Background Clinical and in vitro studies suggest that subchondral bone is involved in the progression and/or onset of osteoarthritis (OA). Human osteoblasts (Obs) isolated from subchondral OA bone tissue show an altered phenotype and particularly produce higher level of leptin than normal human Obs. It has recently been demonstrated that Hif (hypoxia-inducible factor) signaling could play a role in bone cell physiology. Venous outflow occlusion, intraosseous hypertension, decreased perfusion, and hypoxia have all been reported to occur in the subchondral bone tissue of the OA-like guinea pig animal model and in human OA. Modeling the relationships among perfusion, intraosseous pressure, and pO2 suggests that venous outflow obstruction, stasis, and intraosseous hypertension are associated with physiologically relevant reductions in perfusion and pO2 in OA bone tissue, and the pathophysiological consequences of venous outflow obstruction may lie in its association with diminished perfusion and hypoxia.
Objectives We have investigated whether hypoxia could play a role in the OA-induced leptin production in human Obs.
Methods We prepared primary human subchondral osteoblasts using the sclerotic medial portion of the tibial plateaus of OA patients undergoing total knee arthroplasty. Obs were cultured either in 20% (normoxia) or 2% (hypoxia) oxygen tension in the presence or not of VitD3 (50nM). The expression of leptin, osteocalcin (OCN), Hif-1 and -2 was evaluated by qRT-PCR and leptin production by ELISA. Hif-1 and Hif-2 expression was reduced with siRNA technique, and Hif-2 stabilization measured by western blotting. In addition to Hif signaling, MAPK signaling was also investigated using several inhibitors.
Results We verified that Obs were still responsive to VitD3 under hypoxia by analyzing OCN expression. Although VitD3-induced stimulation was slightly decreased from 220-fold in normoxic to 175-fold in hypoxic conditions, no statistical significance was reached (n=5). Leptin expression was stimulated 7-fold under hypoxic conditions (p<0.05). Whereas, VitD3 stimulated leptin expression 2-fold under normoxic condition (p<0.05), hypoxia enhanced this stimulation 9-fold (p<0.05). We confirmed by ELISA the hypoxia-induced leptin production, particularly in presence of vitD3 co-incubation (p<0.05, n=8). Compared to Obs incubated in the presence of siScramble RNA, siHif-1 had almost no effect whereas siHif-2 inhibited leptin expression by 70% and 55% in presence or not of vitD3 respectively. Under the same conditions, siHif-1 and siHif-2 inhibited their own expression by 67% and 75% respectively. We demonstrated by immunoblotting that vitD3 greatly increased Hif-2 stabilization at the protein level under hypoxic condition. The regulation of leptin by Hif-2 was also confirmed by measuring leptin with ELISA (p<0.05, n=5). Finally, in addition to Hif-2 regulation, the huge increase of leptin expression under hypoxia and vitD3 is mainly controlled by p38 map and PI-3 kinases (p<0.05, n=3).
Conclusions The outflow obstruction and decreased perfusion measured in OA subchondral bone tissue that contribute to the hypoxic condition could trigger Obs to produce leptin, a well known factor involved in the pathogenesis of osteoarthritis and abnormal OA Obs function.
Acknowledgements This work was supported by the Osteoarthritis Chair of Excellence (CNRS/Université de Lorraine). E Abed is recipient of a Post-Doctoral award from the FRSQ.
Disclosure of Interest None Declared