Abstract
Psoriasis is an inflammatory skin disorder that exhibits multifactorial mode of inheritance. In addition to the well-known susceptibility locus PSORS1 many other loci have been shown to be implicated in the genetic predisposition for disease. However, interactions between loci have not been thoroughly explored. Thus, we measured the effect of potential interaction between human leukocyte antigen (HLA)-C, CSTA and D1S236 at PSORS1, PSORS4 and PSORS5, respectively, in the development of psoriasis. Analysis of 130 Caucasian psoriatic families showed that the risk to an HLA-Cw6 +ve individual who carries two copies of the risk allele at both the CSTA and D1S2346 is 105 times the risk to an HLA-Cw6 +ve individual who does not carry any risk alleles at the CSTA or D1S2346. This is the first demonstration of an interaction between risk alleles in three susceptibility loci suggesting possible functional interaction between genes in these loci, which might explain the complexity of the pathogenesis of psoriasis.
Similar content being viewed by others
Introduction
Psoriasis is a chronic inflammatory disease affecting ∼4% of the Caucasian population.1 Twin and family-based studies have suggested that psoriasis is a complex genetic trait associated with multiple factors.2 Despite extensive efforts focused on identifying psoriasis susceptibility loci using genome-wide studies, only the PSORS1 locus, located within the major histocompatibility complex (MHC) on chromosome 6p21.3, has been consistently replicated.3 The PSORS1 locus is known to be by far the strongest susceptibility locus and although 19 loci on 15 different chromosomes have been implicated in total, a meta-analysis of six genome-wide studies revealed evidence for linkage to only two regions.4 Because of the difficulties in replicating and identifying actual susceptibility genes within these identified regions, gene–gene interactions have not been extensively explored.
Although epistatic interactions have been documented in other common complex diseases such as Alzheimer's and Crohn's disease,5, 6 it was almost a decade ago that for the first time an interaction between the PSORS4 locus and human leukocyte antigen (HLA)-Cw6 at the PSORS1 locus was identified7 and it has been suggested that the effect of the PSORS1 locus should be conditioned on in future studies of non-major histocompatibility complex susceptibility loci.4 We have previously shown that the CSTA ‘TCC’ haplotype located at PSORS5 is only associated with psoriasis in individuals harbouring the risk allele HLA-Cw6.8 In this study, we investigated the interaction between three susceptibility loci, PSORS1, PSORS4 and PSORS5 in 130 Caucasian psoriatic families and identified a highly significant close interaction between them and development of psoriasis.
Materials and methods
Psoriasis families
Families with psoriasis were recruited from the Sheffield and Edinburgh regions of the United Kingdom, after obtaining their informed consent. This was approved by the Sheffield Local Ethics Committee and conformed to the guidelines set forth in the Helsinki protocol. The total number of families involved in this study was 130 Caucasian families from Scotland and North England. All of the families recruited from Sheffield and Edinburgh were Caucasian. All of the psoriasis patients and their family members used in this study were diagnosed and managed by a single consultant dermatologist (MJC) in a specialist psoriasis clinic. Overall, the psoriasis patient diagnosis was: 121 chronic plaque; 60 chronic plaque and guttate; 10 guttate; one pustular; one chronic plaque and pustular; and one chronic plaque, guttate and pustular. The mean age at onset for all the patients was 16.26 years (age range: 0–52) and 145 of the 194 cases (74.74%) reported a positive family history of psoriasis.
Genotyping and data analysis
Genomic DNA was extracted from whole blood, obtained from all family members included in this study, using standard protocols (Amersham Biosciences, Little Chalfont, UK) and stored in 96-well microtiter plates, with each well containing 500 μl of DNA (100 ng μl−1). For the restriction fragment length polymorphism genotyping of the PSORS1 (HLA-Cw6) locus, genomic DNA was amplified by PCR using the conditions and primers as described previously.9 PCRs were conducted using fluorescently labelled primers (HEX and FAM dyes) for the microsatellite marker genotyping at PSORS4. Conditions, primers and details for the individual microsatellite markers are listed in Table 1. Alleles at each microsatellite marker were numbered according to their frequency with the most frequent being labelled allele 1. The genotyping methodology for PSORS5 locus has been described previously.8 HLA-Cw6 (PSORS1) and CSTA (PSORS5) genotypes were analysed in 130 and 126 psoriatic families, respectively, using the Transmit10 and FBAT programs.11 Both programs allow the user to test for association/linkage between disease phenotype and genetic markers and can handle both bi-allelic and multi-allelic data while performing the classical TDT and sib-TDT. The genotyping at seven microsatellite markers on chromosome 1q21 (PSORS4) in 130 psoriatic families were also analysed using the Transmit and FBAT programs. Alleles for each individual microsatellite marker were coded by frequency with alleles of <5% frequency combined together.
Statistical analysis
Conditional logistic regression (CLR) was used to model and quantify the association between genetic polymorphisms and disease risk. In a family-based CLR analysis one would expect to use unaffected siblings as matched controls. In the absence of such sibling controls, we used the method of Cordell et al.12, 13 to generate pseudo controls for the affected offspring conditional on the parental genotypes. The genotypes of the pseudo controls represent the untransmitted parental alleles. Several families in our study had more than one affected offspring. In the presence of linkage, nuclear families with more than one affected sib cannot be assumed to be independent. We can allow for this non-independence of observations by choosing a robust variance estimator. To determine the mode of inheritance for each locus (which dictates the coding of the genotypes in the regression analysis), we selected the mode of inheritance that provided the largest log-likelihood. For the D1S2346 locus, we considered to be the risk allele and grouped all the others together. The mode of inheritance for the D1S2346 locus was then determined based on this two-allele configuration. We previously showed that the CSTA −190T/+162C/+344C (CSTA TCC) allele is only associated with psoriasis in those individuals carrying the HLA-Cw6 risk allele.8 As a result of this finding, we stratified our data by presence of the HLA-Cw6 risk allele and performed the analyses in the two strata (presence/absence of risk allele) separately. Within each stratum that we tested, using CLR, for association between the putative risk conferring alleles at the D1S2346 locus and psoriasis. Where an association was seen with the alleles at the D1S2346 locus, we again used CLR to quantify the relationship between the risk of psoriasis and carriage of the risk alleles at the CSTA and D1S2346 loci by inclusion of both terms in the logistic model. We also determined whether there was any evidence of a super-multiplicative (synergistic) effect by inclusion of an interaction term in the model. The P-values for the regression models are calculated using the likelihood ratio test for nested models. This provides a test of the ‘improvement’ in the fit of a model following inclusion of further terms (loci) in the logistic model rather than tests of individual terms (loci) in the model (Wald's tests). Both the generation of sibling pseudo controls and the CLR analysis were both undertaken using STATA 10 (StataCorp LP, College Station, TX, USA).
Results
A total of 130 psoriatic families were genotyped for HLA-Cw6. HLA-Cw6 allele frequencies in these 130 families were found to be 73.2% HLA-Cw6 −ve and 26.8% HLA-Cw6 +ve. Our analysis shows the over-transmission of the HLA-Cw6 +ve allele to diseased individuals with a Z score of 3.575 (P=0.00035) (Table 2). At the PSORS4 locus, seven microsatellite markers were genotyped in all 130 families, whereby the strongest evidence for linkage is at marker D1S2346 (Z score 5.8, P<0.000001). Of the other six markers, only D1S498 produced evidence for linkage (a Z score⩾3) with a Z score of 3 (P=0.0027). When the relationship between the PSORS1 and PSORS4 loci was investigated, six of the seven microsatellite markers showed an increased over-transmission when stratified by HLA-Cw6 (Table 2). The evidence for linkage increased with the presence of the HLA-Cw6 +ve allele except for marker D1S2346, which decreased slightly although still showed evidence for over-transmission with the +ve HLA-Cw6 allele. We showed previously that CSTA ‘TCC’ at PSORS5 is strongly associated with psoriasis (Z score 3.8, P=0.00013) (ref. 9). The single strongest association from these three SNPs was found with CSTA c162T>C (Z score 3.6, P=0.00025). When the relationship between PSORS1 and PSORS5 was investigated, the ‘TCC’ was over-transmitted with HLA-Cw6 +ve families (Z score 3.81, P=0.0001). When the CSTA c162T>C was analysed with HLA-Cw6, the +162C allele was over-transmitted with HLA-Cw6 +ve allele (Z score 3.49, P=0.00054). For the investigation into a possible three-loci interaction, D1S2346 and CSTA c162T>C markers were chosen as susceptibility alleles for PSORS4 and PSORS5, respectively.
As we have noticed previously,9 CSTA significantly increases the risk only in HLA-Cw6 +ve individuals. This is not the case for D1S2346, which significantly increase the risk of psoriasis in both HLA-Cw6 +ve and HLA-Cw6 −ve individuals (Table 3). There does appear to be a difference though in the magnitude of the OR in HLA-Cw6 +ve versus HLA-Cw6 −ve individuals. In a logistic regression model, significance at two loci (when both are included in the model) implies a multiplicative risk across the two loci. Hence, we have multiplicative risk within and between the loci. This means that for HLA-Cw6 +ve individuals, each copy of the risk allele at the CSTA and D1S2346 loci increases the odds of disease multiplicatively by 1.9 and 5.39, respectively. Therefore, this suggests that the risk to an HLA-Cw6 +ve individual who carries two copies of the risk allele at both the CSTA and D1S236 loci is 1.9 × 1.9 × 5.39 × 5.39=105 times the risk to an HLA-Cw6 +ve individual who does not carry any risk alleles at the CSTA or D1S2346 loci.
Discussion
It has long been accepted that psoriasis has a genetic basis, albeit a complex one that involves the interaction of multiple genes and environmental factors.2 In this study, we investigated possible epistatic interactions between PSORS1, PSORS4 and PSORS5 loci and psoriasis. In selecting markers for genotyping, we chose the well-known HLA-Cw6 allele for the PSORS1 locus and CSTA c162T>C for PSORS5, a locus on chromosome 3q21, previously shown to be associated with psoriasis.9, 14 In addition, refinement work for the PSORS4 locus has narrowed down the region to the proximity of the D1S2346 microsatellite marker.15 We have therefore genotyped seven microsatellite markers on chromosome 1q21, four of which (D1S498, D1S1664, D1S305 and D1S2346) have been previously reported to show linkage to psoriasis.8, 15 Three additional markers (D1S2343, D1S2399 and D1S2858) were chosen to fit between the other markers to reduce the inter-marker spacing. Our data shows that the strongest evidence for linkage is at marker D1S2346 (Z score 5.8, P<0.000001). This marker is located distal to the epidermal differentiation complex, a 2.0 Mb region that includes at least 45 genes that become activated in terminal stages of keratinocyte differentiation involved in the development, maturation and cross-linking of the epidermis16 and ∼0.5 Mb away from the recently identified LCE3C_LCE3B-del locus associated with psoriasis.17 Interestingly, this locus is in a distinct linkage-disequilibrium block compared with the other genes of the epidermal differentiation complex reported for association (blocks of low linkage-disequilibrium for rs4112788, a tag SNP for LCE3C_LCE3B-del),17 suggesting a second susceptibility locus on chromosome 1q21 region. Although there seems to be a redundancy in genes with a role in epidermal barrier function contributing to disease susceptibility, recent studies highlight also the synergistic role of the immune system in disease pathogenesis by identifying several new susceptibility loci for psoriasis18, 19, 20, 21 and suggesting a close interaction between epidermal barrier impairment and both innate and adaptive immune system dysfunction for disease development.22
In the present study, data analysis point towards a very close interaction between the three susceptibility loci PSORS1, PSORS4 and PSORS5, suggesting a striking 105-fold risk for an HLA-Cw6 +ve individual who carries two copies of the risk allele at both the CSTA and D1S2346 loci. Such increased risk could translate functional interaction between proteins encoded by genes in these loci. This is in keeping with the recent study by the Genetic Analysis of Psoriasis Consortium and the welcome trust Case Control Consortium showing an interaction between HLA-C and endoplasmic reticulum aminopeptidase 1 (ERAP1), encoding an amino peptidase that regulates the quantity of peptides bounds to class I molecule such as HLA-C.22 Interestingly, cystatins have been shown to modulate lysosomal cathepsin activities in antigen presentation.23 In fact, lysosomal cysteine proteases regulate antigen presentation by both major histocompatibility complex class II and class I within the endosomal/lysosomal compartment.24 In addition, cathepsin activity is decreased in human dendritic cells in response to HIV-1 infection, which enhance HIV-1 survival. It also decreased HIV-1 antigen processing and presentation to T cells.25 Cathepsins are also essential for the progression of CD8+ T-cell-mediated autoimmune diabetes.26 The regulation of cathepsin activity by cystatins seems to be important in antigen presentation and T-cell response conforming the fundamental role of immune system in the pathogenesis of psoriasis.
Our study provides strong evidence of an interaction between three susceptibility loci and incites to explore the possibility of functional interaction between recently identified susceptibility alleles.18, 19, 20, 21
References
Griffiths, C. E. & Barker, J. N. Pathogenesis and clinical features of psoriasis. Lancet 370, 263–271 (2007).
Elder, J. T., Bruce, A. T., Gudjonsson, J. E., Johnston, A., Stuart, P. E., Tejasvi, T. et al. Molecular dissection of psoriasis: integrating genetics and biology. J. Invest. Dermatol. 130, 213–226 (2010).
Sagoo, G. S., Cork, M. J., Patel, R. & Tazi-Ahnini, R. Genome-wide studies of psoriasis susceptibility loci: a review. J. Dermatol. Sci. 35, 171–179 (2004).
Nair, R. P., Stuart, P. E., Nistor, I., Hiremagalore, R., Chia, N. V., Jenisch, S. et al. Sequence and haplotype analysis supports HLA-C as the psoriasis susceptibility 1 gene. Am. J. Hum. Genet. 78, 827–851 (2006).
Roberson, E. D. & Bowcock, A. M. Psoriasis genetics: breaking the barrier. Trends Genet. 26, 415–423 (2010).
Combarros, O., van Duijn, C. M., Hammond, N., Belbin, O., Arias-Vásquez, A., Cortina-Borja, M. et al. Replication by the epistasis project of the interaction between the genes for IL-6 and IL-10 in the risk of Alzheimer's disease. J. Neuroinflammation 23, 6–22 (2009).
van Heel, D. A., Dechairo, B. M., Dawson, G., McGovern, D. P., Negoro, K., Carey, A. H. et al. The IBD6 Crohn's disease locus demonstrates complex interactions with CARD15 and IBD5 disease-associated variants. Hum. Mol. Genet. 12, 2569–2575 (2003).
Capon, F., Semprini, S., Dallapiccola, B. & Novelli, G. Evidence for interaction between psoriasis-susceptibility loci on chromosomes 6p21 and 1q21. Am. J. Hum. Genet. 65, 1798–1800 (1999).
Vasilopoulos, Y., Walters, K., Cork, M. J, Duff, G. W., Sagoo, G. S. & Tazi-Ahnini, R. Association analysis of the skin barrier gene cystatin A at the PSORS5 locus in psoriatic patients: Evidence for interaction between PSORS1and PSORS5. Eur. J. Hum. Genet. 16, 1002–1009 (2008).
Tazi-Ahnini, R., Camp, N. J., Cork, M. J., Mee, J. B., Keohane, S. G., Duff, G. W. et al. Novel genetic association between the corneodesmosin (MHC S) gene and susceptibility to psoriasis. Hum. Mol. Genet. 8, 1135–1140 (1999).
Clayton, D. A. A generalization of the transmission disequilibrium test for uncertain haplotypes transmission. Am. J. Hum. Genet. 65, 1170–1177 (1999).
Rabinowitz, D. & Laird, N. A unified approach to adjusting association tests for population admixture with arbitrary pedigree structure and arbitrary missing marker information. Hum. Hered. 50, 211–223 (2000).
Cordell, H. J. Estimation and testing of genotype and haplotype effects in case-control studies: comparison of weighted regression and multiple imputation procedures. Genet. Epidemiol. 30, 259–275 (2006).
Samuelsson, L., Stiller, C., Friberg, C., Nilsson, C., Inerot, A. & Wahlström, J. Association analysis of cystatin A and zinc finger protein 148, two genes located at the psoriasis susceptibility locus PSORS5. J. Invest. Dermatol. 122, 1399–1400 (2004).
Capon, F., Semprini, S., Chimenti, S., Fabrizi, G, Zambruno, G, Murgia, S et al. Fine mapping of the PSORS4 psoriasis susceptibility region on chromosome 1q21. J. Invest. Dermatol. 116, 728–730 (2001).
Mischke, D., Korge, B. P., Marenholz, I., Volz, A. & Ziegler, A. Genes encoding structural proteins of epidermal cornification and S100 calcium-binding proteins form a gene complex (‘epidermal differentiation complex’) on human chromosome 1q21. J. Invest. Dermatol. 106, 989–992 (1996).
de Cid, R., Riveira-Munoz, E., Zeeuwen, P. L, Robarge, J., Liao, W., Dannhauser, E. N. et al. Deletion of the late cornified envelope LCE3B and LCE3C genes as a susceptibility factor for psoriasis. Nat. Genet. 41, 211–215 (2009).
Stuart, P. E., Nair, R. P., Ellinghaus, E., Ding, J., Tejasvi, T., Gudjonsson, J. E. et al. Genome-wide association analysis identifies three psoriasis susceptibility loci. Nat. Genet. 42, 1000–1004 (2010).
Ellinghaus, E., Ellinghaus, D., Stuart, P. E., Nair, R. P., Debrus, S., Raelson, J. V. et al. Genome-wide association study identifies a psoriasis susceptibility locus at TRAF3IP2. Nat. Genet. 42, 991–995 (2010).
Sun, L. D., Cheng, H., Wang, Z. X., Zhang, A. P., Wang, P. G., Xu, J. H. et al. Association analyses identify six new psoriasis susceptibility loci in the Chinese population. Nat. Genet. 42, 1005–1009 (2010).
Hüffmeier, U., Uebe, S., Ekici, A. B., Bowes, J., Giardina, E., Korendowych, E. et al. Common variants at TRAF3IP2 are associated with susceptibility to psoriatic arthritis and psoriasis. Nat. Genet. 42, 996–999 (2010).
Strange, A., Capon, F., Spencer, C. C., Knight, J., Weale, M. E., Allen, M. H. et al. A genome-wide association study identifies new psoriasis susceptibility loci and an interaction between HLA-C and ERAP1. Nat. Genet. 42, 985–990 (2010).
Hsing, L. C. & Rudensky, A. Y. The lysosomal cysteine proteases in MHC class II antigen presentation. Immunol. Rev. 207, 229–241 (2005).
Honey, K. & Rudensky, A. Y. Lysosomal cysteine proteases regulate antigen presentation. Nat. Rev. Immunol. 6, 472–482 (2003).
Harman, A. N., Kraus, M., Bye, C. R., Byth, K., Turville, S. G., Tang, O. et al. HIV-1-infected dendritic cells show 2 phases of gene expression changes, with lysosomal enzyme activity decreased during the second phase. Blood 114, 85–94 (2009).
Yamada, A., Ishimaru, N., Arakaki, R., Katunuma, N. & Hayashi, Y. Cathepsin L inhibition prevents murine autoimmune diabetes via suppression of CD8(+) T cell activity. PLoS One 5, e12894 (2010).
Acknowledgements
We are grateful to patients and their family members for participating in this study. This work was supported by research grants from the Psoriasis Association UK and Cecil King Memorial Foundation.
Author information
Authors and Affiliations
Corresponding authors
Ethics declarations
Competing interests
The authors declare no conflict of interest.
Rights and permissions
About this article
Cite this article
Vasilopoulos, Y., Sagoo, G., Cork, M. et al. HLA-C, CSTA and DS12346 susceptibility alleles confer over 100-fold increased risk of developing psoriasis: evidence of gene interaction. J Hum Genet 56, 423–427 (2011). https://doi.org/10.1038/jhg.2011.33
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1038/jhg.2011.33
Keywords
This article is cited by
-
Genetic Epidemiology of Psoriasis
Current Dermatology Reports (2014)
-
Cysteine proteases: mode of action and role in epidermal differentiation
Cell and Tissue Research (2013)
-
Genetics of Psoriasis: Evidence for Epistatic Interaction between Skin Barrier Abnormalities and Immune Deviation
Journal of Investigative Dermatology (2012)
