Previous studies have indicated that physiological levels of dynamic mechanical strain produce rapid increases in nitric oxide (NO) release from rat ulna explants and primary cultures of osteoblast-like cells and embryonic chick osteocytes derived from long bones. To establish the mechanism by which loading-induced NO production may be regulated, we have examined: nitric oxide synthase (NOS) isoform mRNA and protein expression, the effect of mechanical loading in vivo on NOS mRNA expression, and the effect of mechanical strain on NO production by bone cells in culture. Using Northern blot analyses, in situ hybridization, and immunocytochemistry we have established that the predominant NOS isoform expressed in rat long bone periosteal osteoblasts and in a distinct population of cortical bone osteocytes is the endothelial form of NOS (eNOS), with little or no expression of the inducible NOS or neuronal NOS isoforms. In contrast, in non–load-bearing calvariae there are no detectable levels of eNOS in osteocytes and little in osteoblasts. Consistent with these observations, ulnar explants release NO rapidly in response to loading in vitro, presumably through the activation of eNOS, whereas calvarial explants do not. The relative contribution of different bone cells to these rapid increases in strain-induced NO release was established by assessment of medium nitrite (stable NO metabolite) concentration, which showed that purified populations of osteocytes produce significantly greater quantities of NO per cell in response to mechanical strain than osteoblast-like cells derived from the same bones. Using Northern blot hybridization, we have also shown that neither a single nor five consecutive daily periods of in vivo mechanical loading produced any significant effect on different NOS isoform mRNA expression in rat ulnae. In conclusion, our results indicate that eNOS is the prevailing isoform expressed by cells of the osteoblast/osteocyte lineage and that strain produces increases in the activity of eNOS without apparently altering the levels of eNOS mRNA.

Responses to mechanical strain in calvaria and limb bone organ cultures were compared by measuring cellular glucose 6-phosphate dehydrogenase (G6PD) activity in situ and prostaglandin release. Normal functional strains were recorded in the ulnae (1000 μϵ) and calvarium (30 μϵ) in vivo in 110 g rats. Organ cultures of ulnae and calvaria from similar animals were loaded to produce dynamic strains (600 cycles, 1 Hz) of 1000 μϵ in the ulna, and 100 or 1000 μϵ in calvaria. In ulnae, both PGE₂ and PGI₂ were released and resident osteocytes and osteoblasts showed increased G6PD activity. Neither response was seen in calvaria. However, exogenous PGI₂ (10⁻⁵−10⁻⁹ M) stimulated G6PD activity in osteocytes and osteoblasts in organ cultures of both calvaria and ulnae. In ulnar cells the response was linear, in calvarial cells it was biphasic with maximum activity at 10⁻⁷ M. Osteoblasts derived from ulnae and cultured on plastic plates subjected to dynamic strain (600 cycles, 1 Hz, 4000 μϵ) showed increased G6PD activity. There was no such response in similarly treated calvarial-derived cells. Calvarial bone cells differ from those of the ulna in that they do not respond to physiological strains in their locality with increased prostanoid release or G6PD activity either in situ or when seeded onto dynamically strained plastic plates. Cells from both sites in organ culture show increased G6PD activity in response to exogenous PGI₂, but their dose: Responses differ in shape. These differences may reflect the extent to which functional loading influences bone architecture in these two sites.