Functional interrogation of putative osteoarthritis causal variants

PROJECT SUMMARY A key priority for the NIH is to limit disability caused by osteoarthritis (OA) and other chronic diseases that emerge with age. Our long-term goal is to catalyze effective strategies for early intervention by establishing the mechanisms that lead to OA. Through genome-wide association studies (GWAS), it is clear that the genetic risk for OA is driven primarily by a large number of non-coding single nucleotide variants (SNVs) that alter the regulation of gene expression in chondrocytes and other cell types of the joint. Our overall hypothesis is that testing the effect of OA GWAS SNVs on enhancer strength will identify the most likely causal variants, and that analyzing transcription factor binding at these variants will provide insight into the mechanisms of genetic risk in OA. We work with cadaveric chondrocytes from donors with no history of joint disease to provide a regulatory environment that represents cartilage homeostasis. In addition to this baseline state, we also stimulate cells with bioactive matrix fragments to initiate gene expression and chromatin accessibility changes that mimic the cellular state during OA. The first aim is to identify expression modulating variants (emVars) in primary human chondrocytes at baseline and in response to an OA-relevant stimulus. We will use massively parallel reporter assays (MPRAs) to assess allele-specific activity of the 1259 variants that reside within 104 known OA GWAS loci. The results from MPRAs and expression quantitative trait loci (eQTL) studies in other cell types and diseases give us the expectation that we will identify 1-3 emVars for each of 10-30 loci and that some of these will be specific to either baseline or stimulus conditions. The second aim is to computationally determine the transcription factors that differentially bind to emVars. We expect to find that emVars preferentially reside in regions that have accessible chromatin (as determined by ATAC-seq) and contain histone marks present in enhancers (as determined by Cut & Run for H3K27ac). Further, we expect that emVars will alter transcription factor binding strength and that stimulus-specific emVars will disrupt the binding sites of transcription factors that we have shown coordinate gene expression changes in response to this stimulus. This proposal is innovative, as it represents the first MPRA in primary human chondrocytes and one of the first “response MPRAs” to test the effect of cell state on allele-specific enhancer function for any disease. These results will have a substantial effect on the field by providing some of the first examples of how specific non-coding variants alter transcription factor binding to mediate the genetic risk of OA.