Genome-wide association meta-analysis of eight independent systemic juvenile idiopathic arthritis collections reveals regional association spanning the major histocompatibility complex class II and III gene cluster

Abstract

Background/Purpose: Systemic juvenile idiopathic arthritis (sJIA) is a rare inflammatory disease that is inherited as a complex genetic trait. While the pathophysiology of sJIA is poorly understood, there is evidence of both innate and adaptive immune involvement. We have undertaken a genomewide association study (GWAS) of sJIA to elucidate pathways and molecular mechanisms that underlie this disease. The power of a GWAS is directly related to the size of the population studied. This confounds the application of GWAS to rare diseases, such as sJIA, where acquisition of even modestlysized patient collections is very difficult. In an attempt to overcome the rare nature of sJIA, we have designed a meta-analytic GWAS to investigate sJIA collections from eight different countries. Methods: We genotyped 823 children fulfilling ILAR criteria for sJIA and 442 healthy children using the Omni1M Quad SNP array. We obtained five in silico collections of control genotypes that were geographicallymatched to five of the case collections, providing an additional 4037 controls for our analysis. After performing quality control operations to exclude low quality samples and markers, we combined case and control genotype data for each of the eight strata, retaining only markers that were present in both the case and control datasets. We used principal components analysis to identify and remove genetic outliers from each population. For each stratum, we created phased haplotypes with the ShapeIT software, we performed SNP imputation with IMPUTE2 software and the HapMap3 reference haplotypes, and we undertook association testing with SNPTEST software. Finally, genomewide association meta-analysis of the eight strata was performed using the GWAMA software. Results: Genomewide association meta-analysis of 1,447,416 markers, applying the additive model to eight case-control strata, identified 49 regions that contained one or more marker with p < 5E-5. Fourteen of these genes reside within a 3 Mb segment of the major histocompatibility complex (MHC), spanning the MHC class II and III gene clusters. This sJIA-associated region also contains the only marker to exceed the genomewide significance threshold (rs615672, p = 2.64E-8; OR = 0.7, 95CI 0.62, 0.79), which lies between HLA-DRB1 and HLA-DQA1. Conclusion: The markers most strongly associated with sJIA were found within the MHC locus, nearest to the HLA-DRB1 and HLA-DQA1 genes. However, these variants reside in a larger 3 Mb interval that contains a range of genes involved in both innate and adaptive immunity. Interestingly, we have observed an almost complete absence of overlap between our 49 candidate loci and the known autoimmune and autoinflammatory loci. If this observation is upheld by additional investigations, then we may expect this study to identify novel pathways and mechanisms involved in sJIA, which may also represent novel therapeutic targets. (Figure Presented).