H Vessel Formation as a Marker for Enhanced Bone Healing in Irradiated Distraction Osteogenesis

Abstract In the setting of bone defects, the injured vasculature and loss of hemodynamic inflow leads to hematoma formation and low oxygen tension which stimulates vascular expansion through the HIf-1α pathway. Most importantly, this pathway upregulates sprouting of type H vessels (CD31hiEmcnhi vessels). H vessels engage in direct interaction with perivascular osteoprogenitor cells (OPCs), osteoblasts, and preosteoclasts of bone formation and remodeling. This angiogenic-osteogenic coupling leads to synchronous propagation of vascular and bony tissue for regenerative healing. A growing body of literature demonstrates that H vessels constitute a large portion of bone's innate capacity for osteogenic healing. We believe that CD31hiEmcnhi vessels play a role in bone healing during distraction osteogenesis (DO). DO is a procedure that utilizes traction forces to facilitate induction of endogenous bone formation and regeneration of surrounding soft tissues such as skin, muscle, tendon, and neurovascular structures. While the H vessel response to mechanical injury is adequate to facilitate healing in normal healthy tissue, it remains inadequate to overcome the devastation of radiation. We posit that the destruction of CD31hiEmcnhi vessels plays a role in precluding DO's effectiveness in irradiated bone defect healing. We aim, therefore, to recapitulate the normal pathway of bony healing by utilizing the regenerative capacity of H vessels. We hypothesize that using localized application of deferoxamine (DFO) will enhance the H vessel-mediated vasculogenic response to radiation damage and ultimately enable osteogenic healing during DO. This discovery could potentially be exploited by developing translational therapeutics to hopefully accelerate bone formation and shorten the DO consolidation period, thereby potentially expanding DO's utilization in irradiated bone healing. Sprague–Dawley rats were divided into three groups: DO, radiation with DO (xDO), and radiation with DO and DFO implantation (xDODFO). Experimental groups received 35 Gy of radiation. All groups underwent DO. The treatment group received injections into the osteotomy site, every other day, beginning on postoperative day (POD) 4 of DFO. Animals were sacrificed on POD 40. For immunohistochemical analysis, mandibles were dissected and fixed in 4% paraformaldehyde for 48 hours, decalcified in Cal-Ex II for 2 days, dehydrated through graded ethanol of increasing concentration, and then embedded in paraffin. Samples were cut into 7-μm thick longitudinally oriented sections including the metaphysis and diaphysis. CD31 and Emcn double immunofluorescent staining were performed to evaluate the extent of CD31hiEmcnhi vessel formation. Bone sections were then stained with conjugated antibodies overnight at 4°C. Nuclei were stained with Hoechst. Slides were also double stained with Osterix and CD31 to study the quantity of H vessel-mediated recruitment of OPCs to accelerate bone healing. Images were acquired with a Nikon Ti2 widefield microscope and analyzed in NIS- Elements Advanced Research 5.41.02 software. The abundance of type H vessels is represented by the area fraction of CD31 + Emcn+ vessel area inside the regenerate sample. OPC concomitant proliferation into the distraction gap is represented by the area fraction of Osterix+ cell area inside of the regenerate sample. There were 6× more type H vessels in DO groups than in xDO groups. Localized DFO significantly increased the abundance of type H vessels of irradiated DO animals compared to xDO by 15× (p = 0.00133531). Moreover, the DO and xDODFO groups with higher abundance of type H vessels also demonstrated better angiogenesis and osteogenesis outcomes. Interestingly, xDODFO groups doubled the quantity of H vessel formation compared to DO, indicating a supraphysiologic response (p = 0.044655055). Furthermore, H vessel-mediated recruitment of OPCs mimicked the described H vessel formation trend in our study groups. Irradiated DO groups contained 3× less OPCs compared to DO controls. DFO treatment to xDO animals remediated irradiation damage by containing 12× Osterix+ cells. Finally, DFO treatment of irradiated animals quadrupled osteoprogenitor recruitment into the distraction gap compared to DO controls. In this study, we developed a novel approach to visualize CD31hiEmcnhi in paraffin sections to study DO regeneration. Normal DO demonstrated a significant upregulation of H vessel formation and associated angiogenic-osteogenic coupling. Radiation severely decreased H vessel formation along with an associated significant diminution of new bone formation and nonunion. DFO administration, however, resulted in vascular replenishment and the restoration of high quantities of CD31hiEmcnhi and OPCs, recapitulating the normal process of bony regeneration and repair. DFO treatment remediated new bone formation and bony union in irradiated fields associated with increased H vessel angiogenic-osteogenic coupling. While further studies are required to optimize this approach, the results of this study are incredibly promising for the long-awaited translation of localized DFO into the clinical arena.