A2A adenosine receptor stimulation regenerates cartilage in osteoarthritis animal model

Abstract

Purpose: Many studies have been shown that obesity along with joint injury is one of the most common risk factor in the development of osteoarthritis (OA). In a previous study we described that intra-articular injections of liposomal preparations of adenosine completely prevent progression and reverse cartilage loss in post-traumatic OA. TGF-β signaling plays dual and opposing roles in cartilage health and chondrocyte life depending on the signals activated downstream. Activation of downstream signaling pathways for TGF-β leading to localization of phospho-SMAD2/3 associated with maintenance of cartilage. In contrast nuclear localization of phospho-SMAD1/5/8 results in chondrocyte hypertrophy. Here we report that intraarticular injections of liposomal adenosine and A2AR agonist reverses OA in a post-traumatic OA model in rat and in an obesity related mice model. We moreover explore the role of TGF-β signaling in this phenomenon. Methods: Obesity-induced OA model: C57Bl6 mice (5-6 for each group, 12 weeks old) were fed a 60% fat diet (HFF mice) for 3months, after which received intraarticular injections (10 μl) of empty liposomes (Lipo) or liposomes containing the A2AR agonist CGS21680 (Lipo-CGS) or Adenosine (Lipo-Ado) into the knee every 10 days for 4 injections. Post-traumatic OA (PTOA) was induced in Sprague Dawley rats following non-surgical rupture of anterior cruciate ligament (ACL). Four weeks later rats were injected in the knee with 100ul of saline, Lipo or Lipo-CGS every 10 days (6 injections). RNA was isolated from chondrocytes in knee cartilage of rats treated as described above (3 from each group X 3 replicates) and subjected to RNAseq analysis. TC28a2 human chondrocyte cell line was used for in vitro experiments. Results: Lipo-CGS and Lipo-Ado reversed OA in the obesity and post traumatic OA model. Mouse knees had an OARSI score of 5.17±1.84 before treatment. Treatment with LIPO-Ado and lipo-CGS decreased OA severity (OARSI score 1.33±0.81 and 1.83±0.98, respectively, p<0.001 vs pre-treatment; figure 1). In the PTOA model the OARSI score significantly decreases after Lipo-CGS and Lipo-Ado treatment compare to the saline group (OARSI: 1.28±0.39; 1.42±0.69; 2.97±0.0.75 respectively; p<0.05). Analysis of the transcriptome suggests that the treatment of Lipo-CGS promotes the up-regulation of genes involved in proliferation process and downregulations of genes responsible of apoptosis, cartilage catabolism and chondrocyte hypertrophy (Figure 2). TGF-β expression was increased in deep layers of cartilage in the Lipo-CGS-treated rats and there was notable nuclear localization of phospho-SMAD2/3 in these chondrocytes. In contrast, phospho-SMAD1/5/8 was expressed in the nuclei of chondrocytes in the saline and LIPO-treated rats but not in the LIPO-CGS treated rats. Identical changes were observed in the knees of obese mice. To determine whether the effect of A2AR stimulation on TGF-β signaling was direct or indirect we studied the effect of CGS21680 on nuclear phospho-SMAD expression in TC28a2 cells and found that CGS21680 increased nuclear phospho-SMAD2/3 and reduced nuclear phospho-SMAD1/5/8, as detected by immunofluorescence. Conclusions: Administration of an A2AR agonist to established OA knees reverses OA in rats and mice and shifts TGF-β signaling from ALK1/SMAD1/5/8 to ALK5/SMAD2/3 in OA chondrocytes after activation of A2AR in 2 OA animal models.