Osteoclastogenesis is repressed by mechanical strain in an in vitro model

Functional loading provides a site-specific signal for the regulation of bone mass and morphology. To determine if strain can inhibit the resorptive component of bone remodeling, osteoclast formation was assessed in marrow cultures plated on flexible membranes subjected to 5% strain for 10 cycles/minute, 24 hours per day. Cultures strained during days 2 through 7 inhibited osteoclast formation to 61+/-7% of control cultures (p < 0.05), a degree of inhibition similar to that observed when the cultures were subjected to strains during only days 2 through 4 but also evaluated on day 7 (67+/-4% of control; p < 0.05). In contrast, straining of cultures during days 5 through 7 had little influence on inhibiting the formation of osteoclasts (94+/-5% of control; no significant difference). The nonuniformly strained substrate was subdivided into three concentric rings. and cultures were used to examine the site-specificity of the inhibition caused by strain. Osteoclast formation in the outermost boundary, which was distended from 3.6 to 5%, was 41+/-7% of that observed in outer regions of control wells. The inhibitory potential of mechanical strain was reduced within the middle ring (73+/-6% of control osteoclasts: p < 0.01), where the strain ranged from 0.2 to 3.6%. The central region, which experienced strains equivalent to those in the middle ring (0.2 to -4% strain), showed inhibition of osteoclast formation to a similar degree (75+/-6% of control). Media harvested from strained cultures failed to inhibit osteoclast formation in unstrained cultures; this implies that the inhibitory effect of strain depended on the direct interaction of the cell with the substrate rather than by a humoral factor. A second device, where a uniform strain was delivered at 1.8% throughout the entire plate, inhibited osteoclast recruitment to 48+/-3.6%, emphasizing that uniform strain in the absence of shear stress constrains osteoclast recruitment. These in vitro experiments can but model the complex environment generated by in vivo mechanical strains: however, they provide the first direct evidence that strain must be considered as inhibitory to osteoclast recruitment.