Harvard University

NIH Research Projects · FY 2025 · 2023-07

Project Summary/Abstract Immune modulation holds tremendous promise for the treatment of cancer, autoimmune disease, metabolic disease, and infectious disease. New ways to generate antigen-specific T and B cells inexpensively and with minimal reactogenicity are badly needed. Certain bacterial strains from the gut microbiome elicit a potent, specific adaptive immune response. The underlying mechanisms could guide new therapeutic strategies in which bacteria-specific immune responses are rationally altered or re-directed. The gut is the site of a wide variety of microbe-microbe and microbe-host interactions. However previous papers have characterized microbial strains of the microbiome under artificial conditions of mono-colonization. This approach can identify strains that are capable of modulating immune cells, but it is unknown how a strain functions in the presence of other members of the complex microbiota. This knowledge gap hinders a logical design of a microbial therapeutic. My long-term research objective is to develop new technologies to understand the “physiological” gut ecosystem at the level of molecular mechanisms so that I can identify immune modulatory bacteria from the microbiome and build new therapeutics. In this proposal, I will establish a “physiological” gut by colonizing germ-free mice with a complex defined gut community (104 strains) and profile T cell responses to each strain individually. In Aim1, I will identify a set of bacteria-reactive TCRs and their stimulatory strain by single cell technologies, so that I can provide a big picture of the strain-T-cell interactions at the single TCR level. In Aim2, I will identify and characterize a bacterial antigen common to multiple strains. In Aim3, the result of T cell profiling will be used to “design” therapeutic bacterial communities in which inflammatory strains will be dropped out for building a tolerogenic community to treat colitis in an IBD mouse model. The successful completion of this project will “decode” a strain-by-strain view of immune modulation by the gut microbiome and provide a molecular basis for “designing” the new therapy that logically modulates immune response to treat IBD and other devastating systemic disorders. Support from the K99 and mentors will complement my expertise in immunology with state-of-art technologies in microbiology and single cell biology. I will accomplish this with training from Dr. Michael Fischbach (primary mentor, bacterial genetics), Dr. Daniel Mucida (co-mentor, single-cell biology and T cell biology), Dr. Justin Sonnenburg (advisory committee, metagenomic analysis), and Dr KC Huang, (advisory committee, a synthetic microbial community). The training and mentorship I receive during my K99/R00 award will provide a critical stepping stone for me to achieve my academic goal of establishing a vibrant independent research program that can answer an important question in the gut ecosystem for establishing a new therapeutic.

NIH Research Projects · FY 2025 · 2023-07

Musculoskeletal simulations that quantify muscle forces during movements, rigorously validated in empirical studies, have great potential to improve life-long mobility for many persons. However, current musculoskeletal simulations generally suffer from physiologically inaccurate muscle models that hinder reliable prediction of time-varying muscle force, which limits their quality and usefulness in the clinic. Although other factors are known to hinder muscle model accuracy, we hypothesize that a fundamental cause is the absence of tissue mass in musculoskeletal models. Inactive muscle mass is most relevant to submaximal activities of daily living (ADL), significantly limiting muscle shortening velocity, work, and power output. Our pilot data show that significant interactions occur between inactive mass, fiber arrangement, and muscle bulging that fundamentally affect muscle contractile properties. This proposal will quantify the effects of muscle size and inactive mass on in situ twitch time, peak shortening velocity, and work for different-sized and -shaped muscles in mice, rats, and goats (1000-fold size range); as well as in comparison to small fiber bundles from these muscles. Our comprehensive contractile property results from animal studies will inform the design of mass-sensitive muscle models, which will be incorporated into computationally efficient musculoskeletal simulations (numbering 19,600 cycles – 104 more than studies previously published) of human movement to test how muscle size, inactive mass, shape, and fiber type affect the activations needed to execute ADL and gait across the lifespan. SA1 addresses how muscle inactive mass and size affect contractile performance via in situ and in vitro studies of parallel-fibered animal muscles; testing [H1a] that more inactive muscle mass, due to submaximal activation (i.e., ADL), yields slower muscle shortening and reduced mass-specific work output, and [H1b] that these effects will be exacerbated for larger muscles and for whole muscles, as compared to fiber bundles. SA2 addresses how fiber arrangement interacts with inactive mass to influence work in different-sized pennate mouse, rat, and goat muscles, with comparisons to parallel-fibered muscles (SA1), testing the hypothesis [H2] that pennate muscles will be less sensitive to inactive muscle mass caused by submaximal activation and show smaller reductions in shortening velocity and work, compared to parallel-fibered muscles. SA3 addresses how muscle size affects activation and function across ADL and gait dynamics via simulations of human movement that build mass-enhanced muscle models into OpenSim simulations with computationally efficient direct collocation to compare differently size-scaled human musculoskeletal models (1 - 1/1000th body mass). These simulations will test the hypotheses: [H3a] that larger muscles generate less work with lower efficiency than smaller muscles, and [H3b] that reduced work with increased mass is more pronounced for fast muscle. Incorporating muscle mass and fiber-types in musculoskeletal simulations therefore stands to predict greater reliance on activations of slower muscle fibers to achieve gait and activities of daily living.

NIH Research Projects · FY 2024 · 2023-07

PROJECT SUMMARY Composite tissue regeneration is very limited in mammals; however, humans and mice can fully regenerate the distal tips of the digits following amputation. This process involves the formation of a blastema, a cellular structure that is the source of the regenerated tissue and is integral to successful regeneration. Proximal amputations beyond the nail do not form a blastema and result in fibrotic wound-healing. This differential behavior makes the mouse digit tip an ideal model system to investigate the cellular and molecular factors driving each wound-healing response and why complex regeneration is so limited in mammals. Specifically, this project will focus on fibroblast subtypes and their role in fibrosis versus regeneration. Fibroblasts are a major contributor to the blastema and play an integral part in fibrosis; thus, they may be a cell population that drives the decision between fibrosis and regeneration. Our single cell transcriptomic (scRNA-seq) analysis of the regenerating blastema revealed an extremely heterogenous fibroblast population and that the subpopulations had distinct population dynamics and lineage trajectories during blastema formation and maturation. I hypothesize that there are specific fibroblast subtypes that promote regeneration, inhibit fibrosis, or both. However, no studies have performed direct lineage contributions by tracing fibroblast subpopulations in regeneration and fibrosis. Additionally, the in vivo functional roles and importance of our computationally defined candidate pro-regenerative genes have not been established. This project will utilize single-cell CRISPR based DNA barcoding for lineage tracing fibroblasts at the subtype resolution (Aim 1) and plasmid electroporation for gene delivery to functionally assess candidate pro-regenerative genes (Aim 2). Together, my two aims will provide important insight into how the fibrotic and regenerative processes are determined in the mouse digit tip and will open additional avenues for more effective clinical treatments for large wounds or amputations in humans.

NIH Research Projects · FY 2025 · 2023-07

Project Summary Early language skills like word learning predict later school, social, and behavioral outcomes. While word learning improves dramatically in year 2, so too do other social, cognitive, and linguistic skills, leading to debate about the factors that support word learning, and why it improves. Understanding early word comprehension (which precedes production) holds promise for clinical applications, where timely diagnosis and intervention is critical. The proposed work ties early word comprehension to other improving skills within infants, across year 2. Its overall objective is to establish specific factors that may make older infants better word learners than younger ones, building the evidence base to support children who struggle with this critical facet of language in future work. Aim 1 is to test whether point comprehension is linked to robust word comprehension. Pointing allows child and caretaker to draw each other’s attention to shared context, much as words do. While prior work links pointing and language, none uses a fine-grained developmental lens with high-sensitivity tasks. Exp. 1 tests the hypothesis that point comprehension, i.e. receptive joint attention, precedes and is correlated with robust word comprehension by testing a longitudinal sample of 10-16 mo’s every 2 weeks on both skills. Aim 2 is to establish the strength of the relationship between linguistic skills and robust word comprehension. Advancing theory on whether and how linguistic skills support each other, Exp. 2-4 test 3 cross-sectional samples of 10-16mo’s on word comprehension alongside their ability to recognize how words sound, and their skill at anticipating the words and sounds in utterances as they unfold. Results will establish whether robust word comprehension is correlated with and thus potentially reliant on these linguistic skills. Aim 3 is to disentangle the roles of maturity and exposure by connecting new word learning to familiar word comprehension. Studies testing familiar word knowledge have a built-in confound between exposure and maturation, since older infants have heard more language, with repercussions for word processing. Studies of new word learning rely on overly simplified learning processes. In an innovative 2-week picture book exposure combined with measures of familiar word knowledge in 14, 18, and 22 mo’s, Exp. 5 isolates maturity and exposure to build a more cohesive theory of word comprehension. The proposed work’s unique multi-task multi-age design ensures scientific rigor in providing insight into exactly what improves over year 2, as infants become better word learners. Successfully completed, this work will establish an important foundation for supporting children with language delays and deficits, with particular relevance for ASD, Developmental Language Delay, and hearing loss.

NIH Research Projects · FY 2026 · 2023-07

Project Summary The goal of this project proposal is to advance our understanding of the mechanisms that underlie longitudinal social bond development in children and their pet dogs. The strength of the social bonds that form between children and their pet dogs is thought to mediate the positive therapeutic effects of having a pet dog, yet the mechanisms involved in the longitudinal development of these social bonds are not well characterized. The overall objective of this study is to determine how social bonds between children and their pet dogs develop, and to examine the anxiolytic effects of these social bonds in order to more effectively facilitate child health and wellbeing. The specific aims of this project are 1) to assess the relationship between social bond strength, oxytocin levels, and functional responses within social reward-related brain circuits in dogs and children, and 2) to determine whether social bond strength is associated with cortisol levels and functional connectivity within stress-related brain circuits in dogs and children. These aims will be achieved by assessing changes in social bond strength between child-dog dyads across four timepoints: within one week of dog adoption, two months post-adoption, five months post-adoption, and eight months post-adoption. At each timepoint, social bond strength will be measured via a behavioral task, and salivary samples will be collected from the child and the dog at the start and end of the behavioral testing session to measure changes in hormone level throughout the session. Brain activity will be measured following behavioral testing using fMRI techniques. Through the integration of behavioral, endocrinological, and neural methodologies, we will obtain a full perspective of the longitudinal process of social bond formation in child-dog dyads that can be generalized across the broader population. The proposed study aligns with the National Institute of Child Health and Human Development’s Child Development and Behavior Branch of Human Animal Interaction’s mission of enhancing the lives of children. The results of this study will directly influence child wellbeing by allowing us to better understand the mechanisms involved in the formation of social bonds between children and pet dogs over time, which will lay the foundation for informing therapeutic outcomes in future clinical populations.

NIH Research Projects · FY 2026 · 2023-07

ABSTRACT Natural metabolic diversity is generated through the evolution of novel function in enzymes (neofunctionalization). Society uses this metabolic diversity to obtain many high-value chemicals, such as microbial and plant-derived pharmaceuticals, but harnessing this chemistry relies on discovery of the underlying biosynthetic machinery. While some biosynthetic enzymes are readily identifiable, there are many metabolic reactions with no defined enzyme family, and this acts as a roadblock to elucidating new metabolic pathways. My lab studies enzymes and chemical reactions from the natural world, with a focus on identifying biosynthetic genes and pathways in medicinal plants. We are particularly interested in finding new enzymes that expand the ‘catalog’ of known metabolic protein families. Recently, we identified several α-carbonic anhydrase (CAH)-like proteins that have neofunctionalized to catalyze novel scaffold-forming reactions in the biosynthesis of neuroactive plant compounds. While these are the first CAH family proteins shown to act as biosynthetic enzymes, we predict that neofunctionalized CAHs (neo-CAHs) have critical, undefined functions in metabolism more broadly. Over the next five years, my lab will advance a fundamental understanding of neo-CAHs by providing a mechanistic basis on their enzymatic function and by investigating the breadth and diversity of neo-CAH enzymes throughout nature. While canonical CAHs are very well-studied, the biochemical properties of neo-CAHs are yet to be defined. We will study the foundational biochemistry and catalytic mechanisms of the neo-CAHs through enzymatic characterization, structural biology, and analysis of native post-translational modifications and sub-cellular localization. This work will provide a mechanistic understanding of neo-CAH enzyme catalysis and will yield basic insight into novel chemistry used to produce bioactive plant molecules. Simultaneously, we will investigate the widespread occurrence of neo-CAHs throughout nature. Each neo-CAH identified thus far has mutations in conserved active site residues that are essential for canonical CAH function. Similar mutations are found in other uncharacterized CAH family proteins within plants, bacteria, and animals, suggesting that CAHs have unappreciated biosynthetic functions in multiple kingdoms of life. We will leverage these distinguishing mutations to identify and functionally characterize other neo-CAH enzymes - including homologs from medicinal plants, microbes, and humans - to better define the metabolic capacity of this protein family. Through this work, we will a) provide insight on a previously unknown class of metabolic enzyme that likely has broader biosynthetic roles in nature, and b) further determine how conserved enzymes can gain new function to yield the striking structural and functional diversity of natural metabolites.

NIH Research Projects · FY 2025 · 2023-06

Project Summary/Abstract Avoiding potential threats before experiencing disastrous events is critical for survival, yet excessive avoidance may lead to maladaptive conditions such as withdrawal or missing rewarding events. Abnormalities in threat- coping may underlie psychiatric conditions including post-traumatic stress disorder and anxiety disorders. Recent studies have shown a critical role for the sensory part of the striatum, the posterior tail of the striatum (TS), in avoidance of a potential threat. These studies have indicated that TS-projecting dopamine neurons are activated by salient threatening stimuli, and animals avoid activation of these neurons. The role of the TS in threat avoidance has been pursued further using a foraging paradigm (“Monster task”) in which mice are presented with a potential threat (a moving monster) while they forage for a reward. In this task, although mice never experienced physical harm, they exhibited three stages of threat-response: initial reactive avoidance, gradually-acquired proactive avoidance, and eventual overcoming of the threat to obtain reward. Lesions of TS-projecting dopamine neurons impaired threat avoidance. Further, preliminary results indicate that, in the TS, medium spiny neurons in direct and indirect pathways (dMSNs and iMSNs) facilitate threat avoidance and overcoming, respectively. Building on these observations, the goal of this project is to elucidate the neural mechanisms by which TS and associated circuits of the basal ganglia function to regulate progression of threat-coping. Aim 1 will test the hypothesis that dopamine in TS represents threat prediction error and regulates threat-coping by dual modes of functioning, acute and learning-based actions. To this end, dopamine release in TS will be monitored using fiber photometry or manipulated optogenetically during the Monster task. Aim 2 will examine the striatal circuit mechanisms by which dopamine regulates threat-coping. The specific hypotheses to be tested are that (1) the balance between dMSNs and iMSNs determines the behavioral output (threat avoidance vs. overcoming), and that (2) phasic dopamine signals modulate the balance between these opponent circuits through both acute and learning-based mechanisms. Finally, Threat-coping can involve at least two distinct processes: action selection (choosing to approach or avoid) and/or changes in sensory processing (adjusting the salience of a potentially threatening stimulus). Aim 3 will aim to identify pathways downstream of TS which are involved in these processes. Specifically, this aim will test the hypotheses that (1) the integration of dMSN and iMSN activities occurs in the substantia nigra pars lateralis (SNL) which then regulates avoidance behavior, and that (2) the TS-globus pallidus-thalamic reticular nucleus pathway modulates sensory representation in lateral geniculate nucleus to attend or overcome a monster threat. Overall, this study will elucidate a role for novel neural circuits (TS and associated basal ganglia pathways) in three stages of threat-coping, initial reactive avoidance, proactive avoidance, and overcoming of the threat.

NIH Research Projects · FY 2026 · 2023-06

Project Summary The identity and spatial context of biomolecules (e.g., protein and RNA) in cancer cells are crucial components of their pathology. Therefore, a scalable and multiplexed imaging platform that can simultaneously map the protein and RNA landscapes tissue would be a vital tool towards deeply profiling and mapping cancer cell types in their spatial context. Extensive efforts have been made toward this end to reveal unprecedented details at both cellular level and molecular level. However, these methods generally lack high multiplexity or suffer low sensitivity as the target abundance decreases. In addition, highly multiplexed methods for co-imaging of protein and RNA at the whole tissue level still lag behind. We here aim to address these limitations by developing a versatile imaging platform for highly multiplexed, rapid, scalable proteomic and transcriptomic mapping of cell line and mammalian tissue samples with high-plex signal amplification. We propose to (Aim 1) develop a simple, highly controllable polymerase mediated iterative in situ DNA extension (ISE) and concatenation. We will demonstrate the application of ISE on the imaging of protein and RNA targets in both tissue and cell sample with high signal-to-noise and signal specificity. We will also (Aim 2) validate the scalability of ISE imaging in cell and tissue samples. We will optimize the technology to achieve spatial mapping of 50 to 100-plex protein and RNA targets in mammalian tissue samples. Finally, we will (Aim 3) validate ISE imaging in both FFPE and thick tissue, both normal and cancerous sample types. We will co-detect both RNA and protein tumor biomarkers in clinical FFPE samples and finally, we will integrate the ISE imaging methods with existing tissue clearing methods (iDISCO, CLARITY, etc.) to enable high-throughput and highly multiplexed tissue imaging from cellular level to molecular level for both normal and cancerous brain tissue. We will establish a platform for 50-plex imaging in hundreds micrometer to millimeter thick mammalian tissue specimens to unveil unprecedented detail in the tissue. The proposed work will deliver a comprehensive imaging toolset including a low-cost, simple design of orthogonal DNA probes for multiplexing imaging on protein and RNA molecules, and a scalable signal amplification method for multiplexed fluorescence imaging in different types of tissues. We envision that our technologies will be widely accessible and seamlessly incorporated into the pipelines, workflows, and coordinate frameworks of the mission for clinical researchers, pathologists, as well as the wider bio-imaging community to facilitate cancer research and clinical practice.

NIH Research Projects · FY 2025 · 2023-06

SUMMARY: Myocardial Physiology of Growth Differentiation Factor Signaling GDF11 and the closely related protein GDF8 (also known as myostatin) can regulate cardiac hypertrophy. We now have new prospective data in a large cohort of coronary heart disease patients showing that low blood levels of subforms of GDF8 and GDF11 powerfully predict future all-cause mortality. These new data point to specific forms of GDF11 and GDF8 as critical factors in heart disease. Furthermore, human loss- of-function mutations in GDF11 have now been identified that cause multi-system disease, including cardiovascular disease, showing the importance of GDF11 in human biology. GDF11 and GDF8 are members of the transforming growth factor β (TGFβ) superfamily of extracellular ligands and were initially thought to serve similar or redundant roles due to protein sequence identity (90% identical) within their mature signaling domains. We recently collaborated with multiple other laboratories to determine that mature GDF11 is a significantly more potent activator of SMAD2/3 dependent signaling than GDF8 in vitro, likely due to better utilization of key signaling receptors. Moreover, through x-ray crystallography-guided biochemical experiments, we identified key amino acids of the two ligands responsible for their differences in potency. These findings support the concept that GDF11 and GDF8 are likely not functionally equivalent, especially when ligand concentrations are low, as exist in vivo. However, it is not yet understood if differences in GDF11 and GDF8 at the molecular level translate to distinct functional outcomes and pathway activation in vivo. Defining the roles of these ligands in vivo can best be addressed by genetically engineered mice. Using CRISPR technology, we have now generated three new lines of mice with specific changes guided by our structural and biochemical studies on GDF11 vs. GDF8 to address this Project’s three Aims. This project will uncover the biochemistry of these ligands in vivo while retaining regulatory structure of the endogenous genomic loci. Importantly, we have already used Targeted Locus Amplification to prove that we have edited only the intended amino acids in all three of the new lines of mice. Using these newly generated mice, we will pursue the following Aims: Aim 1. To test the hypothesis that introducing the mature domain of GDF11 into the myostatin (GDF8) locus regulates cardiac size and function using Gdf8Gdf11swap mice. Aim 2. To test the hypothesis that gain of potency in GDF8 with two specific amino acids from GDF11 regulates cardiac muscle growth in mice (Gdf8G89D/E91Q mice). Aim 3. To test the hypothesis that GDF11 potency is required to maintain cardiac muscle function in vivo under pressure overload using chimeric mice with specific amino acids from GDF8 introduced into mature GDF11 (Gdf11D89G/Q91E mice).

NIH Research Projects · FY 2026 · 2023-05

WRITING AND ERASING O-GLCNAC ON TARGET PROTEINS IN THE BRAIN PROJECT SUMMARY O-Linked N-acetyl glucosamine (O-GlcNAc) is a nutrient sensor that dynamically modifies nuclear, cytoplasmic, and mitochondrial proteins. Dysregulation of O-GlcNAc has been linked to disruptions in sleep and circadian rhythm and several neurodegenerative diseases, including Alzheimer’s Disease (AD). While sleep and circadian rhythm defects are distressing symptoms of AD and other tauopathies, sleep disturbance may be a major risk factor for AD and is thought to accelerate its pathology. Extensive studies on the association of O-GlcNAc to AD have led to the first clinical trials targeting O-GlcNAc for therapy. However, these inhibitors globally alter the O- GlcNAc proteome, where a more targeted strategy may provide greater benefit. A systematic investigation of the connection between the O-GlcNAc modification and sleep regulation and AD pathogenesis would significantly impact the discovery of novel mechanisms to provide new avenues for targeted prevention and therapy. O-GlcNAc is regulated by nutrient availability and the complementary activity of two enzymes: O-GlcNAc transferase (OGT) writes the modification and O-GlcNAcase (OGA) erases it from proteins. Recently, innovations in protein engineering and gene editing tools developed by the co-investigators have provided access to precise tuning of O-GlcNAc on specific neurons and desired target proteins in the brain of Drosophila model systems of sleep and AD. Here, we will capitalize on the joint expertise in the Woo Lab and Walker Lab to facilitate the first systematic study to measure, map, and manipulate O-GlcNAc from desired target proteins and in specific neurons in the brain to yield crucial insights to the pathogenesis of AD and novel chemical strategies for remediation. To meet this goal, we will take a three-pronged approach. We will first systematically examine the relationship between O-GlcNAc in specific neurons of Drosophila models of sleep and AD pathogenesis to identify the brain regions that are most dependent on O-GlcNAc regulation. Second, we will use a targeted writer and eraser of O-GlcNAc, developed through protein engineering, to systematically examine the role of O-GlcNAc on selected target proteins in the brain to identify drivers and potential targets for alleviating sleep disruptions and AD pathogenesis. Third, we will pursue the discovery of small molecules that selectively write and erase O-GlcNAc in vitro and in vivo, which will complement our protein engineering approaches and provide targeted alternatives to global inhibitors that are under evaluation for AD therapy in the clinic. The successful outcome of this proposal will afford validated Drosophila models of sleep and AD with neuron-specific manipulation of OGT and OGA or specific target proteins using target writers and erasers of O-GlcNAc, with associated maps of O-GlcNAc proteins and sites, in addition to new and selective small molecule editors of O- GlcNAc to enable more targeted therapeutic approaches in the long-term. Additionally, the systematic methods and tools to connect physiological measurements to molecular function developed here will be translatable to the study of the connection of O-GlcNAc to other neurodegenerative diseases and beyond.

NIH Research Projects · FY 2026 · 2023-05

The epigenome comprises a critical layer for controlling gene expression and genome function. Cancer mutations often alter the function of chromatin complexes, leading to aberrant epigenomic landscapes frequently observed in tumor cells. Determining the mechanisms controlling chromatin complexes and their interactions will advance our understanding of epigenomic processes, how they are disrupted in cancer, and how they can be pharmacologically modulated for drug discovery. Consequently, our central goals are to elucidate the mechanisms of chromatin complexes and test their promise as therapeutic targets. In this pursuit, this application investigates lysine-specific histone demethylase-1 (LSD1), a transcriptional corepressor that is a drug target for oncology. LSD1 forms complexes with various corepressors and transcription factors (TF), including GFI1B, which are critically involved in development and implicated across various tumor types. Using drug-resistance alleles obtained from a chemical suppressor screen, our prior work showed that LSD1 active site inhibitors exert their anti-proliferative effects by disrupting the LSD1-GFI1B complex, revising prior models of drug mechanism of action. Notably, GFI1B is frequently overexpressed by enhancer hijacking mutations in group 3 and 4 medulloblastoma (MB), and LSD1 inhibitors are effective in GFI1B-driven MB mouse models. Intriguingly, the E3 ubiquitin ligase KBTBD4 is also frequently mutated in group 3/4 MB and was recently reported to mediate degradation of CoREST, LSD1’s obligate complex partner. These observations suggest a possible mechanistic connection between GFI1B and KBTBD4 in group 3/4 MB, mediated through LSD1-CoREST. However, the molecular interactions and interplay between LSD1-CoREST, GFI1B, and KBTBD4 remain unclear and present a major gap in our understanding. To address these gaps, the first specific aim investigates the structure, dynamics, and interactions of the LSD1-GFI1B complex through a multidisciplinary approach, with the goal of revealing an unprecedented view into a chromatin regulator-TF complex. The second aim seeks to elucidate the mechanism of small molecules that degrade LSD1-CoREST by potentiating KBTBD4 activity, providing critical insight into strategies to target LSD1 complexes through new emerging modalities. The last aim studies how KBTBD4 MB mutations promote LSD1-CoREST degradation and their downstream consequences on LSD1- GFI1B and the MB cancer epigenome. Across these aims, the mechanisms and interactions of LSD1 complexes will be explored by using innovative chemical genomic approaches that leverage drug suppressor alleles with cell, molecular, and structural biology. It is expected that the findings from these studies will illuminate biochemical principles governing the function and interactions of chromatin complexes and advance strategies to pharmacologically target them for therapeutic applications.

NIH Research Projects · FY 2026 · 2023-05

Summary We will develop a new spatial-omics platform, Light-Seq, for spatial indexing of intact biological samples using light-directed DNA barcoding in fixed cells and tissues followed by ex situ sequencing. Our light-directed barcoding strategy will enable user-directed, in situ selection of rare, disjoint cell populations for full- transcriptome sequencing based on morphology, location, or protein expression without dissociation. We will develop Light-Seq as a spatial-omic DNA barcoding platform capable of extracting the transcriptomic information from single-cells, scalable to uniquely address thousands of user-defined regions, and can be applied in both fixed and FFPE clinical samples for direct applications in human health. We envision that the Light-Seq platform will be a scalable, cost-effective, and flexible approach to spatial transcriptomics that allows the user to define spatial regions in tissue for NGS sequencing. Light-seq can thus serve as a low barrier-to-entry platform for spatial transcriptomics for many pathologists and researchers, and would be a key driver for a wider adoption of spatial transcriptomic tools.

NIH Research Projects · FY 2025 · 2023-04

Project Summary/Abstract At the cornerstone of human bipedal locomotion are the pelvis and knee, two hind limb skeletal structures for which we know little about their respective development in humans. Indeed, these structures have complex 3-D morphologies whose initial patterns arise during the chondrogenic anlagen stage, when coordinated cellular differentiation and proliferation establish various tissue types and the spatial relationships between different structural components (e.g., between the knee’s distal femoral condyles and proximal tibial platform, or between the pelvis’ ilium, pubis, ischium, and acetabular subdomains). Yet, for developing human skeletal structures, we understand little about these cellular events and their relationships to tissue morphology and function. Moreover, while one can envision that these events are mediated by a pleiotropic or common ‘skeletal growth’ gene set and accompanied regulatory apparatus, how this tool kit is used in developing humans to build each structure, remains a mystery. As biomedicine move towards regenerative therapies for joint tissues and mechanistic investigations into developmental disorders of the skeleton, it is crucially important to gain a better understanding of how cells of the skeleton and joints acquire their functional roles, and it is both timely and critical to establish this at single cell and spatial resolutions. To date, large functional genomics-based consortia, such as an ENCODE or the ROADMAP EPIGENOMICS PROJECT, have not focused on the skeleton due to logistical issues in extracting cartilage cells from developing skeletal elements composed of hard extracellular matrix. However, recent advances on this front by the grant investigators have allowed them to isolate and study individual cartilage cells from developing human skeletons. The focus of this proposal, therefore, is to more deeply investigate how the human knee, pelvis, and hind limb in general form in utero, at the level of individual cells and in understanding how changes in their biology and behavior drive the respective development of each hind limb structure. This will be accomplished via two main aims, one focused on the use of spatial transcriptomics to examine expression dynamics histologically (Aim 1), and another on the use of a single cell (sc) multiomics approach (Aim 2), consisting of scRNA-sequencing (to detect genes) and scATAC-sequencing (to detect regulatory regions) on the same cell. By simultaneously profiling gene expression and regulatory element availability at the individual cell level from many cells of these developing human structures, and spatially, the necessary resolution will be achieved to define small but important nuances in the genetic programs that govern anatomical-site-specific cartilage cell biology and how it links to hind limb morphology. Use of these protocols developed by the grant’s Team will help ensure that an extraordinary resource is provided to the musculoskeletal biology community, and that crucial information needed to develop novel pharmaceutical and regenerative medicine-based therapeutics is made public.

Fonds de recherche du Québec – Nature et technologies · FY 2023-2024 · 2023-04

Volet: Bourses de doctorat en recherche; Domaine: Environnement; Objet: Structure et composition de la Terre; Objet: Glace et neige; Application: Science and Technologies; Mots-clés: DERNIER MAXIMUM GLACIARE, RECONCILIATION DE DONNEES , RHEOLOGIE, NIVEAU DE LA MER, CHANGEMENT CLIMATIQUE, STRUCTURE VISCOELASTIQUE DU MANTEAU

Fonds de recherche du Québec – Nature et technologies · FY 2023-2024 · 2023-04

Volet: Bourses de maîtrise en recherche; Domaine: Recherche intersectorielle; Objet: Design de l'environnement; Objet: Espaces urbains et urbanité; Application: Santé; Application: Santé publique; Mots-clés: URBANISME, ETABLISSEMENTS INFORMELS, VILLES, CONTAGION, MATHEMATIQUES, THEORIE DES GRAPHES

Fonds de recherche du Québec – Société et culture · FY 2023-2024 · 2023-04

Volet: Bourses de doctorat en recherche; Domaine: Gestion des organisations; Objet: Droits et libertés de la personne, droits collectifs; Objet: Systèmes informatiques; Application: Structures et relations sociales; Application: Droits et justice; Mots-clés: INTELLIGENCE ARTIFICIELLE, EGALITE ET NON-DISCRIMINATION, GOUVERNANCE POLYCENTRIQUE, DROIT CONSTITUTIONNEL, THEORIE DU DROIT, DISCRIMINATION RACIALE

Fonds de recherche du Québec – Société et culture · FY 2023-2024 · 2023-04

Volet: Bourses de doctorat en recherche; Domaine: Développement et fonctionnement des personnes et des communautés, et vie sociale; Objet: Contrats; Objet: Vie et production économique; Application: Structures et relations sociales; Application: Culture; Mots-clés: HISTOIRE DE LA MONNAIE, THEORIE MONETAIRE, EMPIRE OTTOMAN, CRISE ET MONNAIE, TEMPORALITE, LINGUISTIQUE INFORMATIQUE

NIH Research Projects · FY 2026 · 2023-04

PROJECT SUMMARY/ABSTRACT This program is focused on the discovery, application, and mechanistic elucidation of catalytic reactions that are stereoselective, environmentally friendly, and useful for the preparation of chiral, bioactive compounds. We seek to develop novel concepts in catalytic reactivity and selectivity, and apply them to important problems in chemical synthesis. The premise underlying our current and proposed work is that new classes of small-molecule, chiral organic catalysts can promote challenging bond constructions, controlling the absolute and/or relative stereochemistry of the reactions through networks of attractive non-covalent interactions. The overarching goal is to identify simple organic catalysts that are readily accessible, inexpensive, and bear the minimal structural features necessary for inducing high levels of stereocontrol in synthetically interesting transformations. We will pursue several distinct catalytic concepts over the next five-year period, with each of the proposed reactivity manifolds based on firm mechanistic hypotheses gleaned from extensive preliminary investigations. We will apply precisely designed chiral ureas, thioureas, and squaramides to catalysis of enantioselective carbon-carbon and carbon-heteroatom bond-forming reactions. These dual hydrogen-bond donors can abstract or bind weakly basic anions, such as halides, sulfonates, phosphate, and carboxylates, to promote concerted substitution reactions or generate chiral ion pairs that remain tightly associated during subsequent enantioselectivity- determining reactions of the prochiral cations. We discovered that the combination of hydrogen-bond donors with achiral Lewis or Brønsted acids generates highly reactive complexes that promote activation of weakly electrophilic substrates to access highly reactive cationic species. This new principle will be directed to creative applications involving atom-economical carbonyl addition reactions and additions to alkenes. The principle of anion-binding catalysis will also be examined in pathways where the catalyst-bound anion acts as the nucleophile in the enantiodetermining bond construction. Activation of polar reagents is applied in desymmetrizing ring- opening reactions and generation of stereogenic-at-phosphorus compounds. We will also pursue a new strategy aimed at applying anion binding by chiral H-bond donors to enhance the reactivity and control the stereochemical outcome of transition-metal catalyzed reactions, and separately in the context of stereoselective and site- selective glycosylation reactions. We have found that precisely tailored bisthiourea catalysts promote stereospecific, invertive reactions of alcohol nucleophiles with glycosyl phosphates via cooperative activation of both the nucleophile and the electrophile. This cooperative mechanism provides a new approach to achieving control over the site of reaction in minimally protected sugars and other polyfunctional substrates. We also aim to uncover completely new classes of chiral catalysts, such as a new class of alkali metal isothiourea-boronate complexes we uncovered unexpectedly and that promote enantioselective, catalytic reactions with highly basic reacting partners.

Fonds de recherche du Québec – Société et culture · FY 2023-2024 · 2023-04

Volet: Bourses de doctorat en recherche; Domaine: Développement et fonctionnement des personnes et des communautés, et vie sociale; Objet: Rapports ethniques et interculturels; Objet: Conditions socio-économiques; Application: Structures et relations sociales; Application: Solidarité sociale; Mots-clés: OPINION PUBLIQUE, ECONOMIE POLITIQUE, IMMIGRATION, RAPPORTS INTERGROUPES, METHODES QUANTITATIVES, METHODES EXPERIMENTALES

NIH Research Projects · FY 2026 · 2022-12

While adoptive T cell therapies (e.g., anti-CD19 chimeric antigen receptor (CAR)-T cells) have demonstrated remarkable outcomes in patients with leukemias and lymphomas, significant variability remains in the potency and durability of the antitumor response, and their success against solid tumors has been limited. Previous studies have identified several key determinants of therapeutic efficacy, including distinct T-cell subpopulations in the CAR-T cell infusion product. CAR-T cell production generally requires ex vivo T-cell activation and expansion, and critical attributes of the CAR-T cell infusion product, including its proliferative capacity, persistence, and antitumor potency, are widely determined during this process. Significant research over the past two decades has established that extracellular matrix (ECM) elasticity, or stiffness, impacts many fundamental cell processes, and impacts various aspects of T cell biology (e.g., synapse formation). However, tissues and ECMs are not linearly elastic materials. The ECM is viscoelastic, with its response to mechanical loading being time dependent. Strong effects of matrix viscoelasticity on stem cell differentiation have been demonstrated, but the interplay of matrix stiffness and viscoelasticity on T cell activation is unknown. This project addresses the hypothesis that matrix viscoelasticity and stiffness during activation will directly impact T cell phenotype and therapeutic efficacy. This hypothesis will be explored via the following specific aims: (1) Assess the effects of matrix viscoelasticity and stiffness on T cell phenotype using ECM mimetic hydrogels with tunable stiffness and viscoelasticity, (2) Explore the mechanism by which matrix mechanics regulate T cell differentiation during activation, and its relation to T cells isolated from patients using scRNA-seq analysis and focusing on the AP-1 pathway, and (3) Elucidate the functional effects of changes in T cell state induced by matrix mechanics both in vitro and in vivo using adoptive transfer studies. Completion of these studies will provide fundamental knowledge regarding the role of matrix mechanics on T cell phenotype and function, with a potential impact on approaches to manufacture T cells for adoptive therapies.

NIH Research Projects · FY 2025 · 2022-09

PROJECT SUMMARY The uterus has the unique ability to support the growth, development, and eventual delivery of offspring. The non-pregnant uterus is no less remarkable: the uterine lining (endometrium) undergoes repeated cycles of shedding during menstruation and subsequent repair, ultimately regenerating approximately 400 times over the reproductive lifespan. This repeated, scarless regenerative process holds immense potential for the identification of new strategies to replace old or damaged tissues, which is a major goal for regenerative medicine. Moreover, understanding endometrial regeneration has important clinical implications, as excessive or insufficient endometrial regeneration gives rise to pathologies that affect the lives of hundreds of millions of women, non- binary people, and transgender men around the world. It is critical to address the longstanding unmet needs of people with endometriosis, adenomyosis, infertility, and to combat the rising incidence of endometrial cancers. The goal of this project is to identify the molecular and cellular basis for regeneration of the menstruating endometrium. A major challenge for the field is that humans belong to a very small group of mammals that menstruate. Thus, although animal models propel many scientific studies, common animal models such as mice and rats have limited utility for studies of menstruation. This project uses the common spiny mouse, which is the only known menstruating rodent, to perform functional tests of the requirements for menstruation. In parallel, we will develop approaches to determine how the human endometrium changes over time. Using these approaches, we aim to understand how the endometrium changes when menstruation begins during puberty and restarts after pregnancy. The proposed studies are an ideal fit for the New Innovator program because our understanding of the menstruating endometrium has lagged far behind more commonly studied regenerative organs. Thus, we must take strategic risks to propel us towards more precise mechanistic understanding of menstruation, a fascinating process that affects the lives of large proportions of the population.

NIH Research Projects · FY 2025 · 2022-09

Many fundamental cellular functions depend on a variety of RNA structures conserved through evolution, and other functional RNA structures are expected to be discovered. A signature of a conserved RNA structure is found in alignments where paired positions display correlated substitutions (covariation) that preserve the base pair. This evolutionary signal can be used both to predict RNA structure and to identify new conserved RNAs. Recent publications and preliminary results have made three important advances: A statistical covariation test that identifies significant covariation over background covariation due to phylogeny. This test, implemented in a method called R-scape (RNA Structural Covariation Above Phylogenetic Expectation), provides information and control over the rate of false positive predictions. A power of covariation calculation, recently published, that identifies “negative” pairs with power (variation) but insignificant covariation, unlikely to form RNA base pairs. A new cascading folding algorithm, named CaCoFold (Cascade covariation/variation Constrained Folding) also recently published, that combines all positive and negative evolutionary information into complex structures including all types of pseudoknots and triplets. In human, the efficacy of these advances has been tested by ac- curately predicting the structures of the human non-coding RNAs MALAT1 and telomerase RNA, and by inferring that the non-coding RNAs HOTAIR and XIST do not have a conserved structure. These three advances give us a competitive advantage to perform unbiased genome-wide screens for con- served structural RNAs in vertebrates with accurate 3D structure prediction. Previous vertebrate screens for structural RNAs have been hindered by thousand of false positive predictions. In contrast, our new covariation statistical test allows for controlling the rate of false positives. R-scape has already been used to find struc- tural RNAs in bacteria and viruses. Our recent eukaryotic pilot screen in fungi has identified 17 novel structural RNAs. We hypothesize that many structural RNAs with implications for human health and disease are still to be discovered, and that we now have the tools to find and characterize these RNAs. This proposal has three specific aims that will advance the study of structural RNA biology, and the discov- ery of novel biological mechanisms involving RNA structures. The first aim proposes systematic genome-wide searches to find novel conserved vertebrate RNA structures in human. The second aim proposes to combine revolutionary 3D structure prediction methods in machine learning with the signals used by CaCoFold into a state of the art RNA folding method for the accurate prediction of 3D RNA structures. The third aim introduces a method to identify RNA structures in ultra conserved vertebrate UTRs where there is no covariation signal, and our current method lacks power. We expect our work will unveil primate-specific novel regulatory mecha- nisms. Novel human RNA structures found to have causal variants associated with disease will be prioritized for experimental verification.

NIH Research Projects · FY 2025 · 2022-09

PROJECT SUMMARY ABSTRACT Ligand-triggered events are central to many processes in neuroscience, endocrinology, virology, immunology, and pharmacology. However, molecular and ultrastructural changes that follow the stimulus are difficult to visualize because they involve rapid nanoscale motions and modifications of proteins and membranes. State-of- the-art techniques are insufficient to capture these spatiotemporal changes. For example, live fluorescence imaging is limited by the spatial resolution (diffraction-limit) and labeling constraints (no antibody access or washing in live cells), while nanoscale imaging methods either lack temporal resolution to capture fast dynamics (e.g., super-resolution optical microscopy) or are incompatible with live-cell imaging altogether (e.g., standard or cryo-electron microscopy; expansion microscopy). Given these limitations, time-resolved cryo-vitrification methods are ideal for capturing cellular processes after a defined wait time post-stimulation by freezing samples in the state of amorphous ice prior to imaging. High-pressure freezing (HPF) is often used for this purpose because of its relaxed sample thickness constraints (<300 µm) as compared to cryo-plunging at atmospheric pressure (<10 µm). However, an HPF device compatible with time-resolved buffer exchange does not currently exist. To this end, we will develop HPF-X – an HPF device with a capability for time-resolved buffer exchange preceding cryo-vitrification. Buffer exchange will allow stimulating the sample with various biological and pharmacological agents including ions, small molecules, peptides and proteins (e.g., hormones, cytokines, antibodies, and nanobodies), and even viruses and cells. Thus, HPF-X will allow cryo-vitrifying cells, tissue samples, or entire small organisms at a series of time points following stimulation with ligands for subsequent interrogation with nanoscale imaging techniques such as electron microscopy and super-resolution optical microscopy. This approach will allow capturing ligand-triggered cellular processes with nanoscale spatial resolution and temporal resolution of <50 ms. Biological applications of this technique include nanoscale imaging of protein-protein interactions, post-translational modifications, and protein-membrane dynamics. Although a fundamentally new HPF instrument design is required to allow buffer exchange, our extensive preliminary data confirms feasibility. In Aim 1, we will develop a high-pressure chamber compatible with buffer exchange and cryo-vitrification and characterize its performance. In Aim 2, we will develop a method for time-resolved cryo- cooling and validate the system using gold-standard biological samples. Development of HPF-X is an emergent technical opportunity given the advent of nanoscale bioimaging. Importantly, this work goes beyond the current method development regime in cryo-vitrification field because all available HPF devices are commercial. Our custom-built HPF-X instrument will allow full control, versatility, and ease of adoption and modification by other researchers based on their project needs, which cannot be achieved with off-the-shelf HPF instruments.

NIH Research Projects · FY 2025 · 2022-09

Project Summary Hox genes serve as critical regulators of developmental processes. Disruption of their function during embryogenesis results in dramatic “homeotic” phenotypes where regions of the body are transformed from one identity to another. In humans, these disruptions can lead to malformation of the face, ears, limbs, and genitalia, as well as neural defects and cancer. In many animal genomes, the Hox genes are found in clusters: in vertebrates, these clusters are compact, while those of invertebrates are more loosely arranged or fragmented. While still poorly understood, the structure of the Hox cluster is hypothesized to be important in regulating their deployment. However, this is difficult to study in vertebrates as their genomes encode multiple Hox clusters that are the result of whole genome duplications. While invertebrates typically have a single complement of Hox genes, many invertebrate Hox clusters are disrupted, including those found in the classic invertebrate model systems like flies and nematodes. To address this deficit, we have developed resources and tools for studying cephalopod molluscs (squid and octopus), including chromosome-scale genome assemblies, extensive transcriptomics, and tools for gene manipulation. Through this work, we have found that cephalopods have a single, intact, but massively expanded Hox cluster. In fact, they encode the largest Hox clusters yet described – the squid Hox cluster is two orders of magnitude larger than those found in humans. Conservation of the Hox cluster in cephalopods is particularly striking given that their genomes are otherwise highly rearranged relative to other animals. Notably, we have found that cephalopod Hox genes exhibit the canonical, collinear nested domains of expression, suggesting that elements of the ancestral regulatory program are retained in cephalopods despite the dramatic increase in cluster size. Surprisingly, our preliminary knockout data suggest that loss of a Hox gene results in the absence, rather than the transformation, of body regions. These results - the first functional analysis of Hox genes in a mollusc - point to a fundamentally different mode of action than the homeotic transformations characteristic of overtly segmented animals like flies and humans. Understanding differences between the massive cephalopod Hox clusters and the more compact arrangement found in vertebrates will provide fundamental insights concerning the regulation of these body plan transcription factors across diverse animal species, including humans. This project is therefore poised to provide transformational insights into the biology of Hox genes, which play key roles in human development and disease, and contribute to our fundamental knowledge of how pattern is established in embryogenesis.

NIH Research Projects · FY 2024 · 2022-09

PROJECT SUMMARY The central process of female reproduction is the formation of oocytes within the developing ovary, known as oogenesis. This process is crucial for the formation of healthy oocytes and proper transmission of genetic and epigenetic information to begin embryonic development. Abnormalities in ovarian development and oogenesis are a leading cause of female infertility and disorders of sexual development, and furthermore are the cause of many developmental disorders in the subsequent generation, such as Down syndrome and Angelman syndrome. However, relatively little is known about the genetic regulation of human ovarian development. This is in contrast to other organisms such as the mouse, where transgenic and knockout lines, and a short reproductive cycle, have allowed much research in this area. An in vitro organoid model of human ovarian development would help fill this gap, and enable an improved understanding of human ovarian development that could lead to treatments for infertility and prevention of developmental disorders. Ovarian development involves interactions between primordial germ cells (PGCs) and somatic cells (granulosa cells). The granulosa cells enclose the PGCs within ovarian follicles, and support their differentiation into oogonia, progression through meiosis, and development as oocytes. Therefore, both lineages will be required to model this process in vitro. Existing methods allow differentiation of induced pluripotent stem cells (iPSCs) into PGC-like cells, but these cells are in an immature state, retaining epigenetic characteristics of iPSCs. For an in vitro model of oogenesis to be successful, improved methods must be developed to generate mature germline cells and granulosa cells from iPSCs. Reprogramming of cellular identity by expression of transcription factors (TFs) is a powerful technique that can allow both reprogramming of somatic cells into iPSCs, and directed differentiation of iPSCs to specific cell types. The Church lab has recently developed computational tools to predict TFs that specify cell identity, as well as screening methods for combinatorial TF expression to find sets that can differentiate iPSCs to a cell type of interest. The currently proposed research will identify TFs that can promote maturation of PGC- like cells and produce granulosa cells from iPSCs. Results will be evaluated by single-cell transcriptomic and epigenetic profiling, and by functional validation of key phenotypes. This research will provide an improved understanding of the genetic regulation of ovarian development, leading to an in vitro model of human oogenesis.