There is an evolving body of evidence that demonstrates the role of epigenetic mechanisms, such as DNA-methylation in the pathogenesis of OA. This systematic review aims to summarize the current evidence of DNA methylation and its influence on the pathogenesis of OA.

A pre-defined protocol in alignment with the PRISMA guidelines was employed to systematically review eight bibliographic databases, to identify associations between DNA-methylation of articular chondrocytes and osteoarthritis. A search of Medline (Ovid), Embase, Web-of-Science, Scopus, PubMed, Cinahl (EBSCOhost), Cochrane Central and Google Scholar was performed between 1st January 2015 to 31st January 2021. Data extraction was performed by two independent reviewers.

During the observation period, we identified 15 gene specific studies and 24 genome wide methylation analyses. The gene specific studies mostly focused on the expression of pro-inflammatory markers, such as IL8 and MMP13 which are overexpressed in OA chondrocytes. DNA hypomethylation in the promoter region resulted in overexpression, whereas hypermethylation was seen in non-OA chondrocytes. Others reported on the association between OA risk genes and the DNA methylation pattern close to RUNX2, which is an important OA signal. The genome wide methylation studies reported mostly on differentially methylated regions comparing OA chondrocytes and non-OA chondrocytes. Clustering of the regions identified genes that are involved in skeletal morphogenesis and development. Differentially methylated regions were seen in hip OA and knee OA chondrocytes, and even within different regions of an OA affected knee joint, differentially methylated regions were identified depending on the disease stage.

This systematic review demonstrates the growing evidence of epigenetic mechanisms, such as DNA methylation, in the pathogenesis of OA. In recent years, there has been a focus on the interplay between OA risk genes and DNA methylation changes which revealed a reactivation of genes responsible for endochondral ossification during development. These are important findings and may help to identify eventual future therapeutic targets. However, the current body of literature is mostly showing the differences in DNA methylation of OA chondrocytes and non-OA chondrocytes, but a true longitudinal analysis demonstrating the DNA methylation changes actually happening is still not available.