Date of Award

Spring 5-15-2023

Author's School

McKelvey School of Engineering

Author's Department

Biomedical Engineering

Degree Name

Doctor of Philosophy (PhD)

Degree Type



The skeleton is innervated by sensory and sympathetic nerves, which are known facilitators of pain and vasoregulation. Recent studies suggest that nerves in bone play an additional direct role in regulating bone metabolism through local release of neuropeptides near bone cells; however, this remains controversial. The overall hypothesis of this thesis is that skeletal nerves are therapeutic targets for increasing bone mass and enhancing bone health. To address the feasibility of nerve-targeted skeletal therapeutics, the work in this thesis explores mapping, modification, and stimulation of nerves in bone and the subsequent effects on bone formation and metabolism.Bone is innervated by sensory and sympathetic axons most densely in the periosteum, followed by the marrow and cortical bone. However, comprehensive mapping of nerves in bone has not been previously accomplished due to a lack of reliable imaging and analysis tools to assess neuroskeletal phenotypes. To address this, we mapped sensory, sympathetic, and non-peptidergic axons in the skeleton (Chapter 2). This required comprehensive frozen histology of the long bones from a novel pan-neuronal Baf53b-Ai9 reporter mouse in conjunction with immunostaining for nerve sub-type markers. Results from this work demonstrated that skeletal innervation is not uniform, especially in long bones like the tibia. Within the bone-lining periosteum, we determined that nerves exist in three distinct neuroskeletal niches based on the presence or absence of axons and their orientation. We hypothesized that these niches play a regulatory role in bone metabolism. Next, we sought to identify the function of local periosteal nerves in bone formation, specifically whether niche modification affects or occurs during skeletal adaptation (Chapter 3). Mechanical loading is required for bone health and results in skeletal adaptation to optimize strength. Previous studies suggest that nerves sprout towards bone during this process; however, the extent of sprouting and whether nerves are required for later remodeling remains unknown. To address this, we loaded the tibia of Baf53b-Ai9 mice to induce lamellar bone formation. Mice were injected with calcein twice to label newly formed bone. Mineralizing surface (MS), bone formation rate (BFR) and periosteal axon density were quantified using RadialQuant, a custom tool for spatial analysis of bone. Global periosteal axon density was unchanged by loading, demonstrating that significant nerve sprouting did not occur 6-days after the last bout of loading. However, a trending but non-significant 28 and 50% increase in periosteal axon density occurred at the site of peak compression and at the lateral surface, respectively. This suggests that non-damaging loading may induce subtle, localized nerve sprouting. We also evaluated the necessity of nerves for bone adaptation by performing unilateral femoral and sciatic neurectomy 1-week prior to loading. Neurectomy reduced total periosteal axon density by 86% and unexpectedly evoked a 3.5-fold increase in MS that localized to the lateral aspect of the mid-diaphysis. Additionally, neurectomy resulted in a spatial redistribution of MS and BFR induced by loading without changing the overall magnitude of bone formation. This spatial shift resulted from local blunting of skeletal adaptation at the posterolateral apex and enhancement at the lateral surface. Overall, these results show that the nervous system may play a role in patterning bone formation after applied load and that the anabolic effects of weight-bearing exercise may be reduced in patients with nerve damage or dysfunction. Results from Chapter 3 indicated that efferent signaling from nerves can regulate skeletal adaptation. We next sought to address whether induction of efferent neural signaling is sufficient to induce bone gain in health and disease. In Chapter 4, we stimulated nerves that innervate the skeleton in healthy animals, as well as in a disease model of concurrent nerve dysfunction and bone fragility, type 1 diabetes (T1D). Our hypothesis was that bioelectric stimulation (eStim) would prevent diabetic neuropathy and bone loss in an STZ rat model of T1D. To test this hypothesis, we unilaterally implanted a wirelessly powered nerve cuff to deliver neuroregenerative, supramaximal eStim to the sciatic nerve of rats with and without T1D. While eStim restored some gait and limb length imbalances caused by the cuff-based device, the neuroregenerative eStim treatment regimen we chose was insufficient to rescue either T1D-associated neuropathy or bone microstructural deficits. In addition, eStim did not affect bone microstructure of healthy controls. Nerve health is an important contributor to bone health, especially with regards to gait patterns and vascular tone. Neurotrophic mechanisms may also contribute to bone health, as indicated by their proximity to bone cells in Chapter 2 and their functional contributions to skeletal adaptation in Chapter 3. However, results from this dissertation suggest that skeletal axons may be modulators, but perhaps not initiators, of bone metabolism. Neuromodulation was insufficient to prevent bone loss or promote bone gain in Chapter 4. Thus, nerve-targeted anabolic therapies for the skeleton may be more probable as adjuvant co-therapies with weight-bearing exercise or other pharmaceuticals, rather than standalone interventions.


English (en)


Erica L. Scheller Alexandra Rutz

Committee Members

Shantanu Chakrabartty, Daniel W. Moran, Simon Tang, Deborah Veis,