Session Information
Session Type: Abstract Submissions (ACR)
Background/Purpose
Recent data suggest that epigenetics, including DNA methylation, contributes to imprinting RA fibroblast-like synoviocytes (FLS) and alters their behavior. To understand how RA-associated risk alleles and differential methylation affect cis-regulatory regions, we compared differentially methylated loci (DML) in RA FLS with fibroblast DNase I hypersensitive sites. The differentially methylated enhancers (DMEs) were then integrated with RA-associated SNPs to identify key regulatory sites. In this study, we found and characterized an RA-associated SNP that colocalized with a DME in a key cancer-related gene that regulates cell growth and differentiation, namely limb-bud and heart development (LBH).
Methods
15,220 RA-associated DMLs identified using Illumina 450k chips from 11 RA and 11 OA FLS were integrated with DNase I hypersensitive sites in the ENCODE database for 125 cell-types/conditions, including lung fibroblasts. These DMEs were then compared with GWAS data. Genomic DNA was isolated and the 1.4kb region with the WT allele (G) or RA SNP (T) of lbh was cloned into minimal promoter pGL4.23-luciferase construct. For methylation studies, plasmids were methylated with the CpG-methyltransferase M.SssI and S-adenosyl methionine. Methylation of all CpGs was verified by bisulfite modification and pyrosequencing. The WT, RA SNP or control plasmids (1ug) were transfected into cultured RA FLS by nucleofection with Renilla construct. Firefly luciferase activity was normalized to renilla.
Results
A DNase I hypersensitive site that is hypomethylated in RA FLS was identified in a 1400 bp region upstream of the LBH gene transcription start site. An RA and SLE-associated SNP (rs906868, G/T) was identified in the same region. We first determined the transcription differences between the WT and RA SNP in transfected RA FLS. Surprisingly, luciferase activity was significantly decreased in the WT group compared with the RA SNP (19±7%, p=0.02, n=9 lines), suggesting that the regulatory region decreases gene expression in WT cells. The RA SNP had no effect on luciferase expression compared with control. We then evaluated whether CpG methylation of the plasmids altered function of the enhancer region. Methylation of the RA SNP plasmid significantly decreased the luciferase activity by 46±4% compared with the unmethylated (p=0.0005, n=5 lines). In contrast, methylation increased the luciferase activity in the WT group (42±17%, p=0.049, n=5 lines) but not in the control group.
Conclusion
We identified a novel candidate regulatory region by integrating ENCODE with RA GWAS and RA DNA methylation databases. The RA-associated allele in lbh eliminated its regulatory function. This is the first demonstration that an RA risk allele in a regulatory region has functional effects. By altering enhancer function of a gene that is intimately involved with cell proliferation (LBH), the risk allele could contribute to the aggressive behavior of RA FLS. Furthermore, the functional effects of methylation are distinctly different between RA-associated alleles and WT. These data provide evidence that combinations of epigenetic and risk allele data can provide new clues to the pathogenesis of RA.
Disclosure:
D. Hammaker,
None;
G. S. Firestein,
None;
W. Wang,
None;
J. W. Whitaker,
None;
A. K. Ekwall,
None.
« Back to 2014 ACR/ARHP Annual Meeting
ACR Meeting Abstracts - https://acrabstracts.org/abstract/rheumatoid-arthritis-ra-associated-risk-allele-lbh-alters-the-function-of-a-differentially-methylated-lbh-enhancer/