Background Vitamin D is a steroid hormone which plays an important role in immune regulation. It has been implicated in the aetiopathogenesis of autoimmune diseases (AID), including rheumatoid arthritis (RA). Vitamin D controls transcription of target genes through its nuclear receptor, (VDR), binding to receptor elements (VDREs). These VDREs have been defined and genetically mapped using chromatin immunoprecipitation and re-sequencing methods, identifying 2776 VDREs in lymphoblastoid cell lines. An investigation of 9 genomic loci associated with RA identified an enrichment of VDREs (1). However, since that study, an additional 22 loci have been reproducibly associated with RA susceptibility. In addition, genetic studies have identified four genes associated with vitamin D levels, one of which has subsequently been associated with type 1 diabetes (T1D) (2).
Objectives To test all confirmed RA susceptibility loci to identify the extent of enrichment of VDREs and to test SNPs associated with vitamin D levels for association with RA.
Methods To identify VDRE enrichment, regions surrounding the 31 confirmed RA susceptibility loci (3) were defined in two ways: first, by assigning the locus to a candidate gene and second, by extending 150kb up and down stream of the position of the most associated SNP in the region. Perl scripts were used to bioinformatically map the position of any VDREs at RA loci, overlaying the VDRE genomic positions identified previously, with the RA susceptibility regions. The number of VDREs identified at RA loci was then compared to the number found in a randomly selected set of genomic loci, identified from ensembl, to determine enrichment. The randomly selected loci were randomised 10,000 and 100,000 times before an average relative risk (RR) was calculated.
Eleven SNPs in the DHCR7 region (r2=1 with rs12785878), associated with vitamin D levels and T1D were genotyped in 3,870 UK cases and 8,430 UK controls. Analysis was carried out in PLINK.
Results A significant enrichment of VDREs was seen in RA associated loci compared to random loci (p=0.0001) when regions were defined either by gene (RR 7.186, 95% CI 2.04-18.33) or position (RR 7.95, 95% CI 2.45-23) with 63.9% of RA associated loci containing at least one VDRE. An even greater enrichment of VDREs at RA associated loci was observed when the regions were narrowed to within 50kb up and down stream of the associated SNP (29 VDREs identified in 14 out of 36 RA genes), relative risk 15.31 (95% CI 2.8-59).
A trend towards association was seen with a SNP in the DHCR7 region associated with vitamin D levels (rs11600569 p=0.01). Indeed the odds ratio observed in RA is the same as that seen in T1D (1.07). Stratification by CCP status, showed increased association in CCP negative individuals (1000 cases, 8429 controls, p=0.01, OR 1.15). No association was seen in CCP positive individuals.
Conclusions The significant enrichment of VDREs at RA associated loci, and the modest association of a SNP previously shown to correlate with levels of circulating vitamin D, lends support to the hypothesis that vitamin D may play a role in the development of RA.
Ramagopalan SV et al. Genome Res 2010 Oct 1;20(10):1352-60.
Cooper JD et al. Diabetes 2011 May;60(5):1624-31.
Stahl EA et al. Nat Genet 2010 Jun;42(6):508-14.
Disclosure of Interest None Declared