Testing of ∼25,000 putative functional single-nucleotide polymorphisms (SNPs) across the human genome in a genetic association study has identified three psoriasis genes, IL12B, IL23R, and IL13. We now report evidence for the association of psoriasis risk with missense SNPs in the interferon induced with helicase C domain 1 gene (IFIH1). The rare alleles of two independent SNPs were associated with decreased risk of psoriasis—rs35667974 (Ile923Val): odds ratio (OR) for minor allele carriers is 0.43, P=2.36 × 10−5 (2,098 cases vs. 1,748 controls); and rs10930046 (His460Arg): OR for minor allele carriers is 0.51, P=6.47 × 10−4 (2,098 cases vs. 1,744 controls). Compared to noncarriers, carriers of the 923Val and/or 460Arg variants were protected from psoriasis (OR=0.46, P=5.56 × 10−8). To our knowledge, these results suggest that IFIH1 is a previously unreported psoriasis gene.
Abbreviations
ORodds ratio
SNPsingle-nucleotide polymorphism
T1Dtype 1 diabetes.
Introduction
Psoriasis is a common, chronic, T-cell-mediated inflammatory disease of the skin (
Nestle et al., 2009
) affecting 2–3% of whites of European descent but fewer Asians and Africans (Campalani and Barker, 2005
). It is considered a multifactorial, complex disease involving both genetic and environmental factors (Bowcock and Krueger, 2005
; Smith et al., 2009
). Genetic contribution to the disease is shown by increased concordance in monozygotic compared to dizygotic twins (72% vs. 15–23%, respectively, for northern Europeans) and the observation that upward of 71% of patients with childhood psoriasis has a positive family history (Bowcock and Krueger, 2005
). Indeed, the HLA variant HLA-Cw*0602 strongly predisposes individuals to psoriasis (Nair et al., 2006
; Fan et al., 2008
), and multiple single-nucleotide polymorphisms (SNPs) in at least seven genes, which include IL12B, IL13, IL23R, STAT2/IL23A, TNFAIP3, TNIP1, and LCE, have been convincingly associated with psoriasis risk (Cargill et al., 2007
; Chang et al., 2008
; Liu et al., 2008
; Nair et al., 2009
; Zhang et al., 2009
; de Cid et al., 2009
). Variants in β-defensin, CDKAL1, and other genes may also affect risk of psoriasis (Lesueur et al., 2007
; Capon et al., 2008
; Hollox et al., 2008
; Wolf et al., 2008
; Li et al., 2009
). However, these variants do not fully account for the genetic contribution to psoriasis, underscoring the necessity of further search for other genetic variants.Results
We recently conducted a large-scale genetic association study testing 25,215 putative functional SNPs across the genome in three independent white psoriasis case–control sample sets. On the basis of these data, we reported results that validated IL12B and provided the first evidence for IL23R and IL13 as psoriasis risk genes (
Cargill et al., 2007
; Chang et al., 2008
).To continue the search for psoriasis genes, we prioritized the remaining markers from our genetic study for additional genotyping and association testing. Excluding markers in the HLA, IL12B, IL23R, and IL13 loci, we individually genotyped and analyzed 337 autosomal markers in one or more of the three sample sets (all markers had Hardy–Weinberg equilibrium P>0.001 in controls). An association test showed that 113 markers were associated with psoriasis risk (allelic P<0.05 in all samples tested) (Supplementary Table S1 online); these markers include one SNP in PADI4, which is involved in the genetics of rheumatoid arthritis (
Suzuki et al., 2003
), and one in FNDC1, which is highly expressed in the epidermis of psoriatic skin but barely detectable in normal skin (Anderegg et al., 2005
).- Supplementary Table S1
The most significant marker was rs35667974 in the IFIH1 gene on chromosome 2 (allelic P=1.74 × 10−5; 1,436 cases/1,379 controls). Another SNP in IFIH1, rs10930046, was also associated with psoriasis risk (allelic P=0.0055). Further testing of these two polymorphisms in a fourth sample set (667 cases/371 controls) replicated rs10930046 (replication allelic P=0.031, Supplementary Table S1 online), and the odds ratio (OR) of rs35667974 was in the same direction as in the other three sample sets (replication allelic P=0.42, Supplementary Table S1 online). Both markers are missense polymorphisms, resulting in Ile923Val and His460Arg variants, respectively, in the IFIH1 polypeptide. The minor allele frequencies of both the 923Val and 460Arg variants were low, 2.2 and 2.1%, respectively, in the control population. Because of the low allele frequency, very few individuals carried two copies of either of the protective (minor) variants; therefore, testing the carrier status of the protective alleles did not substantially alter the association results (Table 1). Compared with noncarriers, carriers of 923Val or 460Arg variant were protected from psoriasis.
Table 1Association of missense variants in IFIH1 with psoriasis
| Count | Carrier freq | CC+CT vs. TT | |||||||
|---|---|---|---|---|---|---|---|---|---|
| RS number | Variation | Samples | CC | CT | TT | Sum | CC+CT | OR (95% CI) | P-value |
| rs35667974 | Ile923Val | Case | 3 | 37 | 2,058 | 2,098 | 1.91% | 0.43 (0.29–0.64) | 2.36 × 10−5 |
| Control | 2 | 72 | 1,674 | 1,748 | 4.23% | ||||
| rs10930046 | His460Arg | Case | 3 | 39 | 2,056 | 2,098 | 2.00% | 0.51 (0.35–0.76) | 6.47 × 10−4 |
| Control | 4 | 65 | 1,675 | 1,744 | 3.96% | ||||
Abbreviations: CI, confidence interval; OR, odds ratio.
1 C allele corresponds to 923Val and 460Arg variants.
2 HWE P>0.05 (exact test) in controls of individual sample sets for both SNPs.
3 Frequency of the carriers of the protective allele.
4 Dominant model, adjusted for sample set; test of OR heterogeneity P>0.05 among sample sets.
These two IFIH1 markers were not correlated with each other (r2=0.001 in the control samples). As shown in Table 2, only one case and two controls carried both the rare variants. Regression analyses showed that both markers remained significant when adjusted for the other (adjusted P<0.05), suggesting that the two rare missense variants independently protected from psoriasis. Because of these findings, we further tested the combined effect of these two variants on psoriasis risk. Compared with noncarriers (i.e., individuals homozygous for both the 460His and 923Ile variants of the IFIH1 polypeptide), carriers of the protective variants 460Arg and/or 923Val were significantly associated with lower risk of psoriasis (OR=0.46, P=5.56 × 10−8 in all samples combined, Breslow–Day P=0.11 for OR heterogeneity). Carriers with at least one protective allele (460Arg, 923Val, or both) were common, comprising 8.1% of the combined controls and 3.9% of the combined patient samples.
Table 2Association of psoriasis with carriers of the IFIH1 460Arg and/or 923Val variants
| IFIH1 variant | Count (frequency) | |||||
|---|---|---|---|---|---|---|
| Protective alleles | Residue 460 | Residue 923 | Case | Control | OR (95% CI) | P-value |
| Carriers | ||||||
| Subgroup | Arg | Ile | 39 (1.86%) | 72 (4.13%) | ||
| His | Val | 41 (1.96%) | 67 (3.85%) | |||
| Arg | Val | 1 (0.05%) | 2 (0.11%) | |||
| Combined | 81 (3.87%) | 141 (8.09%) | ||||
| Noncarriers | His | Ile | 2,011 (96.13%) | 1,601 (91.91%) | ||
| Carriers vs. noncarriers | 0.46 (0.35–0.61) | 5.56 × 10−8 | ||||
Abbreviations: CI, confidence interval; OR, odds ratio.
1 Carriers of protective IFIH1 alleles have at least one copy of the arginine residue at position 460 or a valine residue at position 923. Noncarriers are homozygous for the histidine residue at position 460 and the isoleucine residue at position 923.
2 Protective alleles in boldface.
3 χ2-Test, adjusted for sample set.
Discussion
The above results suggest that missense variants in IFIH1 modulate risk of psoriasis in the white, North American population. The observed statistical evidence are at the level of P=2.36 × 10−5, 6.47 × 10−4, and 5.56 × 10−8 for rs35667974 (Ile923Val), rs10930046 (His460Arg), and both SNPs combined, respectively. In the context of our study that tested ∼25,000 putative functional SNPs, the carrier association of both SNPs with psoriasis remains significant when adjusted for multiple testing (Bonferroni corrected P=0.0014), suggesting that variants in IFIH1 are genuinely associated with psoriasis risk.
Pleiotropic effect of IFIH1 in modulating autoimmunity
A role for IFIH1 in the etiology of psoriasis is bolstered by previously established genetic connections between this gene and other autoimmune and inflammatory diseases and the observation that autoimmune and inflammatory diseases share overlapping genetic factors (
Li and Begovich, 2009
). Definitive evidence for a role of IFIH1 in autoimmunity comes from type 1 diabetes (T1D), where genome-wide significance is observed for at least two distinct IFIH1 SNPs: a common missense polymorphism rs1990760 (Ala946Thr; risk allele frequency=60% in whites) (Smyth et al., 2006
) and the rare missense polymorphism rs35667974 tested in our study (Nejentsev et al., 2009
). The effect of the common SNP in T1D is weak (OR=0.86), whereas that of the rare SNP is stronger (OR=0.51). There is also strong evidence for a role of rs1990760 in systemic lupus erythematosus (Gateva et al., 2009
), inconsistent evidence in multiple sclerosis (Martinez et al., 2008
; Couturier et al., 2009
; Enevold et al., 2009
) and Graves’ disease (Sutherland et al., 2007
; Penna-Martinez et al., 2009
), and no evidence for celiac disease (Smyth et al., 2008
) or rheumatoid arthritis (Marinou et al., 2007
).SNP rs1990760 has not been tested in our psoriasis sample sets for technical reasons. Another SNP, rs3747517 (His843Arg), in linkage disequilibrium with rs1990760 (r2=0.61) and associated with T1D (
Smyth et al., 2006
), is not significant in the DNA pools of our Sample Set 1 (data not shown). However, assuming that the frequency of the rs1990760 risk allele in the unrelated controls is 60% and an allelic OR of 1.15, as observed in the T1D and systemic lupus erythematosus studies (Smyth et al., 2006
; Gateva et al., 2009
), one would require 1,701 case patients and 1,701 controls to observe an association with disease status with 80% power at a P-value of 0.05. Thus, our initial samples may lack sufficient power to determine whether rs1990760 is associated with psoriasis.Rare genetic variants and risk of psoriasis
The effect of the IFIH1 923Val variant (minor allele of rs35667974) on risk of psoriasis and T1D is in the same direction and of similar magnitude (allelic OR=0.45 (this study) and 0.51 (
Nejentsev et al., 2009
)). Carriers of the protective allele 923Val comprise 4.2% of the control population, and, when the other rare missense SNP is also considered, carriers of either of the protective alleles account for 8.1% of the controls. These observations underscore the important contribution of rare variants to disease susceptibility of common complex diseases such as psoriasis. They also have implications in not only further validating IFIH1 as a psoriasis risk gene but also suggesting variants in this gene be reevaluated in appropriately powered studies of other autoimmune diseases. Assuming that the minor allele frequency of rs35667974 in controls is 2.2% (as observed in this and the T1D studies) and effect size of 0.50, the number of samples required to achieve 80% power is estimated to be 1,080 cases with an equal number of controls. Interestingly, under the above assumptions, a given case–control sample set is predicted to have more power to detect this rare SNP (Ile923Val) with its stronger effect than the common SNP (Ala946Thr) with its weaker effect. Therefore, this rare SNP warrants testing in the other autoimmune studies to further clarify its role in modulating disease risk.The other psoriasis-associated SNP, rs10930046 (His460Arg), is not associated with T1D (
Nejentsev et al., 2009
). It is possible that distinct variants of the same risk factor may differentially affect individual diseases. Alternatively, this finding may be the result of type I error. We notice that unlike rs35667974, which has the same frequency in our controls and the T1D controls, the minor allele frequency of rs10930046 is somewhat different between our white, North American controls and the UK T1D controls (2.1 vs. 1.0%). Therefore, further testing of this marker in other psoriasis sample sets is required. Interestingly, residue 460 lies in the helicase ATP-binding domain of the IFIH1 protein, although it remains to be determined whether a change from a histidine residue to an arginine residue (BLOSUM62 score=0) affects protein activity.Biological evidence for a role of IFIH1 in psoriasis
The known biological function of IFIH1 supports a role for this gene in psoriasis. IFIH1 is an interferon-induced putative RNA helicase that affects cell growth, differentiation, and death (
Kang et al., 2002
; Besch et al., 2009
) and has been implicated in the recognition of RNA viruses (Kato et al., 2006
). Virus infection may be one of the environment factors that trigger and/or exacerbate psoriasis, and human endogenous retroviruses have been detected in psoriatic skins (Fry and Baker, 2007
). In addition, IFIH1 expression is increased in epidermal cells and tissues from psoriatic plaques compared to normal controls (Prens et al., 2008
); increased IFIH1 expression may account for some IFIH1 risk variants in T1D (Liu et al., 2009
). Together with these and other biological evidence, our genetic results suggest that understanding how the identified missense variants differentially affect IFIH1 protein function is important, particularly the conserved 923Ile versus 923Val (BLOSUM62 score=3). Elucidation of their functional differences will provide insights into the role for IFIH1 in the pathogenesis of psoriasis and may guide the design of pharmacological interventions either directly or indirectly targeting IFIH1 for the treatment of psoriasis and other autoimmune and inflammatory diseases.Materials and Methods
Samples and strategy
Samples from dermatologist-confirmed psoriasis patients and normal healthy controls were collected from the University of Utah (Sample Set 1: 467 cases and 460 controls), from the Genomics Collaborative Division of SeraCare Life Sciences (Sample Set 2: 498 cases and 498 controls), and from Genomics Collaborative and BioCollections Worldwide (Sample Set 3: 483 cases and 427 controls). All individuals were white, North Americans and were 18 years or older at the time they were enrolled in the sample collections. A detailed description of clinical and demographic information can be found in a previous publication (
Cargill et al., 2007
). A fourth sample set contained 667 cases and 371 controls: psoriasis cases were Caucasians with plaque-type psoriasis recruited at the University of California, San Francisco and Washington University, St Louis. Controls were healthy Caucasians with no history of autoimmune or inflammatory disease recruited from San Francisco. The female/male ratio was 1:1, and for cases the mean age of disease onset was 27 years with a standard deviation of 17. All protocols were approved by national and/or local institutional review boards, informed written consent was obtained from all subjects, and the study was conducted according to the Declaration of Helsinki Principles.The initial testing of 25,215 SNPs was carried out in DNA pools of Sample Set 1 as described previously (
Cargill et al., 2007
). Markers that had allelic P<0.05 were then tested for replication in DNA pools of Sample Set 2. Replicated markers (allelic P<0.05) were then sequentially individually genotyped in these and a third sample set. For this report, a total of 62 SNPs were individually genotyped in all three sample sets, 189 markers in two sample sets, and 86 markers in one sample set. The two IFIH1 markers, rs35667974 and rs10930046, were further genotyped in Sample Set 4 (Supplementary Table S1 online details which SNPs were individually genotyped in each sample set).Genotyping
For Sample Sets 1, 2, and 3, individual genotyping was performed using allele-specific kinetic PCR on 0.3ng of DNA at Celera. Data were hand curated before statistical analysis, without knowledge of case–control status. Genotyping calls were made on >95% of samples. Previous analyses suggest a genotyping accuracy of >99% (
Cargill et al., 2007
). For the fourth sample set, genotyping was performed at a laboratory at the University of California, San Francisco using Applied Biosystems (Foster City, CA) TaqMan assays.Statistics
Deviation from Hardy–Weinberg equilibrium was examined in the control samples using the exact test of . Allelic association of a marker with disease status was determined by the χ2-test; a meta-analysis in all three sample sets was carried out using fixed effects of the Mantel–Haenszel method to combine ORs across the sample sets. P-values were two sided in all samples combined. ORs and 95% confidence intervals were estimated from the allele or genotype counts. Assessment of OR heterogeneity across sample sets was carried out by the Breslow–Day test. Testing the independence of two markers was carried out by logistical regression. The linkage disequilibrium measure r2 was calculated from unphased data with use of the LDMAX program in the GOLD package. Power and sample size for an association study in a case–control study were estimated using the program “PS: Power and Sample Size Calculation” (http://biostat.mc.vanderbilt.edu/twiki/bin/view/Main/PowerSampleSize) with the assumption of independent cases and controls being used at a ratio of 1:1. The type I error probability associated with the test of a null hypothesis was set at 0.05. An uncorrected χ2 statistic was used to evaluate this null hypothesis.
ACKNOWLEDGMENTS
We thank the patients and other individuals for contribution to clinical samples; K Ardlie, J Lemaire, and S Mahan for database and sample management at GCI; A Peiffer and M Dixon for invaluable input; M Hoffman, T Christensen, T Nelson, and B Wong for help recruiting patients and coordinating the project at the University of Utah; M Paul and colleagues at LineaGen for managing the collaboration; and TJ White and JJ Sninsky for scientific advice. The financial support for this study was, in part, provided by the Dermatology Foundation to W Liao, by the National Institutes of Health Grant 1R01AR050266 to AM Bowcock, by a Public Health Services research grant to the Huntsman General Clinical Research Center at the University of Utah, by National Center for Research Resources Grant M01-RR00064, and by generous gifts from the WM Keck Foundation and from the George S and Delores Dore Eccles Foundation.
SUPPLEMENTARY MATERIAL
Supplementary material is linked to the online version of the paper at http://www.nature.com/jid
REFERENCES
- MEL4B3, a novel mRNA is induced in skin tumors and regulated by TGF-beta and pro-inflammatory cytokines.Exp Dermatol. 2005; 14: 709-718
- Proapoptotic signaling induced by RIG-I and MDA-5 results in type I interferon-independent apoptosis in human melanoma cells.J Clin Invest. 2009; 119: 2399-2411
- Getting under the skin: the immunogenetics of psoriasis.Nat Rev Immunol. 2005; 5: 699-711
- The clincal genetics of psoriasis.Curr Genomics. 2005; 6: 51-60
- Identification of ZNF313/RNF114 as a novel psoriasis susceptibility gene.Hum Mol Genet. 2008; 17: 1938-1945
- A large-scale genetic association study confirms IL12B and leads to the identification of IL23R as psoriasis-risk genes.Am J Hum Genet. 2007; 80: 273-390
- Variants in the 5q31 cytokine gene cluster are associated with psoriasis.Genes Immun. 2008; 9: 176-181
- IFIH1-GCA-KCNH7 locus is not associated with genetic susceptibility to multiple sclerosis in French patients.Eur J Hum Genet. 2009; 17: 844-847
- Deletion of the late cornified envelope LCE3B and LCE3C genes as a susceptibility factor for psoriasis.Nat Genet. 2009; 41: 211-215
- Multiple sclerosis and polymorphisms of innate pattern recognition receptors TLR1-10, NOD1-2, DDX58, and IFIH1.J Neuroimmunol. 2009; 212: 125-131
- Fine mapping of the psoriasis susceptibility locus PSORS1 supports HLA-C as the susceptibility gene in the Han Chinese population.PLoS Genet. 2008; 4: e1000038
- Triggering psoriasis: the role of infections and medications.Clin Dermatol. 2007; 25: 606-615
- A large-scale replication study identifies TNIP1, PRDM1, JAZF1, UHRF1BP1 and IL10 as risk loci for systemic lupus erythematosus.Nat Genet. 2009; 41: 1228-1233
- Psoriasis is associated with increased beta-defensin genomic copy number.Nat Genet. 2008; 40: 23-25
- mda-5: An interferon-inducible putative RNA helicase with double-stranded RNA-dependent ATPase activity and melanoma growth-suppressive properties.Proc Natl Acad Sci USA. 2002; 99: 637-642
- Differential roles of MDA5 and RIG-I helicases in the recognition of RNA viruses.Nature. 2006; 441: 101-105
- ADAM33, a new candidate for psoriasis susceptibility.PLoS ONE. 2007; 2: e906
- Unraveling the genetics of complex diseases: susceptibility genes for rheumatoid arthritis and psoriasis.Semin Immunol. 2009; 21: 318-327
- Further genetic evidence for three psoriasis risk genes: ADAM33, CDKAL1 and PTPN22.J Invest Dermatol. 2009; 129: 629-634
- IFIH1 polymorphisms are significantly associated with type 1 diabetes and IFIH1 gene expression in peripheral blood mononuclear cells.Hum Mol Genet. 2009; 18: 358-365
- A genome-wide association study of psoriasis and psoriatic arthritis identifies new disease Loci.PLoS Genet. 2008; 4: e1000041
- The interferon induced with helicase domain 1 A946T polymorphism is not associated with rheumatoid arthritis.Arthritis Res Ther. 2007; 9: R40
- IFIH1-GCA-KCNH7 locus: influence on multiple sclerosis risk.Eur J Hum Genet. 2008; 16: 861-864
- Genome-wide scan reveals association of psoriasis with IL-23 and NF-kappaB pathways.Nat Genet. 2009; 41: 199-204
- Sequence and haplotype analysis supports HLA-C as the psoriasis susceptibility 1 gene.Am J Hum Genet. 2006; 78: 827-851
- Rare variants of IFIH1, a gene implicated in antiviral responses, protect against type 1 diabetes.Science. 2009; 324: 387-389
- Psoriasis.N Engl J Med. 2009; 361: 496-509
- The rs1990760 polymorphism within the IFIH1 locus is not associated with Graves’ disease, Hashimoto's thyroiditis and Addison's disease.BMC Med Genet. 2009; 10: 126
- IFN-alpha enhances poly-IC responses in human keratinocytes by inducing expression of cytosolic innate RNA receptors: relevance for psoriasis.J Invest Dermatol. 2008; 128: 932-938
- Genetic susceptibility to psoriasis: an emerging picture.Genome Med. 2009; 1: 72
- A genome-wide association study of nonsynonymous SNPs identifies a type 1 diabetes locus in the interferon-induced helicase (IFIH1) region.Nat Genet. 2006; 38: 617-619
- Shared and distinct genetic variants in type 1 diabetes and celiac disease.N Engl J Med. 2008; 359: 2767-2777
- Genomic polymorphism at the interferon-induced helicase (IFIH1) locus contributes to Graves’ disease susceptibility.J Clin Endocrinol Metab. 2007; 92: 3338-3341
- Functional haplotypes of PADI4, encoding citrullinating enzyme peptidylarginine deiminase 4, are associated with rheumatoid arthritis.Nat Genet. 2003; 34: 395-402
- Genet Data Analysis II. Sinauer, Sunderland, MA1996
- Psoriasis is associated with pleiotropic susceptibility loci identified in Type II Diabetes and Crohn's disease.J Med Genet. 2008; 45: 114-116
- Psoriasis genome-wide association study identifies susceptibility variants within LCE gene cluster at 1q21.Nat Genet. 2009; 41: 205-210
Article info
Publication history
Accepted:
June 4,
2010
Received in revised form:
May 3,
2010
Received:
February 22,
2010
published online 29 July 2010Footnotes
Authors affiliated with Celera declare their interest in the company for employment and/or stock ownership.
Identification
Copyright
© 2010 The Society for Investigative Dermatology, Inc. Published by Elsevier Inc.
User license
Elsevier user license | How you can reuse 
Elsevier's open access license policy
Elsevier user license
Permitted
For non-commercial purposes:
- Read, print & download
- Text & data mine
- Translate the article
Not Permitted
- Reuse portions or extracts from the article in other works
- Redistribute or republish the final article
- Sell or re-use for commercial purposes
Elsevier's open access license policy
