Washington University Medical Center
Washington University St. Louis: Open ScholarshipNot a member yet
26344 research outputs found
Sort by
Check-the-Box Enrollment Will Limit Participation in Trump Accounts: Lessons from Asset-Building Research
Full text linkAs the U.S. Treasury Department considers the implementation of Trump Accounts, one proposed enrollment method is a “check-the-box” option on the federal tax return, requiring parents to actively grant permission to open an account. This policy brief draws on evidence from prior research to explain why such a check-the-box opt-in model is very unlikely to achieve full participation in Trump Account
Accelerating GNN Inference on Multi-Core Systems
No full textGraph Neural Networks (GNNs) are becoming increasingly popular, with their applications expanding across diverse domains. As the scale of graph data continues to grow, including larger numbers of nodes, edges, and higher embedding dimensions, standardized libraries such as DGL and PyG have been developed to facilitate GNN computation. However, with the rapid increase in the number of processor cores and the evolution of multi-core architectures, these libraries often show poor scalability and fail to execute GNN inference efficiently on the latest multi-core systems, particularly those with upwards of a hundred cores. To address this limitation, we present FGI, a Fast GNN Inference system for large-scale graph data (over 100,000 vertices). FGI employs different parallelization strategies optimized for diverse graph structures, maximizing the utilization of multi-level cache hierarchies in multi- core systems. We evaluate a number of GNN models with FGI on a 128-core AMD EPYC system. Compared to state-of-the-art libraries, FGI achieves up to 3.35× and 3.20× speedups over DGL for GCN and GraphSAGE respectively, and even greater improvements over PyG, across a range of large-scale, high-dimensional graph datasets with diverse structural properties
Financial Shocks, Emergency Savings, and Hardship Among Low-Wage Workers
Full text linkFor many households, particularly those with limited income or financial resources, medical emergencies, car repairs, sudden job losses, and unexpected expenses can quickly precipitate financial hardships. Emergency savings provide a buffer that allows households to manage disruptions and avoid reliance on options that incur high-interest debt. Addressing the lack of emergency savings is not just a matter of individual responsibility; it also poses a broader policy challenge with implications for public welfare. Drawing upon data from the nationally representative Workforce Economic Inclusion and Mobility survey, this brief reports on low-wage households’ experiences with shocks, savings, and hardships. It also discusses ways for employers and policymakers to broaden access to emergency savings
Lysosomes as Gatekeepers: How LIC1-Mediated Trafficking Regulates Angiogenesis and Vascular Morphogenesis
Full text linkDynein cytoplasmic 1 light intermediate chain 1 (LIC1, DYNC1LI1) is a core subunit of the dynein motor complex that plays a critical role in cytoplasmic cargo trafficking, including Rab-mediated endosomal recycling and lysosomal degradation. LIC1 interacts with various cargo adaptors, such as RILPL1 and RILPL2, to regulate these processes, and defects in this gene are predicted to impair dynein motor function, Rab binding, and lysosomal trafficking. Here, we identified a dync1li1 zebrafish mutant with a premature stop codon at the exon 12/13 splice acceptor site, that exhibits increased angiogenesis. In vitro, LIC1-deficient human endothelial cells show elevated cell surface levels of the pro-angiogenic receptor VEGFR2, increased SRC phosphorylation, and enhanced Rab11-mediated endosomal recycling. In vivo, endothelial-specific expression of constitutively active Rab11a in zebrafish mimics the excessive angiogenesis seen in dync1li1 mutants. Similarly, zebrafish harboring mutations in rilpl1/2, which facilitate Rab docking to LIC1 for lysosomal targeting, also display increased angiogenesis. These results indicate that LIC1, alongside RILPL1 and RILPL2, restricts angiogenesis not only by modulating endosomal recycling but also by promoting lysosomal trafficking and degradation of VEGFR2-containing endosomes. Disruption of LIC1- and RILPL1/2-mediated lysosomal targeting shifts the balance toward increased Rab11-mediated endosome recycling activity, driving excessive SRC signaling and contributing to aberrant angiogenesis
Dissecting Macrophage Metabolism in Metabolic Liver Disease
Full text linkMetabolic dysfunction-associated steatotic liver disease (MASLD) is the most common cause of chronic liver disease around the globe; however, the mechanisms that lead to the progression of metabolic liver disease are poorly understood. Liver-resident macrophages, known as Kupffer cells (KCs), are the largest population of macrophage in the human body, playing a critical role in immune sensing and tissue homeostasis. In the context of MASLD, KCs undergo cell death through an unclear mechanism, and monocytes enter the liver where they differentiate into macrophages. Next-generation sequencing and spatial-omics studies have revealed significant, heterogeneity in the recruited macrophage populations in both human and mouse models of MASLD. It is now known that macrophage subpopulations have distinct gene expression profiles and reside in specific niches within the liver. However, the functional role of these macrophage subsets during disease remains to be elucidated. This thesis aims to examine various facets of macrophage biology during MASLD in humans and mice to identify potential diagnostic and therapeutic targets. I first investigated how the early stages of human MASLD influence resident and recruited macrophages (MdMs) using several cellular and molecular approaches on patient tissues. I discovered that human and murine liver macrophages share similar cellular markers. Moreover, the number of human liver macrophages directly correlates with the level of steatosis. Further exploration using mouse models revealed that MdMs can engage in crosstalk with lipid-laden hepatocytes to alleviate hepatic lipid load. Next, I leveraged our understanding of the distinct markers expressed by KCs and MdMs by conducting non-invasive imaging with novel PET tracers target to resident or recruited macrophages. Using these radiotracers in a pre-clinical model of MASLD, I was able to assess the dynamic changes in macrophage populations using PET. The KC-specific probe used for in vivo imaging was further validated with human tissue, supporting the potential for this molecular imaging approach in clinical diagnosis and evaluation. This work has progressed into a clinical study of humans with MASLD which is actively enrolling patients. To further examine the biology of different macrophage subsets in mouse models of MASH, I explored the interaction between liver stromal cells and macrophages. Specifically, we found that hepatic stellate cells (HSCs), which mediate collagen production, closely interact with MdMs and promote their ability to degrade collagen. Additionally, I utilized a novel Trem2-driven Cre system and discovered that Trem2 is expressed as monocytes develop into macrophages. This finding may explain the widespread presence of TREM2+ monocyte-derived macrophages that occurs in response to injury in various tissues. Lastly, I investigated the impact of activating TFEB, a key regulator of lysosomal function and lipid metabolism, on KC fitness and liver pathology during MASLD. We demonstrated that activation of the TFEB pathway makes KCs resistant to oxidative stress and cell death by increasing NADPH levels via a reduction in de novo lipogenesis. These TFEB-induced KCs subsequently improve liver filtration and reduce steatosis during MASLD. Collectively, this thesis leverages patient samples, in vivo animal models, and in vitro mechanistic interrogation to explore several aspects of macrophage biology in MASLD. I confirmed the relevance of the murine model for studying human MASLD pathogenesis by demonstrating that hepatic macrophages have conserved features across species. Based on our understanding of macrophage heterogeneity, I validated novel molecular imaging probes designed to track macrophage subsets during disease. To dissect the cellular interaction in pathogenesis, I further explored how hepatic stromal cells can program macrophage function. Furthermore, I engaged a cell-intrinsic pathway in KCs that promoted their lipid metabolic function, survival, and resistance against oxidative stress, thereby reducing pathology. The data presented in this thesis represents a significant advance in our understanding of macrophage biology in the continuum of MASLD. These findings have the potential to drive the development of novel diagnostic tests and therapeutic targets that will impact the lives of patients with metabolic liver disease
Untimely Epics: The Tradition of Changqing-style Poetry in Late Qing and Republican China (1870s–1920s)
No full textThis thesis focuses on the unique re-emergence of long narrative poems in the classical style in late 19th and early 20th-century China. The noted long poems are written in the genre of Changqing style, 長慶體, an ancient heptasyllabic style featuring grand historical narratives of dynastic change around imperial gardens and prominent heroines/heroes, in the elaborate form of poetic expression. In this dissertation, I closely analyze representative Changqing-style epics by four prominent scholar-poets — Wang Kaiyun 王闓運 (1833–1916), Wang Guowei 王國維 (1877–1927), Fan Zengxiang 樊增祥 (1846–1931), and Chen Yinke 陳寅恪 (1890–1969) — unearthing literary value overlooked in previous scholarship. These untimely epics of the Changqing style manifested the literary classicism and cultural conservatism of Chinese intellectuals in the irredeemable progress of Chinese modernization. This dissertation is one of the first English-language works of scholarship on Changqing-style poetry written in late Qing and Republican China
The 3D Genome: Investigating CO2 Driven Chromatin Dynamics and Epigenetic Reprogramming in Arabidopsis thaliana
No full textWithin the last 60 years, the Earth’s atmosphere has experienced a steady and rapid increase in carbon dioxide (CO2) composition. Plants play a critical role in the carbon cycle as primary consumers of CO2 and the foundation of agricultural output. A timely and open question remains regarding how plants will acclimatize to increasing levels of CO2, a rate-limiting substrate in photosynthesis. While genetic traits alone provide robust currency for the selection of advantageous traits, genetic mutations and polymorphisms rarely revert, are slowly acquired, and can bottleneck breeding practices for crop resilience. Land plants are primarily sessile and among the array of multi-cellular organisms that have evolved epigenetic mechanisms to allow for rapid adaptations to occur in the face of turbulent environments. How can cells orchestrate these mechanisms with such sophistication? The interplay between epigenetic traits such as DNA methylation, histone modifications, and the 3D structure of chromatin underlie the plasticity of gene regulation, providing a molecular blueprint for retaining epigenetic memories of environmental conditions across generations via epigenetic reprogramming. In this dissertation, I investigate the hierarchical structure of epigenetic reprogramming that underlies the initiation of plant development at elevated CO2. First, I investigate the epigenetic properties of chromatin that re-wire transcription to support a heritable accelerated growth phenotype in Arabidopsis thaliana. Using high-resolution methylation-sensitive Chromatin Conformation Capture and sequencing (Hi-C), I reveal that CO2-dependent chromatin decondensation renders more accessible chromatin closer to the nuclear periphery and implicate interplay between key epigenetic mechanisms such as small RNA-directed DNA Methylation (RdDM), asymmetric maintenance methylation, and histone modifications involved in chromatin remodeling. This decondensation is supported by broad shifts in the genome activity of euchromatin and heterochromatin. At high resolution, I annotate Local Chromatin Domains (LCDs) and LCD Associated Loops (LALs) that connect specific genomic regions to macro-level changes in genome structure and transcriptional reprogramming at impacted loci. I identify trans acting factors that direct the formation of LCDs and LALs through the regulation of 5-methylcytosine (5mC) and H3K27me3 acting in cis. At sub-kilobase resolution, I define the reprogrammed transcriptional profile of wild-type (WT) and mutant plants responding to elevated CO2 and implicate epigenetic modifications such as 5mC and histone tail post-translational modifications, that are important in shaping the 3D structure of the genome, with the initiation of CO2-dependent epigenetic reprogramming. Together, my dissertation work reveals that the interface between epigenetic modifications and the 3D genome is critical for initiating and supporting the developmental program at elevated CO2
ACLY Is Required for NK Cell Effector Function by Regulating the Epigenome
Full text linkNatural Killer (NK) cells are innate lymphocytes that recognize virus-infected or malignant cells without prior sensitization. NK cells are activated when germline-encoded receptors identify target cells or through cytokine receptors. Upon activation, NK cells produce pro-inflammatory cytokines, IFN-g, and cytotoxic molecules, such as perforin and granzymes, to activate other immune cells or directly kill target cells. Effector function is closely linked to alterations in cellular metabolism, alongside the recognition of target cells by receptors. This is because metabolites are utilized as co-factors or substrates in epigenetic modifying enzymes, which regulates the gene expression. This mechanism, known as the metabolic-epigenetic crosstalk, has been extensively examined in other immune cells; however, much less is known in NK cells. ATP-citrate lyase (ACLY) is an enzyme associated with the citrate-malate shuttle, which generates cytosolic acetyl-CoA. As acetyl-CoA is a main provider of acetyl groups for lysine acetyltransferase, ACLY plays a major role in the metabolic-epigenetic crosstalk. Previous studies showed that treating ACLY inhibitor (BMS-303141) significantly downregulated the cytokine production in NK cells stimulated with IL-2/12 for 18 hours. In this study, we used a tamoxifen-inducible, NK-specific knockout (KO) mouse model to elucidate whether ACLY regulates cytokine production through modifications in the epigenome. In contrast to the inhibitor study, ACLY KO NK cells showed intact cytokine expression with overnight stimulation. However, IL-15 priming, which is a metabolically demanding stimulation, significantly impaired proliferation, glycolysis, and cytotoxic molecule production in ACLY KO NK cells. Moreover, ACLY KO NK cells showed impaired cytokine expression when stimulated with NKG2D or Ly49H, due to the varying expression of adaptor proteins, DAP10 and DAP12. The expression of genes associated with glycolysis, cytotoxicity, and adaptor proteins was regulated by histone acetylation, demonstrated by CUT&Tag sequencing. Supplementing with acetate, an alternative acetyl-CoA provider, while undergoing IL-15 priming rescued most of the impaired phenotypes of ACLY KO NK cells. In conclusion, this study enhances the understanding of how metabolites rewire epigenetic, which ultimately affects the effector function of NK cells
Micro Medical Device to Prevent Myopia for Teenagers and Children
No full textThe purpose of this project is to find a solution and build a micro medical device to help prevent myopia for teenagers and children. The final product would be easily worn like a pair of glasses which can monitor the using habit of eyes and provide real-time correction to the users. There are four main parts. First, we analyzed the medical reasons that cause myopia for teenagers by doing research online. Based on the research, we decided to use the ultrasonic sensor for detecting the reading distance, the photoresistor for detecting the light condition, and the accelerometer to detect the reading angles. Second, we used the SEEEDSTUDIO XIAO ESP32S3 chip to control the sensors above and create a solution (formula algorithm) that has the highest chance of protecting the eyes. In addition, we used AWS to communicate between the real-time readings from the sensors and the decision-making of the algorithm through WI-FI. For the third part, we used SOLIDWORKS to design the 3D frame of our final product. Finally, we designed a web client that receives the data from the device for parents to view the status. Our product works stably. In the future, we will focus on designing our own chip and finding smaller sensors to minimize our product so that it can be an industrial production
Defining B cell responses to emerging viral pathogens
No full textThe primary series of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) vaccines greatly reduced the burden of the coronavirus disease 2019 (COVID-19) pandemic and saved millions of lives. However, waning levels of circulating neutralizing antibodies and the continuous emergence of SARS-CoV-2 variants with increasing numbers of mutations in the spike (S) protein have decreased vaccine effectiveness and led to a recommendation for booster immunizations. In addition, new vaccines based on circulating variants have been developed to enhance the ability of induced antibodies to combat such variants. However, a critical unanswered question was how prior exposure to the ancestral SARS-CoV-2 strain impacts the subsequent immune response following exposure to variant viruses. Specifically, do variant-based booster immunizations induce de novo B cell responses to that variant, or do they primarily recall preexisting B cells targeted against shared epitopes? Understanding this distinction is essential for the rational design of vaccine strategies that can induce broadly neutralizing antibodies against mutating viruses. Herein, we evaluated antigen-specific B cell responses following SARS-CoV-2 mRNA-based booster immunizations in humans. Using ultrasound-guided fine needle aspiration to serially sample draining axillary lymph nodes, we showed that boosting with either the ancestral SARS-CoV-2 mRNA vaccine or the bivalent B.1.351/B.1.617.2 (Beta/Delta) mRNA vaccine elicited robust S-specific germinal center (GC) B cell responses in all sampled participants. These GC responses persisted for at least eight weeks post-booster immunization, leading to significantly more mutated antigen-specific bone marrow plasma cell (BMPC) and memory B cell (MBC) compartments. We also showed that S-specific plasmablast and GC responses predominantly originated from preexisting clonal lineages, which is consistent with our observation that most MBC-derived monoclonal antibodies (mAbs) recognized the original SARS-CoV-2 S protein. Nonetheless, using a more targeted sorting approach, we isolated mAbs that recognized the BA.1 S protein but not the original SARS-CoV-2 S protein from individuals who received the mRNA-1273.529 booster, suggesting that immunization with the monovalent B.1.1.529 BA.1-matched vaccine could induce rare de novo B cell responses against novel epitopes in the B.1.1.529 BA.1 S protein. Therefore, SARS-CoV-2 booster immunizations in humans induce robust GC B cell responses, and immunization with a distinct antigen can generate de novo B cell responses targeting variant-specific epitopes. Over the course of the COVID-19 pandemic, a dozen mAbs that target the virus S protein have been approved for clinical use in treating those infected by, or exposed to, SARS-CoV-2. These antibodies, which largely target the receptor-binding domain (RBD) of the S protein, were all raised against and efficiently neutralize the original SARS-CoV-2 strain (WA1/2020). While WA1/2020-targeting antibodies can provide a degree of protection against emerging variants, including those with minor mutations, they may lose efficacy against variants with significant changes in their S proteins. In fact, all clinically authorized mAbs are now rendered inactive by the latest Omicron subvariants, raising the question of whether antibodies that are specifically tailored to the unique features of Omicron can maintain efficacy against newer, more divergent Omicron subvariants. Herein, we assessed the ability of Omicron BA.1-specific mAbs to recognize and neutralize recently emerged SARS-CoV-2 Omicron subvariants. Out of six mAbs tested, only one mAb, 1A05, demonstrated broad binding and neutralization capacity against multiple SARS-CoV-2 Omicron subvariants, from BA.1 through HV.1. Prophylactic administration of 1A05 significantly reduced infectious virus titers for Omicron BA.5 in the lungs, nasal turbinate, and nasal wash of infected animals. Structural analysis of 1A05 Fab fragment in complex with Omicron BA.1 RBD using cryo-electron microscopy (cryo-EM) showed that the bulk of the 1A05 footprint overlapped with the receptor-binding motif (RBM), allowing 1A05 to achieve neutralization by blocking ACE2 binding. However, with the continuous evolution of Omicron subvariants and the emergence of antibody-evading JN.1, 1A05 lost efficacy, indicating that Omicron BA.1-specific B cell responses may lose efficacy against newer, more divergent Omicron subvariants, which emphasizes the need for mAbs with enhanced breadth and resilience against the continuously evolving SARS-CoV-2. Overall, the studies I describe here provide crucial insights into how prior exposure to SARS-CoV-2 shapes subsequent immune responses, particularly in the context of variant-adapted vaccines. This knowledge not only informs strategies for combating the ongoing evolution of SARS-CoV-2 but also helps us prepare against future emerging viruses