High-throughput cell culture is an emerging technology that shows promise as a tool for research in tissue engineering, drug discovery, and medical diagnostics. An important, but overlooked, challenge is the integration of experimental methods with information processing suitable for handling large databases of cell-cell and cell-substrate interactions. In this work the traditional global descriptions of cell behaviors and surface characteristics was shown insufficient for investigating short-distance cell-to-cell and cell-to-surface interactions. This problem was addressed by introducing individual-cell based local metrics that emphasize cell local environment. An individual-cell based local data analysis method was established. Contact inhibition of cell proliferation was used as a benchmark for the effectiveness of the local metrics and the method. Where global, summary metrics were unsuccessful, the local metrics successfully and quantitatively distinguished the contact inhibition effects of MC3T3-E1 cells on PLGA, PCL, and TCPS surfaces. In order to test the new metrics and analysis method, a model of cell contact inhibition was proposed. Monte Carlo simulation was performed for validating the individual-cell based local data analysis method as well as the cell model itself. The simulation results well matched with the experimental observations. The parameters used in the cell model provided new descriptions of both cell behaviors and surface characteristics. Based on the viewpoint of individual cells, the local metrics and local data analysis method were extended to the investigation of cell-surface interactions, and a new high-throughput screening and knowledge discovery method on combinatorial libraries, local cell-feature analysis, was developed. PLGA/PCL combinatorial libraries were used as a prototype and a shaper and holder phenomenon involving MC3T3-E1 cells interacting with PCL islands was discovered. In summary, the viewpoint of individual cells casts new light on the study of cell-cell and cell-surface interactions and represents a novel methodology for developing new data analysis and knowledge discovery methods. The results of contact inhibition study and the shaper and holder model provide new knowledge, while the local data analysis method as well as the cell model of contact inhibition suggested novel approaches to study cell-cell and cell-surface interactions.

Articular cartilage on weight-bearing joints experiences three main forces: fluid-induced shear across the surface, perfusion through the cartilage from the surrounding fluid, and compression during motion of the joint. A new bioreactor that employs two of these forces was developed in this lab to study their effect on tissue-engineered cartilage development. The focus of this research and overall hypothesis is that bioreactors that employ both perfusion and shear will improve chondrogenesis and preservation to produce functionally relevant cartilage by modulating shear stress and introducing exogenous preservation factors. Applying both a low shear stress across the surface of cell-seeded scaffolds and perfusion through them in a perfusion concentric cylinder (PCC) bioreactor may stimulate chondrocytes to undergo chondrogenesis. Experimental data showed that the PCC bioreactor stimulated cartilage growth over the course of four weeks, supported by the appearance of glycosaminoglycan (GAG) and collagen type II, which are markers for articular cartilage. Computational fluid dynamics modeling showed that shear stress across the face of the construct was heterogeneous, and that only the center experienced a relatively uniform shear stress of 0.4 dynes/cm^2 when the outer cup of the bioreactor rotated at 38 rpm. When compared to a concentric cylinder (CC) bioreactor that employed only shear stress, the PCC bioreactor caused a significant increase in cellular proliferation, which resulted in a 12-fold increase in cell number per construct compared to 7-fold increase within the CC bioreactor. However, the PCC bioreactor had a less pronounced effect on glycosaminoglycan and collagen content with 1.3 mg of GAG and 1.8 mg of collagen per construct within the CC bioreactor and 0.7 mg of GAG and 0.8 mg of collagen per construct within the PCC bioreactor after 28 days in culture (p < 0.05). Our results led to an important observation that the PCC bioreactor affected cellular proliferation significantly but not extracellular matrix synthesis. The next objective of this study focused on the PCC bioreactor to evaluate the direct role of perfusion and shear on chondrogenesis in vitro and in vivo.

The development of tissue engineered blood vessel mimics for the testing of intravascular devices in vitro has been established in the Cal Poly tissue engineering lab. Due to the prevalence of cardiovascular disease in diabetic patients and minimal accessible studies regarding the interactions between diabetes and intravascular devices used to treat vascular disease, there is a need for the development of diabetic models that more accurately represents diabetic processes occurring in the blood vessels, primarily endothelial dysfunction. This thesis aimed to create a diabetic blood vessel mimic by implementing a high glucose environment for culturing human endothelial cells from healthy umbilical veins (HUVECs) and from diabetic coronary arteries (DHCAECs). The characterization of these BVMs was achieved using immunofluorescence, scanning electron microscopy (SEM), and qPCR gene expression analysis. From this study, it was determined that HUVECs and DHCAECs are robust enough to be cultured in a high glucose environment – analogous to hyperglycemia – and these cells exhibited different characteristics when evaluated under microscopy and qPCR gene expression. The immunofluorescence and SEM imaging showed presence of cells within each blood vessel mimic. The qPCR gene expression analysis demonstrated that mRNA expression of endothelial nitric oxide synthase (eNOS), platelet endothelial cell adhesion molecule (PECAM), and receptor for advanced glycation end products (RAGE) differs between HUVECs and DHCAECs, as well as between cells cultured in v normal and elevated glucose concentrations. These differences in gene regulation indicate the potential of the diabetic BVM to more accurately represent the endothelial response to diabetes and to the implementation of intravascular devices in the future. It was determined that culturing DHCAECs in a high glucose cell media for use in blood vessel mimics results in a model that differs considerably from HUVECs grown in normal glucose media. It was also determined that there was a difference between DHCAECs cultured in high glucose media and normal glucose media, as well as HUVECs cultured in high glucose media and normal glucose media. This study aided the development of a diabetic BVM; however, there are still improvements to be made, namely the inclusion of vascular smooth muscle cells in the model and improving the confluency of the BVM.

Trotz intensiver Forschung im Rahmen der Bänder- und Sehnenerkrankungen gelten bestimmte Fragestellungen hinsichtlich Erkrankungs- sowie Heilungsmechanismen als unbeantwortet. Verschiedenste Konzepte des Tissue Engineerings sollen helfen entsprechende Fragen zu beantworten und moderne Therapiekonzepte zu etablieren. Für grundlegende Untersuchungen zur Biologie der Tenogenese sowie zum Wirkmechanismus applizierter mesenchymaler Stromazellen (MSC), gewinnt die Anwendung von dezellularisiertem Sehnengewebe immer mehr an Bedeutung. Zudem erscheint der Einsatz dezellularisierter Sehnen- und Bandkonstrukte zur Wiederherstellung der betreffenden erkrankten Organe sehr vielversprechend. In der vorliegenden Arbeit sollte der Grundstein zur Entwicklung eines in vitro-Modells auf Grundlage equiner Beugesehnen gelegt werden. Primäres Ziel war es, ein optimales Dezellularisierungsprotokoll für intakte equine Beugesehnen (oberflächliche und tiefe Beugesehne) zu etablieren. Um die Zytokompatibilität der dezellularisierten Sehnen zu überprüfen, erfolgte nach Präparation von Sehnenstreifen die Besiedlung mit equinen MSC mit Kontrolle des Besiedlungserfolges. Materialien und Methoden: Oberflächliche und tiefe Beugesehnen (OBS und TBS) des Pferdes (n = 6) wurden nach vier verschiedenen Protokollen dezellularisiert. In zwei Protokollen (Protokolle A und B) erfolgte zunächst die Anwendung von Gefrier-Auftau- Zyklen mit anschließender Lagerung in hypertoner Lösung. Protokoll A sah danach eine Inkubation in 1 % Triton X 100 und Protokoll B eine Inkubation in 1 % Sodium-Dodecyl-Sulfat (SDS) enthaltender Lösung vor. Die beiden anderen Protokolle (Protokolle C und D) sahen ein Verbringen in hypertone Lösung ohne vorherige Gefrierzyklen vor. Anschließend erfolgte bei Protokoll C die Inkubation in Triton X 100 und bei Protokoll D die Inkubation in SDS enthaltender Lösung. Die Effektivität der angewandten Dezellularisierungsprotokolle wurde durch histologischer Färbung, Zellzählung nach Kollagenaseverdau, DNA-Quantifizierung und transmissionselektronenmikroskopischer Untersuchung ermittelt. Nach Evaluierung der Effektivität der Protokolle wurden oberflächliche Beugesehnen nach den Protokollen A und B dezellularisiert (n=3). Nach Präparation von Sehnenstreifen in definierter Größe erfolgte die Besiedelung mit Eisenoxid-markierten equinen MSC. Der Besiedlungserfolg wurde mit verschiedenen histologischen und Fluoreszenzfärbungen (Fluoreszenzmikroskopie) und MRT-Untersuchung kontrolliert. Die Prüfung auf statistische Unterschiede zwischen den Protokollen erfolgte mit dem Friedman-Test und im Falle eines statistisch signifikanten Unterschieds mit dem Wilcoxon-Rang-Test. Das Signifikanzniveau wurde mit p < 0,05 festgelegt. Die Auswertung des Besiedlungserfolges erfolgte deskriptiv. Ergebnisse: Für alle angewandten Protokolle konnte ein signifikanter Dezellularisie-rungseffekt in beiden Sehnenstrukturen (OBS und TBS) gezeigt werden. Die Anzahl der vitalen Zellen nach Kollagenaseverdau sowie die histologisch ermittelte Zellzahl der dezellularisierten Sehnen belief sich in Abhängigkeit des jeweiligen Dezellularisie-rungsprotokolls und der Sehne (OBS und TBS) auf 1 bis 21 % (Median) des nativen Gewebes. Der ermittelte DNA-Gehalt nach Anwendung der mit Gefrier-Auftau-Zyklen kombinierten Protokollen A und B entsprach < 24 % (Median) des nativen Gewebes. Die Anwendung der Protokolle C und D führte zu einem DNA-Gehalt von < 47 % (Median). Die Auswertung der transmissionselektronenmikroskopischen Untersuchung zeigte ebenfalls eine effektive Dezellularisierung des Sehnengewebes bei Erhalt der Struktur der extra-zellulären Matrix. Nach Anwendung der Protokolle A und B konnte wiederum tendenziell eine bessere Effektivität der Dezellularisierung festgestellt werden. Eine gelungene Besiedlung der Sehnenstreifen mit equinen MSC konnte anhand der mikroskopischen Untersuchung und MRT-Untersuchung gezeigt werden. Das beobachtete Zellwachstum bei beibehaltender Vitalität der Zellen sprechen für eine gute Zytokompatibilität. Die nach Protokoll A dezellularisierten und besiedelten Sehnenstreifen ließen ein besseres Zellwachstum über eine Kulturdauer von 14 Tagen erkennen. In der vorliegenden Arbeit konnte eine effektive Dezellularisierung von intakten equinen Beugesehnen gezeigt werden. Anhand der Ergebnisse der Besiedlung erwies sich die Dezellularisierung nach Protokoll A (Gefrier-Auftau-Zyklen und Triton X 100) als vielversprechende Grundlage zur Entwicklung eins in vitro Modells auf Grundlage dezellularisierter equiner Beugesehnen.

Osteoarthritis (OA), a degenerative disease of articular joints, is the leading cause of chronic disability in the US and affects more than a third of adults over 65 years old. Due to the obesity epidemic and an aging population, the prevalence of OA is expected to rise in both young and old adults. There are no disease modifying OA drugs. Therefore, providing any treatment options that delay the onset or progression of OA is highly desirable. The scope of this dissertation examines two different strategies to promote translational therapies for OA. The first approach investigated whether Smad ubiquitin regulatory factor 2 (Smurf2), an E3 ubiquitin ligase, could be a potential therapeutic target for OA. The second approach examined the incorporation of small chemical residues to enhance the physical and bioactivity of a bioinert scaffold for cartilage tissue repair. Overexpression of Smurf2 in chondrocytes was shown to accelerate spontaneous OA development in mice. We hypothesized that reduced Smurf2 expression could slow the progression of OA and enhance the performance of cells for cartilage repair. By performing surgical destabilization of the medial meniscus (DMM) on Smurf2-deficient mice, loss of Smurf2 was shown to mitigate OA changes in young mice but this protection diminished in older mice. Assessment of Smurf2-deficient chondrocytes in vitro revealed an upregulation of chondrogenic genes compared to wild-type; however, these differences were not seen at the protein level, deterring its potential use for cell-based therapies. During the course of this study, new insights about how age and sex affects different joint compartments in response to DMM surgery were also uncovered. These results broadened existing understanding of DMM-induced OA in mice but also questioned the validity of such a model to identify disease modifying targets that are translatable to OA in humans with advanced age. Due to a lack of innate repair mechanisms in cartilage, damage to cartilage increases the risk of developing OA early. Tissue engineering provides a unique strategy for repairing damaged cartilage by delivering cells in a well-controlled environment that can promote the formation of neotissue. We hypothesized that synthetic chemical residues could enhance the mechanical properties of a bioinert scaffold and promote matrix production of encapsulated chondrocytes. Covalent incorporation of small anionic or zwitterionic chemical residues in a polyethylene glycol-based hydrogel improved its stiffness and resistance to fluid flow, however, the resulting physical environment can also exert a dominant negative effect on matrix production of encapsulated chondrocytes. These results suggest that modulating the biosynthesis of chondrocytes with biochemical signals requires a concurrent reduction in any conflicting mechanotransduction signaling, emphasizing the importance of a degradable system to promote new cartilage formation. In summary, this dissertation establishes Smurf2 as a modulator of OA progression but implies that other factors such as age or protein(s) with redundant Smurf2 functions may play a role in limiting its effect as a therapeutic target. This work also reveals fundamental biology about how chondrocytes behave in response to physical and chemical cues in their microenvironment, which will aid in the design of better scaffolds for cartilage tissue engineering.

Ziel der Dissertation ist die Entwicklung eines Perfusionsbioreaktors, um physiologische Zellnischen nachzuahmen und damit 3-dimensionale (3D) Knochenkonstrukte mit knochenähnlichen Eigenschaften zu erzeugen. Zukünftig sollen diese Konstrukte das klinische Ergebnis der Knochendefektheilung verbessern und die biomechanische Belastbarkeit wiederherstellen. Dazu sind neue 3D-Tissue-Engineering-Entwicklungen notwendig, weil die existierenden Zellkultivierungsgefäße übermäßig vereinfacht und nicht hinreichend sind.. Folglich werden in dieser Arbeit mehrere Technologien kombiniert, um physiologische Nischen zu schaffen und klinisch anwendbare Knochenkonstrukte in vitro abzuleiten.

Osteoarthritis is a prevalent disease that affects the articular cartilage of the joints. Millions of people suffer worldwide and it is a major cause of disability in the United States. Current research for treatments of osteoarthritis are studying tissue-engineered cartilage in vitro generated by articular chondrocytes. A challenge faced in vitro for cartilage tissue engineering is the failure of chondrocytes to produce adequate expression of type II collagen. Surprisingly, the media commonly used in vitro lacks 14 vitamins and minerals present in the physiological environment of chondrocytes. Therefore, studying the interactions between micronutrients and chondrocytes may help in potentially increasing the amount of type II collagen expressed by these cells. This project studied the combinatorial effects of vitamins and minerals in defined chondrogenic media on type II collagen expression. Linolenic acid was determined to have predominantly negative effects on chondrogenesis and Vitamin B7 to have beneficial effects. Multiple vitamins and minerals displayed significant interactions, both positive and negative.

Magnetic resonance imaging is a non-invasive imaging modality that is used to produce detailed images of soft tissues within the human body. Typically, MRI scanners used in the clinical setting are high field systems because they have a magnetic field strength greater than 1.5 Tesla. The high magnetic field offers the benefit of high spatial resolution and high SNR. However, low filed systems can also produce high resolution MR images with the added benefit of imaging stiffer samples. In this study, a low field 0.5 T MR system was used to image various samples to demonstrate the capability of the low field system in acquiring MR images with resolution comparable to high field systems. Furthermore, the MR system was modified to one capable of performing low field MR Elastography (MRE), a technique that can non-destructively measure the mechanical properties of soft samples. Agarose gel phantom of 0.5% wt. and 1.0% wt. were used to validate the MRE system. Additionally, a rat brain was used to assess the sensitivity of the MRE system in measuring the mechanical properties of small tissues. The results illustrated that the low field MR system can acquire high resolution images and provide sufficient tissue contrast (e.g through long TE times (80 ms), which is not possible with high field systems). MRE results on gel phantoms illustrated the capability of the low field system to accurately measure the mechanical properties and the MRE testing of rat brain demonstrated the potential of the system to study biological tissues. Finally, the capability of low field MRI and MRE to assess the growth of tissue engineered bone has the potential to transform the field of tissue engineering.

The development of treatments for neurological disorders such as Alzheimer’s and Parkinson’s disease begins by understanding what these diseases affect and the consequences of further manifestation. One particular region where these diseases can produce substantial problems is the blood-brain barrier (BBB). The BBB is the selective diffusion barrier between the circulating blood and the brain. The barrier’s main function is to maintain CNS homeostasis and protect the brain from the extracellular environment. The progression of BBB research has advanced to the point where many have modeled the BBB in vitro with aims of further characterizing and testing the barrier. Particularly, the pharmaceutical industry has gained interest in this field of research to improve drug development and obtain novel treatments for patients so the need for an improved model of the BBB is pertinent in their discovery. In the Cal Poly Tissue Engineering lab, an in vitro tissue engineered BBB system has previously been obtained and characterized for the initial investigation of the barrier and its components. However, certain limitations existed with use of the commercial system. Therefore, the focus of this thesis was to improve upon the capabilities and limitations of this commercialized system to allow further expansion of BBB research. The work performed was based on three aims: first to design and develop an in-house bioreactor system that could be used to cultivate the BBB; second, to characterize flow and functional capabilities of the bioreactor; third, to develop protocols for the overall use of the bioreactor, to ultimately allow co-cultures of BAEC and C6 glioma cells, and further the progression toward creating an in vitro model of the BBB. The work of this thesis demonstrates development of an in-house custom bioreactor system that can successfully culture cells. Results showed that the system was reusable, could be sterilized and monitored, was easily used by students trained in the laboratory, and allowed non-destructive scaffold extraction. This thesis also discusses the next set of experiments that will lead to an in vitro model of the BBB.

<p>Hemophilia B patients may have frequent, spontaneous and life-threatening bleeds that are currently managed by an invasive and expensive treatment. Mesenchymal stem cells (MSCs) are increasingly being applied to clinically therapeutic strategies and lentiviral gene vectors have been shown to be safe and efficient tools for modifying stem cells for long-term expression of high levels of transgenes. In this study, MSCs were engineered with a lentivirus to express sustained and therapeutic levels of human FIX protein <em>in vitro </em>and in mice. The modified MSCs secreted human FIX protein at levels exceeding 4 μg/10⁶ MSCs/24 h with high FIX coagulant activity of greater than 2.5 mIU/10⁶ MSCs/24 h for 6 week <em>in vitro. </em>Functional FIX transgene was continually expressed by these cells when they were induced to differentiate into adipocyte, osteoblast and chondrocyte lineages <em>in vitro</em>. However, the modified MSCs transplanted via tail vein into NOD-SCID-γ mice expressed low levels of FIX <em>in vivo</em>. The transplantation procedure had an increased risk of death that was more pronounced in mice that received cell doses exceeding 2 million cells. Organ examinations suggested the deaths resulted from entrapment of MSCs in pulmonary capillaries. Modified MSCs encapsulated in alginate-PLL microcapsules and transplanted into the peritoneal cavity of both NOD-SCID-γ and hemophilia B mice at 9 million cells/mouse resulted in therapeutic expression around 100 ng of human FIX/mL of plasma only for a few days <em>in vivo</em> as human FIX expression quickly decreased to basal values by the end of the first week. Cultured <em>ex vivo</em>, human FIX expression by retrieved capsules indicated an innate immune response to the encapsulated cells prevented sustained expression of FIX. These investigations demonstrate that lentivirally modified MSCs have the potential to express therapeutic human FIX for sustained periods <em>in vitro</em>, even after their differentiation. However, they also highlight the challenges to overcome to optimize cell engraftment and survival following transplantation, and to minimize the immune responses associated with the xenogeneic translational<em> </em>models used.</p> / Doctor of Philosophy (PhD)