An astonishing number of patients are admitted to hospitals in the United States each year with musculoskeletal injuries. Of the 6.3 million fractures that occur annually in the U.S., approximately 1 million result in more complicated skeletal defects that require additional intervention to achieve normal healing [1]. Skeletal autografts and allografts remain the current gold standard for treating large bone defects, with over 500,000 being performed in the United States annually at a cost of $2.5 billion/year to the medical system [2]. However, bone grafting techniques suffer from a number of limitations, including low tissue availability, donor site morbidity, insufficient engraftment, and host immune rejection – all of which contribute to the need for multiple revision surgeries in as many as 15% of bone grafting procedures annually [1].

The discovery and pharmaceutical production of osteogenic proteins such as bone morphogenetic protein-2 (BMP-2) has revolutionized the treatment of large bone defects by providing a means to stimulate endogenous repair mechanisms to regenerate damaged bone [3-5]. Despite the promise of this approach, clinical approval for BMP-2 treatment using a collagen sponge delivery vehicle has thus far been limited to a few select applications. Furthermore, the widespread off-label use of BMP-2 coupled with poor dose control in the clinic has contributed to an increase in adverse events, eventually leading to a number of lawsuits being filed again Medtronic for its INFUSE® BMP-2 delivery vehicle between 2006-2012. The side effects associated with BMP-2 have been attributed to both the suprahysiological BMP-2 doses required to initiate healing (≥0.2 mg BMP-2/kg body weight) [6] and use of a collagen sponge delivery vehicle that demonstrates low affinity for BMP-2. Thus, there is a clear need to develop biomaterials to improve the efficacy of BMP-2 delivery while minimizing the negative side effects associated with rapid growth factor release.

The goal of this work was to address the shortcomings of current growth factor treatment strategies for large bone defects by creating a versatile biomaterial delivery vehicle that could (1) improve current delivery strategies for clinical BMP-2 treatment, and (2) integrate novel approaches for the delivery of multiple growth factors for better targeting bone repair. We harnessed the high growth factor binding capacity of the glycosaminoglycan heparin to fabricate an affinity-based biomaterial delivery vehicle capable of accomplishing both aspects of this goal.

First, we developed a method to fabricate pure heparin microparticles from a modified cross-linkable heparin species and demonstrated the ability of these microparticles to retain bioactive BMP-2 (Chapter 3). Following in vitro success, we incorporated these microparticles into a previously characterized tissue engineering construct consisting of an RGD-functionalized alginate hydrogel and polycaprolactone nanofiber mesh tube for investigation of both ectopic mineralization and orthotopic bone healing at low BMP-2 doses (Chapter 4). We next developed a computational model to provide insight into the effect of heparin microparticles on BMP-2 release in vivo and help inform future experiments in the femoral defect model (Chapter 5). Taken together, this work revealed that the high affinity of heparin microparticles for BMP-2 resulted in attenuated BMP-2 release in vivo, resulting in reduced bone formation in the femoral defect when low BMP-2 doses were used.

Consequently, in Chapters 6 and 7, we shifted our focus to evaluating heparin microparticles for high dose BMP-2 delivery, as a method to improve spatial localization of bone formation at clinical BMP-2 doses. We demonstrated that, when high doses of BMP-2 were delivered in vivo using heparin microparticles mixed into the alginate hydrogel, bone formation could be better localized within the defect space and the incidence of heterotopic bone could be reduced (Chapter 6). However, when we incorporated empty heparin microparticles into the PCL nanofiber mesh as a barrier to BMP-2 diffusion, we found that this method was ineffective for reducing heterotopic bone formation (Chapter 7). This led to further in vitro investigation into the effects of competitive binding of serum components on BMP-2-heparin interactions.

Finally, we conducted a series of experiments to evaluate the efficacy of heparin microparticles as a biomaterial vehicle to sequester and deliver multiple proteins from a complex solution (Chapter 8). Since previous studies have demonstrated that embryonic stem cell conditioned media contain a variety of potent biomolecules capable of influencing cell fate [7, 8], we investigated the in vitro delivery of stem cell morphogens as a first step in developing morphogen-laden microparticles as a novel approach to tissue regeneration. The promising effects observed on cell growth and differentiation suggested that the delivery of stem cell morphogens could one day be explored as an alternative method for tissue regeneration. Overall, the studies completed within this thesis broadened current understanding of bone tissue engineering, affinity-based biomaterials, and stem cell-based therapeutics, and provided valuable information that could be used to develop heparin-based biomaterials for other tissue regeneration applications