The inability to monitor the molecular trajectories of whole organs throughout the clinically relevant ischemic interval is a critical problem underlying the organ shortage crisis. Here, we report a novel technique for fabricating manufacturing conformal microfluidic devices for organ interface. 3D conformal printing was leveraged to engineer and fabricate novel organ-conforming microfluidic devices that endow the interface between microfluidic channels and the organ cortex. Large animal studies reveal microfluidic biopsy samples contain rich diagnostic information, including clinically relevant biomarkers of ischemic pathophysiology. Overall, these results suggest microfluidic biopsy via 3D printed organ-conforming microfluidic devices could shift the paradigm for whole organ preservation and assessment, thereby relieving the organ shortage crisis through increased availability and quality of donor organs. / Master of Science / Organ failure is one of the most common cause of morbidity and mortality in humans. Unfortunately, there are not enough donor organs to meet the present demand, often referred to as the organ shortage crisis. To compound the problem, there is lack of understanding of the biological processes occurring in organs during the transplantation interval. Here, we present a method to manufacture a biomedical device using a 3D printing technique to monitor, collect, and isolate diagnostically relevant biological species released during the transplantation interval. This information has the potential to lead to a better understanding of organ health, which ultimately could increase the availability and quality of donor organs.

The pervasive issue of bacterial infections underscores the critical need for advancedpathogen diagnosis techniques. Traditional methods often fall short in sensitivity and specificity, necessitating innovations that can enhance pathogen concentration and detection in clinical samples. This study introduces a novel approach utilizing microfluidic droplet technology to partition bulk samples, significantly improving the concentration of bacteria in minuscule volumes. This method inherently amplifies detection sensitivity, offering a substantial leap forward in diagnostic capabilities. To capture the dynamic process of bacterial partitioning and concentration, we have developed a real-time fluorescence imaging system. This system not only facilitates the monitoring of droplet encapsulation but also enables the quantification of bacterial presence, crucial for applications such as quantitative PCR (qPCR). The efficacy of this integrated dropletbased microfluidic device and fluorescence imaging system was rigorously tested using Escherichia coli (E. coli), a prevalent bacterium responsible for urinary tract infections (UTIs), demonstrating its potential in clinical diagnostics. Looking beyond the immediate scope of this study, the presented system holds promise for extensive future development aimed at addressing a wider array of pathogens. Additionally, its versatility positions it as a foundational tool for a range of generalized applications in the ii fields of microbiology, bioengineering, and diagnostic medicine, highlighting its capacity to significantly impact methods of pathogen detection and analysis.

The discovery of cancer biomarkers is crucial in the clinical setting to facilitate early diagnosis and treatment, thereby increasing survival rates. Proteomic technologies with mass spectrometry detection (MS) have the potential to affect the entire spectrum of cancer research by identifying these biomarkers. Simultaneously, microfabricated devices have evolved into ideal analysis platforms for minute amounts of sample, with promising applications for proteomic investigations and future biomarker screening. This thesis reports on the analysis of the proteomic constituents of the MCF7 breast cancer cell line using a shotgun 2-D strong cationic exchange/reversed phase liquid chromatography electrospray ionization tandem mass spectrometry (SCX/RP-LC-ESI-MS/MS) protocol. A series of optimization strategies were performed to improve the LC-MS experimental set-up, sample preparation, data acquisition and database searching parameters, and to enable the detection and confident identification of a large number of proteins. Over ~4,500 proteins were identified using conventional filtering parameters, and >2000 proteins using a combination of filters and p-value sorting. Of these, ~1,950 proteins had p<0.001 (~90%) and more than half were identified by &#8805; 2 unique peptides. About 220 proteins were functionally involved in cancer related cellular processes, and over 100 proteins were previously described in the literature as potential cancer markers. Biomarkers such as PCNA, cathepsin D, E-cadherin, 14-3-3-sigma, antigen Ki-67, TP53RK, and calreticulin were identified. These data were generated by subjecting to mass spectrometric analysis ~42 µg of protein digest, analyzing 16 SCX peptide fractions, and interpreting ~55,000 MS2 spectra. Total MS time required for analysis was 40 h. Selective SCX fractions were also analyzed by using a microfluidic LC platform. The performance of the microchip LC was comparable to that obtained with bench-top instrumentation when similar experimental conditions were used. The identification of 5 cancer biomarkers was enabled by using the microchip LC platform. Furthermore, this device was also capable to analyze phosphopeptides. / Master of Science

The structural and functional abnormalities of the tumor vasculature generate regions of elevated interstitial fluid pressure and aberrant flow shear stress within the tumor microenvironment. While research has shown that the hydrodynamics of the tumor vasculature reduce transport and uptake of therapeutic agents, the underlying mechanisms by which fluid forces regulate vascular organization are not well known. Understanding the reciprocal interaction between tumor and endothelial cells to mediate angiogenesis, and the role of flow shear stress on this process, may offer insight into the design of improved therapeutic strategies to control vascularized tumors. Instrumental to this is the development of physiologically relevant models that enable tumor-endothelial co-culture under dynamic conditions. By integrating tissue-engineering strategies with cancer biology, micro-scale fluid mechanics, and optical flow diagnostics, the goal of this research was to develop a 3D in vitro microfluidic culture model to investigate tumor-endothelial cross talk under physiologically relevant flow shear stress. This objective was motivated by early findings demonstrating a contact-independent, paracrine-mediated mechanism by which endothelial cells enhance tumor-expressed angiogenic factors during 2D, static co-culture. The 3D tumor vascular model consists of a central microchannel embedded within a type I collagen hydrogel, through which a range of normal (4 dyn/cm^2), low (1 dyn/cm^2) and high (10 dyn/cm^2) microvascular wall shear stresses (WSS) were introduced. Endothelial cells lining the microchannel lumen form a confluent endothelium across which soluble growth factors are exchanged with tumor cells in the gel. Microscopic particle image velocimetry ("-PIV) was integrated within the model to enable noninvasive optical measurement of velocity profiles and quantification of WSS, which were then correlated with angiogenic potential. Results demonstrate that endothelial permeability decreases as a function of increasing WSS, while co-culture with tumor cells increases permeability. This response is likely due to shear stress-mediated endothelial cell alignment and tumor-VEGF-induced permeability. In addition, high WSS (10 dyn/cm^2) significantly down-regulates tumor-expressed angiogenic factors, suggesting flow shear stress-mediates endothelial cross talk with surrounding tumor cells. Collectively, this research demonstrates the utility of the 3D in vitro microfluidic culture model as a versatile platform for elucidating the role of tumor-relevant hydrodynamic stress on cellular response. / Ph. D.

In recent years, space exploration has been driving studies that enable sustained human presence in space. In such studies, fluidics relating to biology have become important. Fluids in biological systems span from large-scale flows relevant to circulatory, digestion, and pulmonary systems, but also involve many micro-scale porous flows. Hence, space exploration is driving a novel need to characterize fluidics in microscales in microgravity conditions. In this work, we study the porous flow network within bones that stimulates cellular growth and has the potential to relate to osteoporosis (including driving osteoporosis in astronauts). To study this effect, computational fluid dynamics (CFD) simulations are performed on a microfluidic device with a hexagon structure and compared to experimental results in both normal gravity (1g) and microgravity (0g) via Blue Origin's New Shepard Vehicle (NS-23 attempt and NS-24 launch). CFD results have been created to predict the transport character of nutrients in the bones. These insights have the potential to lead to preventative measures for osteoporosis in astronauts.

Interleukin-1 (IL-1) is a multifunctional cytokine produced primarily by activated monocytes/macrophages and by a variety of other cell types. IL-1 plays an integral role in the immuno-inflammatory response of the body to a variety of stimuli including infection, trauma and other bodily injuries. Once IL-1 is released from the synthesizing cell, it acts as a hormone, initializing a variety of responses in different cells and tissues. These responses are believed to be crucial to survival and are termed acute-phase responses. NF-κB is a family of dimeric transcription factors that control the expression of hundreds of genes which regulate cellular stress responses, cell division, apoptosis, and inflammation. NF-κB dwells in the cytoplasm of the cell until activation in response to a wide range of extracellular stimuli including signaling molecules such as cytokines. NF-κB regulates transcription and gene expression through nucleocytoplasmic transport. Most previous studies on NF-κB activation have been performed using bulk assays to look at populations of cells. Determining cell variance at a single-cell level is crucial in understanding the full mechanisms of drug response. The goal of this study is to analyze the effects of variant concentrations of IL-1β on the activation of NF-κB in individual cells through use of a microfluidic gradient generator. The gradient generator was adopted from Jeon et al and used principles of diffusive mixing and splitting of flows in order create a solute concentration gradient. A soft lithography procedure was used. Briefly, the design was printed on a transparency using a high resolution printer. A master of the design is then created using an SU-8 photoresist and UV light to imprint the design on a silicon wafer. The master is then used to create a Polydimethylsiloxane (PDMS) mold of the design which can be irreversibly attached to a glass slide through oxidation in order to close off the microfluidic channels. FITC-conjugated β-Casein (a protein with similar molecular weight to IL-1β) was used in order to verify the gradient generated by the design. The concentration gradient was analyzed by measuring fluorescent intensity of images taken under a UV light microscope and found to agree with microfluidic simulations run on COMSOL. A procedure for culturing cells in a microfluidic device was then adapted from Jeon that is explained in detail in Chapter 3. Two main trends were revealed; firstly, as IL-1β concentration decreased, the percent of cells activated also decreased. Secondly, as IL-1β concentration decreased, the activation time of the responding cells increased. Cells were observed to act in a single-cell manner; in which multiple cells subjected to the same concentration would not all respond in the same fashion. No major activation threshold was observed but two minor thresholds were; the first at 0.02 ng/mL IL-1β where activation levels drop from 20% to around 5%. The second around 1 ng/mL, in which all greater concentrations show nearly complete activation of all cells exposed. Of the cells that activated, the activation times were recorded and analyzed as well. In general, a decrease in IL-1β concentration caused cells to take longer to activate. Concentrations greater than 5 ng/mL responded on average in 30 minutes with a significant amount of variation. Between 5 ng/mL and 0.1 ng/mL, activation time increased as IL-1β concentration decreased in a linear fashion when concentration was plotted on a base-10 log scale. Below 0.1 ng/mL, the trend disappears and an average activation time of around 95 minutes is observed that no longer depended on concentration. This is interesting because fewer and fewer cells are activating in this concentration range but activation time follows no trend and remains partially stochastic with times ranging from 80 to 105 minutes. The previous results were all observed with a continuous flow and stimulation of the cells. Experiments were also run by only exposing the cells to the IL-1β for 10 minutes and then replacing the flow with a buffer. These studies yielded interesting results; the fraction of activated cells reported the same trends and values as those that were continuously stimulated. The activation times, however, were delayed between 10 and 20 minutes but otherwise followed the same trend as the continuous stimulation. These results suggest that a brief exposure to an external stimulant is all it takes for the cascade of intercellular events to take place and cause NF-κB translocation. / Master of Science

The advancement of cancer immunotherapy, especially in the manipulation of NK cells, holds promise for targeted cancer treatment. NK cell effectiveness is currently assessed using cell populations in cytotoxicity assays, but these lack the details to observe individual cellular behaviours in real time. Droplet-based microfluidics is emerging as a solution to address these limitations by allowing the encapsulation of cells at specific ratios in controlled microenvironments. This advancement enhances the accuracy of immunotherapeutic assessments by providing a more detailed understanding of cellular interactions. In our study, we employed droplet microfluidics to encapsulate and analyze the interactions between NK cells and K562 cancer cells at predetermined effector-to-target (E:T) ratios. Each droplet served as an isolated microreactor, where individual NK cell interactions with cancer cells could be monitored in real-time. The results of our study revealed that droplet-based microfluidics provide detailed insights into the differential cytotoxic capacities of primary (Pri), suppressed (Supp), expanded (Exp), and post-expansion suppressed (PES) NK cells. Notably, expanded NK cells exhibited not only higher cytotoxic activity at a faster rate but also greater serial killing capabilities across different donors and varying E:T ratios, indicating their potential for effective immunotherapy. Additionally, suppressed NK cells showed reduced cytotoxic abilities, emphasizing the importance of overcoming the suppressive factors within the tumour microenvironment. These findings are pivotal for the field of immunotherapy and hold promising implications for the selection and optimization of NK cell-based treatments tailored to individual patient needs. / Thesis / Master of Applied Science (MASc)

Pathogen surveillance and monitoring is the first line of defense in avoiding diseases and adverse outcomes. Point of care (POC) diagnostic devices have made huge strides to achieve that, however, advancements are still required in order to expand the use of portable devices in environmental, food, and clinical diagnostics. In this work, we address critical challenges in biosensing and pathogen detection through three innovative approaches: (i) enhancing the understanding of the impact of nanostructures in DNA hybridization kinetics, (ii) developing a rapid real-time detection system for Legionella pneumophila using functional nucleic acids as biorecognition elements and DNA barcodes as detection barcodes, and (iii) applying biomimicry in microfluidic designs for uniform velocity and DNA hybridization in multiplexing. We first designed a wash and reagent free in situ electrochemical assay to investigate the role of planar and nanostructured surfaces on real-time DNA hybridization kinetics in buffer and complex media (blood, urine, and saliva). We then conducted continuous measurements to understand how these surface modifications influence electroactive DNA hybridization on the surface under a wide range of probe densities (low, medium, high) and target concentrations (0.01-1 µM). The results show that the effectiveness of nanostructures in enhancing electrochemical sensing depends on the probe/target concentration regime and the medium used in biosensing. Specifically, nanostructures were most beneficial in certain target concentration ranges (0.1-1 µM), with enhancing biosensing in all complex media compared to planar surfaces. We then utilized these nanostructures in engineering a rapid and accurate system for the detection of L. pneumophila in cooling tower water - a key factor in preventing Legionnaires' disease. To overcome the limitations of existing technologies (cell culture, enzyme-linked immunosorbent assay (ELISA), and polymerase chain reaction (PCR)), we designed an RNA-cleaving DNAzyme (RCD) electrochemical assay coupled with magnetic beads, fully housed within microfluidics. This system allows for real-time monitoring by programming RCDs to release an electroactive DNA barcode upon encountering L. pneumophila targets. The barcode is detected by an integrated sensor, achieving a limit of detection of 1.4 × 10³ CFU/mL in buffer and 1.9 × 10³ CFU/mL in cooling tower water in 3 hours. This system meets regulatory requirements and enables precise identification of L. pneumophila among other waterborne bacteria and L. non-pneumophila species. Finally, we leveraged biomimicry to design microchannel systems inspired by the efficient transport mechanisms found in human spinal vertebrae and leaf veins network. By replicating and scaling these natural structures, we developed the bio-inspired microfluidic designs that optimize flow uniformity and DNA capture in Silico. Our optimized designs achieved a coefficient of variation for flow velocity of 0.89% for spine-inspired and 0.86% for leaf veins-inspired microchannels compared to 14.68 % and 59.81 % for the unoptimized designs. Additionally, these designs were compared with a simple branched design for uniform DNA capture, using the kinetics parameters extracted from our first objective. The bio-inspired designs demonstrated high DNA capture uniformity, achieving stabilization up to 10 times faster under varying conditions than a simple branched design. Ultimately, this work offers significant advancements in optimizing three crucial aspects of POC diagnostics i) surface reaction kinetics, by studying and identifying the conditions best suited for planar and nanostructured surfaces in both buffer and complex media, ii) mass transport, by investigating flow effects on biorecognition and detection, and determining the optimal conditions for biosensing, ii) and electrochemical biosensing and microfluidics integration and design, by utilizing the optimized parameters for nanostructured surface and develop a rapid, continuous, and real-time microsystem for L. pneumophila detection meeting the regulatory standards. For the second generation of this microsystem, the two bio-inspired designs will enable multiplexed detection of various pathogens. These contributions collectively are pivotal to the development of next generation POC diagnostics, with broad applications in environmental, clinical, and food safety monitoring. / Dissertation / Doctor of Philosophy (PhD) / Point of care (POC) diagnostics are expected to improve the quality of healthcare by enabling early diagnostics, improved prognostics, and enhanced treatment selection and monitoring. To realize this, POC devices must be integrated, easy to use, sensitive, specific, and cost-effective. Despite research efforts a real-time continuous multiplexed system for bacterial detection is lacking. Therefore, this thesis addresses several key challenges in biosensing and real-time continuous pathogen detection by developing innovative approaches using nano engineering, RNA-cleaving DNAzymes, electrochemical microfluidic integration, and biomimetic microfluidic designs. We first explored the impact of surface structure on real-time DNA hybridization kinetics in complex media, identifying specific conditions under which nanostructures enhance sensitivity. Building on this, we developed a rapid, real-time electrochemical microfluidic system for detecting Legionella pneumophila, a dangerous pathogen found in fresh and potable water systems. Current systems either do not meet the required limit of detection or are limited to specific serotypes, precluding other pathogenic serotypes. The electrochemical microfluidic system performed highly sensitive detection across multiple serotypes, meeting regulatory standards and enabling real-time pathogen identification across a panel of other waterborne species, offering a continuous, real-time detection alternative to slow, traditional culture-based methods. The final objective was to draw inspiration from nature to design microchannels able to deliver uniform flow and molecules to the biosensing areas for multi-analyte detection in silico. Both inspired and optimized designs demonstrated great uniformity in DNA hybridization, confirming the hypothesis that these designs are inherently proficient in equal distribution. Together, these innovations contribute to the future of rapid, sensitive, and multiplexed POC diagnostic platforms.

This study explores the design, testing, and modeling of a bubble injector intended for use in studying bubble dynamics in molten salt reactors using a room temperature surrogate fluid by matching the Reynolds number, Eötvös number, and Morton number defined by the properties of the helium bubbles in the pump bowl of the Molten Salt Reactor Experiment (MSRE). The injector, constructed from polydimethylsiloxane (PDMS) and acrylic, was tested to generate bubbles within a precise size range suitable for simulating conditions in molten salt reactors. Experimental data showed that the equivalent bubble diameter is directly proportional to gas flow rate and inversely proportional to liquid flow rate, with clear trends emerging when data were subdivided into constant flow rate plots. The study applied and adapted the bubble size control model proposed by Lu et al. (2014), revealing limitations in existing models under modified conditions such as an elongated two-phase channel. A novel model was developed to better predict bubble size, incorporating dependencies on both flow rate ratios and the capillary number of the microchannels. The injector's design facilitates convenient modifications in channel geometry to achieve target bubble sizes, and future improvements in pressure monitoring and imaging are recommended. This work contributes to the advancement of microfluidic bubble injection technology. / Master of Science / This study focuses on developing and testing a device that creates tiny bubbles to help us better understand bubble behavior in advanced nuclear reactors, specifically molten salt reactors. These reactors use a special type of liquid fuel, and understanding how bubbles move within them is important for improving their efficiency and safety. To simulate the conditions inside these reactors without using the actual molten salt, we used a substitute fluid that has similar properties to the molten salt and built a bubble injector made from clear, flexible materials. Our experiments showed that the size of the bubbles depends on the flow rates of the gas and liquid: larger bubbles are formed when more gas is injected, and smaller bubbles are created when the liquid flow is increased. We tested existing models that predict bubble size and found that they didn't always work well under the conditions we used, like longer channels where the gas and liquid mix. As a result, we developed a new model that better predicts bubble size by considering both flow rates and the Capillary number of the microchannels. The design of our injector allows for easy adjustments to make bubbles of different sizes, and we suggest future improvements in pressure measurement and camera equipment to enhance data accuracy.

Optical manipulation covers a wide range of techniques to guide and trap cells using only the forces exerted by light. Another optical tool is photoporation, the technique of injecting membrane-impermeable molecules using light, which has become an important alternative to other injection techniques. Together they provided sterile tools for manipulation and molecule delivery at the single-cell level. In this thesis, the properties of low Reynolds fluid flows are exploited to guide cells though a femtosecond Bessel beam. This design allows for high-throughput optical injection of cells without the need to individually target cells. A method of 'off-chip' hydrodynamic focusing was evaluated and was found to confine 95.6% of the sample within a region which would receive a femtosecond dose compared to 20% without any hydrodynamic focusing. The system was tested using two cell lines to optically inject the membrane-impermeable dye, propidium iodide. This resulted in an increase of throughput by an order of magnitude compared to the previous microfluidic design (to up to 10 cells per second). Next optical trapping and photoporation were combined to create a multimodal workstation. The system provides 3D beam control using spatial light modulators integrated into a custom user interface. The efficiency of optical injection of adherent cells and trapping capabilities were tested. The development of the system provides the groundwork for exploration of the parameters required for photoporation of non-adherent cells. Finally optical trapping is combined with temporally focused multiphoton illumination for scanless imaging. The axial resolution of the system was measured using different microscope objectives before imaging cells stained with calcein. Both single and a pair of recently trypsinised cells were optically trapped and imaged. The position of the trapped cells was manipulated using a spatial light modulator in order to obtain a z-stack of images without adjusting the objective position.