Rockefeller University

NIH Research Projects · FY 2024 · 2023-09

Project Summary CRISPR-Cas are prokaryotic adaptive immune systems that protect bacteria and archaea from invading mobile genetic elements, such as phages and plasmids. CRISPR-Cas systems acquire immunological memories during infection by integrating short fragments from the invader’s genome into the CRISPR locus of the host. These fragments, called “spacers”, are later transcribed into CRISPR RNAs that are loaded on Cas nucleases and guide them to recognize and cleave infecting nucleic acids. Depending on their genetic composition, CRISPR- Cas systems are classified into six types (I-VI). While spacer acquisition has been extensively studied in type I and II systems, type III systems are just now starting to be explored. The overall goal of this application is to define the molecular mechanisms that govern spacer acquisition by the prevalent, yet less studied, type III-A CRISPR-Cas system, and understand its implications during CRISPR-Cas defense and tolerance. Preliminary work on the type III-A system of Staphylococcus epidermidis revealed that this system preferentially acquires new spacers by two independent modes. The first mode acquires spacers from some, but not all, highly transcribed genes, and spans their entire transcribed region. The first aim of this proposal is to elucidate how the acquisition machinery recognizes specific genes as substrates for preferential acquisition. This will be achieved by dissecting the DNA sequences that recruit the spacer-integrase complex to specific genes, finding host factors that mediate gene-specific spacer acquisition, and test for the physiological relevance of this process during the CRISPR-Cas immune response. The second mode of acquisition by the type III-A system is similar to the previously studied type I and II systems, where spacers are acquired from free dsDNA ends at the bacterial chromosomal terminus, in a manner that is dependent on the cell’s DNA-repair machinery. Such self-targeting spacers are expected to induce autoimmunity and be negatively selected, however we found them to be stably fixed in the bacterial population, suggesting the existence of unknown mechanisms that inhibit targeting by Cas nucleases at this site, thus preventing CRISPR autoimmunity. The second aim of this proposal will define the genomic context that allows self-targeting spacers to be tolerated, analyze the temporal dynamics of CRISPR- Cas immunity at free DNA ends, and explore the genetic components needed for CRISPR-tolerance and accumulation of self-targeting spacers. This proposed work will not only transform our conceptual understanding of the spacer acquisition process, but also could lead to CRISPR-based technological developments in molecular biology and diagnostics. To achieve these goals, I have assembled a team of experts in the fields of transcription, DNA repair, bioinformatics, biochemistry and biophysics. Their guidance, along with the continued mentorship of Prof. Luciano Marraffini and the scientific environment of the Rockefeller University, will allow me to perform the proposed research, as well as to develop writing, mentorship and communication skills, that will support my successful transition to an independent career.

NIH Research Projects · FY 2024 · 2023-09

Project Summary/Abstract Eukaryotic cells require the tight regulation of global gene expression to maintain homeostasis and respond to environmental stimuli. DNA spools around histone proteins form this vital structure, chromatin, and provide a platform for the sophisticated tuning of gene expression through physical and chemical regulation. Unsurprisingly, the disruption of these chromatin regulatory mechanisms is particularly prevalent in cancers as a driver of disease. Completion of the proposed projects will shed light on the mechanisms of healthy chromatin regulation and its disruption in disease, providing the insight necessary to develop improved therapeutic interventions in a variety of cancers. In the F99 phase of this proposal, I study disrupted chromatin signaling by Hepatitis B Virus (HBV), a leading cause of hepatocellular carcinoma worldwide. HBV maintains chronic infections within hepatocytes by establishing an independent minichromosome, termed covalently closed circular DNA (cccDNA), that largely evades immune detection and conventional chromatin regulatory mechanisms. Further contributing to this evasion is the viral protein HBx, which has documented roles redirecting numerous chromatin effectors, including transcription factors, degradation machinery, and epigenome modifiers. So far in my thesis work, I have developed a platform to reconstitute cccDNA in vitro for biochemical and biophysical studies, determined that histone occupancy in cccDNA is required for HBx expression, and shown that HBx binds directly to nucleosomes. The remainder of my thesis work will be spent testing the biochemical effects of other known interactors on cccDNA compaction and gene expression and illuminating the cccDNA landscape in cells using locus-specific proteomic and epigenomic studies. The K00 phase shifts focus to ATP-dependent chromatin remodeling enzymes, a class of proteins shown to be mutated or overexpressed in more than 20% of cancers. In particular, I intend to study the CHD family of remodelers, which have been implicated as drivers of a variety of cancer types. I will apply my expertise in chromatin biochemistry and expand my technical repertoire to include cryo-electron microscopy as a means to study the structure and function of CHD chromatin remodelers. In parallel, I will develop skills in gene editing techniques to knockout wild-type enzyme and introduce clinically-relevant CHD mutants into cells for epigenomic analyses of remodeler dysfunction in disease. Combining these new approaches with my background in biochemistry, chemical biology, and biophysics will position me to address pressing questions in chromatin and cancer biology throughout the rest of my career as I pursue an independent, cancer-focused faculty position.

NIH Research Projects · FY 2024 · 2023-09

Project Summary/Abstract Platelets play a pivotal role in hemostasis and thrombosis as they are required for platelet aggregation which contributes to both the arrest of bleeding and the development of arterial thrombi. The platelet receptor integrin αIIbβ3 plays a non-redundant role in supporting platelet-platelet interactions via binding of fibrinogen (aggregation) while its interaction with polymerized fibrin stabilizes the clot by the process of clot retraction. A fundamental challenge in thrombosis research is to understand precisely how platelets interact with fibrinogen versus polymerized fibrin and, how the latter dynamically retracts the clot. The aim of this proposal is to apply state-of-the-art and novel experimental approaches to obtain new insights and understanding of fibrin-αIIbβ3 binding and signaling mechanisms leading to clot retraction and thrombus stabilization. To that end, I developed a novel assay for assessing the interaction of platelets with fibrin, independent of platelet-fibrinogen interactions, and I recently developed a high-throughput screening assay to identify inhibitors of clot retraction. I screened 408,724 compounds, from which I identified 580 confirmed inhibitory compounds. With colleagues, I have also screened 301 murine monoclonal antibodies (mAb’s) made in response to immunization with human platelets or αIIbβ3 and have identified 4 antibodies that inhibit clot retraction but not platelet adhesion to fibrinogen. Using structural, biochemical, and functional approaches, I now propose to: i) determine the mechanisms of action of the different inhibitors of clot retraction, with the goal of identifying novel fibrin-specific targets for antiplatelet therapy. ii) employ cryo-electron microscopy to obtain structural information on the binding sites on αIIbβ3 of the fibrin-specific mAb’s and thus their mechanisms of interfering with clot-retraction. iii) study the unique αIIbβ3- fibrin cellular interactions and signaling pathways with methods to capture and analyze the protein complexes that form in response to fibrin-αIIbβ3 interaction. My mentor, Dr. Barry Coller is an expert in platelet and αIIbβ3 translational research, having developed the first FDA approved αIIbβ3 antagonist and with a second in a Phase 3 study. My co-mentor, Dr. Alisa Wolberg is a leader in platelet-fibrin interactions and clot retraction. During this award I will continue my technical and scientific education by training in several outstanding collaborating laboratories in techniques that will serve as building blocks for an RO1 research proposal I plan to submit in year 3 of this award, leading to scientific independence at the end of this award. Rockefeller University provides an outstanding research environment, with a wide range of lectures, seminars, and symposia, and access to state- of-the-art equipment and resource centers, led by senior scientists who are charged with training junior scientists. Dr. Coller and Rockefeller University are committed to a more diverse, equitable, and inclusive scientific community and so are delighted to support Dr. Buitrago’s application, because of her demonstrated commitment to mentor and serve as a role model for women scientists and scientists from underrepresented minority groups.

NIH Research Projects · FY 2026 · 2023-09

Project Summary DNA replication is performed by numerous proteins that act as a dynamic machine, termed a replisome. The core components of the eukaryotic replisome consist of 1) an 11 subunit “CMG” helicase that separates the parental DNA strands, 2) The leading and lagging strand DNA polymerases (Pol), Pol d and Pol e, respectively,3) The PCNA sliding clamp that encircles DNA and tethers both Pols to DNA for high processivity, 4) the RFC clamp loader pentamer that loads PCNA onto DNA and 5) Pol a-primase that makes a hybrid RNA- DNA primed site for the Pols to initiate DNA synthesis. In addition to these “core” components, there are several ancillary proteins including RPA, Tof1, Mec1, Csm3, FACT, Mcm10, Ctf4, Ctf18-RFC. There are a host of proteins that assemble two CMGs around duplex DNA at origins. The CMG dimer unwinds the closed duplex in an unknown reaction and scaffolds enzymes to assemble replisomes. We have purified these proteins in the yeast (Saccharomyces cerevisiae; S.c.) system. In this proposal, we will extend our studies on the structure/function of the eukaryotic replisome. We will use biochemical and single-molecule methods to determine if PCNA accumulates on the lagging strand as expected, and whether PCNA may periodically be left on the leading strand for mismatch repair and assembly of naïve nucleosomes. We have solved numerous structures with our collaborator, Huilin Li (VanAndel Institute, MI), and have many more structures in progress and planned. We have purified the several factors of the ATR DNA damage checkpoint signaling system of which many replisome proteins are targets of this pathway. We plan biochemical studies that will clarify targets and their effect on replisomes. We will determine the mechanism of nucleosome inheritance during replication. In metazoans, epigenetic inheritance of nucleosomes, gone awry, can lead to cancer and other diseases. In yeast, cell studies have shown that a Mcm2 histone binding mutant prevents epigenetic transfer to lagging strands, and Pol e lacking the Dpb3/4 subunits does not transfer epigenetic marks to the leading strand. We plan to visualize nucleosome transfer during replication in real time using single-molecule studies with our newly acquired Q trap) in collaboration with Dr. Shixin Liu (Rockefeller University)). We have various nucleosome mobility factors and yeast nucleosomes having different fluorescently tagged histones for these studies. Replication occurs in nuclear foci, having “replication factories” with many DNA replication forks. Our recent biochemical and structural studies have defined the composition and atomic structure of the most basic unit of a replication factory, a dimeric replisome. We will employ super resolution microscopy to validate if our reconstituted factory is the same as that inside cells. We have insight into how duplex DNA at origins is opened into single strands from our recent finding that the twin CMG helicases encircling duplex DNA at an origin are directed inward, opposite the “outward” direction thought for a decade. We find that two inward directed CMG can shear DNA apart. We have plans to further this line of investigation.

NIH Research Projects · FY 2024 · 2023-08

Project Summary/ Abstract The long-term goal of the proposed research is to uncover molecular mechanisms driving non- apoptotic cell death in vertebrate development and disease, specifically the role of nuclear and chromatin organization in this process. Programed cell death functions to sculpt organs, remodel tissues, regulate cell number, and remove defective cells. While apoptosis is the most studied type of cell death, it does not account for all cellular destruction during development. Studies in C. elegans have uncovered a novel developmental cell death program, referred to as linker cell-type death (LCD), which is morphologically and molecularly distinct from apoptosis. Cell death with LCD features is commonly observed during vertebrate development and in neurodegenerative polyglutamine diseases, aggressive cancers, aging, progeria and laminopathies. During my postdoctoral training, I uncovered striking nuclear and chromatin changes that occur during LCD. This K99/R00 proposal seeks to identify the molecular regulatory processes underlying these nuclear and chromatin changes during LCD in C. elegans, and to test the functional conservation of LCD in a vertebrate context through the following specific aims: 1) Elucidate the role of nuclear lamin and its regulation in the process of LCD (K99); 2) Establish a mouse model system to study LCD in a developmental mammalian setting using degenerating Mullerian duct as a model (K99/R00); 3) Identify molecular processes that govern chromatin dynamics during LCD in C. elegans and mammals (R00). This proposal builds on my doctoral studies in mammalian epigenetics and chromatin, as well as postdoctoral training investigating non-apoptotic cell death in C. elegans. During the mentored phase, I will gain essential training in mouse organ culture and management, which will set me up to establish a robust and innovative independent research program studying the contribution of chromatin and nucleus to the non-apoptotic cell death in vertebrate and invertebrate systems, in addition to human disease. The outstanding environment at the Rockefeller University, with mentorship from Dr. Shaham and my Advisory Committee, will provide crucial expertise to facilitate my transition to independence. The studies proposed here will lead to the development of new markers to distinguish among different types of cell death in vertebrate development and disease, while uncovering the molecular underpinnings of LCD. Because LCD is a prevalent type of cell demise, this proposal may not only shed light on basic aspects of development, neurodegeneration, and cancer, but could also eventually uncover in-roads of clinical significance.

NIH Research Projects · FY 2024 · 2023-08

Project summary The highly specialized intestinal immune system is charged with maintaining tolerance to harmless stimuli from food and commensal bacteria, while providing protective immunity against pathogens. Dysregulation of this critical balance can lead to food allergy, inflammatory bowel disease (IBD), or increased susceptibility to enteric pathogens. Dendritic cells (DCs) are key players in intestinal homeostasis, finely orchestrating immune responses by presenting luminal antigens and inducing functional differentiation of CD4+ T cells into regulatory or pro-inflammatory subsets. The cellular mechanisms underlying the decision between tolerance or immunity to intestinal antigens remain unknown, largely because the identification and characterization of the exact DCs involved in these processes has been a decades-long technical challenge. To overcome this problem, I established the use of the LIPSTIC (Labeling Immune Partnerships by SorTagging Intercellular Contacts) technology in the gut, which enables proximity-dependent labeling of specific intestinal cell-cell interactions. The main goal of this research proposal is to elucidate the mechanisms that determine whether DCs will induce regulatory or pro-inflammatory T cells in vivo, by combining the LIPSTIC system with gene targeting, functional imaging, intersectional genetics, and interaction-based transcriptomics approaches. This proposal tests the hypothesis that DCs specialized for different functions exist in the intestine and that continuous dynamic reprogramming of DCs subsets by luminal content dictate tolerance or immunity to food and microbes. Specifically, this concept will be tested in mice under physiological (tolerance to food and microbes) and pathological (allergic sensitization, enteric infection, and colitis) scenarios. To this end, the K99 mentored phase (Aim 1) will reveal the exact nature of the DCs that induce tolerance or inflammation to food and the cellular and molecular mechanisms by which food-specific pTregs maintain oral tolerance. Next, the R00 independent phase (Aim 2) will focus on studying immune responses to commensals, with relevance to disorders such as colitis. Elucidating the DC populations, location and molecular mechanisms involved in tolerance to distinct microbes will reveal the nature of host-microbe interactions in health and disease. Together, this research program will lend fundamental insight into the etiology of certain intestinal disorders and will provide foundation for the development of new therapies for food allergy, colitis, and enteric infections, with profound implications for public health. The proposed development plan complements my training in mucosal immunology and cellular interactions with transcriptomics analysis, imaging, and mouse genetics. I will take advantage of the extensive resources of the Rockefeller University, the mentorship of Dr. Daniel Mucida, Dr. Gabriel Victora and the Advisory Committee team that will lend expertise in key aspects of the project and career development. At the end of the mentored phase, I will be equipped with the necessary tools to conduct comprehensive studies at the intersection of cellular communication and immune response to commensals as an independent investigator.

NIH Research Projects · FY 2026 · 2023-06

The combined degeneration of both “lower” motor neurons in the brainstem and spinal cord and “upper” motor neurons (UMNs) in the cerebral cortex is an important hallmark of ALS. Almost all cases of ALS are eventually fatal, and the rapid progression of the disease makes it particularly terrible, with over 80% of patients dying within five years of diagnosis. No cure exists for ALS and the only available treatments slow disease progression by merely a few months. Therefore, a great need exists for more effective and specific therapies that can stop or even reverse neurodegeneration. Innovation for such therapies will only arise from a better understanding of the molecular mechanisms underlying the pathological process, especially since most genes linked to ALS are ubiquitously expressed yet only specific populations of cells degenerate. Understanding why certain cells are uniquely vulnerable and mapping cell type specific pathways that are dysregulated during disease are crucial milestones. This has posed a considerable challenge for the spinal cord projecting UMNs since they are difficult to distinguish from other pyramidal cell types and are therefore often overlooked in preclinical studies. Because of this, the basis for their selective vulnerability to ALS-causing mutations has remained a mystery. The proposed study aims to overcome this by building on recent work that identified two highly similar yet molecularly distinct subpopulations of projection neurons in layer 5b of motor cortex, where UMNs reside. These populations have overlapping projections to pons, but non-overlapping projections to the spinal cord or thalamus. Examining these cells in preclinical models of ALS revealed that the corticospinal projecting neurons (CSTNs) were vulnerable to degeneration, while the corticopontine-only population (CPN) did not degenerate. The selective vulnerability of the CSTNs was likely due to dysregulation of mitochondrial function since a dramatic upregulation of genes related to oxidative phosphorylation and mitophagy was observed at symptomatic stages of disease. Aim 1 of this grant will employ an integrative multi-omics approach to address whether differences in the properties of mitochondria between CSTNs and CPNs drive differential responses to disease using a novel, viral-based strategy to isolate cell type specific mitochondria during disease progression in two preclinical ALS models, SOD1G93A and FUSP525L. Aim 2 focuses on dissecting the cellular role of identified candidate genes that are enriched in CSTNs and cell type-specific bioenergetic pathways, linking them to mitochondrial function and disease vulnerability. To increase the translational significance of this work, Aim 3 will leverage novel markers for CSTNs and CPNs for a detailed anatomical analysis of postmortem tissue from ALS patients to perform transcriptional profiling on cell type specific nuclei isolated by fluorescence activated nuclear sorting (FANS) from postmortem patient tissue. Results from this study will yield novel mechanisms underlying selective vulnerability of UMNs in ALS.

NIH Research Projects · FY 2026 · 2023-05

Abstract The hippocampus has a well-established role in the initial formation and storage of memory. However, little is understood about brain mechanisms that support the re-organization and transfer of memories into longer-term cortical storage. A detailed understanding of hippocampal- to-cortical consolidation is critical to shed light on the regulation of long-term memories, and how they may become too transient (as in Alzheimer’s, Parkinson’s, Traumatic Brain Injury) or too persistent (as in PTSD). One challenge has been the difficulty inherent to tracking the real-time cellular resolution activity of multiple brain regions throughout the weeks-to-months long consolidation window. Toward this, we have recently developed a head-fixed behavioral paradigm, compatible with longitudinal imaging, where mice learn to form and main contextual associations for at least one month. We have also found that learning of this task requires the hippocampus, while memory consolidation requires the entorhinal and prefrontal cortex. Using methods to perform multi-region longitudinal neural activity recordings, together with anatomically-defined optogenetic inhibition during imaging, we propose to dissect the real-time contributions of the hippocampal-entorhinal-prefrontal pathway in memory consolidation. Specifically, we aim to characterize a new behavioral model of memory consolidation (aim 1), and perform longitudinal imaging to identify prefrontal cortex brain activity patterns unique to consolidated memories (aim 2). Finally, we will identify how entorhinal projections stabilize such activity patterns in prefrontal cortex during consolidation (aim 3). Together, we will contribute to our understanding of how some memories are progressively re-organized and stabilized across the brain for long-term storage.

NIH Research Projects · FY 2026 · 2023-05

Many of our most important therapeutics were inspired by bacterial small molecules (natural products, NPs). Although microbial NPs display a wide range of bioactivities, they have offered their greatest utility as anticancer agents and antibiotics. The incredible success of NPs as lead structures for therapeutic development is thought to be due to their unique structural and mode of action refinement from eons of evolutionary selective pressures. Since many drug discovery programs deprioritized NPs due to unacceptably high rediscovery rates, bioinformatic analyses of genomic sequence data, whether from cultured bacteria or metagenomes, has revealed that the biosynthetic diversity accessed by traditional monoculture fermentation studies represents only a small fraction of the NPs that are actually encoded by the global microbiome. Unlocking the metabolites encoded by this large fraction of previously inaccessible biosynthetic gene clusters (BGCs) should provide structurally and mechanistically novel molecules that can serve as inspirations for new anticancer agents. Traditional NPs discovery methods rely on biological processes (i.e., transcription, translation and enzymes) to convert genetic instructions contained in bacterial genomes into novel bioactive small molecules. Unfortunately, with these methods it has not been possible to coax laboratory grown bacteria into producing all the different NPs they are capable of making. We have therefore developed a “biology free” discovery approach where, instead of decoding genetic instructions using biological processes, bioinformatic algorithms are used to predict the chemical structures produced by bacteria and then chemical synthesis is used to build these structures, which we have called Synthetic Bioinformatic NPs (syn-BNPs). This proposal is designed to bring together advanced bioinformatics, total chemical synthesis, and next-generation metagenomic methods to identify syn-BNP antiproliferative agents that are inspired by BGCs which, until now, have remained hidden in the genomes of cultured bacteria and metagenomes. Interestingly, nearly half of all drugs in clinical use today are inspired by nonribosomal peptides (NRPs) or mixed polyketide-NRPs. Fortuitously, NRP biosynthesis is unique in that bioinformatic algorithms have developed to the point where it is possible to predict many NRP structures from primary data sequence alone. Concurrently with these bioinformatic advances, robust methods for synthetically producing NRP-like structure have become simple and economical, making uncharacterized NRP BGCs model targets for syn-BNP discovery studies and a potentially rich source of mechanistically diverse and novel antiproliferative agents. With this in mind, in Aim 1 bioinformatic analysis of NRP BGCs found in publicly available data bases will be used to inspire syn-BNPs that will be screened for differential antiproliferative activity across a panel of diverse cancer lines. In Aim 2, metagenomic BGCs will be sequenced and used to inspire additional syn-BNPs for antiproliferative activity screening. In Aim 3, antiproliferative syn-BNP hits will be mechanistically studied and synthetically optimized to ready them for future more detailed in vitro and in vivo studies.

NIH Research Projects · FY 2026 · 2023-05

Project Summary / Abstract Our brain provides us with a sense of where we are in space. The importance of this sense is clear when we become spatially disoriented, like when one is confused about one’s orientation after exiting a subway station. Central to the understanding of how brains give rise to spatial cognition has been the discovery of place cells in the 1970’s (i.e., neurons that are active when animals are in one location in space), head-direction cells in the 1980’s (i.e., neurons that are active when animals face one compass direction), and grid cells in the early 2000’s (i.e., neurons that are active when animals are in a grid of locations in space). A fundamental next step in our understanding of spatial cognition would be to describe the circuit-level interactions that give rise to such physiological activity patterns and to understand how such signals ultimately influence navigational behavior. We wish to leverage the advanced genetic, behavioral, anatomical and physiological tools in Drosophila, to achieve three broad goals. First, we wish to rigorously characterize neural circuits that explain how navigational signals are built. Second, we wish to improve the tasks that flies perform while we record from their brain, which will allow us to isolate cells and circuits required for the formation of spatial working memories. Third, we aim to reveal molecular, cellular and circuit mechanisms by which such memories are formed and guide behavior. This work should allow us to more rigorously link molecular factors, through their effects on cells and circuits, to their function in spatial-cognition. Our discoveries should ultimately help to inform how humans perform navigational tasks like driving home from work or finding a car in a parking lot, alongside how to approach neurological conditions in which such abilities are impaired, like in Alzheimer’s disease.

NIH Research Projects · FY 2026 · 2023-04

Project Summary Cancer cells require substantial antioxidant capacity to overcome toxic effects of reactive oxygen species (ROS). Despite their exclusive cytosolic production, cellular antioxidants are also abundantly present in organelles, particularly in mitochondria. In particular, mitochondria, as the source of reactive oxygen species (ROS), require substantial availability of antioxidants to protect its critical redox functions. While previous work suggests a link between mitochondrial redox metabolism and tumor growth, molecular mechanisms involved in maintaining mitochondrial redox homeostasis and the role of mitochondrial antioxidants in tumor progression are poorly understood. Among endogenous mitochondrial antioxidants, glutathione (GSH) is the dominant small molecule thiol, existing in millimolar concentrations. GSH is commonly upregulated in cancer cells, enabling metastatic colonization and resistance to chemotherapeutic drugs. In our previous work, using biochemical and proteomics methods, we identified SLC25A39, a mitochondrial membrane carrier of unknown function, to mediate GSH import into mitochondria of cancer cells. SLC25A39 loss strongly reduces mitochondrial GSH levels and its import, without impacting those in the cytosol. Our preliminary work suggests that SLC25A39 expression is associated with poor prognosis and decreased survival of breast cancer patients, and that SLC25A39 is necessary for breast cancer invasion and metastasis. This finding gave us the opportunity, for the first time, to uncouple mitochondrial redox metabolism from that of cytosol during tumor progression. Building upon this evidence, in this proposal, we will test the hypothesis that mitochondrial GSH import by SLC25A39 enables metastatic colonization of breast cancer cells. To address this, using biochemical and genetic experiments, we will first identify the precise mechanism by which mitochondrial GSH enables breast cancer cells to metastasize to lung. We will then test the role of mitochondrial GSH import by SLC25A39 in tumor formation and metastasis using mouse and human breast cancer models. Finally, SLC25A39 protein levels substantially increase during lung colonization, indicating a strong selective pressure to induce mitochondrial GSH uptake during metastasis. Therefore, we will determine how mitochondrial GSH availability and lung environment regulate SLC25A39 protein levels in breast cancer cells. This proposal will reveal the role of mitochondrial GSH homeostasis in breast cancer progression and will identify a compartmentalized sensing pathway essential for metastatic colonization.

NIH Research Projects · FY 2026 · 2023-04

Project Summary CRISPR-Cas systems provide bacteria and archaea with adaptive immunity against their viruses (phages). The hallmark of the CRISPR-Cas immune response is the acquisition of a “memory” of infection in the form of a short DNA sequence from the invading phage genome. This sequence, known as “spacer”, is integrated into the CRISPR locus of the host and then transcribed and processed into a CRISPR RNA (crRNA) guide. CRISPR-associated (Cas) effectors use the crRNA to recognize the nucleic acids of the invading phage through base-pair complementarity and trigger different defense strategies. For Type III CRISPR systems, commonly present in the human pathogen Staphylococcus aureus, target recognition leads to the dormancy of the infected cell, an event that prevents viral replication and propagation. Here, we propose to investigate a central, yet unanswered, question about this mechanism: if there are spacers that trigger a growth arrest in the host, how are they maintained in the bacterial community after they are acquired? Our central hypothesis is that the degradation of the phage DNA eventually eliminates the viral genome from the host, enabling growth and the fixation of the spacer in the population. To investigate this, we will explore several aspects of the Csm6-mediated response required for the defense mediated by type III CRISPR spacers that match late- expressed viral genes. First, we will define whether dormant cells eventually die or are able to exit this state, survive infection and continue growing. Second, we will determine whether spacers that match late-expressed phage genes can provide a selective advantage to the cell that harbors them, even when they trigger host dormancy. Third, we will determine if these spacers are actually acquired during infection. In all these experiments we will test our central hypothesis by using mutant staphylococci lacking in the expression of several nucleases to determine if they are required for the fixation of dormancy-triggering spacers. Finally, we will use a transposon library of mutants to investigate, in an unbiased manner, the impact of host genes that could be involved in the exit from dormancy. Our proposed experiments, aimed at understanding how spacers from dormancy-inducing CRISPR systems are fixed in the host population, will fill in a fundamental knowledge gap in our understanding of the hallmark feature of CRISPR immunity: the generation of a memory of infection. In addition, by directly addressing a fundamental mechanism of phage defense of staphylococci, our proposal can facilitate the success of phage therapies for the treatment of staphylococcal disease. In a more indirect manner, the characterization of the molecular mechanisms of type III CRISPR systems can lead to avenues to repurpose these immune systems for gene editing, particularly for the development of gene therapies to treat genetic diseases.

NIH Research Projects · FY 2025 · 2023-04

PROJECT SUMMARY/ABSTRACT Alphaviruses are arthropod-transmitted viruses that cause acute febrile illness in humans, with some cases causing long-term skeletomuscular and neurological sequelae. Currently there are no FDA-approved vaccines or therapeutics for alphavirus infection, and continued global outbreaks of chikungunya and Eastern equine encephalitis virus highlight the need for medical countermeasures. Effective antiviral treatments target key viral replication processes without host side-effect are built on the foundational knowledge of viral replication as evidenced by the rapid development of RNA polymerase and viral protease inhibitors towards SARS-CoV-2 during the COVID-19 pandemic. While alphaviruses have been studied for decades, there are many foundational gaps in our understanding of their biology. Like all viruses, alphavirus infection of host cells induces a multifaceted innate immune response, of which one result is the activation of the integrated stress response (ISR) through viral double-stranded RNA sensing by host protein kinase R (PKR) and subsequent phosphorylation of translation initiation factor 2 (eIF2a), causing shutoff of global protein translation. However, alphaviruses have evolved a mechanism by which their structural proteins are uniquely still efficiently translated from the viral subgenomic RNA (sgRNA) during ISR activation. While many canonical translation initiation factors have been ruled dispensable for sgRNA translation, the mechanism of sgRNA translation is yet to be described. Interestingly, the resistance of the sgRNA to the ISR is only observed in infected cells, suggesting viral factors are necessary. Other work has shown that viral nonstructural protein nsP3 interacts with components of stress granules, organelles formed during ISR activation to sequester translation machinery until relief of the stress and return to homeostasis, suggesting a role for nsP3 in modulating translation during alphavirus infection-induced ISR. Together, this leads to the hypothesis that alphaviruses evade the ISR through recruitment of non-canonical initiation factors or the ribosome directly, and that ISR resistance is mediated in part through interactions of the nsP3 and host proteins. Using Sindbis virus (SINV) as a model system, this proposal seeks to identify how alphaviruses evade the ISR to maintain translation of viral structural proteins with hopes of identifying druggable targets to prevent alphavirus replication and spread. Aim 1 seeks to identify translation initiation factors responsible for sgRNA translation during ISR activation and eIF2a phosphorylation. Aim 2 will address the role of viral factors, specifically the hypervariable domain (HVD) of nsP3, in modulating the ISR and sgRNA translation through use of deep mutational scanning methods. Successful completion of this proposed work will provide mechanistic insight for an under-characterized phenomenon observed for decades in alphavirus research, allowing for novel therapeutic interventions for alphavirus infections.

NIH Research Projects · FY 2026 · 2023-03

Project Summary Hepatitis B virus (HBV) remains a major global health problem and chronic HBV (CHB) is a major cause of liver cirrhosis and hepatocellular carcinoma. While antiviral therapies achieve long-term viral suppression, they can rarely clear the infection or achieve a state of functional cure where long-term viral suppression is maintained in the absence of treatment. Along with persistence of viral antigens, impaired HBV-specific immunity contributes to the chronicity of infection. Chronic exposure to high levels of HBsAg may render HBV- specific immune cells overly activated and functionally tolerized Thus, decreasing serum HBsAg could be a valuable therapeutic strategy, due to its potential to alleviate functional exhaustion and confer immune control. Passive transfer of antibodies is a potential strategy in CHB for their dual functionality. Antibodies differ from direct-acting antivirals in that they can recruit immune effector functions through their Fc domains to accelerate clearance of viruses and infected cells. In addition, immune complexes are potent immunogens that can foster development of host immune responses. HepB monoclonal antibody (mAb)19 is a human monoclonal antibody to the a-determinant of the extracellular loop of HBsAg and binds the major HBV serotypes. HepB mAb19 showed exceptional in vitro neutralization activity with IC50 in the nanogram range and in vivo antiviral activity in an animal model of infection. The object of this proposal is to conduct a first-in-human dose-escalation study of a long-acting variant of HepB mAb19 in individuals with CHB on antiviral nucleos(t)ide analogue (NRTI) therapy. The hypothesis to be tested is that the administration of HepB mAb19-LS during suppressive NRTI therapy will be safe and well tolerated, will lead to decreased levels of circulating HBsAg, and enhance host innate and adaptive immune responses to HBV.

NIH Research Projects · FY 2026 · 2023-03

PROJECT SUMMARY Cells perceive mechanical cues in their local environments, which must be converted into intracellular biochemical signals to modulate cellular physiology and control gene expression. There is increasing appreciation for mechanical signal transduction’s (“mechanotransduction”) critical role in development and its dysfunction in disease states such as cancer. However, in contrast to canonical signal transduction, cellular force sensing is poorly understood, hampering efforts to define mechanistically distinct mechanotransduction pathways, delineate their specific biological functions, and target them therapeutically. The actin cytoskeleton, a network of dynamic actin filaments, myosin motor proteins, and hundreds of associated factors, enables cells to mechanically interface with their surroundings. The cytoskeleton is classically understood to serve as a force generation and transmission apparatus that indirectly facilitates mechano- transduction through its physical linkages to membrane-anchored sites which mediate force signal conversion (e.g. cell-cell and cell-matrix adhesions). However, we and others have recently reported direct binding of soluble cytosolic proteins containing tandem arrays of LIM (LIN-11, Isl-1 & Mec-3) domains to tensed actin filaments, suggesting that the cytoskeleton itself may have the capacity to transduce forces into biochemical signals. Here I propose to test the hypothesis that force-activated actin binding by distinct LIM proteins is upstream of functionally discrete downstream mechanotransduction pathways. Through cellular assays and biophysical reconstitution, we will investigate how the representative force-activated actin binding LIM proteins zyxin (Aim 1) and FHL1/2 (Four-and-a-Half LIM domains 1/2, Aim 2) mediate distinct downstream functions in cytoplasmic cytoskeletal damage repair and nuclear gene expression regulation, respectively. We will then innovatively interface these approaches with cryo-electron microscopy (cryo-EM) to visualize force-activated actin binding by LIM proteins in structural detail (Aim 3). Our studies will establish how a conserved mechanism of force transduction through LIM domains is linked to distinct downstream signaling outcomes, which is likely to reveal general principles underlying the modular organization of cytoskeletal mechanical signaling networks. In the longer term, this work will enable precision dissection of context-specific biological functions of LIM proteins in vivo, facilitating rigorous evaluation of their potential as therapeutic targets.

NIH Research Projects · FY 2026 · 2023-02

5-methylcytosine (m5C) is an important RNA modification studied mostly for its role in tRNA biology. However, its roles in other aspects of RNA biology remain understudied. Our preliminary results, using bisulfite treatment of RNA followed by high-throughput sequencing, show that the genomes of many RNA viruses are m5C methylated in a site-specific manner, including Sindbis virus (SINV), chikungunya virus (CHIKV) and Coxsackievirus B3 (CVB3). The presence of m5C in diverse viruses, whose RNAs undergo many processes including translation, replication, transcription, and virion packaging, provides an attractive starting point for understanding the broader significance of this modification in regulating RNA function. A single dominant m5C site in SINV allowed us to generate an m5C-null mutant that exhibited cell-type dependent effects on virus replication. The host tRNA methyltransferase (MTase), NSUN2, which is important for host neuronal development and stem cell differentiation, appears to be the “writer” required for m5C modification of SINV. NSUN5, an MTase of ribosomal RNA is required for CVB3 methylation. We hypothesize that m5C plays a role in regulating viral RNA functions impacting virus-host interactions and viral life cycles. In three aims, using virologic, molecular, biochemical, high-throughput sequencing, and small animal model approaches: i) SINV will be exploited to learn how m5C is deposited and how it regulates RNA functions and viral infection and pathogenesis; studies of the related alphavirus CHIKV will allow conserved and virus-specific features to be uncovered, ii) how m5C regulates CVB3 RNA and what effects the modification has on virus replication and CVB3-associated myocarditis will be determined, and iii) SINV and CVB3 will be leveraged to characterize unknown functions of the NSUN2 and NSUN5 MTases, respectively, and as probes to discover novel m5C binding proteins that can exert their effect by direct binding (“readers”) or by removal of the m5C mark (“erasers”). This work will contribute to our understanding of human biology by revealing fundamental principles and functions of this widespread mark in the epitranscriptome with implications for its roles in maintaining cellular homeostasis. Since m5C methylation is a cellular process exploited by numerous viruses, this study could yield new targets for antiviral intervention.

NIH Research Projects · FY 2026 · 2023-01

Mechanosensitive (MS) channels sense and respond to mechanical forces by opening an ion-conducting pathway. MS channels are found in all kingdoms of life, and in humans play essential roles in a number of sensory processes, including hearing, the sense of touch, balance and regulation of blood pressure. The first MS channels likely evolved in early prokaryotes as protection from hypoosmotic stress. Because bacterial MS channels are ubiquitously expressed in bacteria, but not in humans, and because their uncontrolled opening has a deleterious and often lethal effect on the bacteria, presumably due to the loss of important metabolites, bacterial MS channels are intriguing targets for developing novel antibiotics. Bacteria express two types of MS channels, MS channels of large conductance (MscL) and MS channels of small conductance (MscS). Members of the MscL family are highly conserved and MscL has become a paradigm for the understanding of MS channels because of its simplicity and amenability to different experimental approaches. MscS channels are more diverse, and bacteria often express more than one paralog. Both bacterial MS channels are gated based on the ‘force-from- lipids’ principle and respond to the transmembrane pressure profile of the surrounding membrane. However, even though structures are available for MscL and MscS in different functional states, the mechanism by which membrane tension opens these channels has remained enigmatic. We have recently determined cryo-electron microscopy (cryo-EM) structures of MscS in different membrane environments, provided by nanodiscs, including one mimicking a membrane under tension. The structures, complemented by molecular dynamics (MD) simulations and electrophysiological studies, allowed us to visualize the channel in different functional states and to deduce what roles lipids associated with MscS play in mechanosensation. We will continue to use a combination of single-particle cryo-EM, patch-clamp electrophysiology and MD simulations to study the structure and gating of bacterial MS channels. In Aim 1, we will continue to explore the function of lipids in MscS function, in particular whether it adopts a defined open conformation in a native lipid environment, how modulators affect MscS by changing its lipid environment, and whether 16-carbon acyl chains play a specific role in MscS gating. In Aim 2, we will expand our studies to bacterial cyclic nucleotide-gated (bCNG) channels to elucidate how the MscS fold was adapted to make the channel respond to cAMP binding rather than membrane tension. Aim 3 will focus on MscL. We will determine the structure of MscL in a native lipid environment to confirm (or disprove) the existence of lipid-filled nano-pockets that were suggested to play a critical role in gating. Finally, we will determine the structure of MscL opened by different effectors to visualize the structure of this channel in the open state and to test our hypothesis that different effectors result in open conformations with different pore diameters. The results of these studies will not only provide new insights into the gating mechanism of bacterial MS channels, but also help in exploiting these channels for biomedical applications.

NIH Research Projects · FY 2025 · 2022-09

PROJECT SUMMARY Generation of high affinity antibodies in germinal centers (GCs) is a critical step in a wide variety of clinically relevant processes, from protection against pathogens by prior infection or vaccination to the development of allergies and autoimmune diseases. Antibody affinity maturation follows a prototypical Darwinian framework, in which GC B cells introduce random mutations into the antigen-binding portions of their immunoglobulin (Ig) genes, generating variations in affinity within their progeny. Rare B cells that acquire affinity-increasing mutation are then selectively expanded within the GC population, thus increasing the average affinity of GC B cells as a whole, in a process we refer to as positive selection. Despite decades of work, the precise cellular mechanisms of positive selection—in other words, how GCs “pick out” B cells with the highest affinity—remains a topic of debate. More than 10 years ago, we provided the first in vivo evidence in mice for a role for T follicular helper (Tfh) cells as arbiters of this selective process. In our model, Tfh cells would sense how much peptide a B cell could present on its surface (which in turn depended on the B cell’s affinity), providing help selectively to the highest-affinity B cells. However, despite accumulating functional evidence for this model, selective delivery of T cell help to B cells based on their affinity has never been directly demonstrated in physiological settings. To achieve this, we developed LIPSTIC, a method that allows us to directly record T cell help to B cells with great precision in vivo. In Aim 1 of this project, we propose to use LIPSTIC as a means to directly test the T cell help model in classic hapten-carrier induced GC selection models. In Aim 2, we will follow up on this by testing our findings from mouse LIPSTIC in human vaccine-induced GCs. In Aim 3, we use the original LIPSTIC in conjunction with two novel versions on this strategy to investigate the dynamics of multi-antigen driven selection in influenza-induced GCs.

NIH Research Projects · FY 2025 · 2022-09

ABSTRACT This proposal is a multi PI proposal that aims to understand how antiviral proteins limit host range. Prior work has demonstrated that type I interferon (IFN) induced proteins limit primate lentivirus replication in natural and non-natural hosts. Indeed, our previous studies have led to the discovery of antiviral interferon stimulated genes (ISGs), elucidated their mechanism of action and uncovered ways in which primate lentiviruses evolve evade or counteract antiviral proteins. Additionally, we have exploited this knowledge to generate novel chimeric viruses that better represent the HIV-1 strains circulating in humans for use in non-human primate models, including a minimally modified HIV-1 strain (stHIV) that can cause AIDS in pigtail macaques when CD8+ cells are transiently depleted during the acute infection. In this new proposal, we will explore the role of type I IFN and novel antiviral ISGs in limiting primate lentivirus replication in vitro and in vivo. Specific Aim 1 will explore the mechanism of action of antiviral ISGs affecting viral entry. Using a CRISPR screen, we have found novel ISGs whose expression appears to inhibit HIV-1 infection, specifically at the virus entry step. Knockout of one of these genes enhances HIV-1 infection in a manner that varies dramatically according to the particular strain from which the Env protein is derived. Importantly, the magnitude of effect of the ISG on HIV-1 infection mediated by a particular Env variant correlates with the sensitivity of a virus carrying that Env variant to inhibition by type I IFN. We will determine the molecular details of the how IFN/ISGs inhibits HIV-1 entry by elucidating viral and host determinants of inhibition, across cell types and species, and use imaging and other approaches to determine how inhibition affects incoming viron fate. We will also determine how novel ISGs contribute to the differential IFN sensitivity of primary transmitted founder and chronic HIV-1 strains and adapted/unadapted SHIVs. In Specific Aim 2, we will use newly developed, more effective methods to perturb type I IFN in macaques to determine the effect of IFN on primate lentivirus replication therein. In particular, we will determine the effect of type I IFN blockade on stHIV replication and clinical course in pigtail macaques. We will also use SHIV to determine whether HIV-1 Env variants that exhibit differential type I IFN sensitivity in vitro, also do so in vivo. Finally, we will determine how type I IFN affects SHIV dissemination and competitiveness use barcoded SHIVs, bearing IFN/ISG-sensitive and resistant HIV-1 Env proteins, defined in Aim 1, by determining how efficiently each SHIV disseminates in macaques in the presence and absence of type I IFN blockade.

NIH Research Projects · FY 2025 · 2022-09

Project Summary Arthropod-borne flaviviruses and respiratory-transmitted coronaviruses have the potential to cause severe epidemics and pandemics. One strategy to prepare for and respond to viral outbreaks is to develop drugs that target host factors viruses require to complete their lifecycles. Through a series of CRISPR/Cas9 gene disruption screens, we identified transmembrane protein 41B (TMEM41B) and the closely related vacuole membrane protein 1 (VMP1) as critical pan-flavivirus and pan-coronavirus host factors. Both proteins are highly conserved lipid scramblases with roles in autophagy. Our current model is that viruses from both the Flavivirdae and Coronaviridae families hijack TMEM41B and VMP1 for their ability to remodel ER membranes and induce membrane curvature to establish membrane-protected viral RNA replication organelles. Our overall goal for this proposal is to understand how, on a mechanistic level, both proteins support flavivirus and coronavirus infection. Our previous work indicates that TMEM41B is required at a post-entry step at or prior to viral RNA replication. In Aim 1, we will interrogate early events of the virus lifecycle including primary translation, polyprotein processing, and replication organelle formation in WT, TMEM41B and VMP1 knockout (KO) cells to determine how far the flavivirus and coronavirus lifecycles progress in the absence of either protein. We previously showed that lack of TMEM41B and VMP1, induces a heightened innate immune response upon flavivirus infection. We hypothesize that both proteins are recruited to sites of viral RNA replication, and that in their absence, RNA replication initiates and viral double stranded RNA (dsRNA) is produced. However, without a proper replication organelle dsRNA is exposed and triggers an innate immune response. Alternatively, given TMEM41B’s and VMP1’s lipid scramblase activity and function in lipid homeostasis, their absence may induce ER stress, which triggers an unfolded protein response (UPR) that in synergy with dsRNA may cause a heightened innate immune response. In Aim 2, we will test virus infection in double KO cells that lack either protein in addition to genes that are essential for pathogen sensing, IFN signaling, and UPR activation. We will further conduct RNAseq experiments to investigate lack of TMEM41B in stem cells and stem cell-derived primary-like cells representing different tissue lineages in the absence and presence of viral replication. Lastly, in Aim 3, will use a panel of phenotypic and mechanistic assays to characterize naturally occurring SNPs in TMEM41B that we previously found to impact flavivirus replication, and several reported VMP1 loss-of- function mutants. We will further take a deep mutational scanning approach to comprehensively characterize TMEM41B and VMP1 and determine if any domains or amino acids are differentially required for their cellular and proviral functions. This functional characterization will identify mutants that can be studied in detail in mechanistic assays and may identify amino acids or interfaces in both proteins that can be targeted to prevent virus infection with minimal disruption to cellular biology.

NIH Research Projects · FY 2026 · 2022-09

Project Summary: Alzheimer's Disease (AD) and AD-related dementias (ADRDs; e.g. Frontotemporal Dementia, Lewy Body Dementia, etc.) are crippling neurodegenerative disorders. The onset of these diseases is strongly correlated with aging. Cures for AD and ADRDs remain elusive; Alzheimer's has become the 6th most frequent cause of death in the USA. An emerging candidate contributor to Alzheimer's and ADRD is the L1 retrotransposon, which becomes dysregulated during aging and correlates with Alzheimer's / ADRD onset. One hypothetical mechanism by which L1 may contribute to Alzheimer's / ADRD is through its exacerbation of cellular senescence. Senescence is a phenomenon by which normal cells stop dividing; these cells accumulate with advancing age and are found at the locations of dysfunction in age-related diseases. In mice, senescent cells have been shown to shorten life and actively drive age-related neurodegeneration; preventing senescent cell accumulation decreases tau-dependent degeneration and cognitive decline. AD patients exhibit increased indicators of cellular senescence. It is increasingly clear that senescent cells are not inert, but instead drive tissue deterioration via the senescence-associated secretory phenotype - secreting a variety of growth factors and pro- inflammatory cytokines. L1 retrotransposons have recently been shown to drive progression of the senescence- associated secretory phenotype, and thus, L1 is an important agent of cellular senescence. L1 activation is also associated with AD-related Tau pathologies. The L1 encoded ORF2p enzyme (endonuclease and reverse transcriptase) is often flagged as a source of pathological cellular insults, e.g. via new, mutagenic L1 insertions and contributions to chromosomal instability. However, the effects of L1 expression extend beyond DNA damage. Numerous mechanisms may be at play, including the titration of normally homeostatic host factors away from their physiologic functions and into L1 ribonucleoprotein granules, as well as the production of immunity-and-inflammation-triggering cytoplasmic L1 DNA:RNA hybrids; indeed, the latter is now understood to be a key component of L1’s role in cellular senescence. Moreover, L1 also mobilizes Alu and other non- autonomous retrotransposons. Objectives: In Aim 1, we will use targeted mass spectrometry to profile L1 ORF1 protein expression in the cerebrospinal fluid (CSF) and post-mortem brain tissues of patients exhibiting AD/ADRD and cognitive decline, as well as in senescent cells and neuronal iPSCs; in Aim 2 (in vitro cell culture) and Aim 3 (clinical samples) we will profile L1-associated protein-protein and protein-RNA interactions in the same biological samples as Aim 1; and in Aim 4 we will take a candidate-based approach using molecular genetics to dissect the mechanisms of action of L1 in senescent cells.

NIH Research Projects · FY 2025 · 2022-09

Biology and Genetics of Metastatic Disease My laboratory studies the molecular alterations that contribute to metastasis formation, a poorly understood process and primary cause of solid cancer deaths. It has long been thought that metastasis is caused by somatic metastasis driver mutations—postulated alterations that have yet to be identified. By showing that levels of specific microRNAs become altered in metastatic tumors and identifying their target genes, my lab identified and characterized critical pathways and processes underlying metastasis formation. By studying germline variants of one such target metastasis gene, we discovered that metastatic potential can also pre- date tumor formation and be genetically inherited—revealing an unanticipated genetic underpinning for metastasis and opening up a new direction for the field. Specifically, we determined that two common human germline variants of the secreted glycoprotein ApoE promote or suppress metastasis in melanoma, with recent data suggesting this principle may apply to additional cancers. ApoE signaling was found to govern vascular and immune interactions as well as cellular invasiveness that collectively contribute to metastasis formation. These insights have significant translational potential and formed the basis of clinical trials that are providing proof-of-concept for ‘metastasis targeting therapy’, where multiple metastasis regression responses were observed in advanced stage patients for whom standard of care and immunotherapy treatments had failed. Going forward, we will use allelic variants of ApoE as powerful genetic entry points to understand the molecular events underlying metastasis formation, where we will define how ApoE signals are received by cells and how ApoE mediates intracellular events. We will also extend the concept of hereditary metastasis genetics to additional cancers and genes, applying our reverse genetic and mouse modeling approaches to breast and colorectal cancer metastasis. To achieve this understanding, we will employ innovative optical, physiological, genetic modeling and screening methods to interrogate mouse and human metastatic transitions. This award will enable our group to establish the first genetically guided framework for understanding the molecular mechanisms governing metastasis formation—enabling new avenues for its therapeutic treatment and prevention.

NIH Research Projects · FY 2025 · 2022-09

The cohesin complex is a major factor driving the 3-D organization of mammalian genomes at the scale of tens to kilobases to megabases. Recent single-molecule experiments have shown that it can extrude loops of DNA, which are an organizing principle of genome architecture. Together with CTCF, a DNA-binding protein that stalls cohesin’s translocation and defines loop boundaries, and several regulators of cohesin that promote loading, such as NIPBL, or release from chromatin, such as WAPL, cohesin defines interaction domains in chromosomes that affect patterns of gene expression during development and can lead to developmental diseases or cancer when disrupted. Although loop extrusion on naked DNA has been studied, cohesin in cells must navigate nucleosome-packed chromatin fibers that restrict access to binding sites on DNA, self-organize into compartments of similar epigenetic state that are independent of and compete with loop domains, potentially regulate DNA supercoiling, and are decorated with chromatin-associated RNAs. How cohesin and CTCF interact with chromatin in cells is the next frontier in understanding 3-D genome organization. Progress in this arena will require a multi-scale approach, with methods that probe both nucleosome-scale and megabase-scale features. I propose experiments to probe (1) how loop extrusion by cohesin perturbs the local structure of the chromatin fiber; (2) how the local structure of the chromatin fiber, modulated by depletion of linker histones and destabilization of nucleosomes, regulates cohesin’s ability to load and extrude loops; (3) how changes in the balance of supercoiling due to excess cohesin looping affect local nucleosome-nucleosome interactions; and (4) the chromatin-associated RNA interactome of CTCF at RNA-dependent and RNA-independent loop boundaries. To dissect the specific effects of cohesin, its regulators and the chromatin fiber, we will use a combination of stable protein depletion, acute degradation, and pharmacological inhibition in human and mouse cell lines. We will read out changes in chromatin fiber structure and chromatin-associated RNAs using RICC-seq, a method I recently developed for measuring DNA-DNA contacts at sub-nucleosome resolution in intact cells, and using novel technology development to probe the chromatin-associated RNA interactome of specific proteins. These methods will be combined with more established epigenome and transcription profiling tools and with coarse- grained simulations to develop and test multi-scale models for the interaction of loop extrusion machinery with the chromatin fiber. I anticipate that the results of these experiments will shed new light on how loop extrusion and chromatin’s self-association interact in specific contexts, which models for cohesin’s engagement with DNA are relevant to loop extrusion, how supercoiling is disseminated across chromosomes, and how the local molecular context defines loop boundaries. This knowledge may reveal new strategies for compensating transcriptional dysregulation due to mutations in cohesin or its regulators using targetable factors that regulate the chromatin fiber, cohesin’s native substrate.

NIH Research Projects · FY 2024 · 2022-09

PROJECT SUMMARY/ABSTRACT The methyl-CpG binding protein 2 (MeCP2) is a highly abundant chromatin-binding protein that recognizes methylated DNA to coordinate the expression of thousands of genes essential for neuronal function. Mutations in MeCP2 cause Rett syndrome, a severe neurological disorder affecting one in every 10,000 females and characterized by psychiatric and motor regression at 6-18 months. Although several treatments are available for improving some isolated features of Rett syndrome, there are currently no approved therapies that directly address the underlying defects of MeCP2 loss of function. One primary reason for the lack of MeCP2-targeted interventions is an inadequate understanding of how MeCP2 reads methylated DNA within hierarchically organized chromosomes. MeCP2 is known to bind to methylated DNA with a higher affinity than unmethylated DNA, but how it navigates the main structural form of packaged DNA in the nucleus, nucleosomes, to reach its canonical methyl-DNA substrate remains unclear. Over 80% of the MeCP2 protein is structurally disordered, resulting in extensive conformational heterogeneity and binding plasticity that could underlie the regulatory capacity of the protein, but simultaneously hamper detailed mechanistic characterization of the protein’s chromatin-binding behavior. To this end, I used a single-molecule platform combining fluorescence microscopy with optical trapping to directly observe the real-time trajectory and dynamics of individual MeCP2 on DNA and nucleosomes. This approach enabled the visualization of how MeCP2 navigates the chromatin landscape harboring methylated CpG sites. I discovered that MeCP2 exhibits long-range, diffusive behavior on bare DNA, whereas methylation drastically suppresses such motions. Unexpectedly, I also found that MeCP2 preferentially and stably binds nucleosomes irrespective of methylation status, suggesting that nucleosomes may serve to regulate the availability of MeCP2 for its canonical methyl-reader activity. Based on my preliminary data, I hypothesize that nucleosomes regulate the availability of MeCP2 for methyl-DNA recognition and biophysically modify MeCP2 dynamics and binding configurations on DNA. To test this hypothesis, I will 1) characterize the dynamics of MeCP2 on DNA with and without methylation, 2) examine the binding and biophysical interaction of MeCP2 with nucleosomes wrapped with methylated or unmethylated DNA and investigate how Rett mutations impact this interaction, and 3) determine the structural basis of the MeCP2-nuclesome interaction. Successful completion of the proposed aims will provide novel insights into how MeCP2 biophysically navigates through a complex chromatin environment to reach its canonical methyl-DNA substrate at unprecedented spatial and temporal resolution. My report will establish an experimental framework for systematic interrogation of Rett syndrome mutations and their impact on fundamental MeCP2-nucleosome and DNA interactions and dynamics.

NIH Research Projects · FY 2025 · 2022-08

Project Summary Ribosome and proteomic profiling have revealed a large number of small translated open reading frames (ORF) within previously described “untranslated regions” (UTRs) and long non-coding RNAs. While some of the small ORFs depend on the encoded peptide to function in various fundamental processes (e.g., development). Translation of small ORFs in the 5’UTR, known as upstream-ORFs (uORFs), usually represses gene expression, independent of the encoded peptide. Small ORFs have also been reported in 3'UTR, termed downstream ORF (dORF). However, the dORF function and their relationship to human health and disease remain unknown. I characterized dORFs from human and zebrafish using ribosome profiling data. My preliminary data indicates, contrary to uORFs, translation of dORFs (small ORF in the 3’UTR) strongly enhances translation of the canonical ORFs and remains an uncharacterized regulatory mechanism across vertebrates. The objectives are: 1) Dissect at the single molecular level how dORF enhances translation of the canonical CDS. And 2) Determine whether alternative polyadenylation in cancer influences dORF regulation to cause cancer. The rationale for the proposed research is to gain a mechanistic understanding of dORF-mediated regulation and to assess the possible biological importance of dORF dysregulation under disease conditions (e.g. Cancer). This proposal is conceptually innovative as it is based on the exploration of a novel, yet widespread translation regulatory mechanism conserved across vertebrates. Technically, this proposal will combine single molecular imaging, genomic profiles (RNA-seq, Ribosome profiling), and reporter approaches in different human cell lines (including cancer cells) and published patient data. The outcomes from this project will emphasize the role of ribosome as a master gene expression regulator, and shield light on the importance of small ORFs. This translation kinetics work about dORF will provide critical insights into the molecular mechanism of this uncharacterized regulatory pathway. Exploring dORF dysregulation in cancer due to APA will highlight the mRNA itself as disease driver even without any mutation in DNA, and it also indicates possible clinical impact of dORF to detect and even cure cancer. My long-term interest is to study gene expression dysregulation in cancer. This training award will increase my knowledge background of cancer biology, molecular biology and bioinformatics. It will also promote the technical training of single molecular imaging, ribosome profiling, cell biology assays for cancer. Overall, this proposal will help me for future independent cancer molecular/genomic career.