Autism Spectrum Disorder (ASD) is a devastating neurodevelopmental disorder that causes deficits in social interaction and behavior. To date, 1031 genes are linked to ASD, suggesting its strong genetic background. However, how perturbations in these genes contribute to ASD pathology at the cellular and molecular levels remains largely elusive. Outer radial glia (oRG) are a recently identified population of neural stem cells that is highly enriched in the brains of humans and higher primates. oRG generate the majority of human cortical neurons and contribute to cortical expansion and folding. In this project, we hypothesize that oRG are involved in the pathology of ASD based on several observations. The brains of ASD patients during early development show altered growth and folding, which suggests possible defects in oRG proliferation or differentiation. In addition, a study published by the Li lab found that a mutation in a known ASD gene led to increased proliferation of oRG. Here, I aim to employ CRISPR-based knockout screens to investigate the role of oRG in ASD pathology using human pluripotent stem cell (hPSC)-derived oRG in 2D neural cultures and 3D brain organoids. To realize my aim, I will perform CRISPR screens in arrayed and pooled formats to determine which ASD gene knockouts perturb oRG at both the functional and transcriptomic levels. Based on the results of the screens, I will select high-potential ASD genes that induced significant changes in oRG. Then, I will generate 3D cortical organoids from stable CRISPR knockout hPSC lines and examine in detail how ASD gene disruption affects oRG dynamic behavior, oRG-derived neuronal populations, and organoid development. By leveraging human stem cell models, together with the recent advances in CRISPR screens and single-cell transcriptomics, the proposed study provides a targeted large-scale approach to fill a current gap in our knowledge on the role of oRG in ASD pathology. On a global scale, my study will shed light on the different gene networks that converge within oRG for ASD pathology, which is expected to pave the road for potential biomarkers and therapeutic targets that can improve disease outcomes in ASD patients.

Elevated extracellular protease activity is a hallmark of cancer, known to play an important role in shedding and activating receptors and remodeling the extracellular matrix. Technologies that allow us to understand the complexity, dynamics, and heterogeneity of extracellular proteolysis would significantly advance our understanding of how proteases impact tumor growth and metastasis, and inform how we could harness elevated proteolysis to design safer cancer drugs. It remains difficult to characterize tumor microenvironment protease activities with current molecular tools. RNAseq can determine intracellular transcript levels but doesn’t report protease activities or extracellular proteolysis. Protease sensors, such as genetically encoded reporters or small molecule probes, only allow testing of one substrate sequence at a time, which prevents us from a systematic understanding of the proteolysis landscape. To break the technology barrier for a deeper understanding of the protease activities in the tumor microenvironment, I propose to develop proteolysis signaling integrators (PSIs) that can transform a proteolysis event into a transcriptional event within a defined time window. I have designed two systems for the PSIs. The first is based on leucine zippers, alpha-helical dimerization motifs found in transcription factors. The second one is based on Protease-Activated Receptors (PARs), G protein-coupled receptors activated by proteases. In both systems, I have designed to couple a proteolysis event of a substrate sequence to release a Gal4 transcription factor (TF). The design of PSIs is modular so that a large library of substrates can be encoded and tested simultaneously. A small-molecule controllable destabilizing domain (ddDomain) will be incorporated into a PSI to control the time of expression. Tumor cells engineered to express the library of PSIs will be implanted in mice and activated for expression after the tumor grows to a specific size. If a proteolysis event happens, the Gal4 TF will drive transcription of an enhanced green fluorescent protein (EGFP), resulting in a quantitative signal of the proteolytic events. Following isolation of the tumor, cells with low, medium, and high EGFP signals will be separated and sequenced to determine the propensity of proteolysis for different substrate sequences. The development of PSIs has the potential to transform how we characterize extracellular protease activities, and the gained knowledge will inform the development of a new generation of protease-based prodrugs.

Emergent tissue architecture is orchestrated by the mechano-chemical feedback interactions among the cellular and extracellular components, across scales. While the biochemical principles of tissue organization have been well studied, the mechanical and biophysical aspects remain less well understood. In a homeostatic tissue, the constituent cells both deposit and maintain the extracellular matrix (ECM), thereby determining tissue mechanics, and are responsive to the mechanical properties of the ECM. An imbalance in these processes can lead to pathologies such as fibrosis and cancer. Hence, remodeling the spatial and biochemical interactions of cells and the ECM can decide the tissue fate in between health and disease. Moreover, physiological aging is associated with changes in tissue structure and mechanics, which may relate to the development of pathologies. However, a holistic view on the in vivo principles governing tissue fate as a function of the complex feedback loops among cell types, cell states, ECM, and tissue mechanics is yet to be defined. I hypothesize that mechanical heterogeneity within the homeostatic tissue leads to the formation of stiffness-dependent cell states, which prime the cells in stiff regions to disease. Hence, the aims are – 1) to characterize the in vivo spatial arrangement of various cell types and their respective transcriptional states as a function of the surrounding ECM and tissue mechanics, and 2) to identify specific cell-matrix feedback loops that define the tissue fate. Using the homeostatic and fibrotic murine skin as models of health and disease, I will spatio-temporally map their mechano-chemical properties by combining mechanical mapping (nano-indentation) with high-resolution spatial matrisomics (imaging mass cytometry) and transcriptomics (Stereo-seq). With these data, I will uncover the spatio-temporal correlations among tissue mechanics, topology, and biochemical properties, including cellular composition, arrangement, and their transcriptomes, as well as the ECM alignment and stiffness. Next, to determine the causality of these correlations, I will manipulate the cell-matrix loops in silico using agent-based modelling and in vitro using co-cultures with stiff vs. soft matrix-generating fibroblasts. I will then systematically test these theories by using 4D live imaging coupled with nano-indentation in tissue explants. The shortlisted feedback loops will be manipulated in vivo in the homeostatic tissue using mosaic expression systems and gene-level perturbations (Perturb-seq) and assayed for the tissue fate. Finally, I will test whether these cell-matrix loops are conserved in human pathologies, including skin cancer/fibrosis. Altogether, this project will provide a comprehensive understanding of disease emergence and how local mechano-chemical feedback loops orchestrate the macroscopic tissue fate in space and time, and opens new avenues for generating preventive therapeutics.

The development of a multicellular organism requires the precise control of gene expression in space and time so that cells adopt their correct identity. However, genetic mutations can alter this complex process. Recently, a transcriptional adaptation (TA) has been uncovered as one of the mechanisms underlying genetic compensation in zebrafish, mouse cells in culture, and C. elegans1,2. TA refers to the phenomenon by which mutated genes (with mRNA-destabilizing mutations) trigger the transcriptional up-regulation of other genes, called adapting genes. Mutant mRNA degradation, e.g., via Nonsense-Mediated Decay (NMD), has been shown to be required for TA3,4. However, little is known about the spatial and temporal characteristics of adapting gene regulation. Yet, TA would be required to be fast and regulated at all the different steps of gene expression in order for the embryo to develop correctly. Recent technological advances have made it possible to access the dynamics of transcription but only a few have been implemented in vertebrates thus far. This project aims to decipher when and where the TA occurs in the context of the developing zebrafish embryo. Aim 1 will test the hypothesis that TA is initiated during the zygotic genome activation and that there are several possible modes of transcriptional upregulation of the adapting genes. To test this hypothesis, I will use mutant alleles of early expressed genes that have been shown to exhibit TA including vcla, aldh2 and alcama. These genes and their corresponding adapting genes will be tagged with different live reporter arrays (e.g. MS2, PP7 loops) using standard genome editing tools in order to monitor transcription in live embryos with high-end microscopy followed by quantitative image analysis. This approach will allow us to determine the time-scale at which TA occurs and also help us understand the different modes of transcriptional upregulation of the adapting genes. Aim 2 will test the hypothesis that mutant mRNA degradation occurs in specialized organelles that are critical for TA. To test this hypothesis, I will use high resolution microscopy to look at the localization of NMD components as well as membrane-less organelles, such as P-bodies granules, together with the mutated and the wild-type mRNA. To test the importance of these granules, I will use mutants defective in P-granule formation and look at the effects on TA in terms of transcriptional output. Until now, TA has been mostly investigated on pooled populations of cells, therefore we lack the understanding of this phenomenon at a cellular level. This project will be important to fill this gap and get a better understanding of the spacio-temporal characteristics of genetic compensation which help the robustness of vertebrate development. 1. Rossi, A. et al. Nature 524, 230–233 (2015). 2. Ma, Z. et al. Nature 568, 259–263 (2019). 3. El-Brolosy, M. A. et al. Nature 568, 193–197(2019). 4. Serobyan, V. et al. Elife 9, e50014(2020).

Bacterial viruses (phages) outnumber bacteria by an order of magnitude in the oceans, and they are thought to heavily influence the abundance and diversity of bacterial populations. Yet this view is hard to reconcile with the dilute nature of the ocean environment and the extremely narrow host range of most phages, owing to the vast diversity of bacterial anti-phage defenses. Here, I hypothesize that cooperative co-infections by different phages can overcome defenses to effectively broaden their host range. This hitherto overlooked mechanism is suggested by recent observations of frequent recombination among co-occurring phages, even though these are unable to lyse the same bacterial hosts, and may explain how phages collectively overcome the sum of the bacterial anti-phage defense systems. I propose to establish a phage co-infection matrix to (i) identify cooperative lysis of bacteria by phage pairs as well as cooperative defense of different defense genes within hosts, (ii) use the co-infection matrix to computationally predict genes and genetic elements guiding cooperative interactions, and (iii) experimentally test these predictions using molecular genetics. This work is enabled by the Nahant collection, the largest fully genome sequenced cross infection matrix of co-occurring phages and their hosts, consisting of >1,300 marine Vibrio isolates and 251 lytic phages. This matrix, constructed by using individual phage isolates, showed sparse infection due to multiple anti-phage defense systems in each bacterial strain residing on diverse mobile phage defense elements (PDEs), so that solitary phages can only infect one Vibrio strain on average. Importantly, PDEs are absent from lab-domesticated bacteria (presumably due to their fitness costs in the absence of phages), highlighting the value of the Nahant collection in studying native phage-bacteria interactions. Using a high-throughput liquid co-infection assay, I will identify phage cooperation by infecting PDE-containing Vibrios with binary phage-combinations, to find phage-pairs that lyse their hosts upon co-infection (but not independently). I will receive training in analyzing phage genomes and metagenomes, to select syntenic phages from distinct groups. Cooperative infections between syntenic viruses will enable me to computationally pinpoint the cooperation-guiding genes, as the unique genetic fragments should be encoding complementary anti-defense systems. Because these phage-host interactions are important beyond the ocean, my novel approach will not only allow me to reveal cooperation as a main driver of one of the most prevalent ecological interactions, but also to guide rational phage therapy design that already relies on combining multiple different phages for unknown (but perhaps cooperative) reasons.

Insect hearing can match and even outperform their vertebrate counterparts, for instance, a species of moth can hear up to a few hundred kHz or ten times greater than that of humans (1). However, insect auditory organs operate at microscopic scales making it difficult to understand how they work. The site where sound is converted into electrical signals are membrane protrusions termed cilia. They are so small that they are at the diffraction limit of light. The cilium contains mechanically sensitive ion channels (2,3) that open when they are displaced by sound. Not only is the identity of the channels disputed, but the mechanism by which the insect mechano-electrical transduction (MET) machinery operates is largely speculative. These two complementary questions form the two aims of this fellowship proposal. 1. Identify the MET channels using CRISPR-Cas9-mediated mutations of the two channel candidates. There are two contenders for the MET channel in insects: NompC that resides near the ciliary tip (4) and Nanchung-Inactive located at the base of the cilium (5). My host lab has recently established the experimental workflow for CRISPR-Cas9 edited genetic mutants in the desert locust. I will individually mutate Nanchung, Inactive and NompC and measure the resulting transduction current. These will constitute the first transduction current measurements from insect auditory neurons with genetically mutated transduction channels and will finally reveal which of these two channels is the bona fide auditory transduction channel. 2. Characterise sub-micrometre displacements of cilia with a dynamic super-resolution imaging. Sound-induced movements of the cilia are hypothesised to be less than 200 nm (6) putting it below the resolution of light microscopy. Working at the interface of biology and physics, I will characterize internal motions of cilia in the auditory organ of the desert locust by measuring the proportional activation of photosensors from live images of the cilium projected onto a high-speed (10,000 frames per second) and high photon-efficiency (95% photon capture) CMOS-based camera. My host lab has established proof-of-concept of this imaging system down to the resolution of 10 nm. Live imaging of the cilium, in response to sound, will be combined with electrophysiological measurements of the MET currents from individual auditory neurons (a technique pioneered by the host lab (7)). I will combine the two physiological techniques offered by the host lab with my background in the biophysics of hearing to reveal how the cilium is activated through sound and solve an enduring enigma for the field of mechanotransduction. REFERENCES 1. Moir HM, et al. BioLett. 2013, 9:4. 2. Field LH, et al. AdvInsectPhys. 1998. 3. Vavakou A, et al. PNAS. 2021, 118:39. 4. Liang X, et al. Cytoskeleton. 2011, 68:1. 5. Kim J, et al. Nature. 2003, 424:6944. 6. Moran D, et al. PNAS. 1977, 74:2. 7. Warren B, et al. JNeurosci. 2018, 38:15.

The LacI superfamily of Transcriptional Regulators (TRs) are common in bacteria and allow for the sensing of molecules to regulate gene expression. TRs have a DNA binding domain and a solute binding protein (SBP) domain, which binds specifically to a variety of small metabolites. Ligand binding results in conformational change and allosteric control of the TR’s DNA binding affinity, coupling small molecule binding to gene regulation. The recent explosion in genomic and metagenomics sequencing has unveiled a wealth of sequence diversity. However, only a handful of TR sequences have been characterized, leaving the vast majority of the metabolite binding sequence space unexplored; we estimate that >95% of sequences clusters are ‘orphans’, with no closely related (>35% sequence identity) sequences of known function. Thus, there is great opportunity to expand our understanding of TRs and molecular recognition in general, as there is a plethora of undiscovered ligand binding solutions with unknown mechanisms. In this project, I propose to survey the unexplored sequence space, by characterizing and functionally mapping orphan TR sequences, followed by a multifaceted machine learning approach to predict ligand binding throughout the rest of the family. We will first generate a sequence similarity network (SSN) of the ~1 million TRs collected from the NCBI database, which clusters sequences by sequence identity. Using the SSN, we will identify approximately 200 diverse ‘orphan’ TR clusters, then synthesize and clone approximately 200 of these sequences into expression vectors. To characterize each TR in terms of their function as an activator or repressor, each gene will be co-expressed into a library of reporter plasmids expressing an On/Off selection system that contains a binding site with randomized DNA promoter sequence; this system selects (by survival) for transcriptional activation or repression. The TRs will then be screened using a small molecule library (~1,000s of compounds), where each gene/DNA library is screened with and without the small molecule, and co-selected for active or repressed expression of the reporter. Finally, each condition in which activation/repression is observed will be identified with a molecular barcode, pooled and sent for deep sequencing, allowing mapping of all DNA recognition sequences and all responsive ligands for every screened TR by highlighting the surviving DNA/ligand combinations under On/Off selection. We will analyze novel TRs using protein crystallography, structure prediction (alphafold) and phylogenetic analysis. Most importantly, this data will be used to train machine learning algorithms to allow us to accurately predict both DNA binding specificity and ligand binding throughout the entire superfamily. We expect this project to greatly enhance our knowledge of gene regulation and molecular recognition, allowing for further expansion of the fundamental study of prokaryotic gene regulation.

Synaptic transmission is crucial for brain function and defects are associated with many neuropsychiatric disorders. Presynaptic mechanisms regulating neurotransmitter release decode the information stored in neural firing, and functions have been assigned to several of the key proteins involved in synaptic vesicle docking, Ca++ sensing and membrane fusion. However, the mechanisms, regulation and coordination are not well understood. The function of many other synaptic vesicle proteins also remains unknown. Indeed, previous work has focused on the function of individual proteins without considering the larger context, leading to the assumption that membrane proteins are randomly distributed on the synaptic vesicle. However, interactions between several of them suggest organization which would have profound implications for neurotransmission, behavior and disease. The long-term objective of this project is to develop a holistic understanding of the diverse mechanisms involved in neurotransmitter release and their role in information processing. The strategy is to determine the structure and organization of proteins embedded in the synaptic vesicle membrane at near-atomic resolution and test the functional implications of the most salient findings. I hypothesize that the organization of synaptic vesicle membrane proteins serves to coordinate their function in neurotransmitter release. Previous work on individual proteins involved in synaptic vesicle docking and fusion has suggested interactions between several of them. For instance, membrane fusion requires multiple SNARE complexes, and the synaptic vesicle contains many copies of the principal v-SNARE VAMP2. The Ca2+ sensor synaptotagmin 1 has been suggested to link these complexes into a ring, which may confer cooperativity. The synaptic vesicle also contains t-SNAREs that may act in cis to regulate interaction in trans with t-SNAREs at the plasma membrane. Determining the organization of these proteins will help to understand their role in release. This work will also provide a foundation to understand how synaptic vesicles interact with other elements of the release machinery such as plasma membrane docking sites and presynaptic Ca++ channels. Synaptic vesicles isolated by differential centrifugation and gradient fractionation will be used for structure determination by Cryo-electron tomography, in collaboration with the Stroud lab. The issue of heterogeneity will be addressed by selective immunoisolation and/or class averaging. Imaging and electrophysiology in primary neuronal cultures will be used to test the predictions for synaptic vesicle exocytosis and neurotransmitter release. The results will provide a framework to understand the function, regulation, and interactions among synaptic vesicle proteins, providing a broad context to understand neurotransmitter release, its role in information processing, behavior, and dysfunction in disease.

Urinary tract infections (UTIs) are among the most common infections in the world and the primary cause for the prescription of antibiotics. WHO has recently highlighted UTI treatment as a critical exacerbating factor in the global antimicrobial resistance (AMR) crisis. However, the pathophysiological mechanisms underlying UTI and its frequent recurrence remain largely understudied, hindering the development of effective therapeutics. The vast majority of UTI research has been performed in mouse models or cancer cell lines using uropathogenic E. coli (UPEC) as the predominant agent. However, the obtained results often cannot be fully translated to humans due to differences in bladder ultrastructure, physiology/function and immunity. Therefore, the main goal of this project is to unveil the main strategies and virulence factors underlying UPEC colonization and persistence using a physiologically relevant human microtissue bladder model. The following hypotheses will be addressed: i) Does bacterial adhesion, invasion and the formation of intracellular bacterial communities occur in the human model as previously described for the mouse model?; ii) Which virulence factors (e.g. flagella, type I- or P-pili) are critical for the establishment and persistence of UTI?; iii) What is the role of the innate immune response (e.g. diapedesis and phagocytosis) for the outcome of the infection (persistence vs. clearance); iv) Which of these strategies and molecular mechanisms are shared between different UPEC strains? A multidisciplinary approach will be taken by the in-depth characterization of a perfused, urine-tolerant, fully stratified 3D-human bladder microtissue model, that is under development in the host lab at Biozentrum, Univ. of Basel. Bladder biopsies and urine from UTI patients obtained from the University Hospital Basel will be used for validation purposes. The initial assessment of pathophysiological mechanisms will be performed using UPEC strain CFT073 (broadly used in mice infection models), together with mutants in virulence factors identified by a genome-wide CRISPRi screen in the host lab. To depict the general infection strategies (adhesion, intracellular communities, etc.), I will also address mechanistic aspects of the underlying virulence processes, targeting host compartments and bacterial factors (e.g. flagella, pili, etc.) at single cell resolution using 4D spinning disk live cell imaging. Selected clinical UPEC isolates will be used as comparison and immune cells (i.e., neutrophils) will be added to the model to assess UPEC clearance. Overall, this project will allow the study of basic pathways underlying UTIs and mimic recurrence in a powerful platform, serving as a crucial complement (or even alternative) to the current mice models that dominate the field. We will be able to dissect with high-resolution the specific mechanisms and players involved in the UTI establishment and its persistence, which is a current demand.

Background: From a tiny insect to a giant blue whale the diversity in the animal kingdom is truly remarkable. An understanding of this poorly explored variation in size of organisms can provide a more elaborate map of evolution and will be useful in commercial applications such as increasing meat production from specific body parts. Given the complexity of the animal body, it is more practical to start with one organ that can offer a high range of diversity among different species such as the limb. The differences in size and proportions demonstrating their adaptations during evolution open a door to the factors controlling organ size. Two main steps can be distinguished during limb growth: the patterning phase, in which the different limb segments are specified, and the post-patterning phase, where most of the elongation and differentiation take place. Aim: We aim to generate jerboa iPSCs for the first time to study how they form limbs in host mouse embryos. We plan to compare the development of chimeric jerboa-mouse limbs with that of normal mouse limbs so that their intrinsic differences can be found. Central hypothesis: Based on classic and recent studies, we hypothesize that final limb size is controlled by the signal balance during the patterning phase, whereas internal limb proportions are tuned by intrinsic differences in the tempo of the post patterning phase. This predicts that if early limb bud cells are placed within the signal environment of a different species, the limb bud cells will adapt the duration of their patterning phase, leading to a change in the limb size, while the internal proportions will be preserved. This decoupling between intrinsic and extrinsic processes will allow us to identify the key genes and networks involved in the control of limb size. Experiments: The experiments will take place in three stages: 1) Preparation of jerboa iPSCs under the expert supervision of Dr. Rio Tsutsumi, Kyoto University. The jiPSCs will be chemically transformed in four stages and then characterized by RNA-seq and teratome assays for pluripotency. They will be modified for GFP expression 2) Preparation of hindlimb-less mouse model (Hoxb6-Cre; Ctnnb1(flox/flox)) and complementation of its blastocyst with jiPSCs. Litters will be collected to characterize the expression of segment markers Meis1/2, Hoxa11, Hoxa13 by in situ hybridisation, characterize bone length by micro-CT. Morphological and size analysis will be done according to protocols from the Cooper Lab 3) Relative RNA-seq and ATAC-seq analysis of transcriptome and accessible chromatin of limbs. Conclusion: These experiments are designed to generate the first iPSC line from jerboa providing new resources for research. jiPSCs are expected to form most of the hindlimb in chimeric animals. The size and genetic analysis of the developed hindlimb will provide an insight into what aspects of limb growth are controlled by intrinsic mechanisms, and which aspects by extrinsic ones.

The tumor microenvironment is characterized by a dense extracellular matrix in combination with leaky and irregular vasculature, creating acidic hypoxic tumor regions with small influx of oxygen and nutrients. Most therapies, such as radiation therapy, immune therapies, and chemotherapy are ineffective in these regions due to lack of oxygen, limited drug influx, immune exclusion, and low proliferation rate. Effective ways to target the cancer cells in hypoxic regions are lacking and means to address this problem could potentially improve the clinical outcome for many groups of patients. An important step towards finding effective treatments targeting hypoxic regions is to understand their biology. The arrival of spatial transcriptomics has made possible high-resolution studies of the transcriptional landscape in tumors. In spatial transcriptomics, spots containing 1-10 cells are sequenced and associated with a physical coordinate from a thin tissue section, forming a spatial network of gene expression. Hypoxic regions can be identified using immunofluorescence staining for hypoxic markers, enabling identification of biological traits per level of hypoxia. In this project, we will collaborate with the Human Tumor Atlas Network (HTAN) to obtain spatial transcriptomics data overlayed with hypoxia staining from lung and breast cancer patient samples. The first aim is to establish a method to determine the level of hypoxia per spot based on staining for hypoxic markers, but also from the transcriptomic profiles for use with datasets without staining information. The second aim is to identify regions within the hypoxia range with similar cellular behavior. The third aim is to explore and understand the biology and identify potential drug targets in each hypoxic region. To this end, I intend to use methods such as differential expression analysis, gene set enrichment analysis, genome-scale metabolic modeling, and utilize gene essentiality information from the DepMap database. The fourth aim is to collaborate with the PRISM lab at the Broad Institute, which performs high-throughput screening for drug targets in hundreds of cell lines. To validate our biological theories and provide potential drug targets for hypoxic regions, we will develop a hypoxic setup to test previously identified drug targets as well as perform screening of existing drug libraries for the hypoxic regions previously identified. In this project, I will bring my metabolic systems biology knowledge into the Getz lab and learn about spatial transcriptomics and cancer biology. The results are envisioned to be an atlas over the transcriptional landscape in tumors defined per region of hypoxia and a list of potential drug targets for these regions. These could serve to find a complementary treatment to existing therapies, specifically targeting the hypoxic regions, which could lead to combination treatments with an increased survival chance for cancer patients.

Background: Depression is the leading cause of disability worldwide, creating immense psychological and economic burdens for individuals, families and society. Despite depression’s pervasiveness, the pathology and mechanism of action for treatments remain elusive, and roughly half of patients do not improve with antidepressant use. The high global prevalence, poor treatment efficacy and lack of mechanistic understanding has created a large knowledge gap that needs to be addressed to better understand the disease and help patients. The gut microbiota is an emerging contributor to depression onset and recovery, and can regulate brain function, behavior, and mood of its host by altering the metabolome and immune system. Intriguingly, gut microbial composition is altered in depressed individuals, post-antidepressant treatment and by stress. However, the exact pathways by which these microbes act on the brain, including in depression, remain elusive largely due to an inability to mechanistically interrogate microbe-host interactions. Overall aim: Uncover how the gut microbiota contributes to the development and recovery from stress-induced depressive behavior (initially in mice). Hypothesis: The gut microbiota contributes to stress-induced depressive behavior and amelioration through microbe-host signaling (including modulating host serotonin levels and aryl hydrocarbon receptor immune signaling through microbial tryptophan metabolism), and manipulating these pathways is sufficient to augment depressive-like phenotypes in mice. Experimental Approach: To reveal how stress and antidepressants impact the gut microbiome and physiology of mice, I will analyze behavior and physiology (gene expression, corticosterone/neurotransmitter levels, etc) and use contemporary multi-omics methods (metagenomics/transcriptomics) combined with the novel Record-seq technology recently created by the host laboratory (Schmidt, Nature, 2018; Schmidt, Science, 2022). My bioinformatics analysis will focus on understanding microbe-host pathways altered by stress and alleviated by antidepressant (fluoxetine or probiotic) consumption. I will then functionally validate key findings using CRISPR-based manipulation of top identified pathways in the gut microbiome (with host institutional knowledge) and microbiota transplantation to reveal novel mechanisms by which the gut microbiota contributes to anti-depressive or depressive-like behavior. Conclusions: This research has the potential to shift paradigms for depression research by incorporating novel microbiome-based techniques and centering the gut microbiota in the pathophysiology of depression. This project will expand on my background of the microbiota-gut-brain axis and allow me to pursue new scientific disciplines and techniques, including CRISPR, microbial engineering, bioinformatics, and Record-seq. The use of cutting-edge non-invasive methods (Record-seq) also provides an entry point for follow-up clinical studies.

Glioblastoma multiforme (GBM) is the most common high-grade glioma in adults and has an exceptionally dismal prognosis. Neuronal activity has recently emerged as a critical regulator of GBM progression by regulating membrane depolarization of glioma cells via neuron-glioma synapses or secreting neurotrophic factors from neurons. Oligodendrocytes can wrap around axonal segments, enabling fast saltatory conduction of action potentials and supporting axons' energy metabolism. Whether a functional interaction between oligodendrocytes and cancer cells drives GBM remains unknown. To better understand the interactions between tumor and non-tumor cells, the host lab analyzed larger scRNAseq dataset of GBM to uncover the ligand-receptor interactions from the tumor microenvironment (TME) using a recently developed single-cell tumor-host interaction (scTHI) tool and identified that the oligodendrocytes express high levels of NLGN1 that interacts with NRXN3 expressed by neuronal tumor cells. We propose a model wherein oligodendrocytes directly interact with the tumor cells through ligand-receptor interactions and generate oligodendrocyte-glioma myelin sheaths to promote cancer development. This proposal will uncover the functional downstream consequences of the newly identified ligand-receptor interactions and how they drive glioma progression. A key innovative component of our proposal will be to interrogate the ability of glioma cells to form functional myelin sheaths with oligodendrocytes thus hijacking the normal activities of oligodendrocytes. A diverse range of cutting-edge techniques will be used to address these goals. To distinguish and verify the interactions between tumor and non-tumor cell types in GBM, I will deploy spatial transcriptomic technology and multiplexed error-robust fluorescence in situ hybridization (MERFISH). For these analyses, a panel of genes will be designed based on the host lab scRNA-seq data and cell-type/state annotations, to decipher spatial and functional organization. I will also investigate the ultrastructure of oligodendrocyte-glioma myelin sheaths in mouse models and patient tumors by enhanced focused ion beam scanning electron microscopy (FIB-SEM) technology enabling high-throughput, large volume, and high-resolution 3D imaging. Finally, to elucidate the molecular mechanisms of ligand-receptor interactions governing tumor growth or oligodendrocyte recruitment in glioma, I propose to examine the gene expression profiling and gene set enrichment analysis of ligand-null (CRISPR-KO) versus parental ligand-wild-type in cell lines or genetically engineered mouse models of gliomas. The outcome of this work will unravel the regulation and functional impact of the crosstalk between glioma cells and oligodendrocytes in GBM and provide the key knowledge for the design of innovative therapies. It will also provide the paradigm-shift discovery and functional characterization of myelination as driver of GBM progression.

It is increasingly clear that microbes can influence host health and fitness. In particular, microbes affect important developmental transitions like metamorphosis in mosquitoes and flowering time. The timing of these transitions is central for organisms responses in changing environments. However, we do not know which microbes alter such traits and how they do it, or whether it is adaptive for the microbes or a consequence of their metabolism. Through integration of life-history models, microbial physiology, community selection, and plant evolutionary ecology, my goal is to understand microbial effects on plant developmental timing. The ultimate goal is to generate predictive models of when are microbes more likely to impact plant development, which microbial functions are involved, and how are these interactions going to change with an increasingly variable environment. One intriguing hypothesis is that since microbes are indicators of environmental and ecological conditions, they can provide accurate information about optimal phenology to the plant hosts while integrating to noisy variation. I will use flowering time in Arabidopsis thaliana as a experimental system to determine how microbial community function might change with changes in temperature. Recently, Dr. Rebolleda-Gomez developed a predictive model microbial community change across temperatures in simple carbon sources. In collaboration with members of the lab, I will extend these models to more complex communities. Using this data I will be able to analyse the information and reliability of changes in microbial functions over time (compared to the temperature series alone) to evaluate the quality of microbes as cues. These mathematical approaches are new to me, but as part of my postdoctoral formation at UCI I will benefit from the expertise in microbial ecology of Dr. Rebolleda-Gomez and time series analyses of Dr. Symons (a close collaborator). Using recently developed theory on community level selection, I will select from different substrates, microbes that significantly affect flowering time and inoculate A. thaliana lines with different flowering plasticity. I will test the effect of such communities in different thermal conditions to determine when and why plants respond to microbes. Using omics approaches, I will characterize from the selected microbial communities the metabolic functions that are associated with either late or early flowering time. Finally, I will incorporate our insights into predictive life-history models to evaluate possible effects of microbes on plant adaptation to climate change. Altogether, this project will allow me to gain insights into the mechanisms by which microbial communities affect plant phenology, and to hypothesize about the evolutionary implications. This fellowship will allow me to learn the mathematical and experimental tools necessary to generate predictive models of microbiome effects on plant development.

Tissue folding is widespread in morphogenesis. It often arises due to geometric frustration associated with growth mismatch. One of the big questions is how organisms take advantage of the local mechanics, differential growth, and physical and geometric boundary conditions of tissues to drive this folding. One such example may be found in the final stages of Drosophila wing development. During metamorphosis, the wing elongates along the body of the fly as a planar sheet before anchoring to the hard cuticle layer encasing it. From 30 hours into metamorphosis up to 96 hours when metamorphosis ends, the wing has grown by more than an order of magnitude in surface area. The wing manages to continue to grow in the confined space over these 66 hours by folding. During this period, as the wing continues to grow, it also becomes thinner and develops more folds. However, the driving mechanism for this folding phenomenon and the level of variability in the shape change across individual wings are not known. We propose to study how the physical and geometric constraints on the wing and the differential growth of the cellular layer and its attached extracellular matrix (ECM) allows the wing to fold. This project will be an incorporation of live imaging in 3D through the highly robust Gal4-UAS system to identify geometric changes to the wing as it develops folds and will require an analysis of 3D-reconstructed images to identify the patterns of shapes that emerge and the recurrence or variability of these patterns across wings of different individuals. A key piece to better understanding the role of this differential growth between the cells and the ECM as well as the constraints on the folding process will involve genetic and physical perturbation of the wings to further study how making alterations to the development of the wing changes its folding pattern. These include using recently developed optogenetic systems to spatially control changes to local cellular growth rates and the use of genetic mutants to examine how the changing the expression level of genes and the distribution of their resultant proteins affect the boundary or growth conditions on the wing and thus affect the folding pattern. We will also examine how the wings unfold at various time points during development by removing the boundary that encases them and watch how they unravel. We expect these experiments and the concurrent analyses will allow us to understand how boundary conditions that confine the wing and the differential growth and material properties between cell and ECM combine to allow the fly to continue to develop its wings under spatial confinement.

The ecology of bacteria is directly tied to the ecology of the mobile genetic elements (MGEs) that infect them. These elements act as parasites or symbionts, spreading horizontally between bacteria and shaping the genetic repertoire of their hosts. MGEs can be beneficial when carrying traits like metabolic pathways, virulence genes, or antibiotic resistance, which allow their host bacteria to thrive in new environments. Yet, they are also costly due to the expression of the encoded genes. Additionally, evidence is emerging of extensive interaction and competition between MGEs. Many MGEs co-circulate in natural bacterial populations and co-carriage within the same bacterial host is common. However, little is known about “coexistence conditions” for MGEs: how can diverse MGEs be maintained in the same population? And how does this affect the bacterial community in which they circulate? We aim to address these questions for plasmids, a type of MGE frequently implicated in the spread of antibiotic resistance. Plasmids commonly code for genes that prevent further infection by similar plasmids (“superinfection”). We hypothesize this naturally leads to non-transitive interactions between plasmids that compete for the same bacterial hosts, and that this can maintain a diversity of plasmids even in the absence of selection for plasmid-borne traits. Non-transitive interactions work like the game “rock-paper-scissors”: each strain both inhibits and is inhibit by other members of the community. Such interactions have been proposed as a prime driver of species diversity, yet experimental realizations thus far remain restricted to bacteriocins or antibiotics, with limited extension to explaining MGE diversity. We intend to study these dynamics in an experimental system using Escherichia coli strains that differ only in their carriage of two competing antibiotic resistance plasmids: (i) “F0” carries no plasmid and grows fast, (ii) “ML” carries a non-conjugative plasmid with medium cost, and (iii) “SC” carries a costly conjugative plasmid and grows slowest. As a result, F0 outcompetes ML due to growth, ML outcompetes SC due to growth and immunity to superinfection, and SC outcompetes F0 since the conjugative plasmid transforms F0 into SC. We will use pairwise and 3-way competition experiments to study the stable states of this system over multiple growth-dilution cycles. By varying parameters of the experiment, like growth rate, dilution, dispersal, and relative costs, we can probe the conditions for plasmid coexistence over a range of scenarios. Mathematical modelling will help scale these transitivity rules to a broad diversity of co-circulating plasmids and predict their influence on the stability of more complex bacterial communities. Understanding such plasmid diversity and maintenance is directly relevant for the study of bacterial colonization and infection of human and animal hosts, as well as the spread of antibiotic resistance.

Background: The spatiotemporal organization of T-cell receptors (TCRs) is critical to target cell recognition that underlies a novel class of cancer therapies. While dynamic receptor reorganization was observed in canonical T-cells, less is known about the recognition mechanisms of gamma delta (gd) T-cells, which are emerging as attractive candidates. Specifically, the spatiotemporal organization of their receptors upon immune synapse (IS) formation, their mechanosensitivity, and possibly crucial membrane deformations remain largely unexplored. While high-resolution fluorescence microscopy of T-cells contacting supported lipid bilayers (SLBs) presenting target proteins has been crucial to dissect canonical T-cell organization and signaling, SLBs containing select antigens and adhesion molecules cannot be used to probe gd T-cells as not all factors involved in antigen presentation are known for gdTCRs. Aims: Combining my expertise in membrane biophysics and SLB fabrication on non-rigid substrates with the host’s expertise in biophysics, high-resolution fluorescence microscopy and T-cell immunology, we will reveal the spatiotemporal organization of gdTCRs upon immune synapse formation between T-cells and tumor cell-derived SLBs. In brief, we will: A0. Establish experimental platform - mechanically tunable, tumor cell-derived SLB A1. Map the spatiotemporal (re)organization of gdTCRs upon IS formation A2. Assess effect of surface rigidity on organization of gdTCRs and gd T-cell activation Hypotheses: I conjecture that gd T-cell receptors undergo specific spatiotemporal organization to allow antigen recognition in infection and tumorigenesis. Clustering of TCRs in canonical T-cells and exclusion of phosphatases from the IS amplify the signal from antigen-binding events, but dynamic TCR reorganization in gd T-cells remains unexplored. Intriguingly, receptor organization may interplay with membrane deformation and mechanosensitivity, which may both be critical to T-cell activation. I will study this by modulating support rigidity, with which I have ample experience. Experimental approach: E0: Develop polymer-based supports for fluorescence microscopy. Harvest plasma membrane vesicles from tumor cell cultures. Optimize SLB formation. E1: Total internal reflection fluorescence microscopy - single molecule tracking of synaptic components upon gd T-cell contacting tumor cell-derived SLB. E2: Measure force of gd T-cell–SLB interaction as a function of SLB rigidity using atomic force microscopy-based single-cell force spectroscopy and acoustic force spectroscopy. Conclusions: I will observe the spatiotemporal organization of gdTCRs upon immune synapse formation at the molecular level, gaining insight into their activation mechanism – an unresolved question that has stymied progress in gdT-cell application in immunotherapy. I expect my novel, tumor-cell derived experimental platform to open up a range of new avenues in gdTCR research.

Obesity induces chronic inflammation in the intestine and affects the state of non-immune cells such as gut-associated neurons (e.g., enteric and gut-innervating sensory neurons). For example, the intestinal immune system undergoes proinflammatory shifts during obesity, contributing to metabolic dysfunction; while the sensitivity of sensory neurons to satiety-inducing peripheral signals (e.g., distension and nutrients) are disrupted in obese conditions, which leads to food overconsumption. Furthermore, a growing body of evidence has revealed the importance of neuroimmune crosstalk in several pathological conditions, including obesity. Thus, in this proposal, I hypothesize that enteric neuroimmune interactions underlie the cause of multiple obesity-related pathological symptoms, and manipulation of these interactions may critically modulate those symptoms during the development of obesity. To test this idea, first, I will use diverse techniques to characterize the enteric neuroimmune interactions in multiple aspects during the development of diet-induced obesity (Aim 1, 2). With the Victora lab, the Mucida lab (host lab) has recently developed a murine genetic approach that enables the labeling of cell-cell interactions between specific cell types (universal, or uLIPSTIC). I will apply this tool to label the cells interacting with gut-associated neurons during obesity and use flow cytometry to identify the type of labeled immune cells. Next, I will use live imaging and tissue clearing tools in immune cell/neuron double reporter mice (approaches pioneered by the host lab) in lean and obese conditions. I will also use single-cell or single-nuclei RNA-sequencing from the intestinal immune cells and gut-associated neurons, respectively, in lean and obese mice. These approaches will allow us to define the candidate receptor-ligand molecules and gut-associated neuronal subpopulations involved in these interactions. Second, informed by the characterization performed in Aim 1&2, I will investigate the functional consequences of enteric neuroimmune interactions in developing obesity-related symptoms (Aim 3). I will use genetically engineered mice to target ligand-receptor signaling pathways identified in Aim 1 and manipulate gut-associated neuronal subpopulations while monitoring physiological and behavioral parameters related to obesity-related symptoms using metabolic cages. We can also target immune cells that are found to be changed during the development of obesity by using cell-specific depletion strategies. Collectively, by utilizing state-of-the-art approaches, this project will reveal neuroimmune mechanisms possibly modulating obesity-related pathophysiological and behavioral consequences, which is a poorly understood question in this nascent field.

Social insects can undergo dramatic changes in their physiology and behavior to support the colony. In the ant Harpegnathos saltator (Hs), when the queen dies, workers can transition into reproductive pseudo-queens (gamergates) to re-establish hierarchy in the colony. In the absence of queen pheromones, the young workers engage into a dueling behavior by trading strikes with their antennae. Most workers rapidly abandon the fight (win-lose) but dueling continues for months among the winners (win-win) who become reproductive and share social dominance. These gamergates show dramatic physiological and behavioral changes, lay eggs and cease to perform workers’ duties. They also have a 5X lifespan extension, and their brain is remodeled and shrinks by 20%. This transformation is fully reversible when gamergates are placed back in a queened colony. The goal of this proposal is to understand how the absence of queen pheromones triggers dueling; what the signals are among winning workers that sustain dueling; what brain circuits are affected by sustained dueling to cause changes in behavior. We hypothesize that the neural circuits responding to queen pheromones and controlling their production regulate dueling and the establishment of queen-like behavior. The pheromones normally repress dueling, so their absence in an orphaned colony releases the inhibition, and multiple workers begin dueling. The winning asymmetry appears to be created and maintained not due to the strongest fighters prevailing but due to a signaling imbalance between the animals engaged in duels. We hypothesize that this is achieved through the action of a negative feedforward loop – duelers quickly start producing queen pheromone to repress other duelers – and a positive feedback loop – duelers are more likely to engage in dueling again, possibly due to lowered sensitivity to the pheromones. Dueling further triggers remodeling of the brain circuits that control queen-like physiology and behavior. Aim 1: Track the behavior during win-lose and win-win duels, and measure queen pheromones production on the cuticle and in the producing glands. Aim 2: Generate transgenic lines expressing GCaMP7f or CaMPARI2 in the antennal lobe of the brain to evaluate whether losing and winning workers exhibit different sensitivity to the pheromones. Aim 3: Perform scRNAseq to determine how the gamergate brain changes and which circuits are affected, focusing on homologs of known (female) aggression and egg laying circuitry in flies. Experimental system: The host lab has extensive experience in flies and ants, including a deep knowledge of brain genomics as well as metabolomics of the hemolymph. Genetic manipulations in H.s are also possible. Conclusions: This work will elucidate the molecular underpinning of social conflict that establish a new shared hierarchy in ant colonies, and how molecular memories of social experiences establish long-lasting behavioral traits via brain remodeling.

Mammalian development is an intricate process governed by overlapping and often competing cell fate programmes. Although much of the cell intrinsic transcriptomic machinery and pathways responsible for cellular differentiation is well characterised, the contribution and effect of the surrounding tissue niche(s) toward cell identity acquisition remains poorly understood. This is especially the case when we consider that embryonic macrophages and other immune cells infiltrate many organs early on during tissue specification and develop symbiotically alongside the native cells, increasing the complexity of cellular crosstalk within the niche. Furthermore, our understanding of cellular crosstalk has thus far been restricted to extracellular receptor/ligand interactions, despite increasing evidence that cells might be able to communicate directly via gap junctions, tunnelling nanotubes and extracellular vesicles. Consistently, soluble and membrane-bound ligands alone are unable to fully establish and maintain cell identity of ex vivo cultured cells. In this project, I will focus on pancreatic progenitors and their interactions with the surrounding hematopoietic/immune cells during pancreatic development, in the window of time spanning fate specification and differentiation of distinct pancreatic cell types, such as the acinar and hormone-expressing endocrine cells. Although macrophages and other immune cells represent a large fraction of the embryonic pancreatic microenvironment and are key players in diabetes pathogenesis, they are an often-neglected population and understudied in this context. Single-cell transcriptomics have started to unveil the heterogeneity of these populations, but their functional diversity and potential contribution to tissue niche(s) underlying pancreatic lineage decisions remain unexplored. Utilising in vivo mouse models, human induced pluripotent stem cell models and human fetal tissue together with traditional and spatial transcriptomics, we aim to map the cellular communication network between different sets of hematopoietic/immune cells and pancreatic progenitors. Next, we will explore the possibility of cellular crosstalk beyond traditional receptor/ligand interactions, such as direct cytosolic transfer of mRNA and protein subunits via tunneling nanotubes and extracellular vesicles, and validate the functional relevance of these findings in co-culture experimental set ups. By the end of this project, we will not only have a fuller understanding of how cell-cell communication contributes to the pancreatic tissue niches, but also provide evidence towards a novel and alternate form of intercellular communication via direct cytosolic transfer. Finally, beyond the impact on the development and stem cell biology of the pancreas, this study opens new avenues for understanding mechanisms underlying cell-cell communication and how these processes could go awry in diseases.