which the test cannot identify as it detects viable cysts only . However , clinically the differentiation between active NCC and inactive/no NCC is more important than between NCC and no NCC , because the presence or absence of live cysts will influence treatment decisions . Whereas PWE with active NCC should receive antiepileptic drugs , steroids and , depending on number and location of cysts visible on neuroimaging , anthelminthic treatment or even surgery , PWE with inactive or without NCC only need antiepileptic drugs [31] . Positive Ag-ELISA in PWE may help decide in resource-poor settings whether the patient absolutely needs neuroimaging or not , which justifies transportation of the patient to the nearest hospital with neuroimaging facilities . In conclusion , our results indicate that T . solium cysticercosis antigen detection can be of added value for diagnosis of NCC in PWE . As such , besides its use in epidemiological studies [12] , the value of antigen results for PWE can be twofold: 1 ) assist in diagnostic and treatment decisions as it can determine the presence/absence of viable cysts; 2 ) improve the diagnostic potential , especially in areas where neuroimaging techniques are not available/accessible . It is obvious that Ag-ELISA as a stand alone diagnostic technique cannot be sufficient for the detection of NCC; the use of serological techniques alone is insufficient for this diagnosis [32] . However , clearly more efforts should be put into developing a set of revised diagnostic criteria based on multiple diagnostic tools that can be implemented in resource-poor areas . | Neurocysticercosis is a parasitic infection of the central nervous system and a common cause of epilepsy in Taenia solium cysticercosis endemic countries . According to the current diagnostic criteria proposed by Del Brutto and colleagues , the diagnosis of neurocysticercosis is mainly based on neuroimaging and detection of specific antibodies . Unfortunately , especially neuroimaging is rarely available in endemic countries . The authors analyzed the value of a test that detects antigens that are excreted by living cysts in people with epilepsy . Different diagnostic scenarios and cut-off values are discussed with the respective sensitivity and specificity of the test . When using the antigen-detecting test , considerably more people with epilepsy were diagnosed correctly with neurocysticercosis . There are some concerns about possible false positive results in other cases . The test was useful for the detection of people with living cysts ( active neurocysticercosis ) , who need further diagnostic evaluation and specific treatment . The authors recommend the addition of this test in the diagnostic criteria for neurocysticercosis . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"medicine",
"infectious",
"diseases",
"infectious",
"diseases",
"of",
"the",
"nervous",
"system",
"neurological",
"disorders",
"cysticercosis",
"neurology",
"neglected",
"tropical",
"diseases",
"parasitic",
"diseases"
] | 2012 | Added Value of Antigen ELISA in the Diagnosis of Neurocysticercosis in Resource Poor Settings |
Subsets and Splits
No saved queries yet
Save your SQL queries to embed, download, and access them later. Queries will appear here once saved.