The mechanical properties of the extracellular matrix (ECM) have shown to regulate key cellular processes. However, current tools studying cell-ECM biophysical interactions revolve around cell-mediated traction forces, which, as I will show, are not appropriate in natural matrices due to matrix remodeling. I used active microrheology (AMR) to, instead, measure ECM stiffness in order to quantify these interactions in various cell-ECM systems.

In the first system, I evaluated a commonly used 3D cell-culture method in breast cancer research. I show that this model produces a large physical asymmetry in ECM stiffness, which resulted in altered cellular morphology, adhesion-mediated signaling, and phenotype. Importantly, a hallmark result obtained in this culture method was not repeatable once the asymmetry was removed, highlighting the importance of considering biophysical interactions in cell-culture models.

In the second system, my work, in collaboration with Dr. Stephen Weiss, led to the discovery that stem cells are not passive recipients of ECM stiffness signals as previously thought. Rather they can deliberately alter local (pericellular) stiffness with matrix metalloproteinases as a control for cellular functions. In particular, we found that skeletal stem cells competent in their ability to degrade collagen, increased pericellular stiffness via matrix remodeling to activate β1 integrin signaling pathways and thus controlled their own lineage commitment to osteogenic fates. Cells without the ability to degrade their local matrix lost this functionality and were restricted in lineage commitment to adipogenic or chondrogenic fates.

For the third system, I quantified the contributions of cell contractility and matrix metalloproteinases in matrix remodeling for developing a normal mechanical topography in smooth muscle cells. I also provide evidence that it is the distribution of pericellular stiffness rather than a bulk value that instructs cellular behavior. In order to accomplish this task, I automated the AMR system (aAMR) for a tenfold decrease in measurement time. Importantly, aAMR reduces the complexity of AMR to a few mouse clicks, can create stiffness maps over large distances and provides metrics to assess the distribution of stiffness in the pericellular space within the volume of a natural, fibrous hydrogel.

In Type 1 Diabetes, insulin producing cells are destroyed by the immune system, resulting in unchecked glycemic conditions. Different approaches including tissue engineering and continuous analyte monitoring hold promise in providing insulin independence and glycemic control. Tissue engineering aims to transplant and protect pancreatic islets, cells responsible for secretion of insulin. One strategy is to encapsulate the islets inside alginate hydrogels. The encapsulant provides passage to glucose, nutrients and the secreted insulin, while blocking the passage of antibodies. In this study, confocal microscopy is used to study diffusional characteristics of alginate. This approach can quantitatively analyze the structural changes after exposure to physiological conditions. Using this strategy can potentially tune the structure prior to implantation to account for the upcoming in vivo changes. Another approach is to place the islets inside subcutaneous medical devices. Such devices can provide protection to the cells, however due to hypoxic conditions transplanted cells can lose function. In this study vascularization of different types of polymer devices is studied. Oxygen sensitive tubes were fabricated and placed inside devices prior to subcutaneous implantation in nude mice. Using a non-invasive optical technology oxygen partial pressure within the devices is monitored. This technology aims to create a quantitative metric to assess the state of vascularization and readiness of devices for cell insertion. Another promising technology for diabetes management and achievement of tight glycemic control is continuous analyte monitoring. In this technique, different analytes such as glucose and lactate can be continuously measured. The data collected can be used to create a mathematical algorithm that can predict upcoming glycemic changes and in conjunction with an insulin pump can automatically administer insulin. In this work, a new composite material is invented that can accommodate necessary components to detect and report the changes of analyte levels in physiological conditions. This material can be used to create different types of continuous biosensors. Importantly this composite material shows success in preserving sensitivity and activity of biosensors for long periods of storage, it shows fast responses to changes of analyte concentrations and is manufacturable in very small geometries aimed for painless insertion.

How biological fibrous materials respond to forces and changes at the cellular scale is important for understanding what cells `feel' from their extracellular environment. The local mechanical property within a fibrin fiber network is dependent on the local structure of the network. Elastic force transmission through hydrogel depends on the arrangement of the fiber. A novel method for visualizing force transmission through this network is measuring the pixel fluctuation of individual fibers in the presence of an actively applied oscillation using embedded microbeads and optical tweezers. Optical tweezers can also be used for measuring local stiffness through active microrheology. After micropatterned crosslinking, fibrin hydrogels were shown to produce local stiffening directly and indirectly through strain hardening. This stiffening resulted in stiffness anisotropy and fiber alignment which MDA-231 cells were shown to respond to. Finally, a model of elastically connected nodes recapitulates some of the experimental observations such as fiber force transmission, strain hardening, and stiffness anisotropy when strained. Manipulating, measuring, and modelling of local stiffness changes in response to applied forces offers further insight into how cells sense their surroundings.

Continuous measurements of physiological biomarkers enables patients to assess their levels in real-time and can help healthcare professionals determine if treatment is improving patient outcome. To monitor these biomarkers, an invasive blood draw is often required. Unfortunately, frequent blood draws increase the likelihood of anemia, blood infection, and nerve damage, which in many situations, may worsen the patient’s condition. My thesis works addresses the need to alleviate frequent blood draws by replacing such laboratory assays with continuous and implantable optical sensors.

My first project was in collaboration with Dr. Gregory Weiss in the Department of Chemistry at UCI. I have engineered an implantable Förster Resonance Energy Transfer (FRET)-based calcium (Ca2+) sensor that provides continuous, physiologically relevant Ca2+ measurements. The FRET sensor addresses the foreign body response by incorporating a molecular filter and takes advantage of the conformational changes observed when Ca2+ binds to Troponin-C (within FRET complexes) to optically monitor Ca2+. My findings suggest that FRET-based sensing of target analytes using an implantable optical fiber sensor is effective and, in conjunction with protein engineering, is a new option for continuously monitoring physiologically relevant electrolytes.

For my second project, I spectroscopically monitor pH and lactate on an implantable flexible sensor. Together, pH and lactate values and trends can help healthcare professionals discriminate between metabolic and respiratory dysfunctions, helping to guide patient therapy. When implanted in a rabbit, the pH and lactate multi-analyte sensor shows accurate pathophysiological trends with respect to a handheld blood analyzer.

Collectively, these efforts show the feasibility of implantable optical sensors to continuously monitor multiple physiological biomarkers, simultaneously. With libraries of new and selective luminescence dyes and FRET probes, there is potential for implantable optical sensors to displace blood draws for improved patient care.

Type 1 diabetes (T1D) is an autoimmune disease that affects 1.25 million Americans. Although there have been many technological advancements that improve the care and maintenance of the disease, there is currently still no cure. Transplanting islets of Langerhans has shown potential to maintain normoglycemia in patients, and to prolong the effects of this cell therapy, tissue engineering devices have been used to protect the cells from the host’s immune response. However, the mass transport of oxygen and nutrients in these devices for islet survival and insulin secretion has been limited due to the long distances between cells and the host’s vasculature system and slow diffusion rates. Pre- vascularization of the scaffold is a process in which the scaffold is implanted and allowed to vascularize before the cells are transplanted. This may improve success of the implant by providing a closer source of metabolic transport for the cells.

In the first chapter, we demonstrated a technique to optically measure oxygen concentrations non-invasively in subcutaneously implanted PDMS-based tissue scaffolds over eight weeks. Tracking oxygen diffusion rates at the site of implantation over time may provide insight into the vascularization process and the optimal time for pre- vascularization. To further increase reliability of the oxygen measurements, in the second chapter, an optical oxygen sensor is developed that uses the phase shift between the sinusoidally modulated excitation and emission signals. Fewer sampling points are needed to accurately characterize a wave function than an exponential function, which yields more robust and reliable measurements.

An optical fiber senor with a semipermeable membrane is developed and described as a first step toward a continuous insulin sensor. A PEGDMA 2000 formulation was developed that allows insulin to diffuse through, but blocks larger proteins, in order to provide a physical immunoisolative barrier. Next, this material was incorporated into a semipermeable membrane and an optical fiber sensor was created. This sensor was validated with Twitch 2B, a protein with CFP and YFP fluorophores on either end. In the presence of calcium, Twitch 2B will FRET, and the CFP will transfer energy to YFP, causing YFP to emit light instead of CFP. This protein was used as a stand-in molecule for an insulin binding protein currently under development, and an optical fiber sensor was created that responded to increasing concentrations of calcium with increasing FRET. Diffusion tests were conducted on this sensor, to demonstrate that insulin can diffuse through but larger molecules are blocked. Finally, it was shown that changes in FRET from Twitch 2B caused by increasing calcium concentrations can be detected using a photodiode and emission filters, providing an avenue to transition into a circuit sensor design. It was also shown that the Twitch 2B protein can be incorporated into a composite material made up of PEGDMA formulations and PTFE sheets, to allow for a thinner design.

An estimated 1.25 million Americans suffer from type 1 diabetes (T1D), an incurable autoimmune disease with increasing prevalence. Currently, patients manage their disease with combined insulin administration via injection or pump and blood glucose monitoring. For severe cases, pancreatic islet transplantation into the liver via the portal vein has shown to increase patient quality of life; however, this procedure comes with many risks and potential morbidities making it unsuitable for a majority of the T1D population. A bioartificial pancreas provides an attractive advantage over current treatment methods by allowing “hands-free” glucose control mediated by pancreatic islets.

We developed a two-phase approach to islet transplantation with a thin-sheet device perfused by the host vasculature prior to islet introduction. In phase one, the host develops new tissue within the device that is fully integrated into the subcutaneous space, demonstrated by infiltration of mature vasculature via both lectin perfusion and histology as well as nerve tissue via histology. Noninvasive in vivo oxygen dynamics measurements indicate shorter prevascularization periods may be more beneficial. In phase two, we infuse islets with a newly developed loading method into a single file configuration within the device channels such that they are immediately adjacent to host vasculature. Prototype devices were fabricated, modified, and tested in diabetic athymic nude mice. Devices were allowed to vascularize and then re-accessed to load islets in situ. Intraperitoneal glucose tolerance tests, C-peptide measurements, nonfasting blood glucose values, and immunohistochemistry staining results indicate islets transplanted into devices maintain partial function and in a few cases, euglycemia. Device modeling with in vivo perfusion conditions indicates the islet packing fraction and level of perfusion majorly contribute to insulin production by the device and could explain differences in glycemic control.

While tissue stiffness is thought to play a role in regulation of cellular behavior, for the most part, stiffness is measured at the bulk level. The bulk measurement masks microscale dynamics within the fibrous extracellular matrix (ECM) and is insensitive to changes as cells remodel their local ECM. In order to investigate, cell-ECM dynamics I have developed an automated active microrheology (AMR) system and used it to probe the ECM near both single, isolated cells and multi-cell angiogenic sprouts, quantifying the pericellular distribution of stiffness. Additionally, I developed a new technique to modify stiffness within the ECM, at a scale relevant to the pericellular distribution of stiffness.

My work shows that both human fibroblasts and smooth muscle cells establish a complex heterogeneous pericellular stiffness landscape. As expected, cell contraction strain hardened the matrix, but surprisingly, cells must also be competent in ECM proteolysis, which is to say the matrix must be broken down for cell-mediated stiffening. My findings suggest pericellular stiffness distributions should be considered in the study of cell-ECM interactions.

In collaboration with Professor Andrew Putnam, I measured the evolution of stiffness change within a capillary morphogenesis model over time. We applied both bulk rheology and AMR to measure stiffness at different length scales. This data highlighted that bulk rheology was dominated by the activity of supportive stromal cells but blinded to the stiffness heterogeneity found proximal to vessels via AMR. These findings underscore that characterizing ECM mechanics across length scales can provide a deeper understanding of the microenvironment’s role within these complex processes.

Lastly, I developed and evaluated a method to modify stiffness within fibrin matrices at the micron-scale. This method allows for a patterning of stiffness at a spatial scale and magnitude similar to that observed by cell-mediated stiffening. By using ruthenium-catalyzed photo-crosslinking coupled with our laser scanning confocal microscope, we can selectively illuminate and thereby selectively crosslink regions of interest within the volume of a hydrogel. This results in a stiffness increase of up to 25X, with a steep stiffness gradient in the surrounding area. Selective crosslinking could be of great utility in creating more complex patterns of stiffness, which could be invaluable for the investigation of mechanotransduction within a natural 3D ECM context.

Collectively, these works show that the mechanical topography surrounding cells within ECM is varied and must be considered in future study of mechanically driven hypotheses. Microrheology in combination with selective photo-crosslinking provides a new tool to better understand roles for tissue stiffness in cell regulation, and vice versa.

Biomaterials are engaged ubiquitously to regenerate or replace damaged or diseased tissues. Numerous processing techniques aim to impart interconnected, porous structures within biomaterials to support cell delivery, direct tissue growth, and increase the acceptance of foreign materials in the body. Many processing techniques lack predictable control of scaffold architecture, and rapid prototyping methods are often limited by time-consuming, layer-by-layer fabrication of micro-features appropriate for biomaterials applications. Further, scaffold architecture is implicated in the body’s innate ability to isolate foreign substances making mitigation of this foreign body response (FBR) essential to ensuring the longevity of implanted biomaterials and devices. Bicontinuous interfacially jammed emulsion gels (bijels) offer a robust, self-assembly-based platform for synthesizing a new class of morphologically distinct biomaterials. Bijels form via kinetic arrest of temperature-driven spinodal decomposition in partially miscible binary liquid systems. These non-equilibrium soft materials comprise co-continuous, fully percolating, non-constricting liquid domains separated by a nanoparticle monolayer. In this dissertation, fluid incompatibility in bijels is exploited to process biocompatible precursors to form hydrogel scaffolds displaying the morphological characteristics of the parent bijel template. Bijel-derived materials are first used to generate structurally unique, fibrin-loaded polyethylene glycol hydrogel composites to demonstrate a new, robust cell delivery system. Next, bijel-derived materials are investigated as tissue integrating implants with high vascularization and FBR mitigation potential stemming from their uniquely arranged pore morphology, presenting a new paradigm for designing long-lasting biomaterials.