91ºÚÁÏÍø

OXYGEN, NUTRIENTS, AND ANTIBOTICS: HOW THE CHRONIC INFECTION ENVIRONMENT SHAPES THE EVOLUTION OF PSEUDOMONAS AERUGINOSA

Pseudomonas aeruginosa is one of the most concerning opportunistic pathogens and a common cause of nosocomial and chronic infections, where it establishes biofilm communities within oxygen-depleted, nutrient-rich microenvironments. Despite the clinical relevance of these anoxic niches, susceptibility testing and experimental studies are routinely conducted under oxic conditions that fail to reflect the metabolic reality of infection. This dissertation uses experimental evolution to examine how anaerobic metabolism, lifestyle, and nutrient environment shape P. aeruginosa adaptation.

Chapter 2 investigated how arginine, abundant in the host environment, and growth mode together shape evolutionary outcomes. Populations were evolved across five conditions varying oxygen availability, growth mode, and carbon source. Adaptation to arginine consistently selected for competitive fitness gains and increased biofilm formation. Whole-population genome sequencing identified parallel mutations in fleQ across all oxic arginine replicates, providing a genomic basis for convergent motility loss and biofilm increase. Twitching motility loss was most pronounced under the combination of arginine use and biofilm growth. A trade-off in anoxic biofilm capacity emerged and recovered only when selection acted simultaneously on anaerobic metabolism and biofilm lifestyle. This work highlights arginine as a clinically relevant selective pressure contributing to the persistence of chronic P. aeruginosa infection.

Chapter 3 examined how oxygen availability shapes antibiotic resistance evolution across eight conditions varying oxygen level, lifestyle, and tobramycin exposure. Subinhibitory antibiotic exposure alone drove resistance to clinically relevant levels. Anoxic populations consistently reached higher minimum inhibitory concentrations than oxic populations,
reinforcing the concern that standard susceptibility testing underestimates resistance in oxygen-limited infections. Resistance mutations were condition-dependent: amgS dominated during oxic evolution, while fusA1 and ptsP appeared across all antibiotic-exposed conditions. Mutations in mexT, connecting antibiotic resistance, virulence, and quorum sensing, were broadly selected under both tobramycin and anoxic exposure. Anoxic evolution also produced increased biofilm
formation, type IV pilus gene mutations, reduced twitching motility, and high competitive fitness — phenotypes consistent with chronic infection adaptation.

Together, these studies demonstrate that oxygen availability, nutrient environment, and growth mode collectively shape P. aeruginosa evolution in ways that standard laboratory conditions fail to capture, highlighting the need for resistance research that reflects the environment where infections occur.

Storytelling for Social Support in Perinatal Mental Health & Wellbeing

Perinatal mental health concerns, including depression, anxiety, stress, and postpartumdistress, affect many women during pregnancy and the postpartum period. Although professional care and social support are important for managing these concerns, many women do not receive timely or adequate support because of stigma, limited access to resources, uncertainty about symptoms, and the difficulty of disclosing distress during a period often socially framed as joyful. As a result, women often turn to online
communities, digital tools, and peer networks to seek information, reassurance, and connection, often through stories of their own experiences or by reading the experiences of others. While current perinatal mental health technologies can improve access to resources, they often fall short in supporting the emotional, relational, and situated nature of women experiences. This creates an opportunity to design technologies that use storytelling to make lived experiences more accessible while fostering recognition, trust, and meaningful social support.

This dissertation focuses on understanding how storytelling can support social support in perinatal mental health and wellbeing, and on designing a storytelling-based system that makes peer narratives more accessible. I first explore how women seek and exchange support in online postpartum depression communities and how they perceive existing perinatal mental health technologies. Building on these formative studies, I develop peer narratives as a design direction and design MomStories, a web-based storytelling platform that organizes maternal narratives by topic, keyword, and perinatal stage. The platform includes 40 stories: 20 human-authored stories grounded in online peer-support communities and 20 AI-generated stories created from matched thematic prompts. The AI-generated stories were included to explore how generated narratives could broaden thematic coverage in MomStories while also allowing us to investigate how story source shapes perceptions of relatability, emotional resonance, and trust in perinatal peer support. I then evaluate how human-authored and AI-generated stories
are perceived in this sensitive peer-support context, and examine how mothers engage with MomStories through structured story access, video-based narratives, and low-pressure participation. This work contributes to human-computer interaction and social computing by showing how storytelling can be designed as a form of low-pressure support for perinatal mental health, while also identifying design tensions around emotional safety, AI-supported content, relatability, trust, and the limits of narrative-based support.

Visible-Light-Driven C–H Functionalization of N-Heterocycles and C–C Activation of 1,3-Dicarbonyl Compounds

For the past decade, photochemistry, especially visible light photocatalysis, has drawn wide attention in synthetic organic chemistry. In organic synthesis, photocatalysis has been considered as an important and powerful strategy as a photoactive species absorbs visible light to reach to its excited state and then generate reactive radical intermediates via single-electron transfer or energy-transfer events, followed by subsequent transformations to form C–C or C–heteroatom bonds. As such, visible-light driven photocatalysis has become a valuable tool in organic chemistry for structural modification and late-stage functionalization.

This dissertation presents three different research projects on the development of visible-light-driven reactions. In the first project, various types of C–H functionalized 2,2′-bipyridine derivatives were synthesized via light-driven, cobalt-mediated method, and were characterized by
different techniques. The proposed mechanism demonstrated the roles of zinc in the reaction and indicated that the reaction proceeds via a radical pathway. In the second project, we successfully synthesized and characterized different palladium-N-heteroarene complexes. Individual C–H halogenation and thiolation of palladium-N-heteroarene complex occurred under the light, which provided a solid foundation for the development of light-driven C–H difunctionalization in the future. Lastly, we investigated light-driven, metal-free C–C bond cleavage of 1,3-dicarbonyl compounds. C–C bond of b-keto esters was cleaved under the light and a new C–N bond was formed by reacting with amine to afford structurally remodeled products. This study provided us with deeper understanding of how compounds can be edited and modified via bond cleavage and subsequent new bond formation, highlighting the significance of C–C cleavage in organic chemistry.

MECHANISMS OF SKELETAL MUSCLE FIBER TYPE-SPECIFIC ATROPHY AND DYSFUNCTION WITH AGING

Human aging is accompanied by a decline in skeletal muscle mass that is exceeded by the loss of muscle strength and power, resulting in reduced mobility and functional independence. This age-related phenomenon may be attributed to 1) selective atrophy of muscle fibers expressing fast myosin heavy chain (MyHC) II isoforms and/or 2) impaired intrinsic contractile function of slow MyHC I and fast MyHC II fibers. Although previous studies have reported impaired intrinsic contractile function in older adults, technical limitations in measuring fiber size have made it difficult to determine the relative contributions of reduced intrinsic contractile properties versus fiber atrophy to age-related contractile dysfunction. Because healthy fast fibers generate 5-6x greater power than slow fibers, identifying the mechanisms underlying the divergent effects of aging on slow and fast muscle fiber size and function is critical for preserving muscle power in older adults. Therefore, the purpose of this dissertation was to determine whether intrinsic contractile dysfunction and/or fiber type-specific atrophy contribute to age-related reductions in muscle power. First, we coupled single fiber contractile experiments with three-dimensional imaging to obtain accurate cross-sectional area (CSA) measurements and assessed intrinsic contractile function in young and older males. This approach revealed intrinsic contractile function is preserved with aging in both fiber types, and the age-related reductions in fast fiber absolute force and power are primarily explained by their smaller size. We then investigated whether an increased prevalence of myonuclei expressing senescence-associated markers, γH2AX and HMGB1, contributes to fiber type-specific atrophy. The prevalence of senescent myonuclei did not differ between young and older adults and was not associated with CSA in either fiber type. In contrast, fast fiber myonuclear content was lower in older adults and closely associated with fiber CSA. Finally, we developed an imaging-based quantitative approach to measure myonuclear area positive for γH2AX in isolated muscle fibers, providing a novel method to quantify the extent of fiber type-specific myonuclear DNA damage. Collectively, these studies suggest that the agerelated decline in muscle power is due primarily to fast fiber atrophy and identify reduced myonuclear content as a key feature of fiber type-specific atrophy with aging.

LLM and SLM in Higher Ed Administration

This dissertation explores the use of pre-trained language models (PLM) in the domain of higher education, focusing especially on two common tasks: Multi-label text classification (MLTC) of student survey data, and Automated Essay Scoring (AES) of admissions essays. For each task, we investigate open critical questions, sensitive to the higher ed context. With MLTC, we explore novel architectures using small language models (SLM), including both ensemble and model blending approaches and the fusion of tabular and text data, finding that additional tabular
data does indeed improve classification accuracy. As data can be quite limited and generally cannot be shared across institutions in higher education, we also explore large language models (LLM) for synthetic data generation for MLTC, evaluating uses and contexts in which synthetic data is most effective. Finally, to encourage adoption within higher ed, a domain that prioritizes fairness and transparency, we benchmark common eXplainable Artificial Intelligence (XAI)
methods when explaining the results of our MLTC task. For automated essay scoring, we benchmark various SLM on common admissions-essay-evaluation tasks, and we develop novel approaches to AES based on feature engineering. We also offer a detailed discourse and stylometric analysis of both student-written and AI generated essays and combine discriminative and generative approaches to automated essay scoring, suggesting that a blend of traditional machine learning and small, local generative models can produce highly accurate, task-independent essay evaluation pipelines.

ENGINEERING THE MICROENVIRONMENT OF SINGLE-ATOM CATALYSTS THROUGH SURFACE MODIFICATION FOR CROSS-COUPLING REACTIONS

Single-atom catalysts (SACs) have emerged as a promising class of heterogeneous catalysts capable of bridging the gap between homogeneous and heterogeneous catalysis. By distributing isolated metal atoms across a solid support, SACs offer uniform active sites, near-total metal utilization, straightforward separation and recyclability. These features make SACs compelling candidates for reactions traditionally catalyzed by homogeneous transition-metal complexes, such as carbon–
carbon cross-coupling reactions. However, unlike homogeneous catalysts, SACs lack molecular ligands that can be rationally tuned to control the environment around the active metal center. This dissertation demonstrates that organic ligands bound to the support surface can be used to modify the microenvironment around the active sites and enhance SAC performance in cross-coupling catalysis.

Chapter 1 provides a brief review of the SAC field, covering its historical
development, synthesis and characterization techniques, and catalytic applications. Chapter 2 introduces the surface-modification strategy by coating Pd/CeO₂ SAC with benzoic acid-based ligands and systematically investigating their electronic properties. The study then evaluates the substituent effects on the aryl bromide and organoboron coupling partners for both uncoated and coated Pd/CeO₂. We find that coating the support with benzoic acid-based molecules increases the yield of Suzuki reaction several-fold, an effect that holds across a broad range of reactants and substituted benzoic acids.

Chapter 3 extends this surface-modification strategy to Sonogashira coupling, demonstrating that carboxylate-based ligands similarly enhance the activity of Pd/CeO₂ for this distinct reaction class. The chapter first establishes how the effect of the organic layer varies with solvent environment, then examines several possible origins of the activity enhancement by comparing aromatic carboxylate ligands with varying tether lengths and nonaromatic carboxylate ligands. Notably, both aromatic and nonaromatic ligands improve Sonogashira coupling activity regardless of tail group identity, suggesting that the carboxylate anchoring group contributes to the observed enhancement.

This is the first demonstration in the literature that support-bound carboxylate ligands can enhance the activity of SACs in carbon–carbon cross-coupling reactions. This work opens a path toward rationally designed SACs in which support–ligand interactions are deliberately engineered to create tunable microenvironments around isolated metal
active sites.