Fragility fractures, also known as osteoporotic fractures, result from systemically low bone mass and degraded bone microarchitecture. Osteoporosis is a major global burden that is only expected to escalate in significance with an increasingly aging population. While anabolic and antiresorptive bone therapies, like romosozumab and denosumab respectively, have been developed and approved to significantly improve bone strength, they are not without key limitations in use or an elevated risk of adverse events. Recent studies have shown the safety and efficacy of mechanical loading for improving bone strength and reducing fracture risk in postmenopausal women and in older men. However, there are still gaps in understanding the specific mechanisms of bone adaptation at the microstructural level. Importantly, osteoporosis is a “silent” disease as it often goes undetected until a fracture occurs.

To this end, advanced in vivo micro-computed tomography (µCT) imaging and analysis techniques are used in this study to quantify trabecular bone adaptations under mechanical loading. This study combines the use of µCT-based 3D dynamic histomorphometry, individual trabecula segmentation (ITS), and the interface surface area metric (iSAM) to delineate the longitudinal trabecular microarchitectural adaptations. Furthermore, the murine axial compression tibial loading model and transgenic mouse models are used to investigate the underlying mechanisms of trabecular bone mechanoadaptation.

In the first aim, we used the mouse tibial loading model to establish how loading increases trabecular bone volume. Using µCT-based 3D dynamic histomorphometry, weekly changes in formation and resorption with loading were identified. These results were combined with ITS analyses to separately quantify the individual trabecular responses by type (plate or rod), orientation (axial, oblique, or transverse), and region with respect to the surrounding cortical shell (outer or inner). We found that there are distinct responses to loading at the microarchitectural level. Plate formation significantly increased with loading, while rod resorption significantly decreased with loading. There also may be differential responses to loading in terms of the trabecular region relative to the surrounding cortical shell, with increased formation in outer trabeculae and decreased resorption in inner trabeculae. We also found that iSAM results follow the regional analysis with an increased interface surface area in response to loading.

In the second aim, the established methods for quantitatively analyzing trabecular microstructural dynamics are applied to mice with depleted sclerostin and RANKL. We found that in wild-type (WT) mice, trabecular bone responds to loading even with depletion of RANKL by OPG treatment. However, in mice with depletion of sclerostin by genetic knockout of SOST, we find that there is no overall response to loading when compared to control limbs. Compared to WT mice, SOST knockout mice have reduced relative amounts of bone volumes formed and resorbed. Treatment with OPG reduced resorption to an even greater degree, and these effects were mostly consistent across the trabecular microarchitecture.

Finally, in the third aim, the same techniques are applied as in the previous aims but using a transgenic mouse model to ablate osteocytes, the primary mechanosensor and major cellular source of sclerostin and RANKL. The findings in this aim build on the previous aims, where there is a significant trabecular bone volume response to loading in control mice, but no loading effect in osteocyte-depleted mice over 2 weeks of tibial loading. Like the effects of sclerostin and RANKL depletion, osteocyte-depleted mice had increased trabecular formation and decreased resorption. We also found that the effects on formation and resorption were more robust in plates and rods, respectively.

Taken together, these studies demonstrate how trabecular microstructural adaptations and dynamics can be quantified longitudinally over the course of an applied intervention like in vivo tibial loading. Important insights into the mechanoregulation of trabecular bone are gained by using transgenic mouse models to significantly disrupt trabecular bone adaptation to mechanical loading. Future work will clarify the underlying mechanical and biological mechanisms of trabecular bone mechanosensing and mechanotransduction.