Recent human studies support a positive benefit of exercise in improving subjective clinical outcomes in OA. These studies however have mainly been focussed on improving joint mobility or reducing pain scores14, and have not comprehensively interrogated changes in joint tissues and bone microarchitecture to evaluate the consequences of exercise regimes on joint physiology. Animal studies have yielded contradictory results, with some indicating that repetitive load-bearing exercise is detrimental and rest improves cartilage damage scores, whilst others report findings that contradict this view12,13,15. This divergence can be explained partly by differences in the OA models used, but importantly also by the differences in the type and nature of exercise regimes employed in these studies; i.e., high-load bearing10,13 to mild voluntary exercise12. The consensus from these previous reports and the present study is that low load bearing exercise exerts a positive effect on the OA joint. In the present study, we used a moderate form of exercise requiring mice to walk 850m a day, 5 days/week, which, when extrapolating the distance walked by a mouse in a week, the exercise protocol increased said distance by 1.5 fold. It is also important to note that this protocol allowed for recovery from surgical intervention before the start of forced exercise unlike other protocols which start shortly after intervention or later when disease is stablished. Despite the moderate protocol, exercise resulted in a decrease in weight gain reflected in the loss of WAT mass, indicating that this form of exercise, although moderate, is having a beneficial physiological effect. This is a notable observation, as it is well established that weight loss reduces risk of OA, as well as improving outcomes in established OA16,17. Importantly, our induced moderate form of exercise resulted in protection against cartilage damage and synovitis after 7-weeks of exercise.
This supports the concept that exercise can have indirect benefit on overall physiology that could in the long-term lead to decreased progression of OA pathology. Further studies are needed to fully interrogate the long-term benefits of this form of moderate exercise, and to elucidate molecular and/or biomechanical mechanisms underlying the observed protection.
In addition to the significant changes in cartilage or synovial tissue pathology, evaluation of bone in the exercise DMM group revealed a more plate-like micro-structure with increased connectivity. This type of microarchitecture offers higher bone strength18. Moreover, there was an early, albeit temporary, improvement in subchondral bone osteosclerosis in the exercise group, expressed by the significantly smaller increase in bone density of the subchondral bone. There was also an initial increase in osteophyte formation, which may indicate an acceleration of the subchondral bone expansion19 that ensues in the bone adaptation phase of the DMM model to dissipate the increased load. These bone features may indicate an improvement in the way the damaged joint is loaded in the exercised group. Bone adapts to changes in mechanical loading and the DMM model substantially changes the way the joint is loaded. In essence, instead of the meniscus dissipating the load in the joint, this is transmitted primarily through cartilage and subchondral bone20. Notably, the delay in subchondral osteosclerosis we report in the exercise group, together with the positive change in the microarchitecture of the metaphyseal trabecular bone, suggest that exercise changes the way in which the load is dissipated throughout the joint. A possible explanation is that in this scenario, load is shifted to the metaphysis rather than subchondral bone. This delay in osteosclerosis might underpin the 8-week cartilage protection we observed. Indeed, we have observed previously that subchondral bone changes occur rapidly, preceding significant cartilage damage in this OA model9, and others (reviewed21). Prior studies demonstrated that DMM reduces muscle function at 4 and 8-weeks post-surgery22. It is therefore feasible that exercise-induced improvement in muscle strength results in joint stabilisation and altered load23,24. However, we did not observe any macroscopic changes in muscle mass (data not shown), thus further studies are required to definitively address this question.
It is important to note that the murine exercise protocol used simulates the situation of a human exercising shortly after sustaining a joint injury of a type likely to induce OA onset. This study does not, however, address how this type of exercise regime would affect established OA; this is a key question that future studies should address. Furthermore, longer-term studies would indicate if this form of moderate exercise affords long-term or merely transient benefit to the joint tissues.
In summary, this study demonstrates that early moderate exercise reduced body weight, cartilage degradation, synovitis and temporarily osteosclerosis, with no indication of any detrimental effects to the joint. In addition to inducing weight loss, exercise reduces inflammation17 and has an important positive psychological impact in humans25. This, taken together with the data presented here, supports the use of moderate low load-bearing exercise as a therapeutic regime following joint injury and/or diagnosis of early OA, based on its collective and overriding physiological benefits.