Our previous rat study demonstrated an ex vivo-created “Biomimetic Hematoma” (BH) that mimics the intrinsic structural properties of normal fracture hematoma, consistently and efficiently enhanced the healing of large bone defects at extremely low doses of rhBMP-2 (0.33 μg). The aim of this study was to determine if an extremely low dose of rhBMP-2 delivered within BH can efficiently heal large bone defects in goats.

Goat 2.5 cm tibial defects were stabilized with circular fixators, and divided into groups (n=2-3): 2.1 mg rhBMP-2 delivered on an absorbable collagen sponge (ACS); 52.5 μg rhBMP-2 delivered within BH; and an empty group. BH was created using autologous blood with a mixture of calcium and thrombin at specific concentrations. Healing was monitored with X-rays. After 8 weeks, femurs were assessed using microCT.

Using 2.1 mg on ACS was sufficient to heal 2.5 cm bone defects. Empty defects resulted in a nonunion after 8 weeks. Radiographic evaluation showed earlier and more robust callus formation with 97.5 % (52.5 μg) less of rhBMP-2 delivered within the BH, and all tibias were fully bridged at 3 weeks. The bone mineral density was significantly higher in defects treated with BH than with ACS. Defects in the BH group had smaller amounts of intramedullary and cortical trabeculation compared to the ACS group, indicating advanced remodeling.

The results confirm that the delivery of rhBMP-2 within the BH was much more efficient than on an ACS. Not only did the large bone defects heal consistently with a 40x lower dose of rhBMP-2, but the quality of the defect regeneration was also superior in the BH group. These findings should significantly influence how rhBMP-2 is delivered clinically to maximize the regenerative capacity of bone healing while minimizing the dose required, thereby reducing the risk of adverse effects.

Treatment of segmental bone defects remains a major clinical problem, and innovative strategies are often necessary to successfully reconstruct large volumes of bone. When fractures occur, the resulting hematoma serves as a reservoir for growth factors and a space for cell infiltration, both crucial to the initiation of bone healing. Our previous studies have demonstrated very clear ultrastructural differences between fracture hematomas formed in normally healing fractures and those formed in segmental bone defects. However, there is little information available regarding potential differences in the underlying gene expression between hematomas formed in normal fractures, which usually heal by themselves, and segmental bone defects, which do not. Therefore, the aim of this study was to identify differences in gene expression within hematomas collected from 0.5 mm (normal fracture) and 5 mm (segmental bone defect) fracture sites during the earliest stages of bone healing.

Osteotomies of 0.5 and 5 mm in the femur of Fisher 344 rats were stabilized with external fixators (RISystem AG). After 3 days the rats were sacrificed, and the fracture hematomas were collected for RNA-sequencing. Ingenuity pathway analysis (IPA) was used to identify upstream regulators and biological functions that were significantly enriched with differentially expressed genes from the RNA-sequencing analysis. Animal procedures were conducted following the IACUC protocol of the UT Health Science Center San Antonio.

Key upstream regulators of bone formation were less active (e.g. TGFB1, FGF2, SMAD3) or even inhibited (e.g. WNT3A, RUNX2, BMP2) in non-healing defects when compared to normally healing fractures. Many upstream regulators that were uniquely enriched in healing defects were molecules recently discovered to have osteogenic effects during fracture healing (e.g. GLI1, EZH2). Upstream regulators uniquely enriched in non-healing defects were mainly involved in an abnormal modulation of hematopoiesis, revealing evidence of impaired maturation of functional macrophages and cytokines (e.g. IL3, CEBPE), both essential for successful bone healing. In addition, the enrichment pattern suggested a dysregulation of megakaryopoiesis (e.g. MRTFA, MRTFB, GATA2), which directly affects platelet production, and therefore fracture hematoma formation. Remarkably, the organization of the ECM was the most significantly enriched biological function in the normally healing fractures, and implies that the defect size directly affected the structural properties within the fracture hematoma. Conversely, genes encoding important ECM components (e.g. BGN, various collagens, IBSP, TNC), cell adhesion molecules, MMPs (MMP2), and TIMPs were all significantly downregulated in non-healing defects.

Our most recent findings reveal new important key molecules that regulate defect size-dependent fracture healing. Combined with our previous results, which identified structural differences in fracture hematomas from both types of defects, current findings indicate that differential expression of genes is dictated by the structural properties of the hematomas formed during early fracture healing. Consequently, creating a bioscaffold that mimics the structure of normal fracture hematomas could be the first step towards developing new orthoregenerative treatment strategies that potentiate healing of large bone defects and non-healing fractures.