Osteoblasts Have a Neural Origin in Heterotopic Ossification.

BACKGROUND: Heterotopic ossification (HO) is the process of bone formation at a nonskeletal site. Recently, we showed that the earliest steps occur in sensory nerves. We now extend these studies by identifying unique osteogenic progenitors within the endoneurial compartment of sensory nerves. QUESTIONS/PURPOSES: We asked: (1) What is the nature of the osteoprogenitor in the endoneurium of peripheral nerves? (2) How do osteoprogenitors travel from the nerve to the site of new bone formation? METHODS: HO was induced by intramuscular injection of Ad5BMP-2-transduced cells in mice. Osteoprogenitors were identified through immunohistochemistry and then quantified and further characterized by fluorescence-activated cell sorting and immunocytochemistry. The kinetics of the appearance of markers of extravasation was determined by quantitative reverse transcription-polymerase chain reaction. In each experiment mice were injected with bone morphogenetic protein-2 (BMP-2)-producing cells (experimental) or with cells transduced with empty vector or, in some cases, a group receiving no injection (control). RESULTS: Induction of HO leads to the expression, within 24 hours, of osteoblast-specific transcription factors in cells in the endoneurium followed by their coordinate disappearance from the nerve at 48 hours. They reappear in blood also at 48 hours after induction. During vessel entrance they begin to express the tight junction molecule, claudin 5. The cells expressing both the osteoblast-specific transcription factor, osterix, as well as claudin 5, then disappear from circulation at approximately 3 to 4 days by extravasation into the site of new bone formation. These endoneurial osteoprogenitors express neural markers PDGFRα, musashi-1, and the low-affinity nerve growth factor receptor p75(NTR) as well as the endothelial marker Tie-2. In a key experiment, cells that were obtained from mice that were injected with cells transduced with an empty vector, at 2 days after injection, contained 0.83% (SD, 0.07; 95% confidence interval [CI], 0.59-1.05) cells expressing claudin 5. However, cells that were obtained from mice 2 days after injection of BMP-2-producing cells contained 4.5% cells expressing claudin 5 (SD, 0.72%; 95% CI, 2.01-6.94; p < 0.0015). Further analysis revealed that all of the cells expressing claudin 5 were found to be positive for osteoblast-specific markers, whereas cells not expressing claudin 5 were negative for these same markers. CONCLUSIONS: The findings suggest that the endoneurial progenitors are the major osteogenic precursors that are used for HO. They exit the nerve through the endoneurial vessels, flow through vessels to the site of new bone formation, and then extravasate out of the vessels into this site. CLINICAL RELEVANCE: The biogenesis of osteoblasts in HO is very different than expected and shows that HO is, at least in part, a neurological disorder. This could result in a major shift in orthopaedic methodologies to prevent or treat this disease. The fact that nerves are intimately involved in the process may also provide clues that will lead to an explanation of the clinical fact that HO often occurs as a result of traumatic brain injury.