Heparin microparticles loaded with bone morphogenetic protein-2 (BMP-2) induce bone regeneration in a critically sized rat femoral defect model

摘要

Introduction: Bone morphogenetic protein-2 (BMP-2), an osteoinductive growth factor, is delivered clinically to stimulate endogenous healing after severe bone loss. However, existing biomaterials exhibit limited affinity for BMP-2, resulting in rapid diffusion of growth factor away from the injury site. Heparin, a negatively charged glycosaminoglycan, binds growth factors electrostatically and has high affinity for BMP-2. We have previously fabricated heparin microparticles (HMPs) that can retain large amounts of bioactive BMP-2 (＞300 μg/mg HMPs). Thus, here we hypothesize that HMPs can improve BMP-2 delivery in vivo and enhance healing in a critically sized bone defect. Methods: HMPs were fabricated from heparin methacrylamide as previously described. HMPs were loaded with BMP-2 overnight at 4°C, and bioactivity was evaluated via C2C12 myoblast growth and alkaline phosphatase (ALP) activity. For in vivo delivery, 2% RGD-alginate gels within polycaprolactone mesh tubes were cast with or without HMPs and/or BMP-2. BMP-2 release from constructs was evaluated in vitro at 37°C. Constructs with 1 mg of fluorescent HMPs were implanted subcutaneously in rats for 6 weeks to evaluate HMP retention in vivo. Constructs with 2.5 μg of BMP-2 loaded onto 1 mg of HMPs or mixed directly into the alginate were implanted in rat femoral defects to evaluate bone healing (Fig 2A). X-ray and micrc-CT were performed at 4,8, and 12 weeks post-surgery, histology was completed on 2,4, and 12-week femurs, and biomechanical testing was conducted at 12 weeks. Results: C2C12 cells (20,000 cells/well) exposed to 50 ng of total BMP-2 demonstrated increased ALP activity and cell proliferation when 210 ng of BMP-2 were loaded onto the HMPs (Fig 1A-B). Alginate constructs with HMP-bound BMP-2 exhibited lower BMP-2 release over 21 days in vitro compared to alginate-bound BMP-2 (Fig 1C). Fluorescent imaging of subcutaneously implanted constructs revealed persistence of ～55% of the HMPs after 6 weeks in vivo (Fig 1D-E). Bone volume within femoral defects increased over 12 weeks in response to both HMP-bound and alginate-bound BMP-2 (Fig 2B). Bony bridging occurred in all alginate-bound BMP-2 samples and 60% of HMP-bound BMP-2 samples (Fig 2D). Functional restoration of maximum torque and stiffness was comparable (Fig 2C). New bone formation (Fig 2E-F, blue) was observed in close proximity to HMPs and alginate (red) at 4 and 12 weeks. Proliferating Ki-67+ cells (Fig 2G, brown) were localized with HMPs (red) at 2 weeks, suggesting HMP-mediated proliferation similar to that observed in vitro. Conclusions: BMP-2-loaded HMPs induce functional regeneration of large bone defects; however, optimization is required to achieve consistent bridging of newly formed bone. Since bioactivity assays have demonstrated the benefits of HMP-bound BMP-2, ongoing experiments are being performed to deliver a combination of alginate-bound and HMP-bound BMP-2 in vivo, in order to provide both rapid release and sustained presentation of growth factor. Overall, HMPs may provide a more effective method of spatially controlling the delivery of BMP-2 for in vivo tissue regeneration applications.