Glioblastoma (GBM) is the most common and lethal type of brain cancers with median survival of 12-15 months. The poor patient prognosis is partially due to the highly infiltrative nature of GBM. GBM cells invasion through brain parenchyma is the main reason for secondary tumor formation. We posit that the unique brain extracellular matrix (ECM) facilitates GBM invasion. Specifically, we focused on the perivasculature niche surrounding the GBM tumor and the biophysical properties of the tumor. To study these, we fabricated a brain-mimetic, three-dimensional hydrogel platform based on hyaluronic acid (HA) with orthogonal control on biochemical and biophysical properties. First, we used this platform to study the crosstalk between GBM cells and the perivasculature. We found that cells derived from perivascular niche of GBM patient tumors induced a migratory phenotype in patient-derived GBM cells. Second, mimicking biophysical properties of a GBM tumor within the HA hydrogel platform enabled us to demonstrate that the ECM stiffness directly caused a switch in GBM metabolism pathway. By using this novel in vitro platform, we were able to uncover effects of the ECM on GBM progression which have not been observable with traditional methods.

Glioblastoma (GBM) is the most lethal and malignant cancer originating from the central nervous system. Even with intense treatment involving surgery and radio-chemotherapy, median survival after prognosis remains within 12 months, as GBM constantly develops resistance to common therapies. Many novel therapies developed for GBM have shown promising results in in-vitro studies, but unfortunately failed in actual clinical practices, partially because traditional model systems failed to recapitulate the microenvironment surrounding GBM tumors. Therefore, we posit that unique brain extracellular matrix (ECM) facilitates therapeutic resistance in GBM. To study this problem, we investigated ECM deposition in GBM patient samples and fabricated brain-mimetic, orthogonally tunable hydrogel system in which to culture patient-derived GBM cells in 3-dimensional manner. To validate our novel ex-vivo culture system, genomic sequencing and gene expression profiling were performed for comparison with traditional in-vitro culture and animal xenograft models. At the same time, cell viability, proliferation and markers for cancer stem cell were assessed. Our model system was used to study the therapeutic response of GBM cells to commonly used therapeutics and to investigate resistance mechanisms. We found GBM cells displayed drug response kinetics comparable to in-vivo xenograft models. We also found novel molecular mechanisms describing how unique brain matrix facilitates therapeutic resistance through corresponding receptors in our 3D culture models. By utilizing novel engineered platforms to study drug resistance, we are able to uncover mechanisms that could not be observed through traditional methods.

The central nervous system (CNS) consists of neurons and glia. In the healthy CNS, neurons form connected networks and oligodendrocytes, a type of glial cell, insulate and support the neuronal functions. When a spinal cord injury (SCI) occurs, a second type of glial cell, astrocytes infiltrate the injury site and cause a cascade of changes1. As a result, SCI often leads to permanent motor and sensory loss below the injury site2. In addition, the individual is faced with a lifetime of medical care ranging from $2.3-4.7 billion3. Around 300,000 people are currently living with SCI in the United States, with around 17,7000 new cases per year3. This creates a need and opportunity for restorative, rather than assistive, care.

iii

To repair the spinal cord, there are several considerations, including but not limited to, prevention or reversal of the glial scar formation, introduction of exogenous neural cells and stimulation of the repair process of severed axons. To produce lineages of cells that can be used for introduction at the SCI site, it is necessary to culture them first in vitro. Neural stem/progenitor cells (NS/PCs) are pluripotent precursor cells that can differentiate into neurons and glia and be cultured outside of the body4. Cells have been cultured in 2D environments for over a century5; however, technological advances of the late 19th century have provided new techniques to culture cells in a more biomimetic, 3D microenvironment6,7. These 3D microenvironments provide a medium to reliably culture cells into desired lineages, such as populations of neurons and oligodendrocytes. This research focuses on creating hyaluronic acid (HA)-based hydrogels that serve as 3D microenvironments for cultured NS/PCs in which they can proliferate and differentiate. HA is very biocompatible since it is naturally abundant within the human body, including in the spinal cord, as a key component of the extracellular matrix (ECM)8. One benefit of HA-based hydrogels is their tunability for different applications9. As shown in this research, the chemical and mechanical properties can be manipulated while maintaining the structural integrity of the hydrogel. Specifically, the HA concentration of the hydrogels can be independently assessed to better understand the interactions of HA with the encapsulated cells at different concentrations. NS/PC mechanostranduction of their microenvironment has been shown to produce signaling pathways within the cells with potential consequences on cell survival and growth10 so, another tunable parameter explored in this research was how NS/PCs respond to the stiffness of the hydrogels. HA has also been shown to reduce glial scar thickness by decreasing the number of astrocytes present after SCI11. As indicated, these HA-based hydrogels demonstrate several

iv

advantages as a vehicle for application of differentiated neural cells to the SCI site. HA provides a hydrated matrix in which cells can migrate, differentiate and communicate with cell membrane receptors on neural cells, such as CD44, RHAMM and ICAM-1 to elicit cascade effects such as morphogenesis, inflammation, wound repair and metastasis9. When HA receptors are bound they also cause a relay of biochemical and mechanical information about the microenvironment to effect cascades of intracellular changes12. Thus, it is important to understand and dissociate the effects of mechanical (stiffness) and chemical (HA concentration) properties of the hydrogel microenvironment to elucidate the distinct and/or synergistic effects. The hydrogels produced from this research aimed to test the effects of three different HA concentrations (0.1 wt%, 0.5 wt% and 1 wt%) on the survival, proliferation and differentiation of encapsulated NS/PCs. Another aspect of this project was to test varied mechanical properties of the hydrogels to determine the effect on encapsulated cells. Soft (200 Pascal), medium (400 Pascal) and stiff (800 Pascal) environments were compared, with values closely matched to the elasticity of biological nerve ECM which ranges from 0.5-1 kPa13. A 2D control was also compared to the hydrogel conditions to determine relative cell survival, proliferation and differentiation. Ultimately, this research is useful because it provides the opportunity to understand how the 3D environment alone (relatively no HA), the concentration of HA, and a softer vs. stiffer environment affects NS/PC proliferation and differentiation. The long-term goal of this research is to produce a robust and biocompatible in vitro microenvironment that directs differentiation of an enriched population of specific neural cell types, neurons and oligodendrocytes. Future research can then assess the efficacy of injecting the encapsulated neural cells into the SCI site as an approach to therapeutic repair.

Biomaterial scaffolds provide platforms for biological study and support tissue engineering and regenerative medicine applications by acting as 1) physical scaffolds which can mimic biological properties to interact with cells or tissues and 2) localized reservoirs for controlled release of therapeutic biomolecules and drugs. These scaffolds have been found to benefit from incorporating macroporous networks which facilitate cell infiltration, proliferation, and delivery of therapeutic factors including gene therapy vectors. While several techniques have been established to create non-injectable scaffolds with cell-scale macroporosities to improve integration with host tissues or facilitate gene delivery, only recently have biomaterial scaffolds been developed that are both macroporous and injectable. Macroporous injectable scaffolds would enable minimally invasive delivery to irregularly shaped defects and thus increases their potential for clinical translation.

We describe development and characterization of several such scaffolds based on hyaluronic acid and their respective abilities to improve efficacy of transgene delivery to the mouse mammary fat pad and spinal cord and tissue regeneration after spinal cord injury (SCI). We compare nanoporous scaffolds to two general techniques to achieve macroporous injectable scaffolds: 1) hydrogels encapsulating degradable microparticle porogens and 2) scaffolds formed from crosslinked microparticles. Our findings indicate that while both injectable, macroporous, scaffold strategies are capable of in vivo administration, crosslinked microparticle scaffolds are significantly more effective at improving cell infiltration and transgene expression both in the mouse mammary fat pad model and the mouse clip-compression SCI model. We further investigated regeneration after SCI when utilizing crosslinked microparticle scaffolds in conjunction with viral vectors encoding for brain derived neurotrophic factor (BDNF) and neurotrophin 3 (NT3). We found significantly greater axon density, myelination, and functional recovery when scaffolds were delivered in conjunction with BDNF-over expressing virus but not NT3. This work establishes significant potential in the strategy of crosslinked microparticle scaffolds for developing injectable tissue engineering therapies for SCI and in broad regenerative medicine applications.

Cell-matrix interactions mediate complex physiological processes through biochemical, mechanical, and geometrical cues, influencing pathological changes and therapeutic responses. Accounting for matrix effects earlier in the drug development pipeline is expected to increase the likelihood of clinical success of novel therapeutics. Biomaterial-based strategies recapitulating specific tissue microenvironments in 3D cell culture exist but integrating these with the 2D culture methods primarily used for drug screening has been challenging. The development of methods which can incorporate the flexibility and utility of biomaterials will improve therapeutic development for diseases like Glioblastoma (GBM). Whose unique physiological characteristics have rendered traditional therapies ineffective for treatment. First, I adapt previous hydrogel-based technologies to be compatible with high-throughput (HT) drug screening platforms. Second, I further improve upon the underlying technique by developing a system capable of HT screening of 3-dimensional cultures and demonstrate a proof-of concept personalized medicine type application. Third, I developed techniques to quantify and characterize the diffusivity of solutes into hydrogels as a way to understand how physical pore size of hydrogels may affect penetration of therapeutics to encapsulated cells. Altogether, this work demonstrated the utility of including hydrogel based technologies in screening platforms for drug development and the establishment of fundamental techniques to characterize the effectiveness of future therapeutics for GBM.

Glioblastoma (GBM) is the most common and lethal primary central nervous system (CNS) tumor with a median survival of 12 – 15 months post-diagnosis. GBM aggression is, in part, due to high invasion of tumor cells into the CNS parenchyma which leads to therapeutic evasion as well as tumor recurrence. Mechanochemical cues within the GBM tumor microenvironment create environments conducive for cell survival and migration. Bioengineered in vitro systems which mimic features of the peritumoral niche can recapitulate in vivo GBM tumor cell behavior ex vivo. Thus, we developed three-dimensional (3D), hyaluronan (HA)-based hydrogels of varied HA concentrations and similar poroelastic properties corresponding to the peritumoral environment. Patient-derived spheroids encapsulated in our 3D hydrogels each displayed unique degrees and morphologies of invasion independent of conventional GBM molecular classification. We discovered strengths of cytoskeletal engagement, mediated by ezrin-radixin-moesin (ERM) complex, determined the propensity for cellular invasion, while cluster of differentiation 44 (CD44) expression densities determined the amounts of invasion. Furthermore, blocking the HA binding domain of receptor for hyaluronan-mediated motility (RHAMM) resulted in increased invasion, suggesting further studies on the roles of extracellular RHAMM are critical for further understanding how HA within the TME affects GBM cell invasion.

Insults to the central nervous system (CNS), including multiple sclerosis (MS) and spinal cord injury (SCI), drastically decrease the quality of life by debilitating native function of CNS cells. SCI affects approximately 276,000 people in the U.S with 12,500 new patients each year, while MS affects approximately 400,000 people in the United States and 2.5 million worldwide. Oligodendrocytes play a fundamental role in the CNS by myelinating axons and greatly impact CNS function. Replacement of oligodendrocytes either by neural stem cell transplants or targeting of endogenous stem cells is an attractive potential treatment for CNS tissue repair. However, there is no current consensus on how to efficiently differentiate oligodendrocytes. Developing a standardized protocol is important to better employ cell therapies for axon remyelination and CNS regeneration.

We aim to monitor oligodendrocyte differentiation by tracking reporter gene expression of neural stem cells transduced with a high-throughput array of genetic reporters. In parallel to optimizing a differentiation protocol by comparing previously established methods, we will construct lentiviral reporters for proteins produced at various stages of neural stem cell differentiation to track the stages of live cells. In this system, the same live cells can be tracked over time to achieve single cell data, which is often more representative of the underlying biological processes when investigating heterogeneous stem cell populations.

This thesis describes the successful development of a lentivirus reporter for neural stem cell differentiation, and in particular the neuronal lineage verified using bioluminescence and immunostaining. In addition, a direct comparison of the effects of multiple soluble factors reported to mediate early-stage oligodendrocyte differentiation on human neural stem cells was performed. Neural stem cells were differentiated for three and five weeks in various combinations of epidermal growth factor (EGF), platelet derived growth factor (PDGF), retinoic acid (RA) and purmorphamine (PM). Cells cultured in EGF with PM or RA proliferate and express oligodendrocyte specific markers making EGF, PM and RA attractive factors for future studies.

Preliminary data from these studies will be used to set up upcoming experiments for the construction of additional lentivirus reporters and optimizing a protocol for oligodendrocyte differentiation. In the future, this knowledge can be applied to develop effective neural stem-cell based therapies for neuroregeneration.