(684g) Dual Delivery of BMP-2 and Vancomycin From Polyurethane Scaffold for Infected Bone Wound Healing

Bone regeneration is required for healing of open fractures, and healing is often complicated by chronic infection. To reduce the healing time of the patient, it is desirable to promote bone fracture healing and control infection through one surgical procedure. Bioresorbable polyurethanes (PUR) have been used extensively in tissue engineering to serve as both a supportive scaffold and drug delivery system due to their biocompatibility and biodegradability (1,2). In the present study, we successfully delivered both bone morphogenetic protein (BMP-2) for the acceleration of new bone formation and antibiotic (vancomycin) for the control of infection from preformed PUR implants which are demonstrated to be bioactive. The release profile of both drugs can be controlled. The PLGA microencapsulation approach resulted in a lower burst and more sustained release profile of BMP-2; and conversion of vancomycin hydrochloride into vancomycin free base before incorporation into polyurethane eliminated the burst release and achieved a more sustained release. The PUR scaffolds have been tested in a rat femoral plug model, and the release of BMP-2 promoted significantly more new bone formation at weeks 2 and 4 relative to the negative control as evidenced by histomorphometry. The scaffold disks containing vancomycin were bioactive as demonstrated by Kirby-Bauer assay.

Methods: PUR scaffolds were synthesized by one-shot reactive liquid molding of hexamethylene diisocyanate trimer (HDIt) with a polyester triol (900-Da) (2). PLGA microspheres were prepared using the double emulsion technique (3) at two different sizes: 110 and 1 µm. Vancomycin free base was prepared by precipitation at high pH. Drug powders or PLGA microspheres were incorporated into PUR composite scaffolds through mixing with the polyol before the foaming reaction. In vitro release studies were carried out in ¦Á-MEM cell culture medium containing 1% BSA or PBS at 37 °C. The scaffold discs at the size of 6 x 2 mm containing vancomycin were placed onto the Mueller-Hinton agar plates spread with Staphylococcus aureus bacteria (ATCC 25923). Plates were turned upside down and incubated at 37oC.  Zones of inhibition (ZI) were measured after 24 hours. The scaffolds were cut into 3x6 mm cylinders containing 2 µg BMP-2, treated with ethylene oxide gas overnight for sterilization, and implanted into rat femoral defects. The implants were then harvested at weeks 2 and 4 respectively, fixed by formalin, scanned by µCT, decalcified with EDTA treatment, and then processed for histological analysis.

Results: The polyurethane scaffolds are porous and the pores were interconnected as evidenced by SEM imaging. The pore size was ~150 - 600 µm, and the thickness of the pore walls was 5 ~95 µm. BMP-2 release profiles from PUR scaffolds suggest a lower burst release by adopting the PLGA microsphere encapsulation strategy compared with directly incorporating BMP-2 into PUR scaffold as a powder, and a more sustained release when decreasing the PLGA microsphere size from 110 µm to 1 µm (Figure 1).

Figure 1, in vitro BMP-2 release from PUR

The vancomycin release profiles from the PUR scaffolds are shown in Figure 2. By converting vancomycin salt into free base, the burst release is suppressed and a more sustained release if achieved for up to six weeks. The results from the Kirby-Bauer assay for vancomycin efficacy are shown in figure 3, and both vancomycin treatment groups had zones of inhibition similar to those of the positive control discs. 

Figure 2, in vitro release & activity of antibiotic

The cylindrical PUR implants were implanted into rat femoral defects. At week 2, H&E staining images for all treatment groups show that obvious new bone formation within the scaffold for BMP-2 sample groups compared with the control, and the new bone tissue was mainly unmineralized woven bone. While at week 4, significant mature bone formation occurred for the sample groups, the new formed tissue integrated and reorganized with surrounding tissue (mature bone and marrow) extensively, and cortex bone started to close for PUR/BMP-2 treated samples.

Figure 3, H&E staining of BMP-2 implants

Conclusions: For the femoral plug defect model and selected dosage of 2 µg BMP-2 per implant, PUR containing BMP-2 powder works best which may suggest a critical role of burst release in attracting precursor cells in early stage. The fact that PUR containing small PLGA particles is able to promote new bone formation significantly between week 2 and 4 shows that sustained release of BMP-2 is also favored in order to sustain tissue neogenesis. The present study suggests that both burst release and sustained release are desired in terms of promoting new bone tissue formation. The combination of growth factors with antibiotic incorporated within polyurethane scaffold and the controlled release of both  is expected to promote infected bone wound healing through the synergistic effects of promoting cell differentiation by BMP-2 and anti-bacteria ability by the antibiotic.

References: 

1.      Guelcher SA. Tissue Engineering 2007:13:2321-2333.

2.      Li B., Davidson J. M., Guelcher S. A. Biomaterials 2009:30:3486-3494.

3.      Utada, A. S., Lorenceau E., Link D. R., Kaplan P. D., Stone H. A., Weitz D. A. Science 2005:308:537- 541.

Acknowledgements: This work was funded by the Orthopaedic Trauma Research Program and the Armed Forces Institute of Regenerative Medicine.