Revitalization of Bone Allografts by Murine Periosteal Cells Expressing BMP2 and VEGF

Revitalization of Bone Allografts by Murine Periosteal Cells Expressing BMP2 and VEGF

Nick van Gastel,1,5 Maarten Depypere,2 Scott Roberts,3,5 Ingrid Stockmans,1 Riet Van Looveren,1 Jan Schrooten,4,5 Frederik Maes,2 Frank P. Luyten,3,5 Geert Carmeliet,1,5

1Laboratory of Experimental Medicine and Endocrinology, 2Department of Electrical Engineering (ESAT/PSI), 3Laboratory for Skeletal Development and Joint Disorders, 4Department of Metallurgy and Materials Engineering and 5Prometheus, Division of Skeletal Tissue Engineering, Katholieke Universiteit Leuven, Leuven, Belgium

To date, autologous bone transplantation remains the therapy of choice to treat large bone defects. Nevertheless, the use of autografts is restricted by major drawbacks, including limited availability and donor site morbidity. Bone allografts lack these shortcomings and provide the strength and flexibility that are required for load-bearing applications. However, allografts are not osteogenic, due to their acellular nature, which prevents the integration of the graft into the host bone. Here, we show that murine periosteal cells, which transiently express bone morphogenetic protein 2 (BMP2) and vascular endothelial growth factor (VEGF) and are seeded on a segmental allograft, can substitute for the missing periosteum and improve callus formation and allograft incorporation.

We first established a protocol to isolate periosteal cells from long bones of adult mice. Flow cytometric analysis and colony-forming unit assays revealed that abundant mesenchymal stem cells (MSCs) are present in the periosteal population. In addition, periosteal cells showed trilineage differentiation potential (osteogenic, chondrogenic and adipogenic) and when implanted ectopically in mice they formed a substantial amount of bone.

To enhance their therapeutic potential, periosteal cells were transduced with adenoviral vectors encoding BMP2 or VEGF. Increased secretion of growth factors in the culture medium persisted for at least 14 days, as demonstrated by ELISA. Cells expressing BMP2 and VEGF were seeded onto 4mm cortical allografts and implanted into femoral defects in mice. Autografts, empty allografts and allografts with GFP-transduced cells served as controls. After 4 weeks, bone healing was analyzed using µCT and histology. Defects treated with empty allografts or grafts with GFP-transduced cells showed minimal callus formation and little to no remodeling of the graft. Grafts seeded with BMP2- and VEGF-transduced periosteal cells however induced the formation of a large callus, comparable to autografts, which was accompanied by evident graft remodeling.

In conclusion, the successful isolation of adult mouse periosteal cells could help to further explore the role of this cell population using existing genetic mouse models. Moreover, establishing an artificial periosteum consisting of periosteal cells expressing BMP2 and VEGF considerably improves allograft incorporation into large bone defects in mice.