Permanent URI for this collection
Click here to search the contents of this collection
Full text for dissertations and theses included in this collection dates back to 2011. For older dissertations, check the library’s One Search catalog or ProQuest Dissertations and Theses.
Administrator note: Records with locked files are embargoed for up to one year. To request a copy or learn of the specific embargo date, please request a copy of the file directly on the site or email ArchiveHelp@hshsl.umaryland.edu for more information.
Browse
Recent Submissions
Item Spatial atlas uncovers a unique proteomic and transcriptomic landscape at the human gingival barrier(2024) Theofilou, Vasileios Ionas; Moutsopoulos, Niki M.; Schneider, AbrahamThe oral mucosa is a unique tissue microenvironment, constantly exposed to environmental stimuli, including a rich and diverse community of commensal microbiota, second only to the small intestine. Yet, while host-microbiome interactions have been much studied in other barrier sites, how the local epithelial barrier and immune cell network maintain the balance between microbes and host is not well understood at this mucosal interface. Importantly, an imbalance between microbiota and host leads to a common oral inflammatory disease, periodontitis. A recent human single cell atlas of oral mucosal tissues has provided important insights into the cell populations and states within the oral mucosa in health and periodontitis but lacked spatial distribution of cell subsets. Nonetheless, we expect that location-specific functions of distinct cell subsets will be important both in tissue homeostasis and in the pathogenesis of periodontitis. Hence, our current project aimed to generate a spatial atlas of human oral mucosa in health and severe/untreated periodontitis, using single cell resolution 450-plex spatial transcriptomics (10x Genomics, Xenium), and 18-plex spatial proteomics (iterative bleaching extends multiplexity, IBEX) platforms, which we optimized for oral tissues. Our work provides spatial context for the diverse epithelial, vascular, stromal, and immune cell populations at this mucosa, and defines specialized niches within the oral tissue microenvironment. Within the epithelium, we focus on two distinct niches. The gingival (oral) epithelium displays marker expression consistent with structural integrity and innate defenses, as well as a myeloid-cell rich immune microenvironment. In contrast, the tooth-associated crevicular epithelium serves as a permeable barrier site with functionality indicative of antimicrobial defenses and active neutrophil recruitment. In the crevicular subepithelial space we document unique lymphoid structures which become expanded and obtain features consistent with tertiary lymphoid aggregates in the setting of periodontitis. Within these formations we find a zonation of T cells, APCs, B cells and immune-activated fibroblasts, with presence of plasma cells deeper in the tissue, in the setting of periodontal disease. Collectively, our work compiles a spatial atlas of gingival tissues, uncovering unique structural niches with inferred functionality linked to maintenance of tissue integrity, and to the onset of inflammatory pathology.Item Establishment of a Highly Prognostic Hypoxic Memory Metagene Signature in Luminal Breast Cancer(2024) Moriarty, Aidan; Yu, MinMetastasis is the main cause of tumor-related mortality in breast cancer patients; therefore, advancing our understanding of the metastatic cascade is imperative to improve patient outcomes. Prior research indicates cancer cells with increased metastatic potential have acquired adaptive properties induced by both tumor intrinsic factors and extrinsic factors in the tumor microenvironment (TME), including hypoxia. The cellular mechanisms that enable adaption and survival to low oxygen during normal developmental processes are often hijacked by cancer cells as their rapid proliferation and irregular angiogenesis produce regions of hypoxia within the TME, which has been extensively examined in its association with tumor progression. However, previous hypoxia research has been primarily focused on signaling changes that occur when cancer cells are in the hypoxic TME and does not account for long-term changes that may occur once cancer cells re-enter normoxia, such as when they enter circulation as circulating tumor cells (CTC) and metastasize. In this study, we analyze the lasting effects of hypoxia in luminal breast cancer cells and investigate their clinical impact. RNA-Sequencing of luminal breast cancer cell lines indicates a broad transcriptional hypoxic memory is maintained in post-hypoxic cells. Functional assays to measure well-established hypoxic-mediated phenotypes including metabolic reprogramming, proliferation, migration, chemotherapy resistance, and CTC clustering reveals that these post-hypoxic cells also maintain certain pro-metastatic phenotypes after reoxygenation. Additionally, we established a highly prognostic hypoxic memory metagene signature that was validated in vivo. This hypoxic memory signature is associated with more aggressive tumor characteristics in estrogen receptor-positive (ER+) patients and has greater prognostic potential compared to hypoxic-specific metagene signatures. Moreover, relapse free survival data suggests the hypoxic memory signature may be predictive of patient responses to chemotherapy in ER+ patients, which was further supported by the observed resistance to capecitabine in luminal breast cancer cell lines in vitro. This hypoxic memory signature could provide a novel biomarker for long-term hypoxic burden and patient outcomes in ER+ breast cancer patients.Item Factors Impacting Physical Function in Adults with Fibromyalgia(2024) Lashley, Heather; Resnick, BarbaraBackground: Resilience has been identified as a potential contributor to improved physical function in individuals with fibromyalgia (FM). However, research remains sparse regarding many physical, psychological, and socio-demographic factors impacting physical function in adults with FM and the direct and indirect impact of these variables on resilience and physical function. Purpose: The purpose of this dissertation is to (1) employ a scoping review to better define the concept of physical function, explore measures that quantify physical function, and identify factors associated with physical function; (2) examine the psychometric properties of the CD-RISC-10 in an adult fibromyalgia population, and (3) test a structural equation model of factors (including age, race, sex, comorbidities, pain intensity, work status, resilience, and depression) that may impact physical function. Methods: The scoping review used the Joanna Briggs Institute (JBI) method. A survey was distributed through the Autoimmune Registry to 200 adults with fibromyalgia. Rasch and differential item functioning analyses were used to examine the psychometric properties of the CD-RISC-10. Finally, a structural equation model was constructed to test the relationships between selected factors (age, sex, race, comorbidities, work status, pain intensity, resilience, and depression) and their impact on physical function. Results: The scoping review findings revealed resilience, depression, and multi-morbidity were associated with physical function. The Rasch analyses found the CD-RISC-10 to be a reliable and valid measure of resilience in a national sample of adults with FM. Finally, a revised structural equation model of the factors associated with physical function supported the significant direct impact of resilience on depression, pain intensity, and physical function. Resilience was also found to impact physical function indirectly through pain intensity. Age and comorbidities were indirectly associated with physical function through resilience. Conclusions: This work provided information about several social, physical, and psychological factors that were associated with physical function. There was additional support for the reliability and validity of the CD-RISC-10 measure. Further research on factors associated with physical function using psychometrically validated measures of resilience ensures more accurate measurement of critical attributes and informs interventions to optimize resilience and promote physical function.Item Molecular Strategies to Combat Acquired Resistance to the Anti-Leukemic Drug Venetoclax(2024) Goodis, Christopher; Fletcher, StevenA key hallmark of cancer is the ability to manipulate homeostatic pathways within cells to promote tumor growth. A significant example of this is intrinsic apoptotic pathway, which is regulated by members of the B-cell lymphoma 2 (BCL-2) protein family, that comprises two sub-populations of anti-apoptotic proteins (i.e. BCL-2, MCL-1, BCL-xL, BFL-1 and BCL-w) and pro-apoptotic proteins (which are further divided into BH3-only proteins, such as BIM and PUMA, and effector proteins, e.g. BAK and BAX). Under normal conditions, these proteins exist in similar ratios, interacting with one another via an a-helical BH3 “death” domain on the pro-apoptotic protein and a conserved BH3-binding groove on the surface of the anti-apoptotic proteins. In this state of affairs, homodimerization of the effector proteins is not possible, and apoptosis is prevented. However, when stress occurs in the cell, pro-apoptotic proteins are upregulated. This shift further activates BAK and BAX causing apoptosis. In BCL-2-dependent cancers, one or several of the anti-apoptotic BCL-2 proteins are overexpressed resulting in the sequestration and neutralization of pro-apoptotic proteins bound, disabling the ability to induce apoptosis. Inhibitors seeking to restore apoptotic function are designed to target the BH3 binding groove by mimicking the BH3 “death” domain. These inhibitors, dubbed “BH3 mimetics”, function by masquerading as pro-apoptotic BCL-2 proteins, which themselves become bound by the anti-apoptotic proteins thereby releasing the bona fide pro-apoptotic proteins, and restoring apoptosis in cancerous cells. A breakthrough utilizing BH3 mimicry came with the development of the BCL-2 selective inhibitor venetoclax (VEN). Gaining FDA approval in 2016 for patients with chronic lymphocytic leukemia (CLL), and then later for acute myeloid leukemia (AML), VEN has cemented itself as the current standard-of-care drug for treating these diseases. Unfortunately, patients administered VEN develop resistance to the therapeutic after ~17 months of treatment. While these resistances can manifest in multiple ways, the focus of this dissertation was to tackle the compensatory upregulations of sister anti-apoptotic BCL-2 proteins, particularly BCL-xL, MCL-1, and BFL-1. However, targeting these sister anti-apoptotic proteins presents their own challenges. Direct inhibition of BCL-xL causes thrombocytopenia, while multiple clinical trials for MCL-1 inhibitors have been halted or terminated due to cardiotoxicity, leading to the deaths of at least two patients. It has been stated that BFL-1 is the underdog of the BCL-2 family, having received comparatively far less attention than its sister proteins towards the discovery of new anti-cancer drugs. Although herein we have not identified any potent BFL-1 inhibitors, we anticipate our foundational work will provide a springboard for future work.Item Human Claustrum Function in Cognitive Control Brain Network Recruitment in Health and Disease(2025) Stewart, Brent; Seminowicz, David A.Cognitive control, the coordination of executive functions including working memory, response inhibition, error monitoring, and decision-making, is critical for navigating human life, and cognitive control impairment is a feature of myriad debilitating conditions. Cognitive control is hypothesized to arise from interactions among large-scale brain networks, but mechanisms underlying network dynamics are largely unknown. Accumulating functional studies in model animals suggest the claustrum, a subcortical brain structure with unparalleled cortical connectivity, supports the initiation of multiple brain networks. However, few tests of human claustrum function have been performed despite the neuroimaging field’s established network analysis tools and the opportunity to incorporate clinical samples. A potential barrier to acquiring valid claustrum BOLD signal with standard MRI hardware, partial volume effects, was addressed, and a novel method to protect against partial volume effects in task fMRI was developed. In publicly available datasets of healthy participants, increased claustrum BOLD signal was associated with increased brain-wide network activity in multiple memory tasks evoking a wide range of network states. Structural and effective connectivity findings were consistent with excitatory claustrum projections to recruited cognitive control network regions, and results distinguished claustrum from other regions implicated in cognitive control network mechanisms. Comparisons between healthy participants and patients with chronic pain, which is associated with cognitive control deficits and altered network activity, observed claustrum activation at the onset of a difficult cognitive task and acutely painful stimuli, with patients exhibiting increased right claustrum activity in both conditions. Patients recruited an additional, pain-sensitive cortical region during cognitive task processing, and structural and effective connectivity analyses were consistent with increased excitatory influence of right claustrum on this region in patients than controls. The collected results provide empirical support for a claustrum role in cognitive control network initiation and raise the possibility of the claustrum as a future therapeutic target for neuropsychiatric disorders characterized by cognitive impairment and cortical network abnormalities.Item Differential Impact of Conventional and Novel Risk Factors on Early and Late Onset Stroke(2024) Nguyen, Kevin; Mitchell, Braxton D; Kittner, Steven JBackground: Ischemic stroke (IS) is the leading cause of disability and the 4th leading cause of death in the US. It is a multifactorial disease with genetic and environmental components that influences an individual’s risk of stroke. Thanks to increased awareness of stroke risk factors as well as advancement in intervention and treatment, there has been a substantial fall in stroke incidence and improvement in patient outcome in high-income countries. However, numerous studies have highlighted a divergent temporal trend among young adults; stroke incidence rate among young adults has steadily risen thus far in the 21st century. In tandem to this concerning trend, young adults are experiencing a higher prevalence of conventional stroke risk factors due to sedentary lifestyle and other behavioral changes. Objective The goals of this study were to compare and contrast the impact of conventional and novel risk factors on early (EOS, onset 18-59 years) and late (LOS, onset > 60 years) onset stroke. This study has two specific aims: 1) To demonstrate and compare causal associations of five modifiable stroke risk factors (blood pressure, body mass index, type 2 diabetes, hyperlipidemia, and smoking) within EOS and LOS. 2) To assess causality between higher genetic predisposition of inflammation to risk of stroke and identify potential differential effects in EOS and LOS. Methods: Two-sample MR design was employed to assess the causal relationships between various exposures (previously observed stroke risk factors) and outcome (stroke onset). From publicly available genome-wide association studies’ (GWAS) summary results, genetic variants were identified to be used as instrumental variables to proxy levels of the stroke risk factors. Variant-Stroke Onset associations were derived from GWAS performed on the Early Onset Stroke Consortium (n = 40492) and the Stroke Genetics Network (n = 34396). Causal estimates were calculated respectively in the two groups and then odds ratios between EOS and LOS, as well as stratified by TOAST subtypes. Results: Results from this study suggest that some genetically determined levels of risk factors play a causal role in the increased risk of EOS and LOS. While conventional risk factors do not impact stroke subtypes uniquely, there may be a differential effect attributed to inflammatory biomarkers in EOS and LOS subtypes.Item The role of calcium in activating RhoA and ROCK in stationary and collectively migrating fibroblasts investigated using a novel single-color FRET sensor(2024) Mancini, Allison; Rizzo, Megan ACollective cell migration (CCM) drives physiological processes such as embryogenesis, vascular sprouting, and wound healing, but is also a major contributor to cancer metastasis. Ras homolog family member A (RhoA) drives CCM by modulating actomyosin cytoskeletal activity. RhoA activity is tightly spatiotemporally controlled during CCM, but whether these activation dynamics are shared by its downstream effectors is unknown. To investigate the role of RhoA's downstream effector Rho-associated kinase (ROCK) in stationary and collectively migrating fibroblasts, we developed a new single-color FRET-based ROCK sensor, the Rho Kinase Activity Reporter (RhoKAR) sensor. We observed calcium-dependent activation of ROCK using the RhoKAR sensor following ionomycin stimulation. With modified scratch assays that we developed, we quantified ROCK activity in parallel with CCM. Scratch wounding caused an increase in ROCK activity. We also observed blunting of CCM in response to pharmacological inhibition of ROCK or depletion of intracellular calcium with EGTA. To study the role of calcium from different sources in driving ROCK activity during CCM, we applied carbenoxolone, a gap junction blocker, to collectively migrating fibroblasts. Carbenoxolone decreased ROCK activity and intracellular calcium at corresponding time points and slowed CCM. These results indicate that gap junction-mediated intercellular calcium signaling drives ROCK activity during collective cell migration.Item Perceptions of dyspnea and quality of life before and after lung-sparing surgery for pleural mesothelioma: A pilot mixed methods study(2024) Culligan, Melissa; Klinedinst, N. JenniferBackground: Pleural mesothelioma (PM) is a rare, incurable cancer. One of the most common symptoms experienced is dyspnea. Lung-sparing surgery for PM is investigational and offered to select patients. Despite being fit for surgery, some patients continue to experience dyspnea after surgery and do not return to baseline quality of life (QOL). Purpose: The purpose of this dissertation was to (1) review and synthesize the existing scientific literature of physical, psychological and situational factors influencing QOL before and after lung-sparing surgery for PM, (2) describe factors influencing perceptions of dyspnea and QOL before and after lung-sparing surgery for PM, and (3) compare and integrate qualitatively derived themes with QOL and dyspnea measures, before and after lung-sparing surgery for PM. Methods: Guided by the Theory of Unpleasant Symptoms, the first manuscript is a synthesis of the current state of evidence of factors influencing QOL before and after lung-sparing surgery for PM. The second manuscript is the results of a qualitative study of perceptions of dyspnea and QOL before and after lung-sparing surgery for PM. The third manuscript is the results of the pilot mixed methods study comparing qualitatively derived themes and quantitative data (Dyspnea-12 scores, McGill Quality of Life Questionnaire) to identify areas of convergence and divergence. Results: The scoping review identified 15 studies. Dyspnea was quantified in 78% and individually measured in 14% of the studies. Physiological and psychological concepts were identified. The Interpretive Phenomenological Analysis identified Group Experiential Themes, Mind supports body in the world; Body fighting, enduring, adjusting, and adapting in the world; Others sharing and supporting body in the world; Facing an uncertain future of body living in the world. During the mixed methods analysis, convergence occurred with physical and psychological distress corresponding to lower QOL scores and higher Dyspnea-12 scores. Divergence occurred in the psychological domain, where improvements in scores did not reflect continued mental burden after surgery. Conclusions: This study identifies interconnections between physical, psychological, social, and existential perceptions of dyspnea and QOL. It highlights how a symptom, like dyspnea, can have broad-reaching effects on multiple aspects of QOL before and after lung sparing surgery.Item CaV1.1 gating mechanism for L-type Ca2+ current and Excitation Contraction Coupling(2024) Bibollet, Hugo; Hernandez-Ochoa, Erick EOHO; Schneider, Martin MFSAP(s) and (2) to track CaV1.1 individual VSD rearrangement in muscle fibers. We first demonstrate the physiological relevance of AP(s) optical measurement for ECC study. AP-like depolarization applied to muscle fibers reveals differences in the amount and kinetics of VSD recruitment when compared to long depolarization; and also shows a decrease in RyR1 Ṙ and occurrence of CaV1.1 ICa within AP train, both in a frequency dependent way. Next, to characterize individual VSD activation, we developed a variation of functional site-directed fluorometry (FSDF) allowing individual VSD tracking in dissociated muscle cells. This gave us the ability to detect optically individual VSD motion in response to AP. Through this technique, we demonstrated individual VSD activation heterogeneity. We then combined this approach with voltage clamp to perform voltage clamp fluorometry (VCF), allowing characterization of individual VSD, IQ, Ṙ, and ICa in response to imposed depolarization. This is a first in the implementation of VCF in native cells. We showed that VSDs have independent voltage dependence, and that VSD-II is implicated in RyR1 Ṙ while VSD-I and -IV are involved in ICa gating. Overall, this work provides valuable insights into ECC that could be the basics for CaV1.1 therapeutic regulating ICa and Ṙ independently.Item People Living with HIV (50+) and Nutritional Risk: A Mixed-Methods Study of Health-Related Quality of Life, COVID-19, and other Risk Factors(2024) Viviano, Nicole; Gruber-Baldini, AnnBackground: The population of people living with HIV (PLWH) who are 50 years of age and older will increase significantly by 2030. Older adults are at an increased risk for malnutrition, but because PLWH experience age-related conditions 10-15 years sooner than those living without HIV, understanding what predicts nutritional risk, what is associated with nutritional risk, and exploring how PLWH (50+) understand and experience their diet and nutritional risk is essential for maintaining quality care for PLWH, especially as they age. Purpose: The purpose of this dissertation was to: (1) identify predictors of nutritional risk, including (1a) common malnutrition biomarkers in an urban, community dwelling sample of PLWH aged 50 and older; (2) examine the relationship between physical and mental health-related quality of life (HRQoL) and nutritional risk in an urban, community dwelling sample of PLWH (50+), to (3) explore how urban, community dwelling, Black/African American PLWH (50+) understand and experience their nutritional risk and the factors that shape that risk, as well as (3a) explore how the COVID-19 pandemic influenced their understanding and risk. Methods: Secondary data analysis using baseline data from A) the Strengthening Therapeutic Resources in Older patients aging with HIV (STRONG) Study, B) the Frailty Study, and C) retrospective medical record data. A total of n = 184 were included in the quantitative sample. Linear regression models were used to identify predictors of nutritional risk and to examine the relationship between HRQoL (physical and mental) and nutritional risk. Prospective qualitative interview data, analyzed using the qualitative descriptive approach, were used to explore how urban, community dwelling, Black/African American PLWH who are 50 years of age and older understand and experience their diet and nutritional risk. Results: Depressive symptomology, not BMI, was positively associated with nutritional risk in PLWH (50+), even after control variables were added to the multivariable regression model. None of the blood biomarkers assessed (total cholesterol, triglycerides, albumin, and vitamin D) were associated with nutritional risk. Both physical and mental HRQoL were associated with nutritional risk in PLWH (50+); however, sex did not modify these relationships. Three themes were developed that describe how Black/African American, urban, community-dwelling, nutritionally at-risk PLWH 50+ understand and experience their diet: 1) control, choices, and influence around consumption, 2) the cycle of physical and emotional health and diet, and 3) compensating for barriers related to eating a healthy diet. Finally, a mixed methods analysis was provided in Chapter 6 that integrates qualitative and quantitative findings in the current dissertation. Conclusions: Research examining PLWH (50+) is important to the field of gerontology as it adds to the general knowledge of the aging through a uniquely stigmatized, younger population with compromised immune systems and complex care needs. Future research should continue to elevate the voices of people with lived experience to enhance their care experience and minimize nutritional risk.Item A Study to Assess Gender and Racial Differences in Unmet Dementia-Related Care Needs Among Spousal and Non-Spousal Informal Care Partners of Community-Dwelling Persons Living with Alzheimer’s Disease and Related Dementias(2024) Tucker, Gretchen; Orwig, Denise; Eckert, J. KevinBackground: Research on informal care partners (ICPs) for persons living with dementia (PLWD) has primarily focused on spousal care partners, with limited research on non-spousal ICPs. This study examines the differences in unmet ICP dementia-related care needs between spousal and non-spousal ICPs by gender and race. Methods: A cross-sectional data analysis using baseline data was conducted on 595 demographically diverse ICPs from two community-based studies. Home visits were conducted to collect demographic, clinical, and behavioral data for ICP and care recipient dyads. The Johns Hopkins Dementia Care Needs Assessment 2.0© evaluated the presence of unmet dementia-related care needs across six domains and 18 items. Linear and logistic regression models were used to examine differences in type and total percent of unmet needs among ICPs. Results: Overall, having unmet dementia-related care needs was common, with more than 68% of ICPs having unmet needs each for dementia education, community resources, and ICPs skills – behavior. Gender was not a significant predictor of the total percent or types of unmet needs. The relationship between non-spousal ICPs' race and total percent unmet needs was statistically significant (p <.0001). Being a Black non-spousal ICP was associated with having higher odds of having unmet needs for dementia education, behaviors, patient advocacy, and being physically inactive compared to White non-spousal ICPs. White female non-spousal ICPs had the lowest mean total percent of unmet dementia-related needs, while Black male non-spousal ICPs had the highest mean total percent of unmet needs. Discussion: The study found that diverse ICPs experienced significant unmet needs, especially in dementia education, knowledge, and access to community resources. It also highlighted disparities in the prevalence of certain unmet needs between Black and White ICPs.Item The Effects of a Reach To Grasp Training Intervention on Balance Responses and Fall- Related State Anxiety(2024) Alissa, Nesreen; Westlake, Kelly PThe objective of this dissertation was to investigate the relationship of falls efficacy and balance confidence with fall-related state anxiety, and the effectiveness of a dual-task reach-to-grasp balance perturbation intervention and the relationship between balance confidence and intervention responsiveness in older adults. First, the relationship between the falls efficacy scale-international (FES-I) and activities-specific balance confidence (ABC) scale with psychophysiological state anxiety, measured through skin conductance levels (SCLs)) during trip perturbations. Subjective fall-related state anxiety was assessed using the Subjective Units of Distress Scale (SUDS)), comparing participants with high (FES-I>23) and low (FES-I<23) fall concerns. Results indicated a positive correlation between changes in SCL from pre- to post-perturbation with FES-I and a negative correlation with ABC. SUDS scores differed between groups based on FES-I classifications. Next, the effects of a dual-task reach-to-grasp intervention on grasp responses to balance perturbations were investigated. The training involved 30 randomized walking perturbations with a handrail positioned laterally beside the dominant (trained) arm. Pre-training assessment included three unpredictable slip perturbations - two accompanied by a cognitive task and one without - with a handrail positioned for the trained reach to grasp response. Post-training assessments replicated pretest conditions and introduced two additional perturbations to evaluate the transfer and generalization of training effects (with the handrail on the non-dominant (untrained) side and without the handrail). Analyses focused on grasp accuracy (frequency of grasp errors) and grasp time (time from perturbation onset to handrail contact) for both the trained and untrained (transfer) arms. To assess generalizability, stability (i.e., distance from the center of mass to base of support at first foot touchdown) was calculated during a perturbation without a handrail. Findings showed improvements in grasp time and accuracy in the trained arm, with only grasp accuracy improvements transferring to the untrained arm. No changes were observed in stability during the no-handrail condition. Overall, results suggest that FES-I and ABC are indicative of fall-related state anxiety during balance perturbations, and that dual task reach-to-grasp training enhances performance in the trained arm, with some transfer to the untrained arm, but no generalization to stepping responses when a handrail is not present.Item Model-Informed Approaches to Tackle Challenges in Therapeutics and Generic Drug Development(2024) Lee, Donald; Gobburu, JogaraoInterest in model-informed approaches across pharma and medicine have dramatically increased in recent years. For the first project in therapeutics, a modeling approach was utilized here to optimize rifampin dosing for orthopedic implant-associated S. aureus infection. With a mortality rate of 10-30%, current rifampin dosing regimens may provide inadequate exposures to infected bone tissue. An integrated approach using in vitro, in vivo, and human data, along with pharmacokinetic (PK) modeling and simulation, was utilized to suggest high-dose rifampin regimens are needed to optimize bone exposures, as bone penetration was shown to be much lower (10-15%) than previously thought. For the second project in generic drug development, a model-integrated bioequivalence (MIBE) framework was proposed to demonstrate the application of MIBE for generic companies. With our MIBE approach, we demonstrate how to prepare for, plan, and implement MIBE in a robust manner that lends credibility to the virtual assessment of bioequivalence from an abbreviated trial. With an example drug of Depo-SubQ Provera 104, we show that with MIBE we can reduce the trial size by 33-50%. Thus, we demonstrate in our work the ability to tackle complex challenges with model-informed approaches to optimize dose and to accelerate generic drug development.Item Using a nationally representative health data repository to improve retention in care among people with HIV in Nigeria: A computational epidemiology approach(2024) Ikpe, Sunday; Stafford, Kristen; O'Connor, TimothyTitle of Dissertation: Using a nationally representative health data repository to improve retention in care among people with HIV in Nigeria: A computational epidemiology approach Background: Antiretroviral therapy (ART) has significantly improved the health outcomes of people with HIV (PWH). However, treatment interruptions remain a major challenge, leading to poor health outcomes and an increased risk of HIV transmission. This study aimed to identify predictors of first interruption in treatment (IIT) in PWH and determine if machine learning offers an alternative analytic approach to predicting IIT among PWH in Nigeria. Objectives: i) To determine the predictors for the first IIT defined as returning for a clinic visit 28 days or more post the next scheduled appointment, in HIV patients recorded in the Nigerian National Data Repository. ii) To determine the usability of selected supervised machine learning classification algorithms in HIV patient retention as measured by the prediction accuracy metrics for predicting the first IIT. Methods: We conducted a retrospective cohort study of over 924,847 PWH in Nigeria who started ART between 2015 and 2021. Data were extracted from the Nigeria National Data Repository (NDR), a comprehensive database of all patients managed under PEPFAR funding in Nigeria. We applied supervised learning classification algorithms (random forest, extreme gradient boosting, decision trees, and logistic regression) models to the data and assessed their ability to predict the first interruption in treatment correctly. Results: The median age of the study population was 34 years, and 65.5% were women. Overall, 54.1% of patients experienced at least one treatment interruption. The most common predictors of first interruption and time to first treatment interruption were being female (male aOR=0.90, 95% CI: 0.89-0.91), younger age of 25-34years (aHR=0.92, 95% CI: 0.91-0.93 for 35-44years, aHR=0.91, 95% CI: 0.90-0.92 for 45-54years, aHR=0.89, 95% CI: 0.88-0.90 for 55-64years and aHR=.89, 95%CI: 0.86-0.92 for 65+years ), secondary school education (tertiary education aHR=0.92, 95% CI: 0.91-0.94), and being unemployed (aHR=1.04, 95% CI: 1.03-1.06). Patients who were enrolled in the ART program after the COVID-19 global lockdown were less likely to interrupt their treatment (aHR=0.17, 95% CI: 0.17-0.17). The machine learning models were able to predict the risk of treatment interruption with an accuracy of up to 90% with machine learning logistic regression along with random forest, XG-Boost, and decision tree algorithms outperforming classical logistic regression with time-based variables being the top predictive features. Conclusions: Our findings suggest that the predictors of the first interruption and the time to the same event are similar and that machine learning has high predictivity for these outcomes. The machine learning model developed in this study, when applied to patient care with electronic medical records, can be used to identify patients who are at high risk of interrupting their treatment and to provide them with targeted interventions to help them stay in care.Item Mutations in dehydrogenase/reductase 3 (DHRS3) result in loss of function and dysregulation of all-trans retinoic acid homeostasis(2024) Williams, Christina; Kane, Maureen A.Vitamin A (retinol) is essential in key biological processes such as embryonic development, immune response, and nervous system function. All-trans-retinoic acid (atRA), an active vitamin A metabolite, requires precise metabolic regulation as an excess or deficiency can lead to serious health defects. atRA homeostasis is regulated by a variety of binding proteins and enzymes, such as the short-chain dehydrogenase (SDR) enzymes. The SDR enzymes retinol dehydrogenase 10 (RDH10) and short-chain dehydrogenase reductase 3 (DHRS3) compose the antagonistically bifunctional retinoid oxidoreductase complex (ROC), one mechanism by which RA is regulated. RDH10 oxidizes retinol (ROL) to retinaldehyde (RAL) whereas DHRS3 reduces RAL back to ROL. Retinaldehyde dehydrogenase (RALDH) then oxidizes RAL to atRA. Elevated levels of atRA are known to lead to congenital defects such as craniosynostosis. Mutations in DHRS3 have been identified in craniosynostosis patients using whole exome sequencing (WES), whole genome sequencing (WGS), and single nucleotide polymorphism (SNP) arrays. These mutations, while not in the active site, are highly conserved and result in DHRS3 loss of function. This thesis focuses on the characterization of the impact of these human mutations in DHRS3 on DHRS3 enzymatic activity, DHRS3 protein half-life, and on the global proteomic profile. To determine the impact of mutations on reductive activity of DHRS3, DHRS3 wild-type (WT) and mutants were stably transfected in HEK293 cells. When treated with 1 uM RAL, cell lines expressing the DHRS3 mutants had significantly higher levels of atRA compared to the DHRS3 WT consistent with a loss of function. Co-transfection of RDH10 along with DHRS3 investigated if mutations effected the ability of DHRS3 and RDH10 to activate and stabilize each other. In the co-transfected lines, RDH was able to exacerbate the production of atRA in DHRS3 mutant cell lines due to its unopposed oxidation of ROL to RAL. To determine the effect of DHRS3 mutations on protein half-life, a targeted liquid chromatography-tandem mass spectrometry (LC-MS/MS) method was developed, using a DHRS3-specific peptide and parallel reaction monitoring to quantify DHRS3 protein levels. Using stable isotope labeling by amino acids in cell culture (SILAC), the half-life of DHRS3 protein for each of the DHRS3 mutant variants was calculated. Lastly, to determine the impact of each DHRS3 variant on the global proteome due to atRA dysregulation, untargeted differential expression analysis using LC-MS/MS was performed; changes to individual proteins as well as canonical pathways were determined. Together these studies will advance our understanding of how DHRS3 mutations contribute to dysregulation of retinoid homeostasis as well as the proteome which will contribute to the elucidation of mechanisms by which DHRS3 mutations contribute to atRA-induced craniosynostosis.Item Human Claustrum Function in Cognitive Control Brain Network Recruitment in Health and Disease(2025) Stewart, Brent; Seminowicz, David A.Cognitive control, the coordination of executive functions including working memory, response inhibition, error monitoring, and decision-making, is critical for navigating human life, and cognitive control impairment is a feature of myriad debilitating conditions. Cognitive control is hypothesized to arise from interactions among large-scale brain networks, but mechanisms underlying network dynamics are largely unknown. Accumulating functional studies in model animals suggest the claustrum, a subcortical brain structure with unparalleled cortical connectivity, supports the initiation of multiple brain networks. However, few tests of human claustrum function have been performed despite the neuroimaging field’s established network analysis tools and the opportunity to incorporate clinical samples. A potential barrier to acquiring valid claustrum BOLD signal with standard MRI hardware, partial volume effects, was addressed, and a novel method to protect against partial volume effects in task fMRI was developed. In publicly available datasets of healthy participants, increased claustrum BOLD signal was associated with increased brain-wide network activity in multiple memory tasks evoking a wide range of network states. Structural and effective connectivity findings were consistent with excitatory claustrum projections to recruited cognitive control network regions, and results distinguished claustrum from other regions implicated in cognitive control network mechanisms. Comparisons between healthy participants and patients with chronic pain, which is associated with cognitive control deficits and altered network activity, observed claustrum activation at the onset of a difficult cognitive task and acutely painful stimuli, with patients exhibiting increased right claustrum activity in both conditions. Patients recruited an additional, pain-sensitive cortical region during cognitive task processing, and structural and effective connectivity analyses were consistent with increased excitatory influence of right claustrum on this region in patients than controls. The collected results provide empirical support for a claustrum role in cognitive control network initiation and raise the possibility of the claustrum as a future therapeutic target for neuropsychiatric disorders characterized by cognitive impairment and cortical network abnormalities.Item Disrupted Regulation of Voltage-Gated Calcium Channels by Pathological Calmodulin Mutants(2024) Hussey, John; Dick, IvyElectrical activity in neurons is shaped by the activity of multiple classes of voltage-gated Ca2+ channels (VGCCs). These channels permit the flux of Ca2+ into cells following membrane depolarizations, and these channels drive essential cellular functions in the brain, including activity-dependent gene transcription, neuronal growth and development, pace-making activity, and neurotransmission. The activity of VGCCs is tightly regulated, and disruption of the normal function of these channels can result in multiple forms of disease, including life-altering neurological syndromes. One key regulator of VGCC activity is the Ca2+ sensor, calmodulin (CaM). CaM confers a remarkable ability to VGCCs to modulate their activity in response to changes in cellular Ca2+ concentration. This form of regulation permits these channels to precisely tune their activity in response to a varied array of cellular requirements in different physiological contexts. Mutations in CaM that lead to deficits in CaM-dependent regulation of VGCCs have been identified as contributing to diseases termed calmodulinopathies, which are frequently characterized by cardiac and neurological impairments. Though prior research has revealed a number of important mechanisms by which these mutations participate in cardiac dysfunction via VGCC dysregulation, the relationship between CaM mutations and neurological symptoms observed in patients is less well understood. In this work, I demonstrate that mutant CaM variants identified in patients with cardiac and neurological symptoms are able to confer dysregulated channel behavior on VGCCs that are essential to healthy neurological function. Using a heterologous expression system, I explore the biophysical properties of these neuronal channels in the context of selected CaM mutants and show that typical Ca2+-dependent regulation via CaM is frequently disrupted by these mutants. This work focuses primarily on a region of CaM that harbors mutations that are particularly impactful on these channel properties, indicating a similar mechanism to that which is known to cause disrupted CaM-dependent regulation of cardiac VGCCs. Additionally, protein binding assays show that many of these pathological CaM mutants readily associate with the CaM-binding elements of the VGCCs found in neurons, implying a physiological relevance for the disrupted regulation these mutations impose. In summary, this work gives a background to help contextualize the importance of VGCC regulation in neurological health and provides evidence that CaM mutants identified in patients exhibiting multiple distinct neurological symptoms including schizophrenia, developmental delay, and autism spectrum disorder, may act through disrupted VGCC regulation to participate in neurological disease.Item Defining the Consequences of Nprl2 Knockout In Vitro and In Vivo(2024) Bruckmeier, Sophie; Iffland, Philip H.The GATOR1 complex, particularly the NPRL2 gene, is thought to play a pivotal role in neuronal development and cellular organization within the cerebral cortex, yet it has been poorly characterized. Dysregulation of this gene has been implicated in focal cortical dysplasia type 2 (FCD2), a neurodevelopmental disorder characterized by mTOR pathway hyperactivation, cortical dyslamination, and epileptogenicity. This dissertation aims to characterize the role of Nprl2 in FCD2 pathogenesis, employing a combination of in vitro and in vivo approaches to elucidate the effects of Nprl2 knockout on cellular morphology and circuit function. Using CRISPR/Cas9 to knockout Nprl2 in N2a cells and primary cortical neurons from in utero electroporated Nprl2 KO mice, I demonstrated that Nprl2 KO leads to pronounced soma enlargement, enhanced dendritic arborization, and increased cellular aggregation—features consistent with hyperactive mTOR signaling. Notably, these phenotypic changes were mitigated by rapamycin, underscoring the therapeutic potential of mTOR inhibitors in FCD2. In vivo, focal Nprl2 KO via in utero electroporation induced cortical dyslamination, mTOR pathway hyperactivation, and soma enlargement, all of which were reversed by rapamycin treatment. However, the reduced seizure threshold observed in Nprl2 KO mice was not completely rescued by rapamycin, suggesting that Nprl2 may affect seizure susceptibility through additional mechanisms. Electrophysiological analysis using whole-cell patch clamp techniques further revealed increased capacitance and decreased action potentials in Nprl2 KO neurons, implicating Nprl2 in the modulation of neuronal excitability. Together, these findings highlight Nprl2 as a key regulator of cortical structure and function, offering insight into potential therapeutic targets for cortical malformations associated with epilepsy.Item Molecular Strategies to Combat Acquired Resistance to the Anti-Leukemic Drug Venetoclax(2024) Goodis, Christopher; Fletcher, StevenA key hallmark of cancer is the ability to manipulate homeostatic pathways within cells to promote tumor growth. A significant example of this is intrinsic apoptotic pathway, which is regulated by members of the B-cell lymphoma 2 (BCL-2) protein family, that comprises two sub-populations of anti-apoptotic proteins (i.e. BCL-2, MCL-1, BCL-xL, BFL-1 and BCL-w) and pro-apoptotic proteins (which are further divided into BH3-only proteins, such as BIM and PUMA, and effector proteins, e.g. BAK and BAX). Under normal conditions, these proteins exist in similar ratios, interacting with one another via an a-helical BH3 “death” domain on the pro-apoptotic protein and a conserved BH3-binding groove on the surface of the anti-apoptotic proteins. In this state of affairs, homodimerization of the effector proteins is not possible, and apoptosis is prevented. However, when stress occurs in the cell, pro-apoptotic proteins are upregulated. This shift further activates BAK and BAX causing apoptosis. In BCL-2-dependent cancers, one or several of the anti-apoptotic BCL-2 proteins are overexpressed resulting in the sequestration and neutralization of pro-apoptotic proteins bound, disabling the ability to induce apoptosis. Inhibitors seeking to restore apoptotic function are designed to target the BH3 binding groove by mimicking the BH3 “death” domain. These inhibitors, dubbed “BH3 mimetics”, function by masquerading as pro-apoptotic BCL-2 proteins, which themselves become bound by the anti-apoptotic proteins thereby releasing the bona fide pro-apoptotic proteins, and restoring apoptosis in cancerous cells. A breakthrough utilizing BH3 mimicry came with the development of the BCL-2 selective inhibitor venetoclax (VEN). Gaining FDA approval in 2016 for patients with chronic lymphocytic leukemia (CLL), and then later for acute myeloid leukemia (AML), VEN has cemented itself as the current standard-of-care drug for treating these diseases. Unfortunately, patients administered VEN develop resistance to the therapeutic after ~17 months of treatment. While these resistances can manifest in multiple ways, the focus of this dissertation was to tackle the compensatory upregulations of sister anti-apoptotic BCL-2 proteins, particularly BCL-xL, MCL-1, and BFL-1. However, targeting these sister anti-apoptotic proteins presents their own challenges. Direct inhibition of BCL-xL causes thrombocytopenia, while multiple clinical trials for MCL-1 inhibitors have been halted or terminated due to cardiotoxicity, leading to the deaths of at least two patients. It has been stated that BFL-1 is the underdog of the BCL-2 family, having received comparatively far less attention than its sister proteins towards the discovery of new anti-cancer drugs. Although herein we have not identified any potent BFL-1 inhibitors, we anticipate our foundational work will provide a springboard for future work.Item Clinical Management and Outcomes After an Initial Trial of a Selective Serotonin Reuptake Inhibitor among Adolescents with Major Depressive Disorder(2024) Lee, Haeyoung; dosReis, SusanIntroduction: Selective serotonin reuptake inhibitors (SSRIs) are first-line, evidence-based treatment for adolescent major depressive disorder (MDD). In randomized controlled trials, over 40% of adolescents have a suboptimal response to SSRIs. In clinical practice, this leads to discontinuation, switching to another antidepressant, or augmentation with another psychotropic medication. Real-world evidence of subsequent antidepressant treatment strategies is limited. Objectives: This research aimed to estimate the proportion of adolescents who experience an antidepressant regimen change following an initial SSRI monotherapy, assess adherence trajectories following a switch or augmentation, and evaluate the outcomes of antidepressant augmentation with an atypical antipsychotic versus bupropion. Methods: Two population-based cohorts were identified from 2010 through 2022 using the IQVIA PharMetrics® Plus for Academics. The first cohort was adolescents with MDD who initiated an SSRI, continued treatment for at least eight weeks and did not have a regimen change in the first 12 weeks of treatment. The second cohort was children and adolescents with MDD who augmented their SSRI with an atypical antipsychotic or bupropion. A time-to-event analysis evaluated the first regimen change in the 14 weeks following the initial 12 weeks. Among adolescents who switched or augmented their antidepressants, adherence was measured over six months using bi-weekly measures of the proportion of days covered. Group-based trajectory models identified adherence trajectory subgroups. The comparative outcome of antidepressant augmentation with an atypical antipsychotic versus bupropion was a composite measure of hospitalization, emergency department visits, or suicide. Hazard ratios were estimated using Cox proportional hazards models with propensity score weighting to adjust for baseline measured covariates. Results: Among 15,678 adolescents, 38.8% discontinued, 6.0% switched to another antidepressant, and 2.9% augmented their SSRI. Of adolescents who switched or augmented the initial SSRI (n=1,402), 48.3% and 24.3%, respectively, adhered to the regimen. Children and adolescents with an atypical antipsychotic augmentation had a significantly higher risk of the composite outcome compared to those with bupropion augmentation. Conclusion: The majority of adolescents who initiate SSRI monotherapy discontinue, switch, or augment their initial regimen. Over half who switched or augmented their antidepressant did not adhere to the treatment. Bupropion augmentation had better outcomes than atypical antipsychotic augmentation.