In order to improve disease models, understanding and manipulating the factors that contribute to cellular phenotype is of utmost importance. The role of genetics in disease, while long appreciated from a clinical standpoint, has become more elucidated by modern sequencing techniques. Mechanical forces include both the environmental forces that act on cells as well as the forces exerted by cells back on their environment. While focusing individually on genetics or mechanics is important, this dissertation aims to highlight the value of utilizing both in combination to improve disease modeling.In chapter one, we first provide an overview of the tools that allow us to manipulate and study genetic risk for disease, including gene editing techniques as well as induced pluripotent stem cell (iPSC) technology, and then provide examples where they have been successfully utilized to improve our understanding of cancer and heart disease. We then describe progress in the field of material based-mechanobiology and the engineered systems for mimicking forces exerted on cells by the extracellular matrix (ECM), surrounding fluid, and neighboring cells. We conclude by highlighting studies that have successfully manipulated both genetic and mechanical factors to improve our understanding of different cancers and heart diseases. In the second chapter, we provide an example of how combining patient derived iPSCs and haplotype editing with biophysical cell sorting can unveil insight into how the non-coding gene locus, 9p21.3, incurs risk for coronary artery disease. Utilizing adhesion-based sorting with a microfluidic device, we show that iPSC-derived vascular smooth muscle cells from patients with single nucleotide polymorphisms (SNPs) at 9p21 (RR) have increased phenotypic heterogeneity compared to those lacking the SNPs and isogenic knockouts, specifically with increased presence of a synthetic, non-contractile phenotype. We identified heterogeneous expression of an alternatively spliced long-non coding RNA within a RR patient population drives a subset of cells towards the synthetic phenotype, exacerbating disease and potentially explaining the incomplete penetrance of the disease locus. The methodology highlighted here has broad applicability to a number of different diseases, highlighting the value of approaches that both manipulate and measure genetic factors and mechanics to investigate disease development.

With age, cardiomyocytes’ regenerative capacity is highly compromised, and among many other maladaptive phenotypes such as hypertrophy, the total number of cardiomyocytes begins to diminish. To compensate for the heavier load and mechanical demand placed on the remaining cardiomyocytes, vinculin has been shown to be upregulated with age in the heart across organisms. In Drosophila Melanogaster, vinculin localizes at the intercalated discs (ICD) and costameres to help sustain the sarcomere lattice structure and generate a higher force production. In Aim 1, we tried to confer a similar improvement in cardiac function through therapeutic intervention. We screened several drugs that have been selected from Ingenuity Pathway Analysis (IPA), such as forskolin, a potent adenylyl

cyclase activator, and drugs that target the nitric oxide(NO)-cGMP pathway. Screening methods included both in vitro screening in S2R+ cells and in vivo treatment in wild-type flies. We have found that neither administrating forskolin nor targeting the NO pathway increases vinculin expression or localization in flies. Next, we wanted to investigate the relationship between cardiac contractility and systemic metabolism. In Aim 2, we validated that we can trace the metabolism of 13C-labeled glucose in the whole-fly. This allows for the investigation of differences in the glucose metabolism between the two genetic lines (vinculin-overexpressing line and the control line) across different ages. In Aim 3, we observed that impairing cellular respiration alone via treatment with a mitochondrial inhibitor is sufficient to bring about drastic cytoskeletal changes in S2R+ cells and, subsequently, mechanical function in dissected fly hearts.

Cell behavior is controlled not only by chemical signals but also by mechanical cues from their environment. Rather than being passive agents, cells sense these forces and in turn transmit forces back to remodel the environment. A number of disease processes, such as cancer and heart disease, arise in part due to an imbalance in force sensing and production. This dissertation aims to utilize materials to induce or measure cellular forces in healthy and disease conditions for improved disease modeling. Specifically, we utilized a methacrylated hyaluronic acid (MeHA) hydrogel that can dynamically stiffen in the presence of cells to recapitulate fibrotic remodeling associated with breast cancer progression and cardiac remodeling post infarction.

Mammary epithelial cells (MECs) form and remain as polarized acini but begin to decompose and resemble mesenchymal morphology upon matrix stiffening. The transcription factor Twist, transforming growth factor β (TGFβ), and YAP activation appeared to modulate stiffness-mediated signaling; when stiffness-mediated signals were blocked, collective MEC transformation was reduced in favor of single MECs transforming and migrating away from acini. These data indicates a more complex interplay of time-dependent stiffness signaling, acinar structure, and soluble cues that regulate MEC transformation than previous models suggest.

We next examined how the non-coding 9p21 gene locus regulates cardiac phenotypes associated with fibrotic remodeling. Induced pluripotent stem cell-derived cardiomyocytes (CMs) from patients homozygous for the risk (R/R) or non-risk (N/N) 9p21.3 locus were cultured on a methacrylated hyaluronic acid hydrogel capable of mimicking cardiac fibrotic stiffening. While CMs contracted synchronously in physiological niche independent of genotype, only R/R CMs exhibited asynchronous contractions due to a loss of connexin 43 expression after stiffening. Locus deletion was sufficient to prevent asynchronous contraction by maintaining gap junctions, demonstrating that stem cell technology can be extended beyond modeling of solely genetic disease to include modeling of acquired disease in response to changing environmental conditions.

In addition to the discoveries made with these studies, the methodology utilized here has broad applicability to a number of different diseases, highlighting the value of understanding the role of force induction and generation on cell behavior in the context of disease.

Glioblastoma multiforme is one of the most aggressive tumors, with an extremely low five-year survival rate of 10%. Vast inter- and intra-tumoral heterogeneity in GBM tumors leads to aggressive phenotypes and poor prognosis in most patients. One source of heterogeneity stems from variance in adhesion strength of cells to extracellular matrix, which leads to differences in migration and invasion. We hypothesized that a cell line with a set of mutations that are common in GBM will have a subpopulation of cells that will drive invasion based on their adhesion strength. In order to identify and characterize the cells that lead to aggressive GBM outcomes, we optimized and utilized a parallel plate flow chamber assay to expose wtEGFR murine astrocytes to shear stress and sort the population based on adhesion strength. Sorted populations were described either as weakly or strongly adherent and analyzed through various downstream assays such as the spinning disk shear assay to quantify adhesion strength and immunofluorescence staining to elucidate focal adhesion formation. While phenotypic differences between sorted populations and unsorted control populations were not easily reproduced between replicates at 48 hours post exposure to shear stress, differences between the populations were more easily identified 16 hours post exposure to shear stress. This indicated that phenotypic differences that are revealed as a result of physical selection are not permanent and that the underlying mechanisms of these differences need to be uncovered.

Given the prevalence of cardiovascular disease (CVD) worldwide and in the US, there is a need for mature cardiac models that recapitulate the disease complexity and function of native cardiac tissue. Conventional two-dimensional (2D) human pluripotent stem cell-derived cardiomyocyte (hPSC-CM) differentiations fail to achieve the maturity levels of more intricate three-dimensional (3D) cardiac models. hPSC-CMs often remain immature, electrically isolated, and may not reflect adult biology. Conductive polymers are attractive candidates to facilitate electrical communication between hPSC-CMs, especially at sub-confluent cell densities or diseased cells lacking cell-cell junctions. Here, we investigated how conductive polymers may be used to facilitate electrical communication between hPSC-CMs in both 2D and 3D platforms to create functionally mature cardiomyocytes for disease modeling and potential therapeutics. First, we summarized past, current, and future engineering methods implemented to direct hPSC-CM differentiations to recapitulate the complexity of native cardiac tissue. Next, we developed conductive electrospun fibrous scaffolds and assessed the structural, functional, and transcriptional changes of hPSC-CMs seeded on a 2D network. Lastly, we translated our 2D findings to 3D cardiac spheroid models to investigate how the presence of conductive microfibers may accelerate cardiac spheroid maturation over standard 3D models. Taken together, these data suggest that conductive polymer platforms are supportive of electrical coupling in hPSC-CMs and promote maturation, which may improve our ability to model cardiac diseases and develop targeted therapies.