Subtypes of endothelial progenitor cells affect healing of segmental bone defects differently.

Purpose: Treating fracture nonunion with endothelial progenitor cells (EPCs) is a promising approach. Nevertheless, the effect of different EPC-related cell populations remains unclear. In this study, we compared the therapeutic potential of early (E-EPCs) and late EPCs (L-EPCs).
Methods: Male Fischer 344 rats were used for cell isolation and in vivo experiments. Bone marrow-derived E-EPCs and L-EPCs were kept in culture for seven to ten days and four weeks, respectively. For each treatment group, we seeded one million cells on a gelatin scaffold before implantation in a segmental defect created in a rat femur; control animals received a cell-free scaffold. Bone healing was monitored via radiographs for up to ten weeks after surgery. In vitro, secretion of vascular endothelial growth factor (VEGF) and bone morphogenetic protein (BMP)-2 was quantified by ELISA for both cell populations. Tube formation assays were also performed.
Results: Final radiographs showed complete (four out of five rats) or partial (one out of five rats) union with E-EPC treatment. In contrast, complete healing was achieved in only one of five animals after L-EPC implantation, while control treatment resulted in nonunion in all animals. In vitro, E-EPCs released more VEGF, but less BMP-2 than L-EPCs. In addition, L-EPCs formed longer and more mature tubules on basement membrane matrix than E-EPCs. However, co-culture with primary osteoblasts stimulated tubulogenesis of E-EPCs while inhibiting that of L-EPCs.
Conclusions: We demonstrated that bone marrow-derived E-EPCs are a better alternative than L-EPCs for treatment of nonunion. We hypothesize that the expression profile of E-EPCs and their adaptation to the local environment contribute to superior bone healing.