Steeve Kwan Tata, Jean-Pierre Pelletiera, François Mineaua, Judith Carona, and Johanne Martel-Pelletier
In osteoarthritis (OA) the progression of cartilage degeneration has been associated with remodeling of the subchondral bone. Human OA subchondral bone osteoblasts were shown to have an abnormal phenotype and altered metabolism leading to an abnormal resorptive process. Bone resorption is suggested to occur, at least in part, through the increased levels of two proteolytic enzymes, MMP-2 and MMP-9, and RANKL, which are mainly produced by osteoblasts. In this study, we investigated in human OA subchondral bone osteoblasts the modulatory effect of strontium ranelate on the above key factors. Human subchondral bone osteoblasts were cultured in a medium containing 0.1, 1 and 2 mM of strontium ranelate for 18 h for mRNA and 72 h for protein determination. The effect of strontium ranelate was evaluated on the expression (qPCR) of MMP-2, MMP-9, OPG, RANKL (total), RANKL-1, and RANKL-3, on the production of OPG (ELISA), membranous RANKL (flow cytometry), and MT1-MMP, ADAM17, and ADAM19 (Western blot). After incubation of osteoblasts with pre-osteoclasts (i.e., differentiated human peripheral blood mononuclear cells), the resorbed surface was measured using a sub-micron synthetic calcium phosphate thin film. Firstly, the expression levels of MMP-2, MMP-9, OPG, and RANKL were determined in normal and OA subchondral bone osteoblasts. As expected, the gene expression of MMP-9 and RANKL were not detectable in normal cells, whereas MMP-2 was very low but detectable and OPG demonstrated high gene expression. Further experiments looking at the effect of strontium ranelate on expression levels, except for OPG, were performed only on the OA subchondral bone osteoblasts. In OA cells, the expression levels of MMP-2 and MMP-9 were significantly decreased by strontium ranelate at 1 mM (p ≤ 0.005, p ≤ 0.02, respectively) and 2 mM (p ≤ 0.003, p ≤ 0.007), and for MMP-9 only at 0.1 mM (p ≤ 0.05). In normal cells, the expression of OPG was increased with strontium ranelate at 2 mM, and in OA both the expression (p ≤ 0.02) and synthesis (p ≤ 0.002) of OPG were significantly increased with strontium ranelate at 1 and 2 mM. RANKL (total) as well as the isoforms RANKL-1 and RANKL-3 were significantly increased by strontium ranelate at 1 and 2 mM. Of note, it is known that the different RANKL isoforms differentially regulate RANKL membranous localization: RANKL-3, in contrast to RANKL-1, prevents such membranous localization. This is reflected by the significant (p ≤ 0.02) reduction in the level of membranous RANKL by strontium ranelate at 2 mM. This latter finding was not likely to be related to a proteolytic cleavage of membranous RANKL, as the enzymes known to cleave it, MT1-MMP, ADAM17 and ADAM19, were unaffected by strontium ranelate. In addition, OA osteoblasts treated with strontium ranelate induced a significant (p ≤ 0.002) decrease in resorbed surface at the three tested concentrations. This study provides new insights into the mode of action of strontium ranelate on the metabolism of human OA subchondral bone osteoblasts. These data suggest that strontium ranelate may exert a positive effect on OA pathophysiology by inhibiting, in these cells, the synthesis of key factors leading to bone resorption, a feature associated with the OA process.