Integrative Omics Analysis of Rheumatoid Arthritis Identifies Non-Obvious Therapeutic Targets

Abstract

Identifying novel therapeutic targets for the treatment of disease is challenging. To this end, we developed a genome-wide approach of candidate gene prioritization. We independently collocated sets of genes that were implicated in rheumatoid arthritis (RA) pathogenicity through three genome-wide assays: (i) genome-wide association studies (GWAS), (ii) differentially expression in RA fibroblast-like synoviocytes (FLS), and (iii) differentially methylation in RA FLS. Integrated analysis of these complementary data sets identified a significant enrichment of multi-evidence genes (MEGs) within pathways relating to RA pathogenicity. One MEG is Engulfment and Cell Motility Protein-1 (ELMO1), a gene not previously considered as a therapeutic target in RA FLS. We demonstrated in RA FLS that ELMO1 is: (i) expressed, (ii) promotes cell migration and invasion, and (iii) regulates Rac1 activity. Thus, we created links between ELMO1 and RA pathogenicity, which in turn validates ELMO1 as a potential RA therapeutic target. This study illustrated the power of MEG-based approaches for therapeutic target identification.

Abnormal PTPN11 enhancer methylation promotes rheumatoid arthritis fibroblast-like synoviocyte aggressiveness and joint inflammation

“…Through genome-wide assessment of DNA methylation, we recently identified a DNA methylation pattern that distinguishes RA FLS from non-RA FLS and contributes to the pathogenic behavior of RA FLS (8, 22). Although we did not identify abnormal methylation signals in the PTPN11 locus in our previous studies, the array platform used to detect methylation was focused on promoter regions and only included a small subset of enhancer sites (23). We thus hypothesized that over-expression of PTPN11 in RA FLS could be due, in part, to abnormal methylation of a PTPN11 enhancer.…”

Network diffusion on multiple-layers: current approaches and integrative analysis of Rheumatoid Arthritis data.

“…RegNet 12 is an R package that uses ND to quantify the impact of gene expression changes on user-defined target genes in a network inferred from gene expression and copy number data. We applied ND to jointly analyse genetic alterations, differentially methylated genes and differentially expressed genes found in RA fibroblast-like synoviocytes from a previous study 2 .…”

“…RA data were obtained from a previous study 2 , in which a total of 88 genes were found significant in a genome-wide association analysis, 2291 genes were differentially methylated and 2831 genes were differentially expressed between RA fibroblast-like synoviocytes and control cells (the reported numbers were obtained after the exclusion of a few deprecated genes). The genomescale network containing direct and indirect interactions between gene products was defined using NCBI interaction data 3 .…”

Network diffusion on multiple-layers: current approaches and integrative analysis of Rheumatoid Arthritis data.

“…RegNet 12 is an R package that uses ND to quantify the impact of gene expression changes on user-defined target genes in a network inferred from gene expression and copy number data. We applied ND to jointly analyse genetic alterations, differentially methylated genes and differentially expressed genes found in RA fibroblast-like synoviocytes from a previous study 2 . For each gene list, ND induces a gene ranking that takes into account also the a priori known interactions between all human genes.…”

“…RA data were obtained from a previous study 2 , in which a total of 88 genes were found significant in a genome-wide association analysis, 2291 genes were differentially methylated and 2831 genes were differentially expressed between RA fibroblast-like synoviocytes and control cells (the reported numbers were obtained after the exclusion of a few deprecated genes). The genomescale network containing direct and indirect interactions between gene products was defined using NCBI interaction data 3 .…”