Independent association of variants in the class I region and the class II gene, HLA-DPB1, at the MHC in a Filipino SLE cohort

Abstract

Objectives: Variants within the MHC demonstrate the greatest genetic risk for lupus in European and Chinese populations. Recent high-density SNP genotyping studies have demonstrated multiple independent signals across the MHC in northern European cohorts. We therefore undertook a high-density SNP study of the MHC in a Filipino lupus cohort in order to further refine association signals in this haplotypically diverse South-East Asian population. Methods: We genotyped 11640 SNPs in 335 cases and 247 controls using a custom Illumina chip. 4743 SNPs were informative for major and European ancestry, 6045 SNPs were located within the MHC region (29 - 33.5 Mb) and 852 SNPs were located in putative autoimmune loci outside the MHC region. Results: Following QC measures 217 cases and 176 controls were put forward for analysis. The overall pattern of association demonstrates that the major signal arises from the class II region of the MHC and therefore differs from that observed in European lupus cohorts where associations are seen in class II and class III. The top SNP is rs9270986 located between HLA-DRB1 and HLA-DQA1 (OR 2.7, CI 2.0-3.6; nominal p=1×10-10). The LD around this signal spans ∼37 kb from HLA-DRB1 to the intergenic interval between HLA-DRB1 and HLA-DQA1. There is also a protective haplotype within the HLA-DRB1 and HLA-DQA1 intergenic interval. Conditional logistic regression analyses on the top SNP, rs9270986, reveals independent signals in the class I region of the MHC between HLA-G and HLA-H (conditional OR∼ 0.48, CI∼0.34-0.68, p∼2.36×10-5) and HLA-DPB1 (conditional OR 0.42, CI 0.25-0.70, p=9.13×10-4). Interestingly, the class III SNPs which show association in European lupus cohorts are rare in this Filipino cohort. These latter data are similar to the recent findings of a Chinese case-control lupus genome-wide association study, where the minor allele frequencies for these class III SNPs were ∼0.001. Conclusion: These observations support a significant role for variants within the MHC class II region but not the MHC class III region in South-East Asian SLE in marked contrast to associations seen in European lupus populations.