PARP inhibitors are a revolutionary precision cancer therapy that inhibits PARP function and causes synthetic lethality in cells with homologous recombination deficiency (HRD) such as cells with BRCA1/2 loss. As such, BRCA1 presence plays a key role in PARP inhibitor sensitivity, providing a basis to predict treatment response based on BRCA1 expression in patient tumors. In addition to protein loss mutations, hypermethylation of BRCA1 gene may also result in the loss of BRCA1 expression, and subsequent HRD and potential PARP inhibitor sensitivity. The NRG-GY005 clinical trial focuses on the development of a clinically useful predictor of PARP inhibitor sensitivity/resistance to spare toxicities for patients unlikely to derive benefit. I will first characterize BRCA1 expression in randomized and blinded tumor samples as a biomarker for response to PARP inhibitors in patients. I will then relate BRCA1 protein expression to methylation status of BRCA1. Combining these two aims will describe BRCA1 function/presence in tumors, and better define overall homologous recombination deficiency to optimize patient-specific treatment between PARP inhibitors or other precision therapies. I conducted immunohistochemistry (IHC), a targeted staining with antibody MS110, towards BRCA1 to identify protein presence in tumor tissue and have compiled and analyzed droplet digital PCR (ddPCR), IHC, and clinical treatment reports. I am currently working through Aim 1 by performing IHC staining on tissues from the first phase of the NRG-GY005 clinical trial and will continue investigating whether a lack of BRCA1 expression is associated with cancer cells that are responsive to therapies. In combining IHC and ddPCR assays, we can compare BRCA1 presence with methylation to analyze tumor cases without BRCA1 expression and contribute toward identifying biomarkers to inform patient-specific treatment for individuals with recurrent ovarian cancer. I anticipate that cases lacking BRCA1 expression have an increased likelihood of hypermethylation in the tumor, causing loss of BRCA1.

Multiple Particle Tracking (MPT) is a powerful technique for studying the behavior of microscopic particles, such as viruses and nanoparticles, by tracking individual displacement and movement. One application of MPT is to measure microstructural changes in the brain extracellular environment (ECM) in development and aging, and in response to disease onset and progression. MPT of nanoparticle probes results in the generation of thousands of individual nanoparticle trajectories, from which geometric features, diffusion coefficients, and viscosities can be extracted. The vast array of trajectories contained within our dataset presents a good opportunity for integration into deep learning models that contains self-supervised learning, equivariant graph neural network, and Equivariant transformer. However, to enable MPT data to be trainable and predictable by deep learning models, we need to curate the data to be readable and useable by these models. To enable this, I have created a database and developed a data architecture that would allow MPT data to be passed into Deep learning models that use various techniques such as transformers. I am currently working on utilizing the data architecture on a Deep Learning model that uses transformers and self-supervised learning to predict trajectories of MPT particles. From this model, my expected accuracy of prediction of the trajectories for the MPT data is around 85%. This can allow us to learn complex features directly from raw MPT trajectory data, improve our predictions, and extract biological insights. The python package with our data architecture, the various SQL scripts, and the model will be provided as an open-source resource, allowing other researchers to expand upon my code and apply their unique modifications based on their own data and trajectories.

Hypoxic-Ischemic Encephalopathy (HIE) resulting from a lack of blood and oxygen to the brain is the leading cause of mortality in term newborns. Extracellular vesicles (EVs) serve as critical transporters of biomolecules between cells, with evidence of alleviating inflammation in models after hypoxic ischemia (HI) injury. Therapeutic efficacy of EVs has only been evaluated in males because males are more susceptible to worse outcomes following HIE injury, yet knowledge about EVs and their behavior when administered to females is still needed. In this study, I aimed to address this knowledge gap by systematically comparing the efficacy of male and female neonatal brain-derived EVs (mEVs, fEVs, respectively) applied on male and female neonatal rat ex vivo brain slices. I first confirmed the purity of isolated EVs with protein assays and immunoblots, and utilized an ex vivo oxygen-glucose deprivation (OGD) model of HI injury, applying fEVs and mEVs to sex-matched OGD-exposed brain slices. I evaluated cell viability after 24h of EV exposure, and my results show that fEVs decrease inflammation and cytotoxicity in OGD models. When compared to previous results using mEV treatment, my results suggest that females have a more robust anti-inflammatory response system to injury. Ongoing work to better understand the therapeutic effect of EVs involves further observing morphological shifts in microglia through confocal imaging, as fEV application will likely result in microglia shifting towards anti-inflammatory phenotypes, similar to what was previously observed after mEV application. I am also quantifying expression levels of various inflammatory and reparative genes through reverse transcription quantitative polymerase chain reactions (RT-qPCR). Overall, I have demonstrated in these pilot studies that fEVs have a different therapeutic effect in OGD injury compared to mEVs. This research is intended to open up pathways for more personalized sex-based treatments for various injuries and therapeutics in the future. 

Since approval by the FDA in 2019, BPaL/ BPaLM regimens are the favored treatment for multidrug-resistant tuberculosis (MDR-TB). Whereas traditional MDR-TB treatment generally includes an injectable and takes at least 18 months to complete, BPaL/ BPaLM regimens are all oral and typically take 26 weeks, suggesting potential for widespread implementation. Here, I aimed to elucidate patient demographic patterns and outcomes following treatment with BPaL/BPaLM to fill current gaps in knowledge. King County, Washington, with large populations of immigrants, refugees, and asylum seekers arriving from TB-endemic countries, sees high rates of TB incidence each year. Using medical data collected from active TB cases in King County, WA since 1993, I compiled demographics and outcomes (resolution of TB symptoms and treatment adherence) from patients receiving BPaL/BPaLM regimens. Of the 70 patients in our King County cohort, eight were given BPaL or BPaLM therapy to treat MDR-TB. At this time, five (63%) have completed therapy. Four (50%) are from Vietnam, two (25%) from China, one (13%) from Myanmar, and one (13%) is from the Russian Federation. Two (25%) are male; one (13%) was previously diagnosed with TB; and median age was 35 (range of 19-81). All eight had pulmonary disease and were HIV-negative. One patient was a household contact of another identified on contact tracing. Limitations of this analysis include a small BPaL/BPaLM cohort (n=8), and missing patient data. As more follow-up time accumulates, I can compare BPaL/BPaLM to the prior 18-month regimens using relapse rates and treatment completion. Based on my current analysis of King County MDR-TB cases, BPaL/BPaLM appears effective and well-tolerated—conducive to better TB outcomes than seen with prior MDR-TB regimens. Moreover, this study can provide data-driven insight to effectively treat MDR-TB patients from diverse populations across the US.

Impairments in memory formation are commonly observed during aging, even in the absence of disease-related neuropathology. However, we currently lack a comprehensive animal model of normative aging. Monkeys have a keen ability to remember pictures they have seen before, and this memory can be quantified through the tracking of eye movements. Previous research from the Buffalo Lab has shown that monkeys, like humans, show distinct patterns of eye movements when they first view an image, compared to when they view it during a second presentation. To investigate whether these behaviors are impacted in normative aging, I analyzed data from 6 young Macaca mulatta (4 female, 2 male, aged 7.5 ± 2.1 years) and 3 aged (female, aged 21.3 ± 2.5 years). In each block of trials, monkeys were shown 12-16 complex full-screen images on a computer monitor for 5-7 seconds of cumulative looking-time. If the monkey looked away from the screen, the picture remained on screen, but this time was not counted towards the cumulative looking requirement. Each behavioral session contained 5 blocks of trials, for a total of 60-80 unique images, each shown twice per session. I quantified attention as the total time the picture was onscreen compared to the required cumulative looking time, creating a ratio (or percent overage) for both novel and repeat images. Thus, the more time the monkey spent looking away from the image, the larger the ratio, representing less attention to the image. The data suggest that aged monkeys exhibit less attention to repeated images compared to novel images and less attention to repeated images than young monkeys. These differences in viewing behavior for repeated images may be indicative of age-related changes in memory processes, and this metric has the potential to better inform mechanisms of normative aging.

Hypoxic-ischemic encephalopathy (HIE), resulting from loss of oxygen and blood flow to the brain, remains a leading cause of death and disability in infants with no cure. Following the onset of HIE, inflammation and oxidative stress can drive ongoing injury in the newborn brain. Microglial and neuronal cell populations are promising therapeutic targets. However, drug delivery to the brain and into disease-mediating cells remains a challenge. Our prior work has demonstrated the ability of poly(ethylene glycol)-poly(lactic-co-glycolic acid) (PEG-PLGA) nanotherapeutics to overcome biological barriers in the brain. PLGA-PEG nanoparticles formulated with polysorbate 80 (P80) can further localize to microglia and neurons after systemic administration. We aimed to develop PLGA-PEG nanoparticles for cell-specific delivery of N-acetylcysteine (NAC), an anti-inflammatory agent, in neonatal hypoxia-ischemia (HI). We first varied formulation parameters to optimize a NAC-loaded PLGA-PEG/P80 nanoparticle platform. PLGA-PEG composition and surface-active agent concentration tuned particle size distribution, surface charge, and encapsulation efficiency characterized by dynamic light scattering and high-performance liquid chromatography. Leveraging an ex vivo rat brain slice model of neonatal HI, we investigated cellular uptake of fluorescently labeled nanoparticles. We observed particle localization in microglia and neurons, demonstrating cell-targeting ability following topical application to brain tissue. This work informs optimal particle design for delivering a therapeutically relevant dose of NAC, which is currently limited in clinical application due to unfavorable pharmacokinetic properties. Future experiments could apply the NAC nanoparticle platform using in vivo models to evaluate therapeutic potential for newborns with HIE.

Hypoxic ischemia (HI), the loss of blood and oxygen to the brain, is a common cause of neurological impairment and mortality. Astrocytes are one cell type that responds to acute trauma like HI and modulate the vascular-brain interface via their role in maintaining the blood-brain barrier (BBB). Therefore, astrocytes can be a potential therapeutic target; however, to screen methods to target astrocytes, there is still more to be discovered about astrocyte behavior in response to different stimuli. Towards this goal, this project aims to (1) create a robust and detailed characterization of cultured astrocytes over time, (2) evaluate astrocyte changes to different culturing conditions, and (3) measure the uptake of polymer nanoparticles, commonly used as drug delivery systems, on stimuli exposed astrocytes. My results show that after isolation, astrocytes are initially evenly distributed and form snowflake clusters, collectively assuming a star-shaped morphology. Over time, individual astrocytes move away from clusters and independently adopt the characteristic star shape. This suggests a dynamic process wherein astrocytes exhibit both collective and individual behaviors, contributing to the intricate architecture of astrocyte growth. Additionally, changes in the ratio of glial cells were observed. While microglia decreased in number and became less branched over time, oligodendrocyte populations remained relatively stable over time. Neurons that were initially sparse in the population decreased rapidly over time. Current studies involve the application of polymer nanoparticles to oxygen-glucose deprivation (OGD)-exposed cells immediately after OGD to evaluate uptake via imaging co-localization of particles with cells; OGD exposure induces HI injury in vitro. I have confirmed astrocyte response to OGD by analyzing cell morphology using imaging and cell viability assessments. These studies will establish an in vitro astrocyte model of HI and enable future studies incorporating additional BBB cells and other therapeutic platforms.

My research explores the metabolic flexibility and longevity of Rhodopseudomonas palustris (R. palustris). This alpha-proteobacterium has become a model organism for studying bacterial survival in non-growing states. R. palustris can endure long-term starvation in a growth-arrested state without forming dormant structures, prompting a comprehensive investigation of the molecular basis of growth-arrest and metabolic modes under these conditions. Recent studies have demonstrated that R. palustris can enter the growth-arrested state due to nutrient limitation but not energy limitation. R. palustris utilizes cyclic phosphorylation to generate ATP, allowing it to sustain viability for an extended period, even in the absence of nutrients including carbon and nitrogen. Earlier research examined the molecular response of R. palustris to growth arrest induced by carbon starvation under light and dark anaerobic conditions. Results indicated that light-incubated cells remained viable for months while dark-incubated cells exhibited a significant decrease in viability following growth arrest. The decline in viability was associated with ATP depletion, which underscores the critical role of ATP in R. palustris’s survival during growth arrest. To further investigate the versatile metabolism of R. palustris, we conducted anaerobic growth experiments using wild-type strain CGA009. We manipulated casamino acids concentrations in nitrogen-rich medium (PM) and nitrogen-free medium (NFM). Results revealed that R. palustris CGA009 utilizes casamino acids as both carbon and nitrogen sources. Our experiments also confirmed that R. palustris CGA009 can grow in the amino acid L-Leucine. Currently, we are researching the capacity of R. palustris CGA009 to utilize diverse carbon substrates through aerobic and anaerobic cultivation on Gelrite medium. Distinct growth patterns provided insights into specific concentrations of carbon substrates tolerated by R. palustris. This ongoing research aims to identify additional carbon substrates supporting R. palustris’s growth, with implications for harnessing its unique metabolic capabilities and expanding our understanding of R. palustris’s metabolic versatility.

In the US, there is an average of 69,500 Traumatic Brain Injury (TBI) related deaths, 223,135 TBI-related hospitalizations, 326,600 inpatient stays, and 801,700 Emergency Department visits per year. The Centers for Disease Control report the annual cost of treating non-fatal TBIs to be over $40B. Currently, there is no pharmacological treatment for TBI, and 138 clinical treatment trials were completed since 2004 with a 100% failure rate. A rigorous screening model in vitro is needed to increase the probability of successful clinical trials. TBI is complex with many possible modalities of injury. The primary insult to brain tissue may result from compression or shear stress and strain, followed by swelling that compounds into the secondary insult. The cascade of TBI causes additional neuronal death and dysfunction to complicate injury and treatment further. The range of unknown potential injury to the brain during a TBI makes a single TBI model too simplistic to represent the full extent of injury accurately. I have developed a set of living-tissue organotypic whole hemisphere (OWH) brain slice models to mimic compressive damage with a whole slice and novel partial slice compression. The models simulate mild, moderate, and major TBI representing primary and secondary insult inflammation and cytotoxicity propagation across multiple brain regions. Future work will model shear strain damage and the neurochemical response to injury. This set of robust models will be used to screen treatments for TBI before in vivo and clinical trials to study how the compounds affect damaged tissues at a cellular and molecular level.