The objective of this thesis was to expand the current understanding of how biomechanical factors mediate a variety of processes in articular cartilage, using an approach that focused on single cell biomechanics and mechanobiology. Single chondrocytes were mechanically tested, to derive salient biomechanical parameters that could aid in more accurate descriptions of the in vivo cellular mechanical environment. Building upon these results, single chondrocytes were then subjected to static and dynamic mechanical forces, and the resulting changes in the expression of key genes was measured using single cell real-time RT-PCR. These studies yielded several major findings relevant to chondrocytes and the nature of their responses to mechanical forces. Superficial zone chondrocytes were significantly stiffer than cells from the deeper layers of cartilage. This suggests that cells adapt to their mechanical environment by altering their properties, and that these zone-dependent differences could lead to varying cell responses to the same externally applied mechanical load. Chondrocytes were also found to have strain-dependent recovery properties. Specifically, the residual strain, volume fraction recovered, and recovery time after the cell was compressed were dependent on compressive strain. The most intriguing finding was that the dependence on compressive strain increased at approximately 25-30% strain, suggesting that this range of strain causes a fundamental change in cell biomechanical behavior. Furthermore, this strain range may represent an important threshold for discriminating whether a given mechanical stimulus has a beneficial or deleterious effect on chondrocytes. Finally, dynamic compression was shown to increase type II collagen and aggrecan gene expression compared to statically loaded single chondrocytes. This result was very exciting, as it demonstrated that studying the effects of mechanical forces on single cells was a viable approach. It was also shown that gene expression in single chondrocytes appears to be lognormally distributed. Thus, tests examining populations of cells may be biased by a small fraction of cells with very high levels of gene expression. These findings reinforce the notion that a single cell approach offers significant advantages over existing techniques, and may allow researchers to answer questions that were previously intractable with traditional methodologies.

The photocrosslinking of poly(propylene fumarate) (PPF) using the photoinitiator bis(2,4,6-trimethylbenzoyl) phenylphosphine oxide (BAPO) and low levels of ultraviolet light irradiation has been investigated as the basis for a bone tissue engineering scaffold. The photocrosslinking mechanism as well as the final network structure was studied, showing that a single phosphinoyl radical derived from BAPO was primarily responsible for the photoinitiated crosslinking of PPF. A technique to fabricate PPF into porous scaffolds using a photocrosslinking/porogen leaching strategy was developed, with characterization studies showing that the presence of the leachable porogen did not affect the initiation of the PPF crosslinking reaction in this system. An in vitro degradation study of both solid PPF networks and porous PPF scaffolds in phosphate buffered-saline was performed. The results indicated that porogen size and content could be selected to formulate the photocrosslinked PPF scaffolds with a degradation rate, porosity, and mechanical properties to match target values for a specific tissue defect. The soft and hard tissue response to photocrosslinked PPF scaffolds was studied, with the results indicating that the scaffolds were biocompatible within both soft and hard tissue. The ability of these photocrosslinked PPF scaffolds to act as a carrier for an adsorbed protein in order to promote bone formation was also examined. The results indicate that transforming growth factor-beta1 did induce significant bone formation in these porous PPF scaffolds. Finally, an in vivo study of the effects of a degradable biomaterial upon wound healing and bone formation within a tooth extraction socket was undertaken. The results show that the implantation of the hydrophobic and degradable PPF biomaterial did not significantly alter this process, while the negative control group, a hydrophilic, degradable biomaterial, significantly reduced bone formation. The effect of biomaterial's surface properties upon bone formation most closely parallel the fibroblastic growth factor-2 localization results, indicating its critical role in the initial phases of wound healing to facilitate later bone formation. These results indicate the great potential of photocrosslinked PPF scaffolds in bone tissue engineering applications.

A novel oligo(poly(ethylene glycol) fumarate) (OPF) was synthesized and used to prepare a biodegradable and biomimetic crosslinked network covalently modified with bioactive peptides for guided bone regeneration. First, we examined the cytotoxicity of each component of the OPF hydrogel formulation and the resulting crosslinked network. Marrow stromal cells (MSCs) exhibited different viability with respect to OPF synthesized with poly(ethylene glycol) (PEG) of various molecular weights, crosslinking agent (PEG-diacylate (PEG-DA)), and the redox initiator pairs. Once crosslinked, the leachable products from the OPF hydrogels had minimal adverse effects on the viability of MSCs. Next study investigated the effects of the concentration of incorporated model peptide, Arg-Gly-Asp (RGD) and macromolecular structures of hydrogels on attachment of MSCs. The network structure was varied by changing number average molecular weight of PEG in OPF from 930 to 6090. The cell attachment on the peptide-modified hydrogel was increased with increasing the peptide concentration. In addition, higher molecular weight PEG compared to peptide spacer length reduced cell attachment. In vivo bone and soft tissue behavior of the OPF hydrogels were also assessed using a rabbit model. The results indicated that the OPF hydrogels are biocompatible as evidenced by an uniform thin circumferential fibrous capsule formation following cranial as well as subcutaneous implantation. In addition, histological analysis suggested that the in vivo degradation can be controlled by tailoring the macromolecular structure of the OPF hydrogels. Finally, osteopontin-derived peptide (ODP) was covalently incorporated to the OPF hydrogels and modulation of MSCs on the peptide-modified hydrogels was investigated. The results showed that OPF-based peptide-modified hydrogels can modulate cell proliferation and migration by altering the specific ligand and its concentration in the hydrogels. Furthermore, peptide-modified hydrogels promoted differentiation and mineralization of osteoblasts as characterized by measuring alkaline phosphatase activity, osteopontin expression, and calcium deposition for in vitro culture. This work demonstrated that the OPF hydrogel modified with bioactive molecules is a promising material for a biodegradable and biomimetic substrate in tissue engineering.

The temporomandibular joint (TMJ), commonly known as the jaw joint, can cause a great deal of suffering for those afflicted with TMJ disorders. Everyday activities like chewing, yawning, and sometimes even talking and laughing, can be agonizing, and personal life and work life often suffer. Approximately 3--4% of the population seek treatment for TMJ disorders, and almost 70% of these patients suffer from displacement of the TMJ disc. The TMJ disc is a poorly understood and scarcely studied structure in comparison to other musculoskeletal tissues. Prior to this thesis, a gap existed between the tissue engineering community and the TMJ characterization community. The objective of this work was therefore to perform the characterization studies vital to tissue engineering efforts, and to provide the initial steps toward an engineered TMJ disc construct. An argument was formed that provides a rationale for tissue engineering the TMJ disc, citing deficiencies in the current treatments for advanced stages of internal derangement. The TMJ disc has been shown to be mechanically non-homogeneous and highly anisotropic, which has been attributed to non-homogeneous extracellular matrix and cell sub-population distribution and anisotropic collagen orientation. The greatest variation by region in the TMJ disc is between the intermediate zone and the bands (anterior and posterior). The intermediate zone contains a higher proportion of fibrochondrocytes and higher quantities of type II collagen, chondroitin sulfate, keratan sulfate and dermatan sulfate proteoglycan compared to the anterior and posterior bands. Moreover, the intermediate zone is over an order of magnitude softer and weaker under mediolateral tension compared to these bands. Pioneering efforts in TMJ disc tissue engineering have been made, exploring growth factor effects and exploiting bioreactor technology. Insulin-like growth factor-I was selected from a group of four growth factors as the most promising for TMJ disc tissue engineering, most notably for its benefits associated with collagen synthesis. Moreover, a rotating bioreactor was shown to influence morphology and structure of engineered constructs, accelerating scaffold contraction and producing a much more heterogeneous matrix distribution compared to static culture.

The use of oligo(poly(ethylene glycol) fumarate) (OPF) hydrogels as carriers of growth factors for articular cartilage repair has been investigated. In vitro release studies examined release of transforming growth factor-beta1 (TGF-beta1) directly from OPF hydrogels and also from gelatin microparticles encapsulated within these OPF networks. These studies showed that hydrogel mesh size and microparticle content can be altered to control growth factor release rates. In particular, sustained delivery of TGF-beta1 was achieved by utilizing microparticles as a secondary drug carrier within OPF hydrogels. An in vitro degradation study demonstrated that these microparticles also serve as digestible porogens to enhance material degradation in the presence of collagenase. Furthermore, in this environment, microparticle loading and crosslinking extent were shown to influence the rates of TGF-beta1 release and composite degradation. When utilized in the repair of rabbit osteochondral defects, OPF scaffolds were shown to undergo biocompatible degradation and to support healthy tissue in-growth. However, the incorporated TGF-beta1 was not shown to greatly influence tissue repair. Accordingly, further investigations examined these hydrogel systems as carriers of TGF-beta1 and/or insulin-like growth factor (IGF-1) since these growth factors have been shown to synergistically promote chondrocyte proliferation and cartilage extracellular matrix synthesis in vitro. Surprisingly, individual delivery of IGF-1 appeared to enhance cartilage repair in rabbit osteochondral defects when compared to untreated defects, but delivery of TGF-beta1 with or without IGF-1 had no effect. These findings illustrate that the in vitro effects of growth factors, including the synergistic actions of multiple factors, may not directly translate to the wound healing environment. Furthermore, this research demonstrates the utility of these hydrogel systems in studying the effectiveness of various growth factor delivery regimes in soft tissue repair.

Clinical applications of adenoviruses as gene delivery vectors are limited by their propensity to invoke strong immune responses and toxicity. In addition, the inherent tropism of adenovirus prevents them from reaching the desired cell targets in vivo. This thesis evaluates the ability of chemically modified adenoviral vectors to evade innate immune responses, binding to blood cells, and to target specific cell types with cell-specific ligands. Previous studies have shown that polyethylene glycol (PEG) modification can protect vectors from pre-existing and adaptive immune responses by reducing protein-protein interactions. In this work, we have optimized PEGylation methods and have compared the induction of innate immune responses between modified and unmodified first generation and helper dependent adenoviruses in mouse models. The levels of interleukin-6, a cytokine induced during acute immune response, were found to be significantly lower in the most heavily PEGylated viruses in the murine models. We also observed that the uptake of PEG-modified vectors by macrophages and hepatic Kupffer cells were significantly reduced in vivo. Besides immunogenicity of the vectors, we also explored the binding affinity of chemically modified Ad to blood cells in vitro. PEG-modified Ad not only had reduced binding to erythrocytes and platelets, but also provoked reduced in vivo thrombocytopenia and prevented in vitro hemagglutination. To achieve clinically relevant gene transfer in cell types not susceptible to adenoviral transduction, we conjugated a wide array of cell-specific ligands onto adenoviruses via PEG crosslinkers to retarget the vectors to new receptors. Specifically, we conjugated epidermal growth factors (EGF) and anti-CD59 antibodies to Ad. Conjugation of these new ligands increased transduction on epidermal carcinoma and acute myeloid leukemia cell lines 5--10 fold over PEGylated vectors in vitro. However, the ability of targeted vectors to transduce cells varied greatly and is dependent on receptor densities, ligand functionality after conjugation and size of the conjugated vectors. Nonetheless, this strategy of incorporating cell-specific ligands to PEG modified adenovirus is valuable in creating a safer vector, thereby improving the overall safety and efficacy of adenoviral vectors for future cancer and metabolic disease gene therapy treatments.

Epifluorescent video-microscopy was used to evaluate the effects of a novel antiplatelet agent, IC, on platelet adhesion and aggregation in the formation of mural thrombi. Studies demonstrate that compound concentration and shear rate influence the drug effectiveness. Whole blood, treated with that compound, at concentrations varying from 25 $\mu$mol/l to 200 $\mu$mol/l, and at 100/s and 1000/s shear rates, was perfused over collagen coated coverslips in a parallel plate flow chamber. At 1000/s wall shear rate, the compound inhibits total platelet accumulation from 34% to 81%, depending on the drug concentration used. Whereas at 100/s shear rate and 100 $\mu$mol/l, the inhibition varies from 34% to 57% depending on the flow time. Platelet aggregation and adhesion are affected by the drug at all concentrations and shear rates; however results show that adhesion is less affected than aggregation and that platelet accumulation is kinetically much slower at the lower wall shear rates.

This study is devoted to a quantitative characterization of the functional properties of the photopic electroretinogram (ERG) in the normal human eye. During this analysis, the relationship between the conventional flash ERG and the response to more complex input patterns is investigated. In order to meet these goals, a mathematical model is developed which simulates the behavior of the photopic system. First, the flash ERG is expressed as the summation of five simple shapes, thus parameterizing the complex waveform. The theoretical development of a model based on these parameters is then presented. This model consists of five similar branches, one to account for the behavior of each component of the ERG. Instaneous model output is calculated on a digital computer by implementing the following discrete elements for each branch: (1) an element D which delays each input sample as a function of its energy, (2) a linear element G, whose output g represents the amplitude memory of the system, (3) a static nonlinearity Y, which acts upon g to determine the instantaneous amplitude gain factor y, (4) a model operator which multiplies y by the delayed instantaneous input intensity in order to compute the amplitude of the component response to each input sample, and (5) a shaping filter H whose impulse response is determined by the component being calculated and by the previous exposure to light. This model structure is then used to successfully simulate the response to various light patterns, including flashes, steps, and sinusoids. In this manner, the functional properties of the ERG can be analyzed under different conditions of light adaptation.

Exposure of normal erythrocytes to super-physiologic but subhemolytic shear stresses results in decreased red cell deformability. This model defect has helped pinpoint possible factors effecting similar cell damage from currently used medical devices. Red cell ATP is found not to be a primary determinant of this phenomenon. Intracellular calcium, as measured by a modification of Harrison and Long's procedure, is increased by 35 and 55% above control levels following exposure to shear stresses of 1000 and 1300 dynes/cm('2) for two minutes, and is associated with decreasing deformability. The disparate magnitudes of shear-related sodium influx and potassium efflux indicate genesis of a hyperpolarized membrane during shear, which apparently enhances calcium uptake. No significant changes are found for intracellular magnesium, 2,3-diphosphoglycerate or mean cell volume. Major results of the model studies are verified clinically by measurements on red cells from patients with chronic renal failure receiving hemodialysis or for patients with prosthetic cardiac valves. A strong correlation (r = .91, p < .001) has been found for our deformability indicator with intracellular calcium for the prosthetic heart valve patients; other correlations for this group are found with patient's values for hematocrit, serum lactate dehydrogenase, and percentage of abnormal cells. These findings imply that deformability is indeed an important pathophysiological quantity. Erythrocyte deformability for this work is determined by the pressure drop during constant volumetric flowrate filtration of a red cell suspension through the 3 (mu)m diameter pores of a Nuclepore filter. The zero time pressure drop from the linear portion of this curve is the deformability indicator P(,0). Pressure vs. filtration time curve characteristics of initial pressure drop, transient response, and a slight steady-state positive slope are attributed to membrane viscosity, the experimental setup, and pore plugging, respectively. Theoretical models lead to an estimate of the coefficient of surface viscosity for the red cell membrane (6(.)10('-3) dyn sec/cm) and to a curve fit for pressure as a function of filtration time.

The work herein examines in vitro platelet aggregation in response to fluid shearing motion. Our specific aim is to characterize shear-induced aggregation by means of kinetic measurements. In doing so we consider plausible physicochemical mechanisms for platelet activation in the shear field. Besides resolving some questions concerning the activation of platelets by shear forces, this study further implicates fluid mechanical factors in thrombosis and arterial disease. Specific results may also apply to the design and evaluation of blood-contacting artificial devices. The experimental procedure centers on the use of a rotational viscometer to apply a controlled shearing motion to platelet suspensions for prescribed times. We quantify aggregation through changes in particle size histograms and associated measures (e.g. total number of particles). Additional insight into the aggregation response comes from interpreting kinetic data using the coalescence equation, a population balance specific for particle aggregation. The coalescence equation gives rise to so-called "population balance measures" of aspects of particles pertinent to their aggregation behavior. For example, one measure indirectly assesses the adhesiveness of platelets during aggregation. Kinetic data from shear-induced aggregation indicate two population balance measures are important: the particle collison efficiency, (epsilon), and the particle void volume fraction, (phi). These and other kinetic measures indicate rapid activation (within ten seconds) of platelets in the shear field above a threshold shearing rate of about 2000 sec('-1) but diminishing platelet adhesiveness upon continued shear stress exposure. Moreover, aggregation in the shear field disappears at sufficiently low platelet concentrations (below about 60,000/mm('3) plasma), suggesting that chemical release or leakage from platelets may mediate shear-induced aggregation.