Session VI - Fracture Repair

The Effects of Delayed Callus Vascularization on Fracture Repair: A Murine Gene Knockout Model

Eytan Alpern, DDS, PhD; Theodore Miclau, MD; Thiennu H. Vu, MD; Zena Werb, MD; Jill Helms, MD; University of California, San Francisco, CA

Introduction: Clinicians have long recognized the importance of revascularization in fracture repair. Over the past decade, in an effort to improve healing, indirect reduction and minimally invasive techniques have been developed to better preserve blood supply to the fracture during fixation. Interruption of this blood supply is thought to be a major factor in the development of fracture non-union. The role of neovascularization, however, is incompletely understood. Gelatinase B (Gel B), a member of the matrix metalloproteinase (MMP) family, degrades components of the extracellular matrix and is thought to be important in vascularization and ossification. In the absence of this gene, mice exhibit normally developing hypertrophic chondrocytes but have delayed vascularization and ossification. The purpose of this study was to investigate the role of this gene in adult mouse fracture repair.

Materials and Methods: Adult 10-12 week gel B -/- knockout mice (n=48) and their wild type litter mates were anesthetized and underwent closed, standardized unstabilized fractures produced via three-point bending. The fractures were radiographed postoperatively. The animals were sacrificed at post-fracture days 14, 17, 21, 24, 28, and 42. The callus tissues were decalcified, dehydrated, and embedded. The tissue was analyzed histologically using Safranin O-Fast Green staining. In-situ hybridizations, sensitive methods for visualizing gene expression patterns, were performed using cDNAs of the mouse genes collagen type II (col2), and collagen type X (col10).

Results: Gel B -/- mice displayed abnormal fracture healing. At 14 days post-fracture, the gel B -/- fracture calluses were larger than those in the wild type mice and contained abnormally large areas of hypertrophic cartilage. The presence of hypertrophic cartilage in these areas was confirmed by the col10 staining. This pattern persisted through days 17 and 21. By 24 days after fracture, no cartilage was visualized in either the gel B -/- or wild type calluses.

Conclusion: Gel B -/- mice exhibited delayed fracture healing relative to the wild type mice with increased areas of hypertrophic cartilage. The appearance of this cartilage callus bears resemblance to that of a clinical nonunion which demonstrates interruption of the soft callus stage of fracture healing. These results suggest that the breakdown of the extracellular matrix is important in the neovascularization of the fracture callus. In addition, the histological appearance of the abnormal ossification that occurs in the gel B -/- callus resembles that seen post-natally in the gel B growth plate. These findings support the hypothesis that embryonic genes are reinduced during adult fracture repair.