Organ development depends on the integration of local cell-to-cell interactions with long range signalling throughout the body. We will investigate mechanisms by which long-range signalling via the cerebrospinal fluid (CSF) regulate body axis formation and spine curvature. The CSF is produced by the choroid plexus in the brain ventricles and flows down the central canal in the spinal cord. The CSF instructs brain development by delivering age-dependent grow-promoting factors to target cells. CSF circulation also contributes to the curvature of body axis (embryo) and to the spine (juveniles). However, the mechanisms regulating the flow and content of CSF remain poorly understood. Progress has been impeded by a historical lack of tools and challenges inherent to studying fluids in small, developing organisms. The advent of innovative tools and approaches now provides an unprecedented opportunity to overcome previous limitations. The transparency of zebrafish and the accessibility to mouse CSF provide powerful models for testing our driving hypothesis that the physical and biochemical properties of CSF impact body axis formation and spine curvature. We propose an entirely new, interdisciplinary collaboration of three leaders in their field: (1) physicist Francois Gallaire (EPFL, Switzerland), expert in the theory & modelling of complex fluid dynamics; (2) developmental neurobiologist Maria Lehtinen (Boston Children’s, HMS, USA), expert in analysis of the choroid plexus-CSF system; (3) biophysicist Claire Wyart (ICM, France), expert in imaging and sensory physiology in the spinal cord. Our multi-tiered approach will unravel the principal parameters driving CSF composition and flow. We will map CSF flow in the developing fourth ventricle and central canal (Aim 1), elucidate mechanisms regulating protein secretion into the CSF (Aim 2), and investigate mechanisms of active transport of instructive signals along the anteroposterior axis controlling organogenesis (Aim 3). The proposed studies will transform this historically understudied area of neuroscience into a robust field spanning CSF-based signalling in brain and spine. As studies of paracrine signalling and fluid dynamics lag far behind cell-intrinsic studies of signalling, our techniques and concepts should provide a roadmap for future studies of fluid niches throughout the vertebrate body.

Biofilms are surface-adhered communities of bacterial cells embedded in a polymeric matrix, and they constitute a major form of bacterial life. Biofilms originate in one of two ways: by clonal proliferation from a single founder cell, or by aggregation of many migrating cells. What are the respective benefits and challenges of these two modes of biofilm growth? To answer this question, we need to understand the physical principles that underlie both modes of biofilm formation. Specifically, for clonal proliferation, it is unclear how cell-level processes, such as directional growth and division, cell adhesion, and heterogeneous matrix secretion, give rise to the evolving macroscopic organization and properties of the mature biofilm. For aggregation, the mechanics of how collective cell migration leads to a biofilm remains controversial. Here, under Prof. N. Wingreen’s supervision, I propose an interdisciplinary approach that will address biofilm formation in two such contrasting systems. Collaborating with Prof. B. Bassler, I will develop an active liquid crystal model to understand the growth and internal ordering of Vibrio cholerae biofilms. Working with Prof. J. Shaevitz, I will investigate whether the aggregation of Myxococcus xanthus constitutes a phase separation or a dewetting transition. Their research programs, respectively, provide single-cell resolution of developing biofilms and traction-force maps of collective cell migration. These data call for active matter models — first to elucidate general principles of self-organization that apply broadly across species, and second to understand the evolutionary selection of species-specific modes of biofilm growth.

The idea of combining man-made and natural systems has fascinated humans for centuries. However, despite the explosion of life science advances, the integration of electronic and biological systems remains woefully underdeveloped. To address this, we will apply principles of Synthetic Biology to develop biological cells controlled by electronic signals.

We propose to create synthetic cellular receptors selectively activated by electrochemically activated peptide. In order to endow the system with the ability to activate a selected ligand:receptor pair in the presence of other polypeptides, we propose to construct peptide-selective bio-electrodes. Electro activation of the peptides decreases their affinity for the electrode and leads to its relocation to the two component cellular receptor that utilizes intramolecular proteolysis to activate transcription of the reporter genes. Using the team’s expertise in protein, cellular and electronic engineering we will design an integrated system where the activation of the reporter expression can be both controlled and monitored using electrode array in a few or even in a single mammalian cell.

This proposed approach enables the construction of a potentially unlimited number of orthogonal electrode/ peptide/receptor systems that allow multichannel information transfer between computing devices and biological cells. While the first embodiment relies on the relatively slow gene expression readout, the same approach can be used to control practically any biochemical process in real time. Such synthetic signaling pathways capable of performing logical operations will exert complex and precise control on cellular biology, leading to a first generation of bioelectronic hybrids.

Organizing microtubules (MTs) into arrays is fundamental for many cellular functions and critically depends on MT nucleation by the gamma-Tubulin Ring Complex (gammaTuRC). MT Nucleation occurs from the centrosome as well as from dispersed non-centrosomal sites and is temporally regulated. Although the role of gammaTuRC in MT nucleation is widely accepted, the tools for measuring this activity are severely limited. Current methods rely on indirect readouts using MT plus end tracking proteins, based on the rationale that nucleation generates growing MT plus ends. However, polymerization of MT ends generated by severing enzymes or MT rescue can also result in growing plus ends and may be mis-interpreted as nucleation. To address this, I propose to develop a biosensor for the specific detection of gammaTuRC-dependent nucleation within the cells. I will explore several alternative approaches designed to detect interaction between gamma-tubulin molecules in the gammaTuRC and alpha-beta-tubulin oligomers, as it occurs during nucleation. I will use tagged gamma-tubulin in combination with chemical or protein-based probes to generate a fluorescence signal upon formation of the gammaTuRC/MT interface. The signal will be based on FRET, SPLIT-GFP fluorescence complementation, or aggregation-induced emission by a “turn on” chemical probe. I will then use the biosensor to characterize MT nucleation in cycling cells and in post-mitotic neurons differentiating in vitro. This work will provide unprecedented insights into gammaTuRC-dependent nucleation and its regulation in space and time, and will significantly advance our understanding of how cells assemble and remodel MT arrays.

The ability to associate sensory stimuli with appropriate emotional meaning while computing an optimal behavioral response is essential for survival. One key regulator of this emotional learning is the neurochemical dopamine (DA) acting in multiple brain areas such as the medial prefrontal cortex (mPFC) and basolateral amygdala (BLA). DA’s modulation of neural plasticity during emotional learning has traditionally been studied using electrophysiology in individual neurons, but how it occurs on the level of large neuronal networks (ensembles) remains unknown. To address this gap, I will investigate DA’s influence on dynamic ensemble activity within the mPFC and BLA during pavlovian fear conditioning, a classic model of emotional learning. Using in vivo calcium imaging and optogenetic control of DA release in mice, I will test DA’s role in establishing ensemble representations of conditioned stimuli. The long-term goals of this project are to (1) reveal key principles by which DA modulates neuronal ensemble coding and (2) lay the groundwork for future studies investigating how abnormal changes in ensemble coding relate to DA-sensitive anxiety disorders such as post-traumatic stress disorder.

The brain is a ceaseless worker, endlessly integrating the information it receives into a continuous stream of conscious thoughts. Interestingly, even falling asleep does not completely suppress conscious thoughts, as exemplified by dreams. This is all the more surprising since wakefulness and sleep differ so greatly in terms of neural activity and phenomenology. What is then the core neural substrate enabling the generation of conscious thoughts? I will focus here on spontaneous thought generation across wakefulness and sleep. Spontaneous thoughts will be here defined as conscious representations that are not triggered by a specific external event (e.g. an image, a word). Mind-wandering is a classical example of spontaneous thoughts during wakefulness, and dreams during sleep. My goal will be to identify the neural correlates of spontaneous thought generation, i.e. the minimal set of brain regions that are jointly sufficient to spontaneously produce conscious representations even in the absence of sensory stimulation. I will use both correlational and interventional approaches. By directly stimulating the human brain, I will seek to understand how brain activations constrain the content of thoughts. Comparing the neuronal correlates of spontaneous thoughts in wakefulness and sleep will be quite powerful because spontaneous thoughts arise without external stimuli, removing confounds due to sensory stimulations. Furthermore, the resulting neural correlates of spontaneous thoughts will, by construct, generalize across different conscious states, allowing us to isolate purer neural correlates of consciousness than previously proposed.

Over a 100 years ago, Élie Metchnikoff’s visionary work laid the foundations to our understanding regarding the importance of the gut microbiota for the physiology of the host. Recently, technological breakthroughs sparked a renaissance in the research of the human microbiome. While many observational studies identify clinically relevant correlations between the structure of the microbiome and states of disease, our true aim is to uncover the molecular mechanisms and functional roles of host-microbe interactions. Here I describe an innovative sequencing-based functional metagenomic approach, termed FunMet-Seq, which utilizes transcriptional profiling of single human cells towards the discovery and characterization of microbially produced effector molecules. To this end, a library of microbiome-derived DNA, extracted from stool samples, is heterologously expressed in E. coli. Next, culture broth from the library is sampled, sterilized, and used to treat genetically barcoded HEK293-TN cells. Finally, massively parallel transcriptional profiling of single cells enables to identify culture broth samples which contain an effector molecule, and to trace its origin to a specific clone in the array of metagenomic DNA. While the realization of the FunMet-Seq pipeline is an ambitious goal, it will catapult forward our ability to understand the molecular mechanisms underlying the he complex microbial ecology in the human body. Moreover, as the role of the microbiome in human physiology becomes evident, deciphering the interkingdom crosstalk within the human body is critical to reveal the etiology of microbiota-related diseases, and is expected to produce therapeutic applications.

Circadian rhythms are present in almost all light-sensitive organisms from cyanobacteria through plants, flies, and mammals. Human hormone levels and metabolism are tied to circadian rhythms and affect sleep and wakefulness timings. Disruption of the synchronous action of circadian clocks leads to several human diseases, mainly sleep and metabolic disorders such as non-24-hour sleep wake disorder and diabetes. Most known circadian clocks stem from a transcription-translation feedback loop. Protein and mRNA levels oscillate with a 24 hour period, matching the day-night cycle. An intriguing exception is circadian oscillations in cultured red blood cells (RBCs); since RBCs are anuclear, the circadian rhythm must be non-transcriptionally controlled. RBCs are a major blood component, so their circadian rhythm might affect some of the many metabolic signals carried by the blood stream. Furthermore, the RBC circadian rhythms in individuals with hematological disorders may be out of sync or even cause the disorders’ phenotypes. I aim to use methods that complement each other (e.g., proteomics, metabolomics, and interactomics) to investigate in depth the molecular mechanism of the RBC circadian rhythms, which will advance our understanding of the rhythm’s role in metabolism signaling and hematological disorders. This strategy is suitable for resolving this question since the circadian rhythm is non-transcriptionally controlled and not much is known about it. Revealing the mechanism of the RBC clock will be the first time a eukaryotic non-transcription-translation based circadian rhythm is characterized.

How does an animal perceive the world and execute the appropriate behavior? This task requires sensory organs such as the eye to detect features of the environment, then for the brain to integrate this information and produce a pattern of motion directed at a goal. However there is currently little information about how animals encode features from their natural environment. Indeed, most of our understanding about visual neural processing in the fish brain comes from the use of artificial stimuli that lack the complexity found in the natural visual environment. Fish are ideal for investigating visual processing because their diverse behaviors and compact brains are amenable to modern neuroscience technologies. How does the brain process information? Most current understanding of fish vision relies on the common lab model zebrafish (Danio rerio), but other species have likely evolved distinct solutions for visual processing based on environmental and behavioral differences. To bridge these knowledge gaps, we propose to thoroughly characterize the natural visual environments of two species, zebrafish and Astatotilapia burtoni (an African cichlid), by collecting a large dataset of natural underwater videos in India (zebrafish) and Lake Tanganyika, Zambia (cichlid). We will then use these videos in laboratory experiments with a custom-built immersive viewing arena to examine which visual features from the natural environment elicit behavioral responses in the two fish species. To examine brain responses to natural stimuli, we will transfer existing technology for recording brain activity in larval zebrafish to cichlids. We hypothesize that there will be species-specific brain function adaptations underlying several behavioral differences between these species in response to visual motion. We will then expand from studying basic visual features to analyzing the appearance of other fish during social behavior. What drives fish to school or fight? How does the surrounding environment influence how fish detect the presence of conspecifics and other fish? Our integrative and comparative approach will provide a novel perspective on how vertebrate visual brain areas evolved to encode and adapt to the natural world.

Adolescence is a unique transitional period marked by deficits in impulse control and behavioral inhibition, often leading to high-risk behaviors. It has long been believed that adolescent impulsivity is a product of a delay in the maturation of the brain circuit supporting decision-making, but due to technical limitations, this link remains elusive. While research has correlated adolescent changes in isolated brain regions with changes in decision-making, no study to date has probed how these brain regions act as functional circuits to modulate adolescence impulsivity. Here we will take advantage of recent technological advances to (1) investigate the pattern of brain circuit activity during live decision-making in adolescent and adult animals, and (2) determine the microcircuit dynamics underlying this behavior to identify the circuit basis of adolescent impulsivity. This will pave the way for future experiments in which we will manipulate circuit function in vivo to modulate adolescent impulsivity in behaving animals. Approaching behavior from the standpoint brain circuit maturation will consolidate a major conceptual and technical shift in systems neuroscience. This shift will shed light on the biological mechanisms underlying behavioral changes during early life and adolescence, but will also fuel a novel approach to the study of the etiology of mental and neurodevelopmental disorders with early onset, thereby informing novel, more effective circuit-specific therapeutic strategies.

Plant development and growth are linked to cellular shape changes, which are controlled by genetic programs but also by perception of environmental signals, including mechanical cues. While both genetic regulation and mechanical control of morphogenesis were studied independently, there is a need to explore how cellular shape-associated strain and stress can mechanically regulate gene expression during differentiation. In animals, mechanical stimuli are known effectors of differentiation. They involve propagation of mechanical forces through the cytoskeleton to the nucleus, leading to chromatin remodeling and modification of gene expression. Thus, nuclear envelope proteins that control nuclear shape and transmit forces to chromatin play a key role in rapid triggering of gene expression. In plants, less is known about mechanotransduction from cell surface to the nucleus.
Using a systems biology approach and an interdisciplinary network, we propose to investigate how mechanical cues affecting cellular shaping are sensed at the nuclear envelope to drive chromatin remodeling in Arabidopsis. We will sudy a unique cellular model, the single root hair in an epidermal tissue context, with well-defined morphogenetic programs linked to cytoskeleton and nuclear dynamics. We will analyze root hair formation and growth in WT and mutants affected in either root hair development or nuclear shape. Combining in vivo live imaging and micro-mechanical measurements (rheometry), we will evaluate mechanical properties of cells and nuclei during root hair development and their dependence on cytoskeleton and nuclear dynamics in relation to gene expression. We will also determine how these mechanical, structural and biological properties are modified when a controlled mechanical stress is applied to the root hair cell during development. Our data will highlight proteins involved in mechanosensing, and we will evaluate their interaction with the nuclear envelope network. Live imaging and rheometry data will be correlated to finite element modeling to estimate strain and stress in the system for predicting chromatin remodeling following cellular and nuclear shape changes.
Altogether, this will highlight the molecular networks involved in mechanosensing at the nucleo-cytoplamic interface and reveal how gene expression is robustly regulated during cellular morphogenesis in higher plants.

Melanin is a unique molecule synthesized by all domains of life (bacteria, archaea, and eukarya), but it is most commonly known as a UV screening pigment in humans and other animals. However, surprisingly little is known about the secondary or tertiary roles of the molecule. The recent discovery that melanin-containing fungi located at the ruptured Chernobyl reactor display enhanced growth in the presence of gamma radiation, which is usually thought to be harmful, has hinted at the diverse functions of the molecule. A defining feature of melanin is the chemical and structural variability found across and within microbial species. This feature along with the ability to broadly absorb over the UV and visible spectrum has led to the hypothesis that microbial cells could have evolved to capture and utilize this energy. Low resource environments such as those found in deserts are the most probable habitats in which these adaptations could exist. The main objective of this proposal is to assemble a diverse team of scientists to apply new single cell and single molecule technology to studying divergent biology in one of the most extreme environments on the planet, the Atacama Desert.
Here we test the hypothesis that melanin is a key player in producing biochemical energy in microorganisms that are exposed to high amounts of radiation in the UV range. As part of the project, our international team, consisting of engineers, biochemists and microbiologists, will carry out a comprehensive analysis of the microbes living in the Atacama Desert, one of earth’s most UV-irradiated regions. We will identify alternate functions of melanin using state-of-art technologies to 1) sort out and identify melanin producers; 2) sequence their DNA for melanin pathway analysis; 3) characterize how melanin is being synthesized; and 4) link melanin to the photochemical conversion of UV light. Since microbial communities and melanin represent mixed populations, technology is needed which can probe individual cells and molecules. Only once this is accomplished, can sub-populations within each group be thoroughly examined. The impact of understanding melanin not only extends the fundamental knowledge of a commonplace pigment, but also can also prove useful for searching for life on other UV-irradiated environments such those on other planets.

The cell cortex is a dynamic network of actin filaments (F-actin), myosin motors, and accessory proteins coupled to the cell membrane, which mediates both the maintenance and changes of animal cell shape. Cell shape is determined by a balance between resistance to mechanical perturbation and the mechanical stresses produced internally by the actomyosin cytoskeleton. The precise architecture of the F-actin network and the dynamics of F-actin assembly and disassembly determine the accumulation of myosin-generated stresses, resulting in diverse behaviors from orchestrating local pulsatile contractions to inducing cell-scale polarised flows. For instance, in xenopus oocytes and embryonic cells of echinoderms and frogs, positioning of the actomyosin cytokinetic ring is preceded by travelling waves of F-actin assembly and disassembly1. Similarly, local pulsatile actomyosin contractions lead to persistent constriction within the drosophila mesoderm2, or global polarization and flow of the cortex of the developing C. elegans embryo3. Thus, contractility often follows the maintenance of a dynamic but steady cytoskeleton. However, how the spatial and temporal balance of actomyosin contractility, F-actin architecture and biochemical signaling determine homeostasis or polarized flows in the cytoskeleton remains an outstanding question in molecular cell biology.
Here we propose a unique integration of ex vivo, in vitro and in silico studies to uncover the molecular design principles for cell polarization. Ex vivo, we will modulate protein-protein interactions and use optogenetic control of Rho and light inactivation control of myosin to determine the extent to which cortical stability entrains force generation to impact polarization in neutrophils. Guided by such experiments, we will spatiotemporally modulate protein-protein interactions within an in vitro model cytoskeleton to alter F-actin architecture, adhesion, and myosin activity. In silico, we will systematically vary these parameters to generate testable predictions for cortical stability versus polarization. We envisage that our integrated approach will decipher the general principles underlying the assembly, stability and dynamics of cytoskeletal structures and the regulation of eukaryotic cell shapes.

During infection with bacteria or viruses, our immune system becomes activated to fight these foreign invaders and thereby prevent us from getting ill. Dendritic cells are an important immune cell type that control this process. When they sense the presence of infection, they trigger an immune response against any simultaneously captured bacterial or viral molecules by displaying the molecules on their surface for immune cells to recognise. However, they can also capture normal “self” molecules from our organs during infection, and it is unclear how they focus the immune response upon the foreign invaders without triggering an inappropriate response against “self” molecules that could inflict damage upon our organs. As we currently have only a limited ability to detect and visualise self and foreign molecules on the dendritic cell surface during an infection, our understanding of this process is incomplete.
We propose a bio-engineering approach to address this problem. Nature has already designed a system for biological molecule detection on the dendritic cell surface: immune cells called T cells are exquisitely sensitive at recognising specific surface molecules on dendritic cells. We will leverage the key features of this interaction to develop novel staining materials that imitate the natural way that T cells recognize molecules on dendritic cells. Using an engineering technology based on DNA origami to control the size, shape, and function of small particles, a “synthetic T cell” staining reagent will be assembled on a DNA scaffold. By adding fluorescent signals and experimenting with a variety of engineering blueprints, we can increase the sensitivity and signal intensity of our particles. Combining these synthetic T-cells with state-of-the-art microscopy techniques that can image individual molecules, we will precisely pin-point the exact location of self and foreign molecules on the DC surface during infection to determine if their organisation helps to either inhibit or promote an immune response. Overall, this project will thus employ expertise from a range of different disciplines to dissect how the immune system discriminates between “self” and “foreign” molecules.

Memory is typically treated as representing past experiences, including salient events and stimuli. A key purpose for the retention of these memories is to provide predictive information that guide the animal’s future behavior - where it will go and what it will do. However, the world is rarely static, and memory therefore requires continuous updating from sensory information. While sensation and memory have been well studied separately in tightly constrained tasks, little is known about how they interact to enable complex behaviors. This proposal aims to understand how these three processes – behavior, sensation, and memory – predict each other while animals freely move within an environment, and how these predictions are altered based on an animal’s experience and the predictability of its environment.
We will test three general hypotheses about the interactions between sensation, memory and behavior: (1) Memory evolves from being learnt from (i.e. instructed by) sensation and behavior in a novel environment, to being predictive of (i.e. instructing) sensation and behavior in a familiar environment. (2) The direction of predictability between processes (i.e. which process instructs another) are dynamic and can be manipulated by changing the context of the environment. (3) The direction of predictability is dependent on the predictability of the environment, with memory instructing sensation and behavior for predictable features, but sensation instructing the other processes for unpredictable features.
We will test these hypotheses in an environment with static (predictable) and mobile (unpredictable) features. We will combine long-term recordings from large populations of neurons in the visual cortex and the hippocampus of freely moving animals, high-resolution behavioral measurements, and statistical modeling of neural and behavioral dynamics. Our experiments will provide unprecedented measurements of the neural representations that underlie natural behavior. They will show how these animals, and potentially ourselves, use representations of the past to predict the future.

Sequences of place cells, hippocampal neurons active when a subject enters a particular location, encode spatial trajectories while navigating a given environment. During periods of restful wakefulness and non-REM sleep, condensed sequences corresponding to prior navigation recur during sharp-wave-ripple events in the hippocampus, the disruption of which disturbs memory. Enabled by the use of large-scale neural imaging, recent work has revealed that these sequences are in fact composed of multiple discrete neural assemblies connected together. Understanding such assembly activity during various behaviors is critical as they may represent the basic unit upon which memories are encoded. However, the functional organization of the microcircuits, including assembly activity, recruited during REM sleep (REMs) remains unknown despite the recently confirmed role for REMs in the formation of spatial memory. The goal of the research outlined in this proposal is to address this issue using large-scale neural imaging in the hippocampus of fully habituated head-fixed mice. Specifically, this research aims to investigate the occurrence and characteristics of neural assemblies during REMs under baseline conditions as well as following learning of a spatial task. As a final step, local microcircuits within the hippocampus will be manipulated optogenetically during REMs following spatial learning in order to probe for a direct relationship between place cell (assembly) activity during REMs and subsequent memory performance. Cumulatively, these experiments will provide a novel characterization of the recruitment and mechanistic role of neural assemblies in the hippocampus during REMs.

Non-invasive transcranial stimulation and imaging of targeted neurons deep in brain are a central goal of neurobiology and related engineering science. Optical techniques, such as imaging and optogenetics, have revolutionized experimental neuroscience but are limited by the inability of visible light to penetrate deep into brain tissue. Here I propose an approach for transcranial near-infrared (NIR) deep brain stimulation and imaging by taking advantage of upconversion nanoparticles (UCNPs). UCNPs are lanthanide-doped nanocrystals unique in their ability to absorb tissue-penetrating NIR light and convert into emission in both visible and NIR spectral ranges. First, I will demonstrate UCNP-mediated optogenetics for deep brain stimulation, where UCNPs will be employed to convert transcranial NIR irradiation into visible emission to stimulate rhodopsin-expressing neurons in mouse brain. Second, I will develop blood-brain barrier-crossing UCNPs for targeted imaging of molecular markers in deep brain. For proof of concept, UCNPs will be conjugated to curcumin, a specific binder to beta amyloid - a molecular marker of Alzheimer’s disease, therefore allowing transcranial imaging of amyloids by UCNPs’ NIR emission. Finally, I will integrate the dual functions of UCNPs in deep brain imaging and stimulation, by taking advantage of their simultaneous NIR and visible emission, to achieve transcranial diagnostic imaging and optogenetic treatment of neurological diseases in model mice. The new techniques developed here will allow for non-invasive brain stimulation and imaging without the need for implants, and hold potential for next-generation remote neural therapeutics.

The human brain gains much of its computational abilities from the trillions of connections made between cells by synapses. Molecular changes in synapses (a process called synaptic plasticity) are considered to underlie learning and memory. An important component of the synapse is the postsynaptic density (PSD). This specialized structure has been studied extensively to understand its function, in particular because >100 neurologic disorders (such as autism spectrum disorder and schizophrenia) have been associated with PSD dysfunction. Information regarding the overall molecular architecture of the PSD, however, is largely incomplete, in part because the PSD is extremely complex, containing hundreds of individual components connected in a dense network. The PSD is also highly dynamic and asymmetrical—two properties that render protein structural analyses challenging. While solving the structure of the entire PSD seems insurmountable, half of the PSD’s mass is composed of only ten classes of linker proteins. We hypothesize that these highly-abundant proteins form a molecular scaffold, a network we term the ‘postsynaptic scaffold’ (PSS). We aim to develop an approach we call the ‘Thermophile-Assisted Postsynaptic Architecture Strategy’ (TAPAS) to solve the PSS structure. Given the delicate nature of proteins, structural biologists like to work with resilient, temperature-resistant proteins that can be obtained from thermophilic organisms. For brain-derived proteins, there is one animal known to be comfortable at temperatures >50 °C: a worm that lives on hydrothermal vents in the Pacific Ocean. We plan to sequence this worm’s genome (Alvinella pompejana) to find the PSS proteins that it shares with humans. We will determine the structures of the thermophilic proteins by X-ray crystallography. As much of the PSS architecture consists of filamentous structures that do not crystallize, we will use cryo-electron microscopy to define these larger structures. Finally, we will integrate several other methods (biochemical, proteomic, bioinformatic and advanced electron tomography methods) to build a model of the overall PSS. Achieving this ambitious goal will inform studies into learning and memory, lead to new treatments for devastating brain disorders, and help explain how the synapse contributes to human cognition.

Pattern recognition receptors (PRRs) of the innate immune system recognize structures that are shared by a large group of microbes. Originally posited as the initiators of immunity recognizing pathogen-associated molecular patterns (PAMPs), it was realized that PRRs also detect structures on symbionts which were therefore termed microbe-associated molecular patterns (MAMPs). In addition to initiating inflammation, MAMPs are also involved in the maintenance of symbiosis and in the development and regulation of the immune system. Given the very large microbiota resident in the intestine, a significant amount of intestinal MAMPs penetrates the circulation and modulate diverse aspects of immunity and physiology. For this project, we have developed a transgenic mouse system to control the expression and delivery of a MAMP into the host and assess its impact on immunity and physiology. We will assess the molecular mechanisms governing MAMP detection in the host and impact on development of immunity both basally to maintain homeostasis and during disease. As early life exposure of MAMPs likely modulates effects later in life, we will determine if fetal and neonatal exposure to MAMPs imprints the host. Moreover, as the penetration of MAMPs into the host is normally controlled by the innate immune system, we will use transgenic mice lacking this compartment to assess how innate immunity modulates physiology through the control of MAMPs. We thereby aim to determine the impact of MAMPs on the host, and the threshold levels of MAMPs beyond which immunity and physiology cross into pathology. This study will open the way for the use of MAMPs as modifiers of human physiology.

While microbial pathogens evolve during infection to adapt to the human environment and overcome antibiotic treatments, little is known about how adaptation is affected by other co-infecting species. Recent advances in whole-genome sequencing and phenotyping of microbial pathogens derived from human hosts during infection reveal that bacterial pathogens can diversify quite rapidly within a single patient, accumulating mutations that allow them to adapt to challenges presented by the human environment, as well as to overcome antimicrobial treatments. Yet, pathogens within the human body often do not evolve alone. Indeed, many complex infections, especially long-term chronic infections, involve multiple co-infecting pathogenic species within the same patient. In such polymicrobial infections, interactions between pathogens can determine virulence and alter the nature of evolutionary adaptation for each species. As an HFSP Fellow, I will investigate how interspecies interactions influence pathogen evolution in patients. First, I will focus on the cystic fibrosis lung, a canonical site of polymicrobial infections. Second, I will expand to a wide range of infections, leveraging clinical data from >2 million patients. This research will improve our ability to predict how pathogens will evolve in the human body and to design new strategies to thwart their evolution.