Abstract
Introduction and Objective
Osteoarthristis (OA) has been associated with many genes and yet the genetic basis for this disease has never formally been established. Recent realization that epigenetic changes could be the underlying pathological mechanisms has helped to explain many complex multifactorial diseases with no clear genetic cause. We therefore asked whether epigenetics could also play a role in OA. We have previously shown that the DNA epigenetic modification, specifically the hydroxymethylation on cytosine (5hmC), undergoes a fivefold increase on OA-associated genes which are activated at OA onset. In this study, we further uncovered a set of 5hmC-mediated gene targets and their mechanistic link to OA progression.
Materials and Methods
We surgically induced OA on 4 to 6 months old Tet1−/− mice (Tet1tm1.1Jae, the Jackson laboratory) and wild-type littermates by performing destabilization of the medial meniscus (DMM) surgery. Joints were collected for histological assessment through blinded grading with the OARSI scoring system. Human articular chondrocytes were harvested from OA cartilage samples obtained during total knee arthroplasty or from grossly normal cartilage pieces obtained during notchplasty or debridement from patients undergoing anterior cruciate ligament (ACL) reconstruction with no history of OA symptoms, under approved Human subjects Institutional Review Board protocols. Bioinformatic analyses of RNA-sequencing and CCGG sequencing (reduced representation 5hmC profiling) were performed to identify TET1 target genes associated with OA progression. Several measurements were used to assess the effect of TET1 ablation on the phenotype of mouse cartilage tissue and human chondrocytes including, histological evaluation, and quantitative bone assessment by micro-CT imaging and multiplex cytokine analyses in the serum of mice in vivo (mouse 39-plex assay) and in the supernatant of human chondrocyte cultures (human 62-plex assay).
Results
We used a mouse model with surgically induced OA and found that OA onset was accompanied by a gain of ∼40,000 differentially hydroxymethylated sites prior the notable histological onset of the disease. We additionally revealed that these changes are mediated by the ten-eleven-translocation enzyme 1 (TET1), since Tet1−/− mice lost 98% of 5hmC sites upon OA induction. Remarkably, Tet1−/− mice were protected from OA development including degeneration of the cartilage surface and osteophyte formation. Silencing of TET1 expression in human OA chondrocytes reduced the expression in a set of genes, which may represent the pathological gene targets that exacerbate OA including MMP3 and MMP13 and several inflammatory cytokines. Therefore, our study reveals the unexpected beneficial role of TET1 inhibition in blocking OA progression. In fact, intra-articular injections of a dioxygenases’ inhibitor, 2 hydroxyglutarate, on mice after surgical induction of OA stalled disease progression. Furthermore, treatment of human OA chondrocytes with the same inhibitor also phenocopied TET1 loss, implicating a therapeutic potential of TET inhibition in OA patients.
Conclusions
Collectively, our study not only demonstrate the role of TET1 in OA; the 5hmC-mediated gene targets acting on multiple OA pathways were identified and can be modulated as therapeutic intervention to treat OA.
