For people living with epilepsy (PWE), anti-seizure medicines (ASMs) are the primary treatment option. However, 30% of PWE are unable to control their seizures with ASMs because they have drug-resistant epilepsy (DRE). Pathological mechanisms that contribute to DRE are not currently understood. Nevertheless, both clinical and preclinical data indicate potential involvement of changes in the architecture of neuronal networks. I used a clinically relevant rat model of temporal lobe epilepsy, and novel medication in food delivery system to confirm DRE, or failure to reduce their baseline seizure frequency by 50% with two or more clinically used ASMs. I hypothesized that the DRE animals would have lowered neuronal cell density compared to the those with drug-sensitive epilepsy (DSE). All rats were euthanized at the end of a 6-week treatment period to process the brains for immunohistochemical labeling of mature neurons with the antibody, NeuN. Total percent staining area was quantified in the hippocampus, piriform cortex, and somatosensory cortex of brains. No differences in NeuN immunoreactivity were observed between DSE and DRE animals in any brain region. However, NeuN levels in animals with epilepsy, regardless of treatment outcome, trended lower than naïve animals without epilepsy in the CA1 and dentate gyrus regions of the hippocampus. Together, these data suggest that neuron density may not be driving pharmacoresistance. However, it is possible that the ratio between excitatory and inhibitory neurons may be disrupted in DRE. This underscores the need for future studies to quantify neuronal subtypes, providing a more nuanced understanding of the underlying mechanisms of pharmacoresistance. These studies play a crucial role in guiding future research into novel treatments designed for DRE.

Androgen deprivation therapy (ADT), the pharmacologic reduction of testosterone (T), is a critical treatment for prostate cancer patients, improving cancer-related outcomes but markedly increasing the risk of cardiovascular disease development. Our recent findings suggest that patients with prostate cancer subjected to ADT have increased hypothalamic gliosis (the activation of astrocytes and microglia), a hallmark of central nervous system injury. Similarly, castrated mice fed a high-fat, high-sucrose with added cholesterol diet (HFHS) develop hypothalamic astrogliosis and atherosclerosis, which can be prevented through T replacement at the onset of castration. In this experiment, we tested whether established astrogliosis in hypogonadal mice can be reversed by restoring healthy T levels. Using a gonadotropin-releasing hormone antagonist (acyline) as ADT, we treated 3 groups of HFHS-fed C57Bl/6 wild-type mice with: 1) vehicle, 2) acyline for 4 weeks, and 3) acyline for 4 weeks followed by a 4 week period of reversal to vehicle. Hypothalamic brain sections were used for immunohistochemistry analysis to quantify levels of glial fibrillary acidic protein (GFAP) expression, a marker for astrogliosis, and neurokinin B (NKB) expression, a marker for reproductive function. As anticipated, ADT resulted in a significant reduction in testes weight and NKB expression, serving as surrogate measurements of low T production. ADT also induced a significant increase in hypothalamic GFAP expression, confirming that gliosis is exacerbated in hypogonadal conditions. Following the discontinuation of ADT, testes weight and hypothalamic NKB expression rebounded to normal levels, however, hypothalamic astrogliosis remained significantly elevated. Together these data suggest that once astrogliosis is established, the restoration of T via 1 month of ADT cessation is insufficient to reverse it. This finding is a critical step forward in clarifying the pathways between androgen signaling and cardiometabolic regulation, and raises an important clinical concern given the high incidence of morbidity and mortality associated with ADT.

Chimeric Antigen Receptor T (CAR T) Cells are receptor proteins that can be modified to allow T cells to target specific antigens. CAR T therapy has shown promise in preclinical experiments in patients with solid tumors, such as glioblastoma, however limitations in distribution of CAR T cells in the brain limit the effectiveness of this treatment. Most commonly, CAR T cells are administered intraventricularly through a surgically implanted device and allowed to diffuse throughout the cerebrospinal fluid filled cavities and pathways to reach the tumor. However, there is little evidence supporting the effectiveness of current methods of administration, potentially due to a lack of target engagement. We hypothesize that the glymphatic system of the brain could be used to optimize delivery of CAR T cells to solid tumors. The glymphatic system is a network of perivascular pathways that facilitates the anatomically distinct movement of cerebrospinal fluid (CSF) into the interstitium of the brain, helps distribute solutes such as glucose, lipids, and neurotransmitters, and serves as a solute clearance system in the brain. Physiologically, glymphatic function is mediated by different factors such as arterial pulsation, vasomotion, and heart rate – which can be manipulated with anesthetics, pharmaceuticals, or even sleep. We proposed exploration of the difference in parenchymal distribution of CAR T cells when injected in the cisterna magna versus intraventricularly in mice. Non-tumor bearing mice were injected via the cisterna magna or intraventricularly with fluorescently labelled CAR T cells. We observed that at 1-, 4-, and 24-hours post-injection, CAR T cells were localized in the sub-ventricular regions similarly, regardless of injection site. In follow-up experiments, we will employ the same technique in tumor bearing mice, with and without pharmacological intervention to define the effect of glymphatic function on distribution and effectiveness of CAR T cells.

Alzheimer’s disease (AD) is the most common neurodegenerative disease, with over 6 million Americans suffering from the illness and prevalence increasing each year. My work was conducted as part of an NIH-funded multi-institutional network called TREAT-AD (TaRget Enablement to Accelerate Therapy Development for AD) that aims to find potential therapies for AD. The bioinformatics core of this network identified genetic targets of interest using RNA-sequencing and proteomic analysis of post-mortem tissue from participants with AD. We hypothesized that manipulating expression of these target genes in a relevant human model would influence levels of AD-related biomarkers. To manipulate genetic expression efficiently, I used shRNA technology in human induced pluripotent stem cell (hiPSC) derived neurons. I then analyzed these hiPSC-derived neurons for AD-relevant readouts, such as soluble amyloid beta secretion and intracellular phosphorylation of Tau protein, relevant to the two main neuropathological hallmarks of AD. I ran quantitative polymerase chain reactions (qPCR) to measure neuronal expression levels of each gene target, and compared amyloid beta and phosphorylated tau outputs to control samples using MSD ELISA assays. I found four gene targets that have substantial neuronal expression and found that each affected AD-related output levels when gene expression was knocked down with shRNA. My findings provide direct molecular genetic evidence that links these genes to AD pathways, suggesting that these genes could serve as promising targets for therapeutic development.

The ability to spatially pattern three-dimensional culture opens new avenues for modelling the heterogeneity of physiologically relevant tissue models, allowing for investigation of complex cell signaling questions. Our lab has developed a method for generating suspended tissues with spatial patterning using open microfluidic principles that we call suspended tissue open microfluidic patterning (STOMP) to improve the opportunities in patterning specific, multi-region suspended engineered tissue for biomedical applications. Within our patterning setup, open microfluidic channels allow hydrogel precursor to flow via spontaneous capillary flow, allowing any material that transitions from a fluid to gel-state (including native extracellular matrix such as collagen and fibrin), to be patterned in three dimensions with a standard pipette. Patterning devices guide the flow throughout the device, enabling us to create interesting, biologically relevant geometries as determined by the location of the channels. Additionally, pinning ridges can be implemented to stop the flow of gel through the device and create multiple regions. The multiple regions divide different cocultures within one contiguous tissue. My research project aims to pattern distinct regions of 3D cell culture within one suspended tissue by creating a multi-region STOMP device which fits within a well plate, commonly used in cell culture. The tissues can be patterned in a two region coculture using IMR-90 lung fibroblast cells, human umbilical vein endothelial cells, and A549 adenocarcinoma human alveolar basal epithelial cells. By patterning the cells in different regions, we can generate a model for angiogenesis and cell signaling, whereby the adenocarcinoma cells patterned in one region can signal for angiogenesis from the endothelial cells patterned in another region. This research project develops a new mesoscale co-culture method to enable precise regional positioning of cells to investigate cell signaling for biomedical applications in an affordable and accessible way.

Soluble factor signaling between immune cells and fibroblasts is critical in regulating biological processes. However, it is often dysregulated in diseases and leads to physiological changes, including airway inflammation in asthma and allergies. One immune cell type that can be attributed to airway inflammation is eosinophils (EOS). When activated by interleukin-3 and heat-aggregated immunoglobulin G, EOS release certain soluble factors associated with the activation of lung fibroblasts. To investigate the interactions between human lung fibroblasts (HLFs) and EOS, we used the open microfluidic coculture device. This device has two chambers, in which two types of cells can be cocultured in the shared media while being physically separated by a half wall. We found that HLFs in coculture with activated EOS had the highest levels of proinflammatory gene expressions and proinflammatory cytokines. However, the exact mediators responsible for promoting these biological processes are still uncertain. We hypothesize that EOS secrete a protein, transforming growth factor alpha (TGFa), to be consumed by HLFs, triggering proinflammatory responses of HLFs. The goal of this study was to elucidate the roles of TGFa in airway inflammation. HLF-EOS cocultures are seeded in the microfluidic coculture device, then TGFa and their respective cellular receptors are neutralized using antibodies. Then, reverse transcription quantitative-polymerase chain reactions are used to quantify gene expression levels relevant to proinflammatory responses of HLFs, in addition to multiplex immunoassays to analyze the secreted soluble factors from both cell types. We anticipate that HLF-EOS cocultures treated with neutralizing antibodies have lower expression levels of proinflammatory genes than cocultures without antibodies. Findings from this study will help us better understand the key regulators that promote proinflammatory behaviors of HLFs in airway inflammation.