Prosthetic joint infection (PJI) poses a significant risk to implanted patients, requiring multiple surgeries with high rates of reinfection. The primary cause of such infections is otherwise innocuous bacterial species present on the skin that have survived sterilization protocols. Antibiotic drugs have significantly reduced efficacy due to the lack of vasculature in the newly implanted site, allowing microbes to form biofilms with even greater resistance. Graphene oxide (GO) is known to have good biocompatibility while providing drugless antimicrobial properties. The focus of this study is on the development and characterization of a robust coating for titanium alloy implants to promote bone regeneration while inhibiting microbial biofilm adhesion to the implant surface. The novelty of this study is the use of proprietary coal-derived graphene oxide (c-GO) in a biomedical application. c-GO has been demonstrated to have a greater number of functional oxygen groups to promote cell adhesion, while also maintaining thinner layers than possible with graphite exfoliation methods. As an alternative to powerful antimicrobial drugs, it was hypothesized that an advanced coating of graphene-oxide would provide a defensive, passively antimicrobial layer to a titanium implant. While GO is typically quite expensive, the newly developed process provides an economical and environmentally friendly method of producing GO from coal (c-GO). The result is a coating that is inexpensive and capable of halving the biofilm formation of MRSA on titanium-alloy surgical screws in addition to providing improved bone cell adhesion and hard tissue compatibility. / Master of Science / Any time a patient receives implantation surgery, there is a chance of microbes entering the body. These are typically naturally occurring skin flora, harmless but opportunistic. On the surface of implants within the body, these bacteria can form colonies called biofilms, leading to severe and potentially deadly infections, called prosthetic joint infection (PJI). PJI often requires multiple surgeries to remedy, but rates of reinfection are relatively high. As with any surgery, patients are given antibiotic drugs, but implants to not receive blood flow as the body normally would, reducing the effectiveness of antibiotics. Once biofilms are formed, the bacteria become even hardier and resistant even to powerful antibiotics. Graphene oxide (GO) is a carbon material known to have good biocompatibility (i.e., non-toxic) while providing antimicrobial properties. The focus of this study is on the development and characterization of a robust coating for titanium alloy implants to promote bone healing while reducing microbial biofilm colonization on the implant's surface. The novelty of this study is the use of proprietary coal-derived graphene oxide (c-GO) in a biomedical application. c-GO has been demonstrated to have a different chemical makeup than graphite-derived GO, which may improve its efficacy as an antimicrobial coating. As an alternative to powerful antimicrobial drugs, it was hypothesized that a coating of graphene-oxide would provide a defensive, passively antimicrobial layer to a titanium implant. While GO is typically quite expensive, the newly developed one-pot process provides an economical and environmentally friendly method of producing GO from coal (c-GO). The result is a coating that is inexpensive and capable of halving the biofilm formation of MRSA on titanium-alloy surgical screws in addition to providing improved bone cell adhesion and hard tissue compatibility.

In vitro differentiation of stem cells in biocompatible scaffolds in a bioreactor is a promising method for creating functional engineered tissue replacements suitable for implantation. Basic studies have shown that mechanical, chemical, and pharmaceutical stimuli enhance biological functionality of the replacement as often defined by parameters such as cell viability, gene expression, and protein accumulation. Most of the assays to evaluate these parameters require damage or destruction of the cell-scaffold construct. Therefore, these methods are not suitable for monitoring the development of a functional tissue replacement in a spatial and temporal manner prior to implantation. Bioluminescence imaging is a technique that has been utilized to monitor cell viability and gene expression in various in vivo applications. However, it has never been applied in an in vitro setting for the specific purpose of evaluating a cell-scaffold construct. This research describes the design of flow perfusion bioreactor system suitable for bioluminescence imaging. In the first experimental chapter, the system was tested using MC3T3-E1 cells transfected with a constitutive bioluminescent reporter. It was found that bioluminescence imaging was possible with this system. In the second experimental chapter, MC3T3-E1 cells transfected with BMP-2 linked bioluminescence reporter were cultured by flow perfusion for a period of 11 days. Bioluminescence was detectable from the cells starting at day 4, while peaking in intensity between days 7 and 9. Further, it was also found that bioluminescence occurred in distinct regions within the scaffold. These results indicate that these strategies may yield information not available with current assays. / Master of Science

The overall goal of this study was to identify mineralized scaffolds which can serve as potential alternatives to bone graft substitutes intended for cleft palate repair. The aim of this preliminary study was to evaluate the role of fibrinogen (Fg) and nano-hydroxyapatite (nHA) in enhancing mineralization potential of polydioxanone (PDO) electrospun scaffolds. Scaffolds were fabricated by blending PDO:nHA:Fg in the following weight ratios: 100:0:0, 50:25:25, 50:50:0, 50:0:50, 0:0:100 and 0:50:50. Scaffolds were immersed in different simulated body fluids for 5 and 14 days to induce mineralization. The inclusion of fibrinogen induced sheet-like mineralization while individual fiber mineralization was noticed in its absence. Modified protocols of alizarin red staining and burn-out test were developed to quantify mineral content of scaffolds. After mineralization, 50:50:0 scaffolds were still porous and contained the most mineral. 50:25:25 scaffolds had the highest mineralization potential but lacked porosity. Therefore, it can be anticipated that these mineralized organic-inorganic electrospun scaffolds will induce bone formation.

La reparation de defauts osseux par les techniques de l’ingenierie tissulaire osseuse (ITO) est consideree comme une alternative aux greffes conventionnelles. L’objectif de ce projet de these fut de developper des materiaux destines a servir de scaffolds pour le comblement et la regeneration osseuse, ces derniers etant sous la forme d’hydrogels injectables d’une part, et d’eponges, d’autre part. Ces deux types de materiaux ont ete obtenus par melange de chitosane (CHT, cationique), et de polymere de cyclodextrine reticule par l’acide citrique (PCD, anionique), interagissant via des liaisons ioniques et formant des complexes polyélectrolytes. La premiere partie de la these a ete consacree au developpement et caracterisation d’une eponge CHT/PCDs qui a ete chargee avec le facteur de croissance de l’endothelium vasculaire (VEFG) dans le but de favoriser sa vascularisation. Le second volet de la these a eu pour objectif d’optimiser la formulation d’un hydrogel injectable destine a la chirurgie mini-invasive, compose de CHT et de PCD sous sa forme soluble (PCDs) et insoluble (PCDi) [CHT/PCDi/PCDs]. L'etude a ete concentree sur l’optimisation et la caracterisation des proprietes rheologiques de l’hydrogel. Enfin, une etude prospective sur le developpement de l'hydrogel/eponge composite en ajoutant une phase minerale - l'hydroxyapatite (HAp) dans la formulation a ete realisee afin d'ameliorer les proprietes mecaniques et osteoconductrices.L’eponges CHT/PCDs a ratio 3 :3 a ete obtenue par lyophilisation des hydrogels et a subi un traitement thermique (TT) afin d’ameliorer leur stabilite par la formation des liaisons covalentes. L’eponge CHT/PCDs avec un TT a 160°C a montre des proprietes de gonflement eleve (~600%) et une biodegradation ralenti induite par le lysozyme (~12% perte masse dans un mois). Sa microstructure, ses proprietes mecaniques de compression et sa cytocompatibilite avec deux types de cellules (pre-osteoblastes (MC3T3-E1) et endotheliales primaires (HUVECs) ont ete etudiees. Une porosite elevee (~87%) avec des pores interconnectes a ete observee par microtomographie de rayons X, ainsi qu’une bonne adhesion et colonisation cellulaire au sein de l’eponge par microscopie electronique a balayage (MEB). Le VEGF a ete incorpore dans l’eponge, et son profil de liberation a ete suivi, ainsi que la bio-activite du VEGF libere. La liberation du VEGF a ete rapide pendant les trois premiers jours, puis ralenti jusqu'a devenir non-detectable par la methode ELISA jusqu’a 7 jours. Le VEGF libere pendant les deux premiers jours a montre un effet pro-proliferation et pro-migration significatif sur les HUVECs.Les hydrogels injectables de CHT/PCDi/PCDs a differents ratios ont ete optimises et caracterises en fonction de leurs proprietes rheologiques, leur injectabilite, et leur cytotoxicite. L’impact de l’ajoute du PCDi dans l’hydrogel a ete clairement observe par analyses rheologiques Ainsi, l'hydrogel CHT/PCD, compose a parts egales de PCDi et de PCDs, a demontre le meilleur compromis entre stabilite structurelle, proprietes rheofluidifiantes et autoreparantes, et injectabilite. En plus, l’hydrogel a montre une excellente cytocompatibilite vis-avis les cellules pre-osteoblastes MC3T3-E1.Bases sur la formulation optimisee, l’HAp a ete incorporee a differentes concentrations dans l’hydrogel. L’ajout de la phase minerale n’a pas perturbe la formation ni la stabilite structurelle des hydrogels, mais a ameliore les proprietes viscoelastiques. Les eponges composites, elaborees par lyophilisation de ces hydrogels, ont montre que les particules de HAp etaient dispersees de maniere homogene dans la structure macroporeuse de l'eponge. Ces resultats encourageants ont montre qu'il etait possible de fournir un hydrogel injectable ou une eponge composite comme scaffold pour l’ITO [...] / Repair of bone defects by bone tissue engineering (BTE) methods is considered as an alternative to conventional grafts. The aim of this PhD project was to develop two types of BTE scaffolds for bone regeneration: one is in the form of injectable hydrogel, and the other is in the form of sponge. Both scaffolds based on the formation of polyelectrolyte complexes by mixing chitosan (CHT, cationic) and polymer of cyclodextrin (PCD, anionic). Besides developing the sponge scaffold, the vascularization of 3D scaffold (a challenge of BTE) was specially investigated in the first part of the work, for which vascular endothelial growth factor (VEFG) was loaded on the CHT/PCDs sponge to promote the vascularization. The second part of the thesis was dedicated to the elaboration of an injectable CHT/PCD hydrogel, which was intended for minimally invasive surgery. The formulation optimization of hydrogel was performed by tuning the composition ratios of two PCD components: soluble-form PCD (PCDs) and insoluble-form PCD (PCDi), in order to better reach the specific requirement (e.g. rheological properties) of injectable hydrogel for regenerative medicine. Finally, a prospective study on developing the composite hydrogel/sponge by adding a mineral phase - hydroxyapatite (HAp) in the formulation was realized to improve the mechanical and osteoconductive properties.CHT/PCDs sponges were obtained by freeze-drying the hydrogels CHT/PCDs 3:3. The thermal treatment (TT) at different temperatures was further applied on the sponge to improve the mechanical stability. The CHT/PCDs sponge treated at 160°C was opted for further study thanks to high swelling capacity (~ 600%) and moderate lysozyme-induced biodegradation rate in vitro (~ 12% mass loss 21 days). This sponge of choice was further evaluated for the microstructure, the mechanical property (compressive strength) and the cytocompatibility with pre-osteoblasts (MC3T3-E1) and endothelial cells (HUVEC). Results of X-ray microtomography showed a high porosity (~87%) in the sponge with interconnected pores. Good cell adhesion and in-growth (colonization) in the sponge were observed by scanning electron microscopy (SEM). After loading VEGF on the sponge, the release profile of VEGF and the bioactivity of released VEGF were thoroughly studied. It showed that the release of VEGF was rapid (burst) during the first two days, then slowed down up to non-detectable by ELISA method after 7 days. The released VEGF during the first two days showed a significant pro-proliferation and pro-migration effect on HUVECs.For the injectable CHT/PCDi/PCDs hydrogels, optimization of composition ratio was based on evaluating their rheological properties, injectability, and cytotoxicity. The beneficial effect of combining both PCDi and PCDs in the formula of the hydrogel was clearly observed on the properties of hydrogel. Namely, the CHT/PCD hydrogel, composed of equal quantity of PCDi and PCDs, demonstrated the best compromise between structural stability, shearthinning and self-healing properties, and injectability. An excellent cytocompatibility with preosteoblast cells (MC3T3-E1) was also confirmed for the hydrogel with this composition.Based on the optimized formulation, HAp was incorporated at different concentrations, which didn’t disturb the formation or the structural stability of the hydrogels, but improved the viscoelastic properties. The composite sponges, elaborated by lyophilization of these hydrogels, showed that the HAp particles homogeneously dispersed within the macroporous structure of the sponge. These encouraging results showed the feasibility of providing an injectable hydrogel or a composite sponge for BTE scaffold [...]

Los estímulos mecánicos son uno de los factores que afectan a la diferenciación celular en el proceso de regeneración del tejido óseo, por lo tanto, en el desarrollo de andamios para ingeniería de tejidos, se pueden aplicar las cargas mecánicas con el fin de inducir la actividad de las células. Cuando se aplican cargas mecánicas, los estímulos mecánicos específicos transmitidos a las células a nivel microscópico pueden estudiarse mediante técnicas numéricas. El objetivo de esta tesis fue estudiar la mecanoregulación de la diferenciación de tejido en andamios de fosfato de calcio utilizando modelos de elementos finitos basados en micro tomografía axial computarizada.Dos muestras de materiales porosos basados en fosfato de calcio fueron utilizadas. Se desarrollaron mallas de elementos finitos congruentes, discretizando la fase sólida y los macro poros interconectados, con el fin de tener en cuenta la morfología irregular de los andamios.En primer lugar, se estudió la distribución de los estímulos mecánicos. La fase sólida y el fluido intersticial se simularon como material elástico lineal y como fluido Newtoniano, respectivamente. Se simuló una compresión del 0.5% en el sólido y un fluido con velocidades de entrada de 1, 10 y 100 µm/s en los poros. Se encontraron distribuciones de deformación similares en las paredes ambos materiales, con valores máximos de 1.6% en compresión y de 0.6% en tracción. En algunos poros, la velocidad del fluido aumentó a 100 y 1000 veces la velocidad de entrada. Este estudio mostró como estímulos mecánicos macroscópicos pueden causar distintos niveles de estímulos mecánicos microscópicos dentro los andamios, debido a la morfología.A continuación se realizó un estudio en el tiempo de la diferenciación de tejido en un andamio sometido a condiciones in vitro. La compresión y la perfusión se modelaron como en el estudio anterior. Se simularon una compresión del 0.5% y una velocidad de entrada de fluido constante de 10 µm/s o una presión de entrada de fluido constante de 3 Pa. La deformación cortante octaédrica y el esfuerzo cortante del fluido se utilizaron como estímulos mecano-regulatorios basándose en la teoría de Prendergast et al. (1997). Al aplicar velocidad constante, se predijeron fluctuaciones entre los estímulos equivalentes a la formación de tejido y a la muerte celular, debido al aumento en el esfuerzo cortante del fluido cuando el tejido comienza a llenar los poros. Sin embargo, al aplicar presión constante, se predijo estímulo equivalente a la diferenciación de tejido óseo en la mitad del volumen de los poros. Estos resultados sugieren que para permitir la diferenciación de tejido, la velocidad del fluido debe disminuirse cuando el tejido empieza a mineralizarse.Finalmente, se llevó acabo un estudio en el tiempo de la angiogénesis y de la diferenciación de tejido en un andamio bajo condiciones in vivo. La deformación cortante octaédrica y la velocidad relativa del fluido se utilizaron como estímulos mecano-regulatorios. Las fases sólida y porosa fueron tratadas como materiales poroelásticos. Se simuló la actividad individual de las células. Compresiones de 0.5 y 1% fueron simuladas. La mayoría de los vasos crecieron en los poros de la periferia del andamio y se bloquearon por las paredes. Se formaron redes capilares similares independientemente de la magnitud de deformación utilizada. Al aplicar 0.5% de compresión, estímulos correspondientes a la formación de hueso se predijeron en el 70% del volumen de los poros, sin embargo, sólo el 40% del volumen se llenó de osteoblastos debido a la falta de oxigeno. Este estudio mostró el efecto de la falta de vascularización en el centro del andamio en la diferenciación de tejido.Ese tipo de estudios, combinados con estudios in vitro, deberían contribuir a la comprensión del proceso de diferenciación de los tejidos dentro de los andamios y por lo tanto a la mejora de los métodos de diseño de andamios. / Mechanical stimuli are one of the factors that affect cell differentiation in the process of bone tissue regeneration; therefore, in the development of scaffolds for tissue engineering, mechanical loads can be applied in order to induce cell activity. The specific mechanical stimuli transmitted to cells at a microscopic level when mechanical loads are applied can be studied using numerical techniques. The objective of this thesis was to study the mechanoregulation of tissue differentiation within calcium phosphate scaffolds using micro computed tomographed based finite element models.Two samples of porous calcium phosphate based materials were used. Congruent finite element meshes, with the solid phase and the interconnected pores discretized, were developed in order to account for the scaffold irregular morphology.First, a study of the distribution of mechanical stimuli was performed. The solid phase and the fluid flow within the pores were modeled as linear elastic solid material and Newtonian fluid respectively. Compressive strains of 0.5% of total deformation applied to the solid and interstitial fluid flows with inlet velocities of 1, 10 and 100 µm/s applied to the pores were simulated. Similar strain distributions for both materials were found, with compressive and tensile strain maximal values of 1.6% and 0.6% respectively. For the fluid flow models, the fluid velocity in some of the scaffold pores increased to 100 and 1000 times the inlet velocity. This study showed how mechanical loads and fluid flow applied to the scaffolds caused different levels of mechanical stimuli within the samples according to the morphology of the materials.Next, a study of the mechanoregulation of tissue differentiation over time in a scaffold subjected to in vitro loads was performed. The solid phase and the fluid flow were modeled as in the study described above. Compressive strain of 0.5% and fluid flow with constant inlet velocity of 10 µm/s or constant inlet pressure of 3 Pa were applied. Octahedral shear strain and fluid shear stress were used as mechano-regulatory stimuli based on the theory of Prendergast et al. (1997). When a constant velocity was simulated, fluctuations between stimuli equivalent to tissue formation and cell death were predicted due to the increase in the fluid shear stress when tissue started to fill the pores. However, when constant pressure was applied, stimuli equivalent to bone formation were predicted in about half of the pore volume. These results suggest that in order to allow tissue differentiation within a scaffold, the fluid velocity should be decreased when tissue starts mineralizing.Finally, a study of the angiogenesis and the mechanoregulation of tissue differentiation over time in a scaffold subjected to in vivo conditions was performed. Octahedral shear strain and relative fluid velocity were used as mechano-regulatory stimuli. The solid and pore phases were treated as poroelastic materials. Individual cell activity was simulated within the pore domain. Compressive strains of 0.5 and 1% of total deformation were simulated. Most vessels grew in the pores at the periphery of the scaffolds and were blocked by the scaffold walls. Similar capillary networks were formed independently of the magnitude of the mechanical strain applied. When 0.5% of strain was applied, 70% of the pore volume was affected by mechano-regulatory stimuli corresponding to bone formation; however, because of the lack of oxygen, only 40% of the volume was filled with osteoblasts. This study showed the effect of the lack of vascularization in the center of the scaffold on the tissue differentiation.Such kind of studies, combined with in vitro studies, should contribute to the understanding of the process of tissue differentiation within the constructs and therefore to the improvement of scaffold design methods.

Tissue engineering strategies that combine porous biomaterial scaffolds with cells capable of osteogenesis or bioactive proteins have shown promise as effective bone graft substitutes. Attempts to culture bone tissue-engineering constructs thicker than 1mm in vitro often result in a shell of viable cells and mineralized matrix surrounding a necrotic core. To address this limitation, we developed a perfusion bioreactor system that improves mass transport throughout large cell-seeded constructs. Additionally, we established and validated 3-D computational methods to model flow and shear stresses within the microporosity of perfused constructs. Micro-CT scanning and analysis techniques were used to non-destructively monitor mineral development over time in culture. CFD modeling of axial perfusion through cylindrical scaffolds with a regular microarchitecture revealed a uniform flow field distributed throughout the scaffold. Perfusion resulted in a 140-fold increase in mineral deposition at the interior of 3 mm thick polymer scaffolds seeded with rat bone marrow stromal cells. The total detected mineral volume tripled as the construct length was increased from 3 to 9 mm. Increasing scaffold length to 9 mm did not affect the mineral volume fraction (MVF) within the full volume of each construct. Mineral volume, spatial distribution, density, particle size and particle number were then quantified on cell-seeded constructs in 5 different culture environments. The effect of time varying flow conditions was compared with continuous perfusion as well as two different control cell culture methods in an attempt to enhance mineralized matrix within the constructs. Intermittent elevated perfusion and dynamic culture in an orbital rocker plate produced the greatest amount of mineral within 9 mm long constructs compared to low continuous flow and high continuous flow cases. Together, these studies indicate that dynamic culture conditions enhance construct development with regards to cell viability, mineralized matrix deposition, growth rate, and distribution. Furthermore, these techniques provide a rational approach to selecting perfusion culture conditions that optimize the amount and distribution of mineralized matrix production. Finally, the established perfusion bioreactor, in combination with micro-CT analysis, provides a foundation for evaluating new scaffolds and cell types that may be useful for the development of effective bone graft substitutes.

Complexed microspheres of poly(4-vinyl pyridine) (P4VN) and alginic acid were prepared by internal gelation method and subsequent freeze drying. The 4% and 10% microspheres were loaded with Bone Morphogenetic Protein-2 (BMP-2) and Bone Morphogenetic Protein-7 (BMP-7), respectively for in vitro studies and were entrapped in PLGA foams. Foams containing only 4%, BMP-2 microspheres, only 10%, BMP-7 microspheres and both populations were prepared. Control samples of each group were prepared with drug free microspheres. Bone marrow derived stem cells from rat femur and tibia isolated by a surgical operation, were seeded onto foams. Proliferation of cells on foams containing both microsphere populations was higher at all time points regardless of the presence of BMPs. This was attributed to different porosity characteristics. Proliferation was higher at all times in control samples in comparison to their positive samples for all groups, suggesting proliferation attenuation related enhancement in osteogenic activity due to BMP supply. Alkaline phosphatase (ALP) activities were lower at all time points for foams containing both microsphere populations regardless of BMP presence. This was attributed to different physical characteristics of foams confirmed by the inverse correlation between proliferation and osteogenic differentiation. Total and specific ALP activity results demonstrated the significant positive influence of all BMP containing types in enhancing osteogenic differentiation. Best results were obtained with co-administration of sequential delivery performing 4% and 10% microspheres loaded with BMP-2 and BMP-7, respectively.

COLLAGEN IN SITU GROWN CALCIUM PHOSPHATE SCAFFOLDS FOR OSTEOGENIC DIFFERENTIATION OF WHARTON&rsquo / S JELLY AND MENSTRUAL BLOOD STEM CELLS Karadas, &Ouml / zge M.Sc., Department of Biotechnology Supervisor : Prof. Dr. Vasif Hasirci Co-Supervisor: Assoc. Prof. Dr. Gamze Torun K&ouml / se February 2011, 91 pages The importance of developing new techniques for the treatment of bone and joint diseases is increasing continuosly together with the increase of human population and the average life span. Especially bone fractures as a result of osteoporosis are often seen in humans older than 50 years old. The expenses of bone and joint disease operations are very high and the duration of recovery is long. Because of these reasons World Health Organization, The United Nations and 37 countries announced that the years 2000-2010 is the Bone and Joint Decade. Tissue engineering is an alternative approach to clinically applied methods. In this study collagen scaffolds crosslinked with genipin, to improve the stability of foams in culture media, were prepared by lyophilization. To mimic the natural bone structure calcium phosphate mineral phase in the foam was formed by wet chemical precipitation. Collagen concentration (0.75% and 1%, w/v), freezing temperature (-20 oC and -80 oC) of the collagen solution before lyophilization and immersion duration (2x4 h and 2x48 h) of the foams in calcium and phosphate solutions for wet chemical precipitation were changed as process v parameters of foam production. Pore size distribution and porosity analysis as well as compression test were performed for characterization of the scaffolds. The foam with 1% w/v collagen concentration, frozen at -20 oC before lyophilization and immersed for 2x4 h in calcium and phosphate solution was chosen for in vitro cell culture studies. The defined foam had 70% porosity and pore sizes varying between 50 and 200 &mu / m. The elastic modulus and compressive strength of the foam was calculated as 127.1 kPa and 234.5 kPa, respectively. Stem cells isolated from Wharton&rsquo / s jelly (WJ) and menstrual blood (MB) were seeded to foams to compare their osteogenic differentiation. Both cells are isolated from discarded tissues and used in this study as an alternative to the commonly used cells which are isolated by invasive techniques such as bone marrow stem cells. Cells were seeded to collagen foams with and without calcium phosphate (CaP). It was observed that WJ cells proliferated during 21 days on collagen foams without CaP, but MB cell number decreased after day 14. Collagen foams with CaP supported the alkaline phosphate (ALP) activity compared to tissue culture polystyrene (TCPS) and foams without CaP. Contrarily lower cell numbers achieved on CaP containing collagen foams, possibly because of the calcium and phosphate concentration changes in the medium and as the result of osteogenic differentiation. ALP activity of both cell types increased almost 10 times and specific ALP activity (activity per cell) increased 40 times and 150 times for WJ and MB cells, respectively on the CaP containing foams compared to TCPS. Therefore, in this study it was shown that in situ CaP formed collagen foams induce osteogenic differentiation of WJ and MB cells, and these cells isolated from discarded tissues can be used as alternative cell sources in bone tissue engineering applications.

Bone defects and fracture non-unions remain a substantial challenge for clinicians due to a high occurrence of delayed union or non-union requiring surgical intervention. The current grafting procedures used to treat these injuries have many limitations and further long-term complications associated with them. This has resulted in research efforts to identify graft substitution therapies that are able to repair and replace tissue function. Many of these tissue engineered products include the use of growth factors to induce cell differentiation, migration, proliferation, and/or matrix production. However, current growth factor delivery methods are limited by poor retention of growth factors upon implantation resulting in low bioactivity. These limiting factors lead to the use of high doses and frequent injections, putting the patients at risk for adverse effects. The goal of this work was to develop and evaluate the efficacy of BMP-2 delivery systems to improve bone regeneration. We examined two approaches for delivery of BMP-2 in this work. First, we evaluated the use of a self-assembling lipid microtube system for the sustained delivery of BMP-2. We determined that sustained delivery of BMP-2 from the lipid microtube system was able to enhance osteogenic differentiation compared to empty microtubes, however did not demonstrate a significant advantage compared to a bolus BMP-2 dose in vitro. Second, we developed and assessed the functionality of an affinity-based system to sequester BMP-2 at the implant site and retain bioactivity by incorporating heparin within a collagen matrix. Incorporation of heparin in the collagen matrix improved BMP-2 retention and bioactivity, thus enhancing cell-mediated mineralized matrix deposition in vitro. Lastly, the affinity-based BMP-2 delivery system was evaluated in a challenging in vivo bone repair model. Delivery of pre-bound BMP-2 and heparin in a collagen matrix resulted in new bone formation with mechanical properties not significantly different to those of intact bone. Whereas delivery of BMP-2 in collagen or collagen/heparin matrices had similar volumes of regenerated mineralized tissue but resulted in mechanical properties significantly less than intact bone properties. The work presented in this thesis aimed to address parameters currently preventing optimal performance of protein therapies including stability, duration of exposure, and localization at the treatment site. We were able to demonstrate that sustained delivery of BMP-2 from lipid microtubes was able to induce osteogenic differentiation, although this sustained delivery approach was not significantly advantageous over a bolus dose. Additionally, we demonstrated that the affinity-based system was able to improve BMP-2 retention within the scaffold and in vitro activity. However, in vivo implantation of this system demonstrated that only delivery of pre-complexed BMP-2 and heparin resulted in regeneration of bone with mechanical properties not significantly different from intact bone. These results indicate that delivery of BMP-2 and heparin may be an advantageous strategy for clinically challenging bone defects.

Non-healing bone defects have a significant socioeconomic impact in the U.S. with approximately 600,000 bone grafting procedures performed annually. Autografts and allografts are clinically the most common treatments; however, autologous donor bone is in limited supply, and allografts often have poor mechanical properties. Therefore, tissue engineering and regenerative medicine strategies are being developed to address issues with clinical bone grafting. The overall objective of this work was to develop bone tissue engineering strategies that enhance healing of orthotopic defects by targeting specific osteogenic cell signaling pathways. The general approach included the investigation of two different tissue engineering strategies, which both focused on directed osteoblastic differentiation to promote bone formation. In the first cell-based strategy, we hypothesized that constitutive overexpression of the osteoblast-specific transcription factor, Runx2, in bone marrow stromal cells (BMSCs) would promote orthotopic bone formation in vivo. We tested this hypothesis by delivering Runx2-modified BMSCs on synthetic scaffolds to critically-sized defects in rats. We found that Runx2-modified BMSCs significantly increased orthotopic bone formation compared to empty defects, cell-free scaffolds and unmodified BMSCs. This gene therapy approach to bone regeneration provides a mineralizing cell source which has clinical relevance. In the second biomaterial-based strategy, we hypothesized that incorporation of the collagen-mimetic peptide, GFOGER, into synthetic bone scaffolds would promote orthotopic bone formation in vivo without the use of cells or growth factors. We tested this hypothesis by passively adsorbing GFOGER onto poly-caprolactone (PCL) scaffolds and implanting them into critically-sized orthotopic defects in rats. We found that GFOGER-coated scaffolds significantly increased bone formation compared to uncoated scaffolds in a dose dependent manner. Development of this cell-free strategy for bone tissue engineering provides an inexpensive therapeutic alternative to clinical bone defect healing, which could be implemented as a point of care application. Both strategies developed in this work take advantage of specific osteoblastic signaling pathways involved in bone healing. Further development of these tissue engineering strategies for bone regeneration will provide clinically-relevant treatment options for healing large bone defects in humans by employing well-controlled signals to promote bone formation and eliminating the need for donor bone.