A Genome-wide Association Study Of Vitiligo In Chinese Population

Background Vitiligo （MIM193200） is a common depigmentary disorder resulting from selective destruction of melanocytes in the skin and hair, with diverse prevalence rates ranging from 0.1 to 2.9% in different geographic regions and ethnic groups. Clinical and epidemiological investigations indicated that vitiligo might follow a pattern of polygenetic or multifactorial inheritance. A number of genetic susceptibility factors have been identified through linkage and association studies, however, only a few genes/loci, such as NALP1 and several HLA alleles has been replicated in multiple studies. A serial of genome-wide association study （GWAS） for complex diseases have been carried out successfully, which provided new methods for exploring the susceptibility genes of vitiligo. Recently, a large scale GWAS of vitiligo have been performed in European population and identified TYR and several autoimmunity susceptibility loci associated with vitiligo.Object To explore the susceptibility variants for vitiligo, we carried out a GWAS in Chinese populations.Methods We conducted a genome-wide association study of generalized vitiligo in Chinese Han population by genotyping 1,149 cases and 1,468 controls using Illumina Human 610-Quad BeadChips. We took the most promising SNPs for replication in Chinese Han （5,910 cases and 9,916 controls） and Chinese Uygur （713 cases and 824 controls） using Sequenom MassArray and TaqMan Assay.Results （1） After stringent quality control, 493,909 SNPs were analyzed in 1,117 cases and 1,429 controls by using Cochran-Armitage trend test to assess the genotype-phenotype association. （2） We identified two independent association signals within the MHC region (rs11966200: P_{Chinese Han} =8.77×10^-43, P_{Chinese Uygur}=1.15×10^-7, and rs9468925: P_{Chinese Han} =1.11×10^-30, P_{Chinese Uygur}=3.65×10^-4). Further analyses suggested that the strong association at rs11966200 might reflect the reported association of HLA-A*3001, B*1302, C*0602 and DRB1*0701 alleles, and the association at rs9468925 might represent an unidentified vitiligo-related HLA allele. Haplotype analysis showed that two haplotypes were highly associated with vitiligo (H1, P=6.71x10^-8, OR=0.72; and H5, P=1.76x10^-9, OR=1.85). The protective allele A of rs9468925 sits on the haplotype H1, whereas the risk-associated allele A of rs11966200 sites on the haplotype H5. （3） We identified one novel locus at 6q27 associated with vitiligo both in Chinese Han (rs2236313, P=1.60×10^-15) and Chinese Uygur (rs2236313, P=2.00×10^-2), which been reported to confer risk to several autoimmune disease, such as inflammatory bowel disease （IBD） and rheumatoid arthritis （RA）. （4） We also identified one suggestive association evidence in ZMIZ1 at 10q22 in Chinese Han (rs11593576, P=1.82×10^-7), which also been associated with IBD reported by other GWAS. （5） Pair-wise interaction analysis was also performed between 6q27, 10q22 and 2 MHC SNPs, however, no significant genetic interaction was identified （P>0.05 after correction for multiple testing）. （6） We also performed an association analysis for candidate loci/variants of vitiligo and autoimmune diseases in current GWAS. Only a few genes/loci, such as LPP, MBL2, XBP1, CXCR4, CD69 and PTPN2 provided suggestive evidence.Conclusions We have performed a large scale GWAS for vitiligo in Chinese population, with replication in different Chinese population groups. Our study identifies two independent associations within MHC region, one novel susceptibility locus at 6q27 and one suggestive locus at 10q22. In particular, it highlighted potential pathogenic overlap between vitiligo and other autoimmune diseases, especially IBD.