Despite recent advancements in cancer treatment, the 5-year survival rate for glioblastoma is only 22%. The greatest challenge in brain cancer treatment is the blood-brain barrier (BBB), a physical barrier that protects the central nervous system (CNS) from circulating solutes in the blood, but in cancer, prevents therapeutics from entering the brain space. While surgery is the gold-standard treatment, this procedure is high-risk. Thus, we are investigating injectable therapies to treat brain tumors by crossing the BBB. To do this, we are developing nanoparticles (NPs) to cross the BBB via receptor-mediated transcytosis (RMT). In RMT, ligands (keys) bind to receptors (locks) expressed by the BBB to gain entrance into the CNS. By attaching the ligand transferrin, we can trick the BBB into granting our NPs access to the brain. Prior to testing particles in vivo, I am developing a cellular model of the BBB to assess the ability of different NP formulations to cross the BBB in vitro. The transwell model comprises two main compartments: the donor and acceptor compartments, separated by a cellular monolayer that only allows transport between compartments via RMT. Here, I developed a co-culture consisting of brain endothelial cells and astrocytes. This co-culture model is more representative of the BBB and provides higher monolayer integrity than single-cell models. In parallel, I am investigating the use of acidification inhibitors to enhance the ability of NPs to cross the BBB. If successful, these inhibitors will be incorporated into a new NP design. Through this project, I am (i) developing a more representative in vitro model of the BBB and (ii) exploring alternative mechanisms to enhance NP transport through the BBB. Ultimately, these two aims will enable us to better direct NP behavior in vivo and cross the BBB.

 Stroke is the leading cause of long-term disability in the United States. Disabilities can range from a loss of sensory function like touch to motor functions like controlling arm movements due to the damage to the brain’s network. Despite the prevalence of stroke, the underlying network dynamics that lead to functional deficits are not well understood. Due to the physiological similarities between non-human primate (NHP) and human brains, an NHP model is essential for studying the effects of stroke and developing therapies. Here we used the photothrombotic (PT) stroke technique to study network dynamics in the NHP sensorimotor cortex following an ischemic lesion. Using the PT stroke technique, we induced a focal ischemic lesion on the NHP sensorimotor cortex. We collected local field potentials from both hemispheres using an electrocorticographic  array on the cortical surface. As a measure of neural activity, we calculated ipsilesional power in the low gamma band. Channels were then organized into three clusters based on their net change in power (increase, decrease, no change). As expected, the cluster with an overall decrease in power corresponded to the lesion's physical location. We also studied the network connectivity by calculating pair-wise coherence across different frequency bands: theta, beta, low gamma, and high gamma. Overall, we saw that low frequencies were associated with decreases in coherence, while higher frequencies were associated with increases following stroke. Preliminary results from the contralesional hemisphere show similar changes. In this study, we observed local neurophysiological changes up to three hours following an ischemic lesion. The observed increases in power in the perilesional region and coherence at high frequencies suggest compensatory mechanisms immediately following an injury. We can use this study's results to guide future developments in stimulation-based therapy to alleviate the functional deficits from a stroke.

Though renewed efforts in tuberculosis (TB) research have facilitated massive strides towards treating Mycobacterium tuberculosis (M.tb), TB remains a global health problem with an estimated 10 million infections and 1.4 million deaths in 2019. The ability of the pathogen to hide itself inside a granuloma, a mass of immune cells whose precise mechanism of regulation is unknown, prevents both the study of M.tb pathogenesis and the development of drugs or treatments. Current in vivo models have been established to study TB infection using animal models or tissues, limiting its biological relevance as it relates to human disease while current in vitro models lack components of the complex lung microenvironment during infection. We present the creation of a novel microscale infection model, which uses open and suspended microfluidic principles to enable spatial and temporal manipulation of cultures in suspended hydrogel plugs. By stacking together two devices, we demonstrate the ability of a suspended model granuloma consisting of M.bovis BCG (Mycobacterium bovis bacille Calmette-Guérin) and monocyte-derived macrophages to interact with a model vasculature layer consisting of endothelial cells. Analysis of soluble factors for proinflammatory cytokines and characterization of infection-dependent angiogenesis in the vasculature layer are used to verify communication between cultures. In the future, we envision this model expanding to contain multiple immune cell types and to incorporate additional aspects of the lung anatomy to approach a more accurate pathophysiological model as a tool for other researchers.

Glioblastoma (GBM) is a cancer originating in glial cells in the brain that accounts for more than 60% of all brain tumors in adults. The low survival rate can be attributed to high resistance to radiotherapy due to the hypoxic tumor environment which induces signaling networks in cancer cells that lead to the epithelial to mesenchymal transition (EMT). EMT gives rise to mesenchymal cancer stem cells (MSC) with a highly invasive phenotype which resists traditional means of therapy. Phospholipid glutathione peroxidase (GPX4), a selenocysteine-containing enzyme that dissipates lipid peroxides, has been shown to regulate pathways that prevent ferroptosis, a unique iron dependent form of cell death initiated by an increase in reactive oxygen species. Disrupting the GPX4 pathway by siRNA-induced gene knockdown induces ferroptosis. Therefore, NPs as a vector for gene therapy may be able to eliminate mesenchymal state stem cells for a more effective treatment. Using hypoxia to induce EMT to develop a cell model for this work, preliminary results from quantitative real time PCR showed a correlation between GPX4 and EMT markers of human glioblastoma cells in hypoxia. GPX4 siRNA were evaluated using commercially available transfection agents on hypoxic and normoxic cells as proof of concept in vitro over a period of ten days. NP mediated delivery of validated siRNA were optimized using different ratios of NP and siRNA. Incubation time was also optimized. Finally, dual therapy of siRNA knockdown and radiotherapy were performed to evaluate sensitization of cells. The capabilities of NPs, along with concurrent radiation therapy, may provide a means to overcome radioresistance in GBM therapy.

Rapid behavioral procedures to identify hearing-aid users’ preferred amplification settings and to efficiently explore a wide range of device configurations may provide opportunities for delivering individualized hearing-aid fitting remotely and over-the-counter. A behavioral procedure was developed to allow a user to simultaneously make preference-based adjustments to the gains in six frequency bands (i.e. the gain profile) using a two-dimensional control surface. On each trial, the user moves a mouse cursor, which updates the gain profile in real time, to identify a location on the surface with most preferred sound quality. The map from the control to the gain profile is optimized iteratively following each trial using a machine-learning algorithm, which maximizes the anticipated information gain from the following trial. Monte-Carlo simulations were conducted to provide initial validations for the machine-learning algorithm. A large cohort of simulated listeners with individually drawn random preferred gain profiles performed the proposed gain-adjustment procedure with 100 trials each. The errors between the estimated preferred gain profiles and the ground-truth values decreased as the number of trials increased. Furthermore, the procedure implemented with the machine-learning algorithm, which iteratively optimized the control-to-gain maps, demonstrated advantages over alternative procedures with either randomly determined or pre-determined control-to-gain maps.This procedure can provide hearing aid users the ability to adjust their device to their preference without the need of an audiologist. 

Due to COVID-19, people are becoming more concerned about what they come into contact with on a daily basis. The development of low-cost, portable diagnostic technologies would be an important step towards enabling more ubiquitous sampling of relevant environmental factors, such as pathogens. Thus, we seek to use nanopore sensors, which are a simple class of molecular detection devices, to detect disease-causing agents. We will try to classify bacteria and viruses that are found in environmental samples. To do this, we will use Oxford Nanopore’s MinION sequencing device and a library of Nanopore-addressable protein Tags Engineered as Reporters (NanoporeTERs), together with a DNA computing-based approach, to classify samples of multiple disease states. The library of NanoporeTERs can theoretically be designed and read by the nanopore device to allow for the classification of multiple disease states simultaneously. We do this by associating a distinct barcode with a specific disease biomarker. We will analyze the resulting nanopore data to quickly and easily determine whether specific disease markers were present in the environmental sample. The nanopore assay can be designed to provide nearly real-time readout and multiplexed output to have broad use in disease detection or common use in day-to-day life. After initial proof-of-concept testing, our goal is to further expand the nanopore-based reporter library to detect more diseases, as well as port the library to mobile platforms to make it accessible to a wider range of users in the future.

Secondary spinal cord injury occurs after an initial trauma to the spinal cord and can last from several hours to days. It causes further damage to surrounding healthy tissue due to the activation of pro-inflammatory pathways. There have been studies on potential pharmaceutical therapies that target mechanisms in the pro-inflammatory pathway, but many were associated with severe health complications or require a high dose regimen that results in severe off-target side effects. Thus, there is a need for a drug delivery system that can control drug release temporally and spatially for local treatment of inflammation post-spine surgery. I investigated the effects of electrically controlled dexamethasone phosphate, an anti-inflammatory steroid, release on inflamed cells grown in vitro using our polypyrrole (Ppy)-coated microneedle array. The microneedle device was developed using 3D printing technology at the Washington Nanofabrication Facility and designed to puncture the dura mater, the thick membrane that surrounds the spinal cord and brain. The device was coated with the biocompatible conductive polymer (Ppy) to encapsulate the steroid, dexamethasone phosphate. The Ppy coating underwent cyclic electrical stimulation to release the drug. Concentrations of nitric oxide, pro-inflammatory cytokines (TNF-a, MCP-1, IL-1b and IL-6) and cell count were measured to observe the degree of inflammation in the cells. Inflammation reduction was observed with a significant decrease in cell count, nitric oxide, TNF-a, MCP-1 and IL-6 concentrations in the group that was treated with dexamethasone phosphate using the microneedle device compared to the negative controls. Results demonstrate the feasibility of the microneedle device in reducing inflammation caused by secondary spinal cord injury and future work will be directed toward the in vivo release of dexamethasone phosphate using the microneedle device.