It is by now well accepted that gene expression is not only controlled through the interplay of information stored in the DNA sequence, chromatin modifications and the binding of regulatory proteins such as transcription factors, but also through the spatial organization of the genome within the cell nucleus. Indeed, not only cell-type specific transcriptional programs but also cell-to-cell variability in gene expression (i.e. transcriptional noise) seems to be mediated by differential genome topology, probably because it either permits or prevents regulatory sequences (such as enhancers) to find their target regions (such as promoters). In this project, I will investigate how such chromatin contacts are established, which molecular players regulate these interactions, whether some or all of the chromatin contacts can be remembered through cell division and how the subtle differences in genome organization cause variegated gene expression among otherwise identical cells. This work will therefore further elucidate the impact of genome topology on gene expression and will provide mechanistic insight into a novel concept of biological variability and versatility at the single-cell level.

Our knowledge of the world is organized in concepts. Bert and Ernie, for example, are two person-concepts, well known to “Sesame Street” viewers. The ease with which we recognize Ernie and distinguish him from Bert belies the daunting complexity of the computational tasks at hand. This is, first, because the retinal activity patterns generated by an object as complex as Ernie is ever changing. Thus representations need to be computed that lump all the different instantiations of Ernie patterns in the retina, into an activity pattern in the brain that is stable across these different instantiations, yet selective enough to tell one person from another. Second, repeated encounters of the same object need to be grouped together over time and with other information -e.g. the name and the voice- to form a modality-free representation: the concept of “Ernie”. Understanding how the brain achieves these remarkable feats is the goal of this project. Our starting point is the finding of ‘concept cells’ by team member Rodrigo Quian Quiroga, which respond almost like a switch to any representation of one individual (e.g. different pictures, the name), but not to other stimuli. Our central hypothesis is that conceptual knowledge rests on explicit concept cell representations: to understand the neurobiology of knowledge, we need to understand concept cells. Conceptual representation is in fact crucial for perception and memory. We bring in expertise on concept cells in humans (through team member Rodrigo Quian Quiroga), the processing of sensory information in rats (team members Davide Zoccolan and Mathew E. Diamond), face recognition in macaque monkeys (team member Winrich Freiwald), and on theoretical neuroscience (team member Haim Sompolinsky) to tackle the problem of concept representations from different angles. Our goals are to find, localize and understand the functional properties of concept cells in three different species with very different brain sizes; to understand, mechanistically, how concept representations are built from sensory inputs; and to determine how conceptual knowledge might be acquired over time. Exploiting unique advantages of our interdisciplinary approach and the complementary use of three biological systems, our project will be the first concise effort to develop a neuroscience of knowledge.

Many animals can detect and use the Earth’s magnetic field for orientation and navigation functions (Lohmann, 2010; Mouritsen and Ritz, 2005). However, the receptor for magnetoreception in vertebrates remains unknown. Members of our team (Dickman Team) have recently discovered cells in the vestibular brainstem of pigeons that encode the direction, intensity, and polarity of the magnetic field (Wu and Dickman, 2012), as well as a magnetic sense neural pathway that includes regions in the brain known to be involved with spatial orientation and navigation tasks (Wu and Dickman, 2011). In addition, our team has obtained preliminary evidence for biogenic magnetic particles in specialized cellular regions of an inner ear receptor (the lagena) that could function as a magnetoreceptor transduction mechanism. Our preliminary data has suggested that these magnetic particles are made of hematite and maghemite, ferromagnetic compounds with permanent magnet properties. Thus, our primary hypothesis is that the lagena contains cells that perform as a magnetoreceptor. Afferent fibers that innervate those receptor cells are hypothesized to signal magnetic information to the brain. Other team members (Winklhofer Team) have recently developed a magnetic microscope that can identify and isolate cells that contain magnetic particles, then extract these cells for subsequent cellular/molecular and biophysical analyses. In fact, they have identified magnetic cells in the nasal epithelium of fish and performed elemental analyses on the magnetic particles contained within those cells. The primary goal of the proposed project is to determine the nature of the lagena receptor and the neural response properties of primary afferent fibers innervating the receptor cells. The project represents a new collaboration between J. David Dickman (Baylor College of Medicine, USA), a systems neurobiologist and Michael Winklhofer (Munich University, Germany) a biophysicist specializing in biomagnetics. Together, the two research teams will focus on two objectives. The first is to delineate the magnetic receptor mechanism in the avian lagena using state-of-the-art methodologies to isolate and characterize the magnetic particles and their cellular location within the lagena. The second is to characterize the response properties of neural afferents innervating the receptor by using both magnetic and motion stimulation. The findings from these objectives will then be used to develop a novel working model for a new magnetic sense.

A wide variety of post-transcriptional modifications, with over a hundred known so far, decorate RNA molecules from all domains of life. N6-methyladenosine (m6A) is the most prevalent internal mRNA modification. It is essential to cell survival and organismal development, strongly impacting gene expression through unknown mechanisms. Recent mapping of the m6A landscape in human and mouse revealed over 12,000 m6A sites characterized by a typical consensus in transcripts of over 7,000 genes. Remarkably, sites preferentially cluster in three distinct landmarks of telling biological significance: at transcription start sites, around stop codons and within long internal exons - are dynamic, and are highly conserved between human and mouse. The primary aim of this proposal is to elucidate the modes of action of m6A in regulation of gene expression and shed light on this novel post-transcriptional regulatory layer. I will address the major challenges facing this evolving field, employing three chemical biology oriented approaches: 1) Profiling, by generating quantitative, single-nucleotide resolution methylation maps of systems under varying physiological conditions, aimed at correlating the dynamic methylome with observed phenotypes; 2) hypothesis-driven, aimed at probing one of most intriguing, innovative and well substantiated questions that arise from the current state of knowledge, namely the influence on protein translation; 3) open-ended, aimed at identifying interacting cellular machineries, acknowledging that what we know is only a tip of the iceberg of cross-talk between methylation and other regulatory layers.

Communication between cells is vital to the development and homeostasis of multicellular organisms. In vivo, signaling pathways function not in isolation but in combinations and are influenced by the specific cell type’s basal gene expression. A major gap in the field of cell signaling is to elucidate how pathways are integrated in real time. In particular, in most cases when two pathways have been implicated in a process, it is not clear whether they function sequentially, (eg. one pathway activating expression of another’s ligand), or synchronously (eg. regulating a common target). I propose to use Drosophila intestinal stem cells (ISCs), whose behaviors are regulated by multiple signaling pathways including Notch, Wingless, EGFR and Insulin, as a model system to dissect signaling pathway interactions. While some interactions between pathways have been identified in ISCs the full extent and nature of crosstalk is unclear. I will first compare the changes in gene expression following single or combinatorial pathway activation in vitro to characterize concomitant and sequential signaling crosstalk. Then, using target genes that respond to multiple pathways concomitantly, I will examine whether signal integration occurs by independent transcriptional responses or by convergence of upstream signals to regulate a transcription factor’s activity. Finally, I will use this information to develop a screen for regulators of crosstalk in cultured cells and validate the findings in vivo in ISCs. Together, these studies will clarify how multiple pathways are coordinated in the maintenance of stem cell homeostasis and provide mechanistic insights into crosstalk between pathways.

Signal transduction pathways enable cells to process information essential to their survival and proliferation, yet the genesis of these pathways remains poorly understood. These pathways are typically assembled using components from a small number of protein families that are organized in different ways to process diverse extra- and intracellular stimuli. We aim to investigate the evolutionary mechanisms that mold the interactions between pathway components following gene duplication or lateral gene transfer, leading to changes in the pathway structure and the emergence of novel signaling functions.
We specifically focus on the evolution of the two-component signaling pathway that drives sporulation initiation in Firmicutes. Sporulation is an ideal model for investigating signaling evolution: the pathways in Bacilli and Clostridiales arose from a common ancestral pathway, have similar components and engender analogous phenotypes, yet are characterized by different network architectures: a two-component system in Clostridiales and a complex phosphorelay in Bacilli.
We propose to reconstruct the evolutionary steps leading to the divergence of these pathways at the level of the protein, domain, interacting interface and active site. We will employ a multidisciplinary approach, combining computational phylogenetics and evolutionary systems biology with genetic manipulations and experimental assays.
A network evolution model will be developed to connect phylogenetic data on sequence evolution directly with the probabilities of gain and loss of interaction. This will allow us to reconstruct the most likely states of the sporulation pathway in key ancestral species and hence to predict network rewiring events: changes in protein dimerization and phosphotransfer specificities. The evolutionary histories produced will allow us to understand how the rewiring of modular proteins supports functional innovation at the network level, for example through the insulation of different signals by the elimination of crosstalk between related signal kinases.
The inferred interaction specificities will be tested in vitro using FRET and phosphotransfer assays. Using ancestral sequence reconstruction, we will probe the viability of the inferred ancestral pathways and the likely sequence of evolutionary transitions leading from one pathway structure to another.
We anticipate that our integrated approach will serve as a blueprint for probing the evolution of other signaling systems.

The vast majority (>95%) of human proteins is thought to be affected by alternative splicing. In fact, alternative splicing is one of the mechanisms thought to be underlying the increased organismal complexity of higher mammals. Without a doubt, the combinatorial complexity enabled by alternative splicing is mediated by changes on the protein level. However, only few studies thus far have addressed the consequences of alternative splicing on protein function, e.g. in the context of interaction networks. Particularly, large-scale interactome network mapping so far ignores the increased complexity and dynamic modulation of network connectivity by differential splicing. This project aims to fill this gap and identify protein interactions that are regulated by tissue-specific alternative splicing. We will use a bioinformatic modeling approach and analysis of deep-sequencing transcriptional profiling data to identify physiologically relevant splice isoforms that are likely to affect physical protein-protein interactions. Subsequently, we will experimentally identify differential interaction partners of the identified proteins and their isoforms using high-quality high-throughput interactome mapping by screening the different isoforms for interactions against a genome-wide set of ORFs and subsequent systematic verification. In the third part of the project we aim to genetically demonstrate the biological role of experimentally validated differentially-interacting-splice-isoforms using phenotypic screens and biological follow-up studies for selected high-confidence candidates. All information will be integrated with existing network maps and other biological information.

Light is not only important as an energy source for photosynthesis, but also a critical environmental signal for regulating plant growth and development. Plants have sensory photoreceptors that monitor different parameters of the light environment, including wavelength, intensity, direction and duration, and help plants to adapt their growth pattern to diverse and ever-changing environmental conditions. The photoreceptors and downstream signaling components in lower and higher plants, e.g. in the moss Physcomitrella patens and the seed plant Arabidopsis thaliana, are very similar. However, they respond to their surrounding light environment differently to meet their needs. The long-term goal of our project is to understand how species-specific response patterns to light evolved. In Physcomitrella and Arabidopsis, the phytochrome (PHY) family of photoreceptors, the PHY INTERACTING FACTORS (PIFs) and the nuclear transport protein FAR-RED ELONGATED HYPOCOTYL 1 (FHY1) are critical components of light perception networks that promote photomorphogenesis in response to red/far-red light. Although the topologies of light perception networks, i.e. components (e.g. PHYs and PIFs) and wiring (e.g. PHY-PIF interactions), in higher and lower plants appear to be conserved, their response profiles to light are different. One possibility is that the evolution of species-specific light perception networks may have relied in part on changes in the network topology, which in many cases have been shown to be important for network evolution. Alternatively or in addition, the evolution of species-specific light response modes may have relied on modifications of network parameters, such as RNA/protein synthesis/degradation rates and complex association/dissociation constants. By systematic examination of the light perception networks in Arabidopsis and Physcomitrella using biochemical, genetic and systems biology approaches, we will build mathematical models that will help to distinguish between these possibilities. Using a reiterative method of model based guidance of experimental design and mathematical modeling, these models will identify key steps in the evolution of species-specific light perception networks in terrestrial plants.

Synaptic communication relies on neurotransmitter release from a specialized compartment, the synaptic vesicle (SV), followed by formation of new vesicles from the plasma membrane, in what is termed SV recycling. This highly regulated process needs to be homeostatically optimized throughout the organism’s life and implies the continual biogenesis of new proteins and organelles, coupled with the degradation of non-functional components.

Biochemical experiments were able to demonstrate that in culture several abundant vesicle components were degraded rapidly, with a half-life of less than 24 hours (Daly and Ziff, 1997). Surprisingly, although protein aging and degradation are necessary components of the life cycle of any cellular specialization, including the synapse, this finding has received little attention.

How fast are different proteins/vesicles replaced? Are SVs replaced as whole units, or are proteins first segregated and degraded separately? Is the degradation of vesicles related to their inclusion in different functional pools (readily releasable, reserve)? How are soluble (non-vesicular) proteins degraded? Is there a role for endosomes in degradation? Does the accumulation of non-functional synaptic vesicles correlate with aging and/or neurodegenerative diseases? Answering these questions will be essential to increase our basic understanding of synaptic function and will help to test novel hypothesis about neurological disease mechanisms.

I will employ here a host of advanced biochemical and imaging methods, including super-resolution microscopy and novel fluorescent protein-based assays, to address the spatio-temporal regulation of pre-synaptic boutons. I will study the turnover of SVs and important non-vesicular presynaptic proteins, in relation to their function. Both temporal and spatial aspects will be analyzed, in an attempt to understand the dynamic nanoscale organization of SV turnover.

The development of the central nervous system (CNS) is one of the most spectacular processes in biology. Key aspects include the formation of neuronal axons, their subsequent growth and guidance through thick layers of nerve tissue, and the folding of the brain. All these processes involve motion and must thus be driven by forces. However, while our understanding of the biochemical and molecular control of these processes is increasing rapidly, the contribution of the dynamic interplay between cellular forces and tissue elasticity remains poorly understood – mostly due to a lack of suitable measurement techniques. In this highly interdisciplinary project, we will integrate international expertise in optics, biophysics and neurobiology to investigate the role of mechanics in CNS development. We will develop a novel photonic toolbox for in situ, label-free and non-contact tracking of mechanical force and elasticity at cellular and intra-cellular levels with microscopic spatial and real-time temporal resolution. Force sensing will be based on a radically new approach that will enable the continued measurement of cellular forces down to the pN range without the need for detaching cells from their substrate as required in traction force microscopy. To measure the elasticity of cells and extra-cellular matrix material with 3D microscopic resolution, we will develop Brillouin-scattering based ElAsticity Mapping (BEAM). Unlike standard methods for cellular level elasticity measurements (eg atomic force microscopy), which are usually either invasive or limited to surfaces, BEAM will provide a non-contact, non-perturbative probe. These innovative experimental techniques will be applied to illuminate how forces exerted by neurons contribute to axon formation and neuronal guidance. Ultimately, we will investigate the involvement of mechanics in brain morphogenesis. Our approach will allow to look at these crucial biological processes from a new angle and shed new light on CNS development. Beyond this, our novel platform will be broadly applicable to areas far beyond the boundaries of neurobiology, as it provides unique cell-microenvironment metrics to relate mechanical interactions to cell behavior.

It has long been assumed that because early mammalian development relies on almost immediate zygotic transcription, the requirement for translational control during embryogenesis and pattern formation may not be very significant. Our findings have, however, uncovered a surprising role for ribosome-mediated specificity in tissue patterning and gene expression. In particular, our studies show that expression of Homeobox (Hox) mRNAs is tightly regulated at the translation level to promote restricted patterns of protein expression within the developing vertebrate embryo. This strongly suggests that tight regulation in translational control of transcripts, largely at the level of 5’UTR activity, may provide a critical additional layer of regulation to the mammalian genome. At present, there is a strong bias in examining patterns of gene expression within the developing vertebrate embryo at the transcript level and we lack an even basic understanding of which transcripts are being translated into functional proteins. In this proposal, I will apply a new powerful technology that I have optimized for small embryos known as ribosome profiling to delineate the translational landscape of gene expression in mesoderm, one of the most important cell lineages specified during vertebrate development. This will provide unprecedented insight into the classes of mRNAs and the topography of regulatory elements positioned in their 5’UTRs that are regulated at the translation level. I will extend these studies into chick embryos where I will be able to subsequently functionally characterize regulatory elements that impart transcript-specific translation control.

Cells are constantly exposed to factors that damage DNA, either from exogenous sources, such as UV light, or endogenous sources, such as oxidative stress. Moreover, DNA replication itself poses a problem to genome integrity. Not surprisingly, a huge number of proteins work together to faithfully replicate DNA and to detect, signal and repair DNA damage. Failure to do so results in genome instability.

Over the last years, a considerable number of proteins involved in DNA metabolism have been identified to bind to an iron-sulphur (FeS) cluster as a cofactor. Considering that - upon FeS cluster oxidation - free iron atoms can generate reactive oxygen species and damage DNA, the abundance of FeS proteins in DNA replication and repair has come as a surprise, and the function of FeS clusters in these processes has remained elusive to date.

At the same time, their redox sensitivity makes FeS clusters particularly interesting and versatile cofactors that would be uniquely suited e.g. to sense oxidative stress conditions and allow adaption to suboptimal conditions of DNA replication.

I propose to exploit a combination of complementary experimental approaches and systems, such as proteomics, biochemistry, molecular and cellular biology and yeast genetics in order (1.) to identify new FeS proteins, (2.) to characterise new FeS proteins involved in DNA metabolism, and (3.) to gain insight into the role of FeS clusters in DNA replication and repair.

One of the most abundant proteins in that structures cell shape and guides its movement by forming filaments, moonlights in the nucleus. This protein, actin, has been studied for years yet precisely because of its dominant life as a filament, little is known how it acts in the cell nucleus. Recent evidence points to it serving in a non-filamentous or globular (G) form in the nucleus. Globular actin forms a complex with other proteins to help remodel chromatin and binds transcription factors to modulate gene expression. Recent data suggest that it directly influences the integrity of the genome once single and double strand damage is incurred. There has been no specific way to visualize or distinguish the filamentous from globular form of actin. Novel chemistry allows selection of circular peptidyl structures that are highly specifc for different forms of protein structure. These bicyclic peptides can be used to study the location, binding partners and function of actin in the nucleus. This program combines the expertise of a chemist, an electron microscopist, a yeast geneticist and a vertebrate protein expert in actin and actin related proteins, to try to solve this problem.

This proposal seeks to answer fundamental questions about cytoskeletal filaments, which are central regulators of cell biology. Our current understanding of bacteria is that these organisms use dynamic cytoskeletal systems to organize and regulate essential cellular processes. The results of our previous HFSP grant broke new ground by demonstrating that the number of these systems extends beyond homologs of the three primary eukaryotic cytoskeletons (i.e., actin, tubulin, and intermediate filaments). Two representative examples are our discovery of bactofilins, which are a bacteria-specific family of proteins, and of CTP synthase, an enzyme that is widely conserved across all kingdoms of life. These cytoskeletal proteins play key roles in regulating cell shape, metabolism, and intracellular compartmentalization, and their identification has had a transformative impact on the field of bacterial cell biology.
Building on our successful previous approaches, our renewal proposal is designed to characterize these novel cytoskeletons and, in addition, identify new cytoskeletal proteins. Apart from using a wide range of genetic, biochemical, and biophysical techniques, a central theme in this proposal is the identification, characterization, and application of small molecules to analyze protein function in vivo and in vitro. Small molecules can be powerful tools for studying bacterial and eukaryotic cytoskeletons, as they enable the rapid perturbation of protein function in vivo. Nevertheless, only few small molecules are currently available for specifically targeting cytoskeletal elements in bacteria. The discovery of chemical probes is poised to further our understanding of bacterial cell biology, while new classes of small molecules that emerge from these studies may provide starting points for developing new classes of antibacterial agents. Compounds that regulate widely conserved proteins may also inspire the development of therapeutic compounds for cancer and immune research and treatment.
In the first three years of the HFSP grant, we established a successful formula for productive collaboration between our labs. For the renewal, we have modified this formula to match the goals of the next phase of this project by focusing the expertise and capabilities of an interdisciplinary group of young scientists: Zemer Gitai uses high-throughput methods to study the bacterial cytoskeleton, Justin Kollman uses electron microscopy to study the biophysics of cytoskeletal assembly, Martin Thanbichler studies the biochemistry and genetics of the bacterial cytoskeleton, and Doug Weibel uses chemistry and engineering to study bacterial physiology. The salient features of this collaborative and interdisciplinary project will facilitate intellectual and cultural cross-pollination, and the scope and high-risk-high-reward merit of this proposal are designed specifically for an international funding mechanism such as the HFSP.

Given their biomedical significance, transmembrane (TM) proteins are especially desirable, but difficult targets for structure determination. In the past decade advancements in cryo-electron-microscopy (cryo-EM) have resulted in an increased number of low- and intermediate-resolution density maps of TM and soluble proteins. Yet, the number of cryo-EM-derived protein models is considerably lower due to the lack of a general powerful method. Elaboration of such methodology would resolve a bottleneck in proteins structure determination from cryo-EM maps. I will develop a method for proteins structure determination from density maps employing the deformable elastic network method and evolutionary conservation information. Recent studies have demonstrated the power of evolutionary conservation information in structure prediction of TM proteins in ab-initio, template-free modeling. Integration of this information into the interpretation process of low and intermediate-resolution cryo-EM maps will make the latter more accurate and fast. I will apply the resultant method on three systems: the 18 kDa translocator protein TspO, the cystic fibrosis transmembrane conductance regulator CFTR and a V-ATPase. Targets TspO and CFTR are appealing due to their involvement in brain pathological conditions and cystic fibrosis, respectively. ATPases are complex, not yet understood machineries coupling between the membrane electrochemical potential and the synthesis of the main energy carrier ATP, structure of which will shed light on their action mechanism. This proposal paves the way towards a powerful method for structure determination from cryo-EM maps and applies it to fascinating examples.

Bacterial biofilms are integrated communities of cells that adhere to surfaces and are fundamental to the ecology and biology of bacteria. The accommodation of a free-swimming cell to a solid surface is more complex than passive cell adhesion. We propose a multi-disciplinary study to investigate the interplay between motility appendages, molecular motors, exopolysaccharide (EPS) production, and hydrodynamics near the surface environment using tools not usually used by bacteriologists. We address two questions in P. aeruginosa.
How do different bacterial motor systems and EPS interact with solid surfaces? How do flagella/TFP adapt and respond to an interaction with a surface, and how do they couple to near-surface hydrodynamics and EPS? We will examine the transition from swimming to attaching for the two principal motor systems, flagella and type IV pili (TFP), using direct force measurements on single motors and correlate with multicell behavior using single cell tracking techniques. In all cases, the different roles of solid versus viscous friction in this low Reynolds number regime will be examined. With these anchoring measurements it will be possible to examine this important cdiGMP signaling network in a new light.
How do bacterial motors, exopolysaccharides, and guiding signals influence early biofilm formation? We track the entire motility history of every cell and make inferences about biofilm initiation at single cell resolution. Single molecule TFP force measurements will be used to examine the possibility of a TFP and flagella sensing function for EPS and EPS-mediated interactions between cells. Results will be interpreted using theoretical techniques developed to study nonlinear feedback mechanisms and ‘jamming’ transitions. We will perform new experiments exploiting cdiGMP signaling mutants that impact motility and/or EPS. Our physical measurements will be combined with techniques to measure cdiGMP levels at the single cell and sub-cellular levels, so that spatiotemporal phenomena in signaling and motility can be directly connected. Thus, the initial events in the association of a bacterium with a surface will be investigated in unprecedented detail by combining genetic techniques with single cell measurements, providing novel mechanistic insights into the earliest steps in biofilm formation.

Most neurons in the mammalian brain are born during embryogenesis and persist throughout the life of the animal. The mammalian olfactory bulb is a notable exception in that it receives many thousands of newly-born inhibitory granule cells (GCs) throughout adulthood. GCs provide recurrent and lateral inhibition to mitral cells, principal cells in the olfactory bulb, but the precise functional role of adult neurogenesis is unknown. Here we propose a collaborative and innovative project to test the hypothesis that adult-born GCs play a unique and essential role in odor learning, combining optogenetics, intersectional genetics, in vivo two-photon calcium imaging and behavioral tasks.
To enable selective characterization and manipulation of adult-born GCs, we will develop a genetic method to selectively target adult-born GCs. This strategy combines two independent recombinases for intersectional specificity and an inducible recombination system. With this genetic method, we will express a genetically encoded calcium indicator selectively in adult-born GCs and use in vivo two-photon calcium imaging to characterize their odor response properties in awake, head-fixed mice. Next, to test the effect of suppressing the activity of adult-born GCs, we will develop optogenetic tools that suppress neuronal activity acutely and reversibly. The strategy includes modifications of pre-existing opsins to develop novel, inhibitory step-function opsins (iSFOs). iSFO-expressing neurons will be suppressed upon the delivery of a brief pulse of activating wavelength of light for extended periods of time even in the absence of sustained illumination until the delivery of an inactivating wavelength of light. We will express iSFO selectively in adult-born GCs using the genetic method described above. We will use two-photon calcium imaging to monitor in vivo the odor responses of mitral cells when adult-born GCs are suppressed and unsuppressed on alternating trials with iSFO in awake, head-fixed mice. We will determine relative contributions of local dendritic activity and global, action potential-dependent activity of adult-born GCs by targeting iSFO in specific subcellular compartments. Finally, we will test the role of adult-born GCs in odor learning, by training mice for odor discrimination tasks and determining if the suppression of adult-born GC activity by iSFO affects the animal’s learning.
Together, this collaboration combines complementary expertise and cutting-edge techniques of three co-PIs to understand the functional role of adult neurogenesis, one of the most dramatic forms of plasticity in the adult brain.

Humans and animals respond to novel stimuli with distinct orienting and exploratory behaviors. From an evolutionary perspective, these responses seem highly adaptive, since they increase an animal’s chance of finding new resources or mating partners. Novelty-mediated orienting responses have been used extensively in developmental psychology as a non-verbal measure for object recognition in infants. Moreover, disturbances in novelty processing have been linked to human pathological conditions including schizophrenia and autism as well as human personality traits associated with maladaptive behaviors such as addiction. Together, these findings highlight the importance of novelty processing for human cognitive function. Although, previous studies have identified key brain areas involved in novelty processing, the neural mechanisms for how novel stimuli are detected and learned remain unresolved. Furthermore, a mechanistic description of how behavioral responses are initiated is currently lacking. Thus, I propose here to investigate the neural mechanisms for novelty processing using a combination of multielectrode recording and optogenetic techniques in awake, behaving mice. By genetically tagging specific neuronal subpopulations for electrophysiological identification and manipulation, this approach allows for the precise characterization of the firing properties of defined neuron types and for probing their causal involvement in specific aspects of novelty behavior.

Many medically important viruses have RNA genomes. These viruses replicate with an extremely high mutation rate, giving rise to highly diverse populations, and enabling rapid adaptation to changing environments. However, numerous mutations are deleterious as they destabilize the structure and compromise the function of viral proteins. It is not known how many viral variants in any given population are defective and how this affects its infectivity. Here, we propose to determine the relationship between the physical outcome of protein mutations and virus fitness. Through a systematic mutagenesis of the poliovirus genome, we plan to assess the effects of numerous mutations on protein thermodynamics and on viral infectivity. The link between protein thermodynamics and viral fitness will provide understanding of the biophysical origins of the majority of mutations and their effects on fitness.
This analysis will enable us to map the mutational composition of natural viral populations in terms of protein biophysics, and to estimate which variants are defective. Furthermore, we will analyze how the population composition changes in response to varying factors, such as availability of host factors. Thus, our study aims to give a quantitative understanding of viral population dynamics and evolution in light of the biophysical properties of its proteins. It will open avenues to addressing questions about virus pathogenesis, adaptation and molecular evolution.

How do animals process, represent and use multimodal sensory information from the natural environment? Our project will address this fundamental question by mounting ultra-lightweight sensors on free-flying bats to capture auditory, visual and chemical signals received by animals engaged in natural behaviors. The sensor data will lay the foundation for novel perceptual experiments, computational modeling, and biorobotic platforms. Our work will focus on the Egyptian fruit bat, a Neotropical frog-eating bat, a European trawling insectivorous bat, and a North American aerial insectivorous bat. The diversity of bat species and their adaptations to different environments and behavioral tasks make them especially well-suited for this line of inquiry.
Research on sensory processing in bats has, up until now, largely emphasized a single modality, typically echolocation. We will launch a major research effort aimed at uncovering the contribution of multisensory processing to the natural behaviors of free-flying bats. The use of novel on-board sensors will give us access to a bat’s view of the world as it moves in time and space. Sensor recordings will be taken in the field, where bats commute over long distances, negotiate obstacles and find food, and in the laboratory, where animals can be trained to perform tasks in experimentally controlled settings. Lab experiments will include creation of a visual-acoustic virtual reality environment.
Our project is directed at understanding the conditions that give rise to cross-modal enhancement of information, differential weighting of sensory information through separate channels, and behavioral decisions in the face of sensory conflicts. In addition, we propose studies that elucidate the role of spatial memory in a variety of tasks and determine the conditions in which bats switch from sensing modes to memory modes. We will investigate and develop mathematical representations of sensory processing and computational architectures for spatial memory, navigation, learning and adaptive behaviors with a view towards exploiting such architectures in complex robotic navigation systems.
The overarching goal of our project is to advance critical new insights to the multisensory representations and cognitive maps of the environment animals use to make decisions for goal-directed actions. Through integrated field and lab studies, computer modeling and biomimetic robotics, our project will contribute to a deeper understanding of multisensory integration, scene analysis, and spatial navigation in natural settings.