NADPH OXIDASE 4 DEFICIENCY ATTENUATES EXPERIMENTAL OSTEOARTHRITIS IN MICE

Purpose: Osteoarthritis (OA) is a degenerative disease characterized by damage of articular cartilage, alteration of subchondral bone, and inflamed synovium. Articular cartilage is essentially composed of an avascular matrix produced by chondrocytes. In OA, articular cartilage follows a progressive degeneration, where low-grade inflammation plays a pivotal role through an oxidative stress-dependent mechanism and exposure to reactive oxygen species (ROS). In chondrocytes, the only cells of the cartilage, NADPH oxidase 4 (NOX4) is one of the major ROS producers and plays a major role in catabolism regulation. In this study, we evaluated the role of NOX4 during experimental OA in mice.

Methods: Cartilage explants isolated from 9 weeks old femoral heads (wild type (WT) and NOX4^-/- mice) were cultured 72 hours with either PBS or IL-1β to induce experimental OA ex vivo. Cartilage explants were then fixed with paraformaldehyde 4% and prepared for histology analysis. Anabolism, catabolism, NOX4 expression and oxidative stress were assessed by immunohistochemistry (IHC). In vivo, OA was induced by destabilization of the medial meniscus (DMM) in WT and NOX4^-/- mice (n=12 for each group). At day 0, mice were radiographed using a microCT and operated. Eight weeks after surgery, mice were radiographed and sacrificed for histological analysis. Anabolism, catabolism, NOX4 expression, macrophage recruitment and oxidative stress were assessed by IHC. Bone phenotype was also determined by histomorphometry on sections after safranin’O coloration.

Results: Ex vivo, we demonstrated with safranin’O staining that NOX4 deletion protects from the loss of proteoglycans in the cartilage induced by IL-1β. We also showed inWT explants, a decrease of AGG expression and an increase of MMP13 and COL1 expression, but we did not observe any effects on COL2 expression after IL-1β treatment. In NOX4^-/- compared to WT cartilage, we observed a higher expression of AGG, a lower expression of MMP13 and COL1 in the superficial cartilage. To better understand the role of NOX4, we performed NOX4 and 8-OHdG (ROS marker) IHC. As expected, we showed an increase of NOX4 expression in explants treated with IL-1β and no staining in the KO. Even more, we demonstrated that IL-1β stimulation increased cartilage expression of 8-OHdG in WT explants but not in NOX4-/- explants. In vivo, histological analysis of the knee joint, 8 weeks post-surgery, demonstrated that DMM induced significant OA damages in WT mice (OARSI score of 2.8±1.2). A significant improvement was observed in DMM-NOX4^-/- mice with a reduced OARSI score of 1.5±0.5. We observed thicker cartilage and less osteophytes formation in DMM-NOX4^-/- mice as compared to DMM-WT mice. By micro-CT, we observed no differences at day 0 between WT and NOX4^-/- mice. However, 8 weeks after DMM, a slight increase was observed in the trabecular thickness (Tb.Th) and in the trabecular space (Tb.Sp) as compared to day 0, only in WT mice. By histomorphometry we confirmed that 8 weeks after DMM there is a slight increase of the BV/TV, Tb.Th and Tb.Sp. Moreover, we observed that the NOX4^-/- mice had a higher Tb.Th compared to the WT mice. By IHC we first showed that there is a stronger NOX4 expression after DMM surgery compared to the sham. Interestingly, we only observed NOX4 staining in the cartilage and not in the synovial membrane. We then evaluated the effects of NOX4 deletion on cartilage anabolism and catabolism by IHC. As expected DMM surgery increased catabolism (MMP13 and COL1 expression) and inhibited anabolism (COL2 expression). In comparison NOX4 deletion in mice inhibited MMP13 and COL1 expression and increased COL2 expression after DMM surgery. Then we assessed if the effects observed in the NOX4^-/- mice was linked to synovial inflammation and ROS production in the cartilage in vivo (Figure 8). First, we evaluated the expression of NOX4 by IHC. With HE staining, we demonstrated that DMM surgery induced synovial inflammation (increased cellularity and thickness) in the WT mice but not in the NOX4-/- mice. Using synovitis scoring system, we observed an inflammation score of 2.5 ± 0.5 in WT mice versus 0.9 ± 0.4 (p<0,01) in NOX4-/- mice. In accordance with these results, we observed a strong decrease of the expression of the macrophage marker F4/80 in the synovial membranes of the NOX4-/- mice compared to the WT mice. Finally, we demonstrated that NOX4 deletion inhibited 8-OHdG expression in the cartilage and decreased its expression in the synovial membrane.

Conclusions: Our results demonstrated that NOX4 deficiency decreases cartilage degradation and regulate the anabolism/catabolism balance ex-vivo after IL-1β stimulation and decreases significantly experimental OA severity in mice. We also showed that NOX4 increase OA severity via chondrocytes ROS production and synovial inflammation. Taken together these results underline that NOX4 could be a major target to dampen OA progression.