Rockefeller University

NIH Research Projects · FY 2026 · 2026-06

Project Summary The long-term goal of this MOSAIC Pathway to Independence proposal is to understand and characterize the mechanisms by which episomal DNA transcription is silenced and promoted through the lens of hepatitis B virus (HBV). HBV establishes and maintains chronic infection via the persistence of its covalently closed circular DNA genome (cccDNA), and cccDNA serves as the template for all viral mRNAs. Many studies of HBV replication have identified HBV’s regulatory X protein (HBx) as a key viral protein involved in the promotion of viral transcription and numerous other steps in HBV’s life cycle. However, how HBx, the smallest HBV protein, achieves its many functions necessary for HBV replication remains largely unknown. During my postdoctoral training, I have applied and expanded the capabilities of a new RNA-based method for the study of HBV in cell culture, identified host restriction factors that are antagonized by HBx, and developed technologies that could enable comprehensive, low-biased studies of HBx variants. To this end, in the mentored phase of this K99 award, I will obtain a mechanistic understanding of how cccDNA transcription is regulated by a complex network of host proteins and interactions with HBx. This will be achieved through detailed functional characterization of non- SMC element 3 (NSE3) and mitochondrial topoisomerase 1 (TOP1MT), host factors identified as specifically disrupted by HBx. In addition, I will receive training in deep mutational scanning (DMS) to uncover the genetic basis for HBx’s promotion of viral RNA expression and CRISPR interference (CRISPRi) methods to uncover new host factors involved in episomal circular DNA transcription that were previously missed by gene knock-out approaches. Additional aspects of my comprehensive career development plan include attending workshops and seminars on the topics of project management, mentorship, and grant writing. The training phase will be carried out in the laboratory of Dr. Charles Rice at The Rockefeller University (RU), one of the world’s experts in virology and host-pathogen interactions. A critical component of career development will be the close counsel of a highly experienced Advisory Committee, comprised of Drs. Charles Rice, Robert Roeder, and Jeremy Rock (all RU), and Xiaolan Zhao (Memorial Sloan Kettering Cancer Center). The innovative techniques and comprehensive datasets generated in the K99 training phase will establish a detailed understanding of HBx and the host gene network that promote or silence HBV cccDNA and more importantly contribute to viral persistence. This award will be key to my success as a young independent investigator in a highly competitive field. Altogether, the training will help me achieve both my short-term goals of expanding my scientific expertise and systems-biology experimental toolkit and my long-term goal to become an independent investigator at the intersection of systems biology, viral genomics, and host-virus interactions. This proposal will contribute significant insights into our basic understanding of HBV transcription and maintenance of chronic infection and could ultimately serve as the basis for identifying new curative therapies for HBV.

NIH Research Projects · FY 2026 · 2026-06

ABSTRACT The latent HIV-1 reservoir that persists despite the effect antiretroviral therapy (ART) comprises infected cells with intact integrated proviruses that are transcriptionally near silent. The existence and maintenance of this reservoir in essentially all ART-treated individuals is a major barrier to curative interventions in HIV-1 infection, and necessitates the lifelong administration of ART. In the preliminary studies that underpin this proposal we developed a xenograft based model system in NSG mice that reliably generates populations of human primary memory CD4+ T-cells that contain latent proviruses carrying reporter genes. Our methods overcome some of the key limitations of existing approaches to study HIV-1 latency. We determined hundreds of HIV-1 integration sites in active and latent cell populations and uncover relationships between provirus genomic location, infected cell clonal expansion, and the establishment of HIV-1 latency after engraftment. Uniquely, we have recovered and cultivated in large numbers primary memory CD4+ T cell single-cell clones each harbouring a unique HIV-1 integration site in which latency was (or was not) established in vivo. Comparison of these clones with the small number of CD4+ T cell clones that harbor intact HIV-1 proviruses that have been cultivated from ART-treated individuals reveals features in common that suggest that the model system we have developed more accurately recapitulates features of HIV- 1 latency than those deployed heretofore. In Aim 1, we will build on our preliminary studies by deriving a diverse collection of primary CD4+ T-cell single cell clones that harbor latent or active HIV-1 proviruses. Therein, we will investigate the stability, dynamics an mechanisms of transitions between active and latent states at defined integration sites and, whether the new DNA synthesis that accompanies cell proliferation enables these transitions. We will determine relationships between the epigenetic profile of loci containing proviruses and whether proviruses are transcriptionally active or latent. We will further assess how the physiologic status of T-cell clones conspires with epigenetic profile of loci containing proviruses to impact the establishment of latency and transitions between active and latent states. In Aim 2 we will manipulate epigenetic modifications at or surrounding the site of provirus integration in the genome of primary memory CD4+ T-cell clones and determine their effects on HIV-1 latency. We will build a custom, targeted sgRNA library, focused on epigenetic modifiers and gene expression regulators and identify genes that affect HIV-1 latency in various genomic contexts. We will further conduct genome wide CRISPR screens in primary memory CD4+ T-cell clones to identify genes that regulate latency therein. Ultimately, we will aim to provide mechanistic insights into HIV-1 latency and evidence about how manipulation of latency might be accomplished, such that HIV-1 infection might be cured.

NIH Research Projects · FY 2026 · 2026-06

Project Summary Antimicrobial resistance is a growing global threat that is predicted to kill more people than cancer by 2050. The rise in pathogens resistant to our current arsenal of antibiotics necessitates the development of a new generation of therapeutics. Bacterial natural products have historically been the most productive source of mechanistically and structurally novel antibiotics, serving as lead structures for the development of numerous clinically approved therapeutics. It is well established that most of the bacterial biosynthetic diversity on Earth remains locked in microbial dark matter (i.e., unstudied environmental bacteria) that is inaccessible to natural product discovery using existing methods. The inefficiency of previous discovery pipelines results from both the inability to culture most bacteria and the fact that most natural product biosynthetic gene clusters remain silent under laboratory fermentation. In the studies outlined in this proposal, I will couple our new Direct Nanopore Metagenomic Sequencing methods and our Synthetic Bioinformatic Natural Products Discovery methods to overcome both bottlenecks to identify previously inaccessible nature-inspired antibiotics. Direct Nanopore sequencing to acccess previously inaccessble metagenomic biosynthetic gene clusters: One way to overcome issues with culturing bacteria is to sequence DNA directly from environmental samples and identify novel biosynthetic gene clusters from this data. Until now, metagenomic sequencing resulted in assemblies that were too short to fully capture most biosynthetic gene clusters. We have recently developed a robust, direct Nanopore metagenomic sequencing pipeline that readily assembles metagenomic contigs large enough (>100 Kb) to identify complete natural product biosynthetic gene clusters. Accessing antibiotics from metagenomic biosynthetic gene clusters using synthetic bioinformatic natural products: Existing methods for “decoding” biosynthetic gene clusters use biological processes (transcription, translation, biosynthesis). Unfortunately, most biosynthetic gene clusters remain silent under laboratory fermentation. To help address this bottleneck, we have developed a “biology free” biosynthetic gene cluster decoding process where bioinformatics is used to predict the structure encoded by a gene cluster, and then chemical synthesis is used to generate the predicted structure to give compounds we have called synthetic bioinformatic natural products (synBNPs). In our preliminary studies, synBNP antibiotics, like natural products themselves, regularly had novel or rare modes of action and in vivo activity. Coupling these two methods allows us to identify novel compounds from microbial dark matter that has, until now, remained inaccessible using previous natural product discovery methods. This is designed to be an antibiotic discovery proposal. The final output will be a collection of novel, naturally inspired antibiotics that are ready to enter downstream studies to evaluate their potential as clinical candidates.

NIH Research Projects · FY 2026 · 2026-06

PROJECT SUMMARY Mycobacterium abscessus (Mabs) is an increasingly prevalent cause of chronic lung infections, especially among individuals with cystic fibrosis (CF), bronchiectasis, or other structural lung diseases. Treatment outcomes for Mabs are dismal: current regimens are prolonged, poorly tolerated, and only modestly effective. Mabs shows high levels of intrinsic antibiotic resistance, and therapeutic development has lagged far behind progress made for Mycobacterium tuberculosis. This project aims to accelerate the discovery of condition- specific vulnerabilities in Mabs using genome-scale CRISPR interference (CRISPRi) to systematically quantify gene vulnerability under physiologically relevant conditions. We have developed and optimized CRISPRi platforms for Mabs and constructed the first genome-wide CRISPRi library in this pathogen. Using this platform, we will (1) define gene vulnerabilities across three distinct growth conditions, including a validated synthetic CF sputum medium; (2) validate high-priority condition-specific targets and investigate their mechanisms of action; and (3) perform genome-scale CRISPRi screening during macrophage infection to identify genes required for intracellular survival. Our approach will uncover how host-like infection conditions reshape Mabs physiology and gene vulnerability, define new therapeutic targets, and establish broadly applicable methods for functional genomic analysis in Mabs and other nontuberculous mycobacteria. Ultimately, this work will lay the foundation for future drug discovery efforts aimed at a pathogen of rising clinical importance.

NIH Research Projects · FY 2026 · 2026-05

Project Summary Abstract There is increasing awareness of the role of thromboinflammation mediated by the activation of αIIbβ3 through the binding of immune complexes to FcγRIIa in the pathogenesis of a variety of serious disorders, among which are heparin-induced thrombocytopenia/thrombosis, viral infections including COVID-19, autoimmune disorders, and chronic inflammatory disorders, including cardiovascular disease. The primary mechanism proposed to underly the thrombotic phenomena is the binding of immune complexes to the FcγIIa receptor (FcγRIIa) on platelets, leading to activation of αIIbβ3 and platelet aggregation. Both infection and autoimmune diseases can generate immune complexes that interact with FcγRIIa. Activated αIIbβ3 interacts with soluble fibrinogen, leading to platelet aggregation, thrombus formation, and thrombin generation. We and others have found that highly-specific BTK inhibitors (BTKi) can inhibit FcγRIIa-mediated αIIbβ3 activation and confirmed that platelets of XLA patients, who lack BTK on a genetic basis but do not have a bleeding disorder, do not aggregate in response to engagement of FcγRIIa. Thus, the central hypothesis of this grant renewal is that the activation of αIIbβ3 by engagement of FcγRIIa plays an important role in the pathophysiology of many immune-mediated disorders. We propose to create better murine models of these disorders by using mice that have had their entire Fc receptor repertoire and their platelet factor 4 (PF4) genes humanized and then test in these models the safety and efficacy of already-approved BTKi and antiplatelet drugs, as well as new agents targeting FcγRIIa directly that are developed under this grant. Our goal is to improve our understanding of these disorders, improve their current therapy, and develop novel, orally active drugs that directly target FcγRIIa for chronic therapy. Specific aim 1: A) To better understand the pathogenesis of the FcγRIIa-mediated disorders of αIIbβ3 activation by developing optimal murine models in which the entire murine repertoire of Fc receptors has been humanized, along with PF4, and both the Arg131 and His131 variants of FcγRIIa are studied. B) To compare the effects of antiplatelet agents (aspirin, P2Y12 inhibitors, αIIbβ3 inhibitors), and inhibitors of immune complex binding to FcγRIIa or signaling through FcγRIIa, initially focusing on BTKi, on the pathogenesis of the disorders in the murine models. Specific aim 2. To develop a rapid functional diagnostic assay that measures the combined effects of platelet FcγRIIa number, FcγRIIa affinity for the Fc domain of immunoglobulins, and the downstream signaling pathways to platelet αIIbβ3 activation for screening compounds, individual risk assessment, and optimizing therapy by monitoring drug effects. Specific aim 3. To use both high-throughput and in silico screening, along with deep learning-based de novo design, to identify novel specific inhibitors of immune complex binding to FcγRIIa or signaling through FcγRIIa as potential human therapeutics.

NIH Research Projects · FY 2026 · 2026-04

Cells perceive mechanical cues in their local environments, which must be converted into intracellular biochemical signals to modulate cellular physiology and control gene expression. This process of mechanical signal transduction (“mechanotransduction”) is critical for development and frequently dysfunctions in disease states such as cancer. Despite increasing appreciation of its importance for human physiology, the molecular mechanisms of mechanotransduction remain poorly understood, hampering efforts to define mechanistically distinct mechanical signaling pathways, delineate their specific biological functions, and target them therapeutically. The actin cytoskeleton, a network of dynamic actin filaments (F-actin), myosin motor proteins, and hundreds of associated factors, enables cells to mechanically interface with their surroundings. While the cytoskeleton is classically understood as a force generation and transmission apparatus that indirectly facilitates mechanotransduction, our research has provided evidence that actin filaments can also serve as direct molecular force transducers. By developing innovative cryo-electron microscopy (cryo-EM) sample preparation and machine-learning based computational analysis approaches, we have uncovered multiple classes of force- dependent structural transitions in F-actin (elicited by fluid flow and myosin molecular motor forces) that can be discriminated by force-sensitive actin-binding proteins. This work has provided a first direct glimpse at how forces alter protein structure to regulate function. Using biophysical reconstitution and cell biology studies, we have also shown how force-activated F-actin binding by proteins from the LIM (LIN-11, Isl-1 & Mec-3) domain superfamily can coordinate downstream mechanotransduction processes, including repair of physical damage to actin- myosin cables mediated by zyxin and extracellular matrix stiffness-dependent nuclear localization of Four-and- a-Half LIM domains (FHL) transcriptional control proteins. Additionally, we have developed cryo-EM and cryo- electron tomography (cryo-ET) approaches for visualizing crosslinking proteins bridging actin filaments, setting the stage for studying how forces alter the geometry and composition of higher-order cytoskeletal networks in atomistic detail. To build upon this progress, we will now study the structural mechanisms and cellular functions of a ubiquitous class of force-stabilized protein-protein interactions known as catch bonds, which are prominent in cytoskeletal crosslinkers and cell adhesion proteins implicated in mechanotransduction. We will also probe the structural basis of force-activated cytoskeletal engagement by LIM proteins, focusing on both individual actin filaments and cytoskeletal networks in vitro and in cells, as well as scrutinize the nuclear localization and gene regulatory mechanisms of FHL proteins. In the near term, our efforts will provide detailed mechanistic insights into mechanical signaling pathways, facilitating precise dissection of their functions in vivo. In the longer term, our work may guide the development of chemical probes and therapeutics which selectively target mechanotransduction.

NIH Research Projects · FY 2026 · 2026-02

Project Summary Hepatitis B virus (HBV) chronically infects more than 250 million people and puts them at risk of developing liver cirrhosis and hepatocellular carcinoma. Every year over 800,000 people die due to HBV-related complications. The natural history of chronic HBV infection goes through several phases characterized by the presence of HBV e antigen (eAg) in serum. Most chronically infected individuals are initially eAg positive and later in life become eAg negative, which is associated with enrichment of viral mutations in the basal core promoter (BCP) and precore (PC) regions. Some individuals spontaneously control HBV and lose surface antigen, termed a functional cure. There are two approved types of medications for chronic HBV infection, nucleoside analogs (NA) and interferon-alpha (IFNα). While NA are well-tolerated, work across genotypes and irrespective of eAg status, they rarely lead to a functional cure. By contrast, IFNα therapy can functionally cure a subset (<10%) of patients but comes with significant adverse effects. It has long been appreciated that IFNα responses differ by eAg status and by HBV genotype. eAg positive patients are cured at a higher rate than eAg negative patients, and HBV genotype A (HBV-A) is more susceptible to IFNα than HBV-D. In a recent North American trial studying the effects of NA withdrawal after four years of therapy a small number of eAg positive HBV subgenotype A2 (HBV- A2) patients responded very well, with 67% functional cure rate in the monotherapy group and 100% cure in the NA + IFNα group. This contrasted with HBV-A2 eAg negative and HBV-A1 or HBV-D patients, none of whom where cured. These results suggest that eAg positive HBV-A2 is highly susceptible to currently approved therapies and offers an opportunity to understand the mechanisms by which IFNα leads to a cure. IFNα has broad antiviral mechanisms and can induce hundreds of human interferon-stimulated genes, depending on the pathogen and cell type. The antiviral mechanisms of IFNα against HBV are poorly understood. They involve the induction of interferon-stimulated genes in hepatocytes, the liver cell type infected by HBV, as well as effects on immune cells. We here aim to use eAg positive HBV-A2 to understand the antiviral mechanisms of IFNα in human hepatocytes. Our recent advances in launching HBV infection from recombinant DNA and our primary human hepatocyte cell cultures and animal models can now be exploited and combined with multiomics approaches and novel CRISPR validation tools. These systems will be used to investigate the antiviral mechanism of an interferon-stimulated host gene against HBV-A2 and other genotypes, map regions of patient- derived HBV-A2 genomes conferring susceptibilty to IFNα, and compare IFNα responses between isogenic wild type and PC and BCP region HBV-A2 mutants. In all, these studies combine expertise from multiple fields with the goal of advancing insights into the antiviral mechanisms of IFNα, which may aid in the development of more effective therapies against chronic HBV.

NIH Research Projects · FY 2025 · 2025-09

Project Summary Simultaneously monitoring neuronal activity in large populations with single-neuron resolution in vivo is critical for understanding the mechanisms underlying brain function and the generation of flexible, adaptive behavior. Genetically encoded voltage indicators (GEVIs) offer a promising approach for optically capturing electrical membrane potentials, enabling spatially resolved recordings with genetic specificity. However, both the intrinsic properties and the current limitations of GEVIs present significant challenges for the design and implementation of optical systems, particularly of two-photon (2P) imaging systems, that are maximally adapted to GEVIs. These include their millisecond response times, membrane localization, low brightness, and limited signal-to-noise ratio (SNR). Consequently, current voltage imaging techniques have been restricted to small neuron populations, superficial cortical layers, or sparse labeling conditions. In our previous work on calcium imaging, we demonstrated mesoscale volumetric imaging of up to one million neurons in the mouse cortex by implementing the Many-fold Axial Multiplexing Module (MAxiMuM). This project extends that conceptual framework to voltage imaging while incorporating a novel spike detection algorithm. The proposed optical platform is designed in a principled manner to deliver a highly optimized 2P multiplexing system that meets the unique requirements of GEVIs in terms of temporal, spatial, and energy efficiency. This system will drive iterative development of voltage indicators with tailored kinetics, while its versatility enables efficient in vivo screening and diverse biological applications. Among its capabilities, the system will allow imaging from fields of view several hundred microns wide at 750 Hz and depths of ~500 μm. Smaller FOVs will be achievable at higher speeds (up to 2 kHz) which will enable detection of single spikes within bursts. A high-sensitivity mode for detecting sub-threshold potentials will also be implemented. By combining this platform with an innovative excitation strategy and the co-development of advanced GEVIs, we aim to achieve targeted multi-plane population voltage imaging of large populations of neurons in the mouse brain.

NIH Research Projects · FY 2025 · 2025-09

ABSTRACT The proposed project involves focused construction to expand the advanced fabrication facility and capabilities at The Rockefeller University in Manhattan, New York City. The Precision Instrumentation Technologies (PIT) Resource Center provides a centralized and expertly staffed prototyping and fabrication facility that supports more than 50 principal investigators at The Rockefeller University (RU). This project seeks to expand and enhance our design and fabrication capabilities. The project goals are to: (1) Renovate and annex adjacent space to the current PIT to accommodate additional instrumentation, activities, and personnel; (2) Undertake elevator removal and rigging required to add institutionally funded fabrication equipment (CNC mill, CNC lathe, water jet cutter); and (3) Create a new microfabrication facility.

NIH Research Projects · FY 2025 · 2025-09

PROJECT SUMMARY Dysregulation of energy storage in white adipocytes and of energy expenditure in brown or beige adipocytes, respectively, cause lipodystrophy and obesity, which lead to metabolic disorders that include insulin resistance, diabetes, and cardiovascular disease. A favorable metabolic state is achieved through storage of excess energy as lipids by white adipose and dissipation of energy as heat by brown adipose, as well as by conversion of white adipose tissue to brown-like beige tissue. Studies, including our own, have established that adipose tissue physiology is largely regulated at the level of transcription by a web of interactions between nuclear receptors (NRs) that include PPARs and ERRs and their associated coactivators, including the multi-subunit Mediator complex, that are recruited to enhancers of adipocyte-specific genes. Remarkably, our biochemical and genetic studies in mice collectively demonstrate conditional requirements for the two NR-interacting NR boxes of the MED1 subunit of the Mediator and consequent dramatic beneficial effects of attenuating NR-Mediator interactions via the NR boxes. However, the detailed mechanisms underlying these observations remain completely unknown. We hypothesize that differential recruitment of cofactors by unique combinations of NRs and other transcription factors bound to target gene enhancers determines their adipocyte-specific spatiotemporal expression patterns. Our hypothesis further posits that Mediator recruited by these factors is the critical node in establishing the locus-specific higher-order complexes that ultimately determine the amplitude of the transcriptional output. Here, we propose a holistic approach that merges biochemical and cell- and animal- based methods to investigate how adipocyte identity arises from combinatorial effects of enhancer-bound NRs and other transcription factors as channeled through the Mediator. Our specific aims therefore are (i) to uncover using in vitro transcription, cross-link mass spectroscopy, multiomics, and rapid degradation approaches how cooperative interactions of PPARg, C/EBP pioneer factors, Mediator, and other coactivators at adipocyte lineage- specific enhancers give rise to a transcriptional program that establishes white and brown adipocyte identity, as well as how additional interactions among ERRg, NCOA3, and Mediator are specifically responsible for function of brown/beige adipocytes; and (ii) to analyze how transcriptional activation proceeds in mature adipocytes, both when Mediator recruitment to the enhancer is mediated by MED1 NR boxes and when alternate pathways are operative. The underlying cellular and molecular mechanisms of the metabolically favorable phenotypes of mice harboring MED1 NR box mutants under obesogenic conditions will also be investigated. Projected research outcomes thus include both a detailed understanding of enhancer-mediated transcriptional mechanisms that regulate the formation, interconversion, and function of different adipocyte cell types, as well as insights into rationally developing therapeutic modalities for treating metabolic diseases.

NIH Research Projects · FY 2025 · 2025-09

Spatial and temporal regulation of RNA is key to cellular homeostasis and can become disrupted in disease. The lifecycle of RNA can include splicing, modification, transport, and translation, which all require RNA-binding proteins (RBPs), a class of proteins that bind and regulate RNAs. Specifically in the brain, there are many neuronal and non-neuronal cell types that are each comprised of distinctive sets of transcripts with unique RNA regulatory machineries. While there are genetic tools in mice to parse out cell type-specific RNA regulation, a parallel tool does not currently exist to study this in human postmortem brain tissue, limiting the ability to study human diseases. We have developed crosslinking immunoprecipitation-Fluorescence Activated Nuclei Sorting (CLIP-FANS) which will allow us to overcome this hurdle and for the first time, identify cell-type specific RBP mediated RNA regulation in postmortem samples. CLIP-FANS builds on two previous methodologies: fluorescence-activated nuclei sorting (FANS) to achieve cell type specificity, followed by high-throughput sequencing of RNA isolated by crosslinking immunoprecipitation (HITS-CLIP) to analyze gene regulation in sorted populations. Here, we have built an ordered approach and show preliminary data of the similarities and differences in the Hu-binding between excitatory and inhibitory neurons in the human entorhinal cortex in the first iteration of CLIP-FANS. One neurodegenerative disorder with dysregulation of RNAs and RBPs in Alzheimer’s disease (AD), a highly prevalent disease and the leading cause of dementia. For therapeutic interventions, maximum benefit would be achieved by targeting the first brain regions affected in AD, such as the entorhinal cortex, which has few studies on gene expression changes and none on gene regulation changes. Therefore, a combined approach of RNA sequencing of sorted cell types (FANSseq) and CLIP-FANS will generate cell-type specific gene expression and regulatory data of the differences occurring in AD to identify potential targets for therapeutic intervention.

NIH Research Projects · FY 2025 · 2025-09

Project Summary Metabolism encompasses all chemical reactions within a cell. Solute carriers (SLCs), a family of 450 metabolic transporters, play a crucial role in these processes by controlling intracellular and organellar metabolite availability. Dysfunction of these transporters can lead to homeostatic disturbances, contributing to progression of various diseases. Despite extensive research, approximately 30% of SLC family members remain orphan, or lacking annotated substrates. Identifying the physiological substrates of SLCs remains challenging for various reasons that include promiscuity, structural diversity, compartmentalization, and redundancy. Addressing this problem necessitates the development of unbiased, systematic, and efficient methods. Recent Genome-Wide Association Studies (GWAS) of the human metabolome have revealed significant genetic influences underlying blood chemical composition providing an opportunity for linking the genes encoding for transporters to their substrate. This approach has previously identified SLC22A1 as an acylcarnitine carrier and SLC2A9 as a urate transporter. Earlier in my thesis, I leveraged such datasets to develop large-scale platforms that led to the discovery of FLVCR1 and SLC25A48 as plasma membrane and mitochondrial choline transporters. This proposal aims to develop advanced methods for deorphanizing metabolic transporters and uncovering their roles in cellular and organismal metabolism. Leveraging machine learning-driven analysis, Aim 1 will comprehensively investigate the function of SLC25A45 as a methylated basic amino acid transporter, emphasizing its impact on renal physiology. Aim 2 will integrate diverse multiomic databases to not only deorphanize metabolic transporters but also provide an insight into their roles in organismal metabolism, with a particular focus on kidney function. Successfully completing these studies will yield an innovative tool for data integration and serve as a valuable resource for advancing biological discoveries related to metabolism and renal physiology.

NIH Research Projects · FY 2025 · 2025-08

PROJECT SUMMARY/ABSTRACT B cells participate in germinal center (GC) reactions to generate high-affinity antibodies, which represent a crucial line of protection from viruses and other pathogens. To date, the mechanism by which the highest-affinity B cell clones are selected remains elusive. Vital to the development of these high-affinity clones are T follicular helper (Tfh) cells, which stimulate B cells to undergo GC selection through an interaction known as T cell help. While our understanding of T cell help dynamics in single-antigen GCs has significantly increased over the last few years, less is understood about the distribution of Tfh cell help in inflammatory conditions such as viral infections or cancer. Importantly, GC-like clusters referred to as tertiary lymphoid structures (TLSs) have been a subject of recent interest, as chronic respiratory viral infections and many cancers and are accompanied by TLS formation. Notably, the occurrence of TLSs in cancer patients correlates – with some exceptions - with increased responsiveness to immunotherapy. However, most studies are limited to describing the appearance of TLS and how their formation is correlated with patient outcome, rather than providing mechanistic details about their anti- tumoral effects. The F99 phase of this proposal is focused on determining how T cell help varies and is distributed among different viral protein-specific B cell clones co-residing in GCs and TLSs using intercellular labeling and in vivo infection models. Having established novel methods to study immune networks in lymph nodes and virus- induced TLSs in lungs of Influenza A virus infected mice, I am characterizing the properties and directionality in which T cell help is provided to B cells of varying antigen-specificities and investigating how competition for T cell help shapes the antibody repertoire against Influenza A virus. For the K00 phase of this proposal, I plan to leverage my experiences using mechanistic mouse models of B- and T-cell collaboration with my prior work studying cancer immunology to elucidate the dynamics of TLS contribution to humoral immunity in vivo, with a focus on lung adenocarcinoma models. By characterizing the contacts between immune cells taking place in cancer-induced TLSs, I hope to uncover the pathways in tumor- bearing mice that generate beneficial TLSs. Applying my expertise with intercellular labeling approaches and tissue multi-photon microscopy, I aim to identify if and how the immune cell interaction landscape changes in the context of immunotherapy. These insights could then be used to develop ways to reprogram inert TLSs, or produce new TLSs, that are capable of tumor control.

NIH Research Projects · FY 2025 · 2025-08

PROJECT SUMMARY/ABSTRACT Research: Parkinson's disease is characterized by degeneration of dopaminergic neurons in the substantia nigra, leading to reduced dopaminergic input to the striatum and subsequent motor and cognitive deficits. While the loss of presynaptic dopamine input initiates striatal dysfunction, mere replacement of dopamine is insufficient to effectively treat the disease over time. This gap in knowledge has spurred interest in understanding the molecular dysfunction within the postsynaptic neurons of the striatum. D1 and D2 neurons, named for their predominant expression of dopamine receptors D1 and D2, are the principal postsynaptic neurons of the striatum. They exhibit distinct downstream signaling mechanisms in opposing pathways that together influence striatal function. The overall objective of this project is to identify the molecular mechanisms underlying striatal dysfunction in Parkinson’s disease by investigating aberrant glutamate and dopamine receptor signaling separately in D1 and D2 postsynaptic neurons. This study focuses on the cell-type specific role of the neuronal RNA-binding protein ELAVL3, whose levels are altered in Parkinson’s disease. ELAVL3 is a known modulator of alternative splicing of mRNAs encoding proteins critical for glutamate and dopamine receptor signaling in the mouse striatum. The central hypothesis is that altered RNA-binding function of ELAVL3 leads to aberrant splicing in D1 and D2 neurons, ultimately disrupting striatal function in Parkinson’s disease. The applicant will test this hypothesis by leveraging technical breakthroughs in her mentors’ laboratories that enable the isolation of D1 and D2 neurons from postmortem brains to study RNA regulatory mechanisms within specific cell types. The project has two specific aims: 1) To identify the role of alternative splicing in aberrant glutamate and dopamine receptor signaling in Parkinson’s disease 2) To determine the role of ELAVL3 on alternative splicing and downstream signaling in D1 and D2 neurons in Parkinson’s disease. This research is expected to shed light on the molecular mechanisms involved in striatal dysfunction in Parkinson’s disease, potentially leading to the development of novel therapeutic approaches. Candidate: The applicant Dr. Krithi Irmady is a movement-disorders neurologist and Instructor at the Rockefeller University, who aims to become an independent, tenure-track physician- scientist investigating RNA biology in Parkinson’s disease. Dr. Irmady has outlined a five-year period of mentored training where she will develop skills in fluorescence activated nuclear sorting, cell-type specific profiling of RNA binding protein targets, and experimental manipulation of splicing in neuronal culture systems to study RNA dysregulation in Parkinson’s disease. Environment: The proposed research will be done in Rockefeller University, one of the world’s leading biomedical research universities. The University exposes trainees to an exceptionally robust academic research environment with a strong commitment and track record of successfully supporting junior faculty who are seeking careers as independent investigators.

NIH Research Projects · FY 2025 · 2025-07

PROJECT SUMMARY Isolated congenital asplenia (ICA) is characterized by the absence of a spleen at birth without any detectable associated developmental abnormalities (OMIM #271400). ICA is the only known human developmental defect affecting only a lymphoid organ. Patients with ICA are prone to life-threatening bacterial disease. We discovered germline heterozygous variants in the exons of RPSA, encoding ribosomal protein SA, as the first genetic etiology of ICA, in about half of the kindreds studied in our unique international cohort. The observation that variants of a ribosomal protein caused ICA was surprising, as deleterious variants in 20 other ribosomal proteins underlie Diamond-Blackfan anemia (DBA), a complex malformation syndrome with normal development of the spleen. Autosomal dominance at the RPSA locus operates by haploinsufficiency (HI). For unknown reasons, most RPSA variants display incomplete penetrance for ICA. In this context, we hypothesize that (1) incomplete penetrance may be mediated by the genetically determined levels of expression of the WT RPSA allele, (2) ICA in other kindreds may be explained by mutations in other genes, including genes whose translation is impaired in cells with RPSA HI, and that (3) RPSA HI may impair the translation of genes, some of which differ from those impaired in DBA cells. We will test these three related hypotheses. First, we will identify candidate cis-eQTLs governing WT RPSA expression using pre-existing statistical methodologies and databases. We will assign genotypes at these eQTL positions on the WT RPSA alleles of ICA and non-ICA cohort members heterozygous for an ICA-causing RPSA mutant allele, and assess the segregation between WT RPSA eQTL genotype and phenotype (with or without a spleen). Second, we will search for new genetic etiologies of ICA by genome-wide approaches. Third, we will test the hypothesis that RPSA HI leads to ICA by impacting the translation of a subset of mRNAs. We will use genetic and biochemical approaches to identify candidate genes whose mRNA is selectively sensitive to RPSA HI. Our preliminary data are exciting. First, we have already sequenced the whole genome of 63 RPSA heterozygotes from 31 families, including 38 and 25 individuals with and without a spleen, respectively. We have delineated 3 groups of eQTLs that independently modulate RPSA expression. Using these three groups, we have segregated the WT RPSA alleles of 31 families into groups that tightly correlate with the presence or absence of a spleen. Second, we have found 4 kindreds with bi-allelic mutations in C6orf25. Third, we have performed Ribo-seq and started analyzing the data. An understanding of how RPSA HI underlies ICA with incomplete penetrance will provide proof-of-principle that the levels of expression of the WT allele can drive the penetrance of a human dominant trait. Identification of a new genetic etiology of ICA will provide a long- awaited explanation for the unexplained cases. Finally, investigating the translational impact of RPSA HI will provide insights into the mechanisms by which RPSA HI leads to a distinct, limited phenotype, when compared with DBA, thereby laying the groundwork for fundamental studies on the function of human ribosomal proteins.

NIH Research Projects · FY 2026 · 2025-05

2024 R35 Project Summary This proposal presents a series of interrelated new techniques and concepts aimed at discovering otherwise hidden therapeutic targets for neurologic diseases. This treatment-oriented approach combines the urgency of a practicing neurologist with the knowledge and technology that modern molecular biology brings to neuroscience. From the basic science perspective, understanding the fundamental root mechanisms of disease is an uncompromising goal. From the neurologist’s perspective, the perfect should not be the enemy of the good. This leads to several fundamental points: • Genetic (DNA) etiologies of brain disease set the stage for our focus—the downstream manifestations of such defects. • Different cell types contribute to different brain disorders, but the difference between individual cells is unknown. These differences are manifest at the level of RNA, mediated by the stoichiometry, variation and localization of cell-specific regulatory systems, and their consequent effects on proteins within affected cells. • Neurons are spatially complex and temporally dynamic, requiring commensurate focus and tool development that address these aspects of RNA regulation. • Developing and validating new model systems will lead to unexpected discoveries. • Understanding human neurologic disease is complicated; thus, the best model system for understanding neurologic disorders is the human. Therefore model systems must be complimented and validated by studying human neurons. • Building the bridge between model systems and human neurobiology alongside computational biologic analysis of large datasets is necessary to develop new insights and predictions fully. • Providing new strategies for fundamental discovery and focusing on actionable subsets for neurologic disease is our primary strategy. A significant innovation our lab has developed supplements the standard analysis of RNA quantity (RNAseq) to understand the regulation of RNA quality through the development of CLIP, a rapidly expanding technology. Recent work has established the potential of this technology, and we propose to further map the ability to study RNA regulation in single neuronal cell types in the brain, subcellular domains of neurons, and during the rapid changes that accompany neuronal depolarization. Importantly, we propose to integrate data from model systems with a novel concept to study molecular variation in neurons obtained from human autopsy. We will analyze complex molecular datasets with computational biology. Success in the treatment of brain diseases to date emphasizes the importance of focusing on accessible molecules which will guide our approach to big data analysis.

NIH Research Projects · FY 2026 · 2025-04

Project Summary/Abstract Cell competition is a fitness-sensing mechanism where more-fit “winner” cells eliminate less-fit “loser” cells. First described to optimize overall tissue fitness, cell competition can also give rise to cells with enhanced fitness, such as those with oncogenic potential. These dual roles of cell competition may explain why oncogenic cells can initiate tumors (oncogenic winners/wild-type (WT) losers), but also why they often do not lead to overgrowth despite their accumulation in epithelial tissues (WT winners/oncogenic losers). However, it is unclear how losers are eliminated or restrained, nor how winners identify and eliminate losers. Insight into the competitive cell-cell interactions occurring during both homeostasis and premalignancy will thus fill a critical knowledge gap and address a pressing need for preventing oncogenesis at early stages. To address this hypothesis, I have preliminarily studied competitive interactions between a model for premalignancy and their WT neighbors in the embryonic mouse epidermis, which harbors epidermal stem cells (EpSCs) capable of acquiring oncogenic potential and is a well-characterized cell competition model. I observed that sparse, premalignant EpSCs ectopically expressing SOX2 are actively eliminated by WT neighbors and homeostasis is maintained. Interestingly, this loser status seems modulated by the degree of homo/heterotypic interactions (HHIs), or who their neighbors are. In fact, SOX2 EpSCs shed their loser status and initiate pathological overgrowth when SOX2 EpSC abundance in the tissue is increased. Armed with a premalignant model that can behave like a winner or loser depending on their cellular “neighborhood”, I propose to mechanistically dissect how these competitive statuses are assigned and further underlie homeostasis and premalignancy. Aim 1 will address how SOX2 and WT EpSC losers are eliminated. Building upon my preliminary data suggesting that intercellular adhesion is increased between SOX2 EpSCs in loser contexts, I will perform in vitro assays and in vivo lineage tracing to assess resulting differentiation and/or anoikis. Aim 2 will interrogate context- and contact-specific transcriptomic programs to uncover how SOX2 EpSCs eliminate WT neighbors and vice versa. Using a lab-verified, in vivo contact-based tagging and purification system coupled with single- cell RNA-sequencing, I will mechanistically decipher how winners identify and kill losers during both homeostasis and premalignancy. Completion of these aims will increase our understanding of the mechanisms underlying cellular fitness in cell competition; uncover the inherent, tissue-level dynamics used to maintain homeostasis in the face of premalignancy; characterize the very earliest stages of overgrowth initiation; and elucidate the “tipping point” between epithelial homeostasis and oncogenesis.

NIH Research Projects · FY 2026 · 2025-04

Project Summary Ribosomes are molecular machines composed of ribosomal RNAs and up to 80 ribosomal proteins. These large assemblies catalyze protein synthesis in all cells. The long-term goal of this project is to understand how eukaryotic ribosomes are assembled with the help of more than 200 non-ribosomal factors as a series of molecular and mechanistic snapshots of assembly intermediates. Combining genetic, biochemical, mass spectrometry, and AI-based approaches with cryo-EM is an essential step to engineer, trap, isolate and determine atomic-resolution molecular snapshots of transient assembly intermediates of ribosomal subunits. Eukaryotic ribosome assembly can be subdivided into four stages, co-transcriptional assembly events and initial maturation of small and large ribosomal subunit precursors in the nucleolus, nuclear maturation of pre-40S and pre-60S particles, nuclear export, and cytoplasmic maturation. While late events in eukaryotic ribosome assembly are relatively well characterized, the early assembly of ribosomal subunits in the nucleolus is still poorly understood. Especially the co-transcriptional assembly of pre-ribosomal particles of both subunits and the subsequent transitions towards stable post- transcriptional nucleolar assembly intermediates have remained elusive due to their transient nature. This proposal describes new approaches to define the molecular mechanisms that govern co- transcriptional assembly events as well as transitions towards stable post-transcriptional assembly intermediates. My laboratory has developed new AI-based tools and biochemical approaches that now enable us to survey the entire ribosome biogenesis pathway for protein-protein interactions that guide the specific isolation of early eukaryotic ribosome assembly intermediates and study their transitions. The synergistic use of these approaches has allowed us to overcome previously intractable hurdles, thereby enabling the detailed study of essential early assembly intermediates of both ribosomal subunits. Insights from these studies will shed light onto both the mechanisms that are employed during eukaryotic ribosome assembly to coordinate key processing events as well as how defects in eukaryotic ribosome assembly can result in human blood disorders, which are collectively termed ribosomopathies.

NIH Research Projects · FY 2026 · 2025-03

PROJECT SUMMARY Colorectal cancers (CRC) are characterized as having a hierarchical organization requiring proliferating and de-differentiated stem cells to maintain tumor growth and progression. Cellular plasticity underlying colorectal cancer is essential for a process which occurs following selective pressures of the tumor microenvironment and chemotherapeutics. The colonic tumor microenvironment is characterized by extreme hypoxia due to the anoxic lumen. Hypoxia promotes metabolic rewiring, and such processes are utilized by cancer cells to support biosynthesis, cell survival and dynamic alteration in cell fates. A critical feature of cellular metabolism is organellar interaction and coordination, yet how these contribute to CRC plasticity, survival, progression and treatment response are unclear. Endoplasmic reticulum-mitochondria contact sites (ERMCS) are the most abundant inter-organellar interaction. I generated a panel of ERMCS reporter CRC cell lines, and through unbiased high content imaging and CRISPR screens, I have identified essential mechanisms required for ER- mitochondrial interactions in CRC. Moreover, I show a key role of tumor hypoxia in modulating ERMCS. Hypoxia inhibited mitochondrial complex III and IV to decrease ERMCS. Treating cells with the mitochondrial electron carrier, coenzyme (CoQ) rescued ERMCS suppression following hypoxia. I hypothesize that tumor hypoxia regulates ER-mitochondrial contacts (ERMCS) by altering mitochondrial respiration and CoQ redox for metabolic adaptation and survival. In aim 1 (F99 phase), I will focus on identifying the molecular mechanism of hypoxia dependent ERMCS inhibition and expand into in vivo models with our novel ERMCS reporter mouse model. During the K00 phase, I will apply knowledge gained during graduate school in cancer metabolism and organellar interaction to an independent postdoctoral project. The plasticity of colorectal tumor epithelium depends on integration of organellar functions to sustain metabolic demands. Therefore, my goal as a postdoctoral fellow is to understand the dynamic changes and requirement for organellar interactions and metabolic compartmentalization during cell fates alterations in CRC. I plan to use genetic murine and primary patient organoid models of CRC, volumetric electron microscopy, in vivo organellar metabolomics, and functional CRISPR screens to answer these questions. Lastly, in addition to the proposed studies, this training plan includes activities important for career development, mentorship, networking, and scientific communication to prepare me for successful transition to a postdoctoral fellowship and my career as an independent investigator studying cancer metabolism.

NIH Research Projects · FY 2026 · 2025-02

Project Summary Betacoronaviruses (beta-CoVs), including SARS-CoV-1, MERS-CoV, and SARS-CoV-2, have reshaped our understanding of pandemic preparedness. These viruses demonstrate a remarkable ability to mutate and evade defenses, continuing to infect populations worldwide despite extensive vaccination efforts and antiviral therapies. The chameleon-like nature of SARS-CoV-2, particularly its modifications to the Spike protein, consistently outpaces existing countermeasures, necessitating new strategies. This proposal introduces a pioneering class of nanobodies (Nbs), engineered from the immune system of llamas, designed to provide comprehensive protection against all beta-CoVs. These biologics not only advance treatment but also signify a pivotal step in pandemic preparedness, equipping us to outpace the relentless evolution of beta-CoVs. Our innovation lies in developing multivalent, synergistic combinations of broad-spectrum, high-efficacy Nbs. By harnessing these combinations, we amplify their efficacy and scope, concurrently increasing their resistance to viral mutations. Administered intranasally or directly to the lungs, these Nbs serve as both prophylactics and therapeutic agents. Our first Aim is to strategically expand upon our proven repertoires to identify, isolate, and characterize a much larger and more diverse repertoire of Nbs that collectively are strongly neutralizing across the beta-CoVs. We will use cutting-edge methods to produce diverse Nbs from llamas exposed to spike proteins of various beta- CoVs, selecting those with high affinity, specificity, and stability. We aim to discover synergistic, escape-resistant Nb pairs through combination testing and structural analysis. In our second Aim, we will optimize critical parameters important for developing broadly neutralizing Nb combinations and derivatives for human use. We will evaluate the in vivo synergistic potential of Nbs targeting major threats like MERS-CoV and SARS-CoV-2, and engineer Nbs to optimize their properties and efficacy in preparation for clinical trials. Deploying these pre- programmed Nbs at an outbreak's onset will protect first responders and medical personnel, reduce hospital surges, limit transmission, and buy time for new vaccine development and rollout. They will also provide crucial support to immunocompromised individuals, safeguarding the most vulnerable from the start. We hypothesize that our synergistic Nb combinations will introduce new beta-CoV neutralization methods, effectively prevent and treat infections, and maintain efficacy against emerging beta-CoV threats.

NIH Research Projects · FY 2026 · 2025-02

PROJECT SUMMARY: Genes are regulated differently throughout the body, and mutations that may benefit one kind of cell may conflict with the regulatory needs of other tissues. When mutations occur in reproductive cells, particularly ones that provide a “selfish” advantage, these variations are transmitted to all of the offspring’s cells, representing a form of germline selection that often conflicts with genetic processes of somatic tissues in resulting offspring. These germline-somatic conflicts often serve as the primary driver of many spontaneous congenital disorders, and many more variants of unknown significance (VUS) may be similarly influenced by poorly understood evolutionary pressures that are uniquely subjected to the germline. Current approaches to study germline selection are limited in their ability to study the natural emergence of germline mutations in a causal and iterative fashion. This proposal addresses the need for a generalizable and comprehensive model by establishing a novel model of germline selection using the germline restricted chromosome (GRC) of songbirds. This unique genomic element contains reproductively-relevant gene duplications that have subtly diverged over evolution, offering a promising avenue to explore germline selection mechanisms that are not restricted by somatic interference across generations, mirroring the spontaneous processes impacting selfish DNM transmission and germline selective congenital disorders. The project seeks to establish the zebra finch as a unique and comparative model of germline-somatic conflict through three key aims across a mentored and independent research phase. In Aim 1, the candidate will enrich for GRC DNA samples by flow cytometry of mosaic testicular tissue for long-read sequencing that will derive a high-quality GRC assembly. This assembly will then be used to identify sequence differences between duplicate regions of the GRC and the rest of the genome. Aim 2 will utilize a comprehensive collection of single-nuclei multiomic (RNA + ATAC) germline datasets across the vertebrate phylogeny to identify the impact of GRC gene paralog sequence variations uniquely modifying gene regulatory networks in the zebra finch germline, which will be further refined during the independent phase of the award using complementary transcriptional and histological analyses. Aim 3 will apply CRISPR gene editing tools in zebra finch germline cells to clarify the functional significance of GRC genes that maintain and enhance the evolutionary fitness of the zebra finch germline, leveraging the GRC assembly and network analyses developed in the mentored phase. This proposal will benefit from the candidate's expertise in avian genetics and biotechnology, along with training in genome assembly and alignment, gene co-expression and cis-regulatory network analyses, and a foundation in germline genetic processes across vertebrates. The work executed in this proposal will establish the zebra finch GRC as a uniquely informative and broadly applicable model system to understand germline selection mechanisms. Ultimately, this proposal will advance our understanding of genetic regulation, evolutionary processes, and the assessment of complex genomic elements that will broadly impact biomedical research.

NIH Research Projects · FY 2026 · 2025-01

Even though ranked as the third diagnosed cancer, colorectal cancer (CRC) is the second cause of cancer death globally. Over 70% of CRC tumors metastasize to the liver, which is a major factor in mortality. This drastically diminishes the five-year survival rate, leaving current standard of the care (surgery and chemotherapy) ineffective, due to resistance. This underlines the urgent need for new therapeutic options. The Tavazoie lab has identified a critical survival mechanism in CRC cells, where in hypoxic conditions, metastatic cells secrete creatine kinase Brain (CKB), which converts creatine to phosphocreatine using extracellular ATP. Phosphocreatine is then transferred into the cells via SLC6A8, providing an essential energy source. Inhibition of SLC6A8 with the small-molecule RGX-202 has shown encouraging results in preclinical models and early clinical trials. However, resistance to RGX-202 poses a significant challenge. Unraveling the mechanisms that CRC cells use to survive in the hypoxic liver environment and resist treatments is central for developing more effective targeted therapies. This study intends to discover the mechanisms of resistance to SLC6A8 inhibition in CRC liver metastasis using a multi-omic approach. We hypothesize that adaptive genetic and metabolic alterations drive resistance to RGX- 202, which can be identified and targeted. One specific aim with three sub aims will address this hypothesis: Aim 1: To identify mechanisms of resistance to SLC6A8 inhibition in CRC liver metastasis. Aim 1a: Conduct in vivo CRISPR-Cas9 screen in RGX-202-01 sensitive and resistant CRC liver metastasis model cells to identify genes that cause resistance to SLC6A8 inhibitor in hypoxic liver microenvironment. Aim 1 b: Define metabolic alterations in RGX-202-resistant CRC tumors and identify metabolic pathways that contribute to CRC liver metastases. Aim 1c: Establish the molecular corelates of CRC resistance to RGX-202 using unbiased genomic approaches. Significance of this research is to delineate the genetic and metabolic mechanisms by which CRC cells survive in the hypoxic liver environment and become resistant to SLC6A8 inhibition. This study is innovative as it incorporates in vivo CRISPR screens, metabolomics, and transcriptomics to identify resistance mechanisms and establishes new paradigms for treating therapy-resistant CRC.

NIH Research Projects · FY 2026 · 2025-01

Project Abstract: The nuclear pore complex (NPC) is an ancient molecular machine embedded in all eukaryotic nuclear membranes that acts as a physical gatekeeper for macromolecules that enter or leave the nucleus. mRNA is bound by RNA-binding proteins to form messenger ribonucleoprotein particles (mRNPs) that aid in transit across the NPC so that they may exit into the cytoplasm to be translated into proteins. The NPC mediates mRNP export along a pathway that is distinct from transport of proteinaceous or ribosomal cargoes that culminates at a highly conserved cytoplasmic export location termed the mRNA export platform. At the mRNA export platform, mRNPs are remodeled, undergoing conformational and compositional changes, to facilitate release of packaged mRNA for translation into the cytoplasm. Mutations in or mis-regulation of proteins in the mRNA export pathway and particularly at this critical export complex are implicated in developmental defects, cancer, and in particular, aging. Previous studies that combined crosslinking mass-spectrometry and negative-stain electron microscopy with integrative modeling strategies yielded a low-resolution map of the architecture of the complex, but a high- resolution map of the entire mRNA export platform that yields molecular insights into the assembly of the complex and identifies key points of regulation has eluded us. Additionally, recent work has provided substantial evidence that the NPC has a functional role in post-transcriptional gene regulation, with preliminary data indicating that dysfunction in the mRNA export platform leads to disruption in this regulatory role. The goal of this proposal is to understand the molecular determinants of mRNA export platform assembly and elucidate the mechanisms by which aging-associated mutations affect platform assembly and lead to aberrant gene regulation, therefore manifesting in the onset of premature aging. Aim 1: To leverage single particle cryo-EM of the core Nup82 complex, as well as reconstitution biochemistry, crosslinking mass-spectrometry, and negative stain EM to generate a high-resolution integrative model of the mRNA export platform, onto which aging-associated mutations can be mapped to understand how these mutations disrupt platform assembly. Aim 2: To generate yeast strains carrying these aging-associated mutations that can be functionally characterized to uncover the mechanisms by which the mRNA export platform regulates gene expression and elucidate how these mutations affect this critical function and give rise to aging phenotypes. This work would give us an unprecedented understanding of post-transcriptional gene regulation at this critical mRNA export and remodeling site. Furthermore, this study would reveal new structural and molecular insights into the highly documented but poorly understood role of NPC dysfunction in the onset of aging and disease.

NIH Research Projects · FY 2026 · 2024-12

PROJECT SUMMARY/ABSTRACT Circuits of the visual system transform patterns of light impinging on the retina into an understanding of objects and their behavioral significance, achieving robustness and versatility currently not matched by even the most advanced computational vision models. Much of this high-level information extraction is thought to occur in the ventral stream, a pathway of interconnected brain regions which include areas for the processing of faces. The prefrontal cortex, typically associated with cognitive functions, receives highly-processed visual output from the ventral stream and also feeds signals back into it, resulting in a recurrently connected visual-cognitive loop. To address these fundamental questions, this proposal uses face stimuli and the face-selective patches that process them, including one in the anterior temporal cortex (aTC) in the ventral stream and the other in ventrolateral prefrontal cortex (vlPFC). The central hypothesis to be tested is that aTC and vlPFC contain qualitatively different face representations, one emphasizing physical properties, the other meaningful categories, and that this difference shapes feed-forward and feed-back processing between them. The work proposed here leverages the advantages of a small, lissencephalic brain, the use of multi-regional high-density electrophysiology, and mesoscale calcium imaging to determine the spatial organization of face codes with single-cell resolution. Aim 1 will determine how face features are represented in aTC and vlPFC through simultaneous, translaminar recordings using Neuropixels probes. It will test the specific hypothesis that aTC represents faces in a feature- specific and physically accurate manner, while vlPFC represents faces holistically and as exemplars of meaningful social categories. Aim 2 will determine the spatial organization of face representations in vlPFC through mesoscale multiphoton calcium recordings of thousands of cells simultaneously. It will test the specific hypothesis that multiple abstract face-related categories are represented within vlPFC in a spatially segregated manner. Aim 3 will determine the role of vlPFC feedback signals on aTC face representation using dual probe recordings to test the hypothesis that robustness to ambiguous inputs results from recurrent interaction between temporal and frontal circuits and reflects category selectivity. This project will make use of modern advances in neural recording technology to address questions about the precise orchestration of brain activity across regions, whose disruption is associated with debilitating dysfunctions such as schizophrenia. It will develop a model system and approaches for mechanistic dissections of high-level perceptual and cognitive functions and contribute to our understanding of the specific neural circuitry underlying face perception and social cognition that is known to be perturbed in neuropsychiatric disorders such as prosopagnosia, frontotemporal dementia, and autism spectrum disorders.

NIH Research Projects · FY 2025 · 2024-09

Project Summary To understand how the brain develops and accomplishes its unique computational and cognitive capabilities through the dynamics of three-dimensional networks of neurons remains a key goal of contemporary neuroscience. To inform accurate modeling of network dynamics, tools will be indispensable that enable recording of neuronal activity at both, the temporal scale at which neuronal communication and computation is performed, and a spatial scale that encompasses the ensemble of neurons participating in a given function. Recently, increasingly efficient genetically encoded voltage indicators (GEVIs) have been developed that allow optical readout of electrical neuronal activity at a millisecond timescale. However, recording of GEVI activity has remained limited to small fields-of-view and few axial planes, either due to low signal and speed achievable in scanning multiphoton methods or due to the low signal-to-background ratio that hampers existing one-photon modalities. To develop optical recording techniques capable of capturing the full electrical dynamics of large ensembles of neurons therefore remains an unmet need. This project aims to address this need by designing and demonstrating a novel ultra-high-speed volumetric imaging system capable of capturing the full dynamics of genetically encoded voltage indicators from thousands of neurons in the cortex and hippocampus of awake, behaving rodents. This will be achieved through a combination of structured one-photon excitation with advanced light field detection. The proposed technique will achieve a temporal resolution sufficient to capture the timing of single action potentials from thousands of neurons. The voltage dynamics of active neurons will be extracted and demixed from background and crosstalk by an advanced machine learning algorithm that exploits the presence of unscattered and weakly scattered light in the targeted depth range. The proposed approach is designed to deliver a time resolution approaching electrical recordings while also determining the 3D positions of all active neurons in a volume, in a manner that is immediately parallelizable to cover larger lateral areas at minimal invasiveness. If successful, this project will therefore contribute to enabling optical electrophysiology at the level of entire neuronal systems. 1