Session Title: Genetics and Genomics of Rheumatic Disease II
Session Type: Abstract Submissions (ACR)
In a recent fine-mapping study using the Immunochip array, the RUNX1 region was strongly associated with rheumatoid arthritis (RA p=5×10-10), juvenile idiopathic arthritis (JIA p=1×10-8) and psoriatic arthritis (PsA p=6×10-4). In each disease the SNP rs9979383 confers disease protection with odds ratios ranging from 0.8-0.9. This SNP maps approximately 290kb upstream of RUNX1, indicating that it may be involved in regulation of gene expression. RUNX1 encodes a transcription factor highly expressed during chondrogenesis, which has also been shown to interact with the regulatory T cell marker FOXP3. Unlike many loci on the Immunochip, variation in the RUNX1 region is not well covered, therefore further fine mapping is required.
As rs9979383 lies between two points of high recombination, SNPs between these points were selected for fine mapping. 42 common (MAF>0.05) tag SNPs (r2>0.9) from the Utah residents (CEPH) with Northern and Western European ancestry (CEU) 1000 genomes July 2010 release were selected. 2255 RA cases and 1877 healthy controls from the United Kingdom Rheumatoid Arthritis Genetics Group (UKRAG) were genotyped using the Sequenom iPlex MassARRAY platform. SNPs and samples which reached >90% success rate were included in the analysis. Allelic association testing was performed using PLINK v.1.07 and functional annotation of SNPs was performed using ASSIMILATOR. To determine if rs9979383 represents an expression quantitative trait locus (eQTL), Taqman genotyping and gene expression assays were performed in 75 healthy controls from the national repository healthy volunteers study. RUNX1 expression was normalised to 2 endogenous controls (ACTNB and GAPDH) and linear regression performed in STATA v.11.2.
In the UKRAG cohort, association with rs9979383 was replicated (p=0.02, OR = 0.9, CI 95% 0.82-0.98) with an odds ratio of 0.9, identical to the previous RA study. No other SNPs in the region showed evidence for association. Functional annotation of this SNP indicates that it lies within a region of open chromatin with transcription factor binding potential, making it an ideal candidate for gene expression studies. In whole blood, RUNX1 expression was not correlated with genotype at rs9979383 in healthy controls (p=0.92) indicating that rs9979383 may alter expression of another gene or that this eQTL may only be present at a cell specific level.
The results indicate that the RUNX1 association is localised to r99793983 (or SNPs in high LD) in RA but the association requires confirmation in independent JIA and PsA cohorts. Although no eQTL was observed in whole blood in healthy controls, cell specific whole transcriptome studies are currently underway to determine whether this SNP is an eQTL. Combined with the fine mapping this will inform further investigations of the role of the RUNX1 region in the common pathogenesis of inflammatory arthritis.
United Kingdom Rheumatoid Arthritis Genetics Group (UKRAG)
K. J. A. Steel,
« Back to 2013 ACR/ARHP Annual Meeting
ACR Meeting Abstracts - https://acrabstracts.org/abstract/fine-mapping-and-expression-of-a-locus-overlapping-3-types-of-inflammatory-arthritis/