Although a bit surprising, a negative effect of axial mechanical load on trabecular bone of WT mice is consistent with previous studies from our group in aged mice, showing that trabecular bone volume actually decreases in young adult mice upon axial compression using the waveform used here. This is in contrast to findings from other groups who have used tibial compression under different loading regimes. Preliminary results from our laboratory suggest this discrepancy may depend on the waveform and cycle number. On the other hand, BV/TV significantly increased in cKO mice with 1200 me load, further suggesting that Cx43 is involved in response to load, though the trabecular compartment in the cKO may experience greater compressive strains than WT due to the altered architecture. Interestingly, axial load increased periosteal formation rate in cKO loaded at 1200 me as well as in WT loaded at 1900 me; however, periosteal mineral apposition rate did not significantly change in either group. Such a discrepancy implies a higher number of VE-822 osteoblasts depositing new bone, rather than increased activity of existing osteoblasts. Thus, these results are consistent with the notion that axial compression loading promotes osteoblast recruitment and differentiation on the periosteal surface, an effect enhanced by Cx43 deficiency. It is possible that signaling from osteocytes to the periosteal surface is enhanced in the cKO. However, since periosteal cells in cKO mice are exposed to a lower degree of strain for a given force, the fact that periosteal bone formation is increased in Cx43 deficient bones relative to WT bone even in animals living under normal loading conditions strongly suggests that periosteal cells from cKO are more sensitive to mechanical strain than are periosteal WT cells. that 1200 me was sufficient in cKO to elicit the same bone formation response as did 1900 me in WT animals. Notably, a periosteal increase in bone diameter contributes more to load bearing strength than do increases in cortical thickness associated with endocortical apposition. Thus, the increased periosteal bone formation observed in the cKO may be related to the need for generating bone at the diaphysis to withstand loads that are sensed as abnormally higher than they actually are. Increased periosteal bone apposition but decreased endosteal formation after mechanical loading had been previously reported in 9 week old C3H/HeJ mice. Such a paradoxical effect on the endocortical surface may reflect the young age of the mice used in these studies, and the decrease in bone formation rate may represent a slowing of the rapid bone formation rate occurring in growing mice. Since such an effect is more pronounced in mice with an osteoblast/osteocyte specific deletion of Gja1, it can be concluded that Cx43 deficiency also increases the sensitivity of the endocortical cells to mechanical loading. Decreased endocortical bone formation by axial load is in apparent contrast with our previous study showing increased endosteal bone formation after application of a 3-point bending load. Aside the different loading approaches, other differences in the experimental settings may contribute to the discrepancy, including a substantially larger strain and older mice used in the 3- point bending experiment relative to the present study, as well as different promoters used to drive Gja1 ablation. In both cases, however, Cx43 deficiency reduces bone formation at the endocortex upon mechanical loading.