P128 Analysis of DNA methylation patterns in rheumatoid arthritis patient: a system for prioritising meaningful difference. (21st February 2018)

• Record Type: Journal Article
• Authors: Pitaksalee, R; Burska, AN; Rogers, J; Xie, X; Emery, P; Hodgett, R; Ponchel, F
• Is Part Of: Annals of the rheumatic diseases. Volume 77(2018)Supplement 1
• Page Start: A69
• Page End: A69
• Publication Date: 2018-02-21
• Subjects: Rheumatism -- Periodicals; 616.723005
• DOI: 10.1136/annrheumdis-2018-EWRR2018.143
• Languages: English
• ISSNs: 0003-4967

Abstract

Abstract : Introduction: Alterations in DNA methylation patterns have been related to several diseases, including Rheumatoid Arthritis (RA). Objectives: To identify changes in DNA methylation pattern of naïve and memory CD4 +T cells and monocytes in early, drug naïve RA patients to help understand early event in disease pathology. Methods: The methylation patterns of 480, 000 CpGs were analysed in the 3 cell types from 6 healthy control (HC) and 10 RA patients using an Illumina genome-wide array. Standard t-tests were performed to associate p-value to individual CpG-probe. A scoring system was developed to select and prioritise differentially methylated CpGs with potential cumulative effect due to proximity with other significant CpGs. Rules for scoring were designed with respect to the significance of each CpGs and the distance (in bp) between them. Further filtering was applied to initially select the CpGs which highest significant (p-value≤0.0001) which have at least 2 proximal significant CpGs (p-value≤0.01). Rules were coded in R for systematic analysis. Lists of selected CpGs for all three cells for both hypo- and hyper-methylation were generated. Commonality was analysed using Venn diagram. Results: Different changes in methylation patterns were observed between HC and RA in the 3 cell types and thresholds of significance were set at p-value 0.01, 0.001 and 0.0001. The total number of differentially methylated CpGs was enriched in naïve T-cells (18, 020) compared to memory (14, 197) and monocytes (6, 490) (p-value≤0.01). Using our designed rules, we were able to prioritise cluster of differentially methylated CpGs which were then associated to 420, 7, and 48 hypomethylated genes, and 420, 719 and 21 hypermethylated genes respectively in naïve/memory T-cells and monocytes. Venn diagrams of hypermethylated gene showed only 1 genes (ABAT) in common to all 3 subsets. Hypomethylated gene showed no commonalities between the 3 cell subsets, and only 1 gene common between T-cell subsets (SLC43A2). Of note, the TNF gene was second on the priority list for naïve T-cell hypomethylation while no difference was observed in memory/monocytes. The IL-17 gene was de-methylated only in memory T-cells, hypermethylated in RA, but fully methylated in naïve/monocytes. Hypomethylation of socs3 was specific to monocytes. Conclusions: These data suggest quite different types of changes in patterns of DNA methylation affecting the 3 subsets early in the RA disease process. Our scoring system to prioritise clusters of differential methylation highlighted genes known to be related to early RA pathogenesis. Further work remains to explore the relationship between these genes and the biological effect at the transcriptional/translational level resulting from these alterations in DNA methylation. Disclosure of interest: None declared