Dendritic spines are small dendritic protrusions (~0.5 µm in diameter) where most excitatory synapses reside. Spine enlargement is associated with, and essential for, induction of long-term synaptic potentiation (LTP), a cellular model for learning and memory. Following LTP induction, the spine volume increases rapidly by 3-4 times within 1 min and decays over several minutes, followed by a smaller sustained enlargement lasting more than 1 hour. This spine structural plasticity requires re-organization of the actin cytoskeleton, and exocytosis and endocytosis of vesicles. However, the detailed dynamics of membrane trafficking and cytoskeleton rearrangements during spine growth are unknown. This project proposes to develop a technique to image structural dynamics of spines with nanometer resolution using high-speed atomic force microscopy (AFM). By resolving the detailed structural changes including filopodia extension, membrane trafficking, and rearrangement of surface proteins during the spine growth, this project is expected to lead to a better understanding of the mechanisms underlying structural plasticity of the spine and LTP, and ultimately learning and memory.

This project seeks to understand how cells migrate in complex environments. While cell migration on flat substrates has been thoroughly studied, cells migrating in vivo are faced with substrates of varying geometry, molecular composition and mechanical properties. Here, the cells show a much richer phenotype and are known to adapt their internal machinery to the extracellular environment. Different cell types exhibit specific migration strategies such that some of them use predominantly actin polymerization-powered protrusions, while others use motility based on pressure-driven blebbing as a default. Furthermore, cells have been observed to switch between these two different migration strategies upon manipulation of the actin polymerization dynamics within the cell, but also upon modification of such substrate properties as rigidity and adhesiveness. We will employ zebrafish germ cells and mouse leukocytes as in vivo model systems that represent the “professionally” blebbing vs. the actin polymerization driven phenotype. In vitro, these and other model cell types will be subjected to genetic, pharmacological and most importantly environmental manipulations that will uncover their range of plasticity regarding force generation and transduction. We will study how the cellular protrusive and contractile activities respond to variations of the geometry, adhesiveness and stiffness of the external substrate and to the presence of external loads. Readouts will include the dynamic visualization of cytoskeletal components, morphological response of the cell and actual motility, while forces exerted by the cell will be recorded quantitatively. Our data will be analyzed in the light of physical models to understand the mechanisms by which internal and external factors affect the strategy of cell motility, and to identify the key physical parameters governing the efficiency of actin-powered protrusions and bleb-based motility. The proposed research will establish novel techniques to quantitatively measure forces and visualize cytokeletal dynamics in a wide range of engineered as well as physiological settings. The integration of experimental data and theoretical modeling should ultimately lead to the development of a universal phase diagram that allows to predict the migratory strategy of a cell as a function of internal and environmental conditions.

Certain types of memories, especially fear memories, are stored in collections of cells in a part of the brain called the amygdala, in a manner that depends on the activity of the transcription factor CREB.

Experiments in which CREB activity was altered in subsets of neurons in the amygdala revealed that enhancement of CREB activity can enhance memory formation in those neurons and inhibition of CREB can decrease it. However, these experiments controlled CREB levels on time scales of days to weeks, whereas a mouse can form memories as a result of a single event lasting a few seconds. Is CREB required precisely during this training event? Soon after? Just before?

Memories, once formed, are not static but may be reconsolidated or undergo extinction depending on the nature of further experience. These processes also depend on CREB, but again it is unclear when CREB is required.

Using a combination of computational protein design, genetic selection via phage display, and structural biochemistry we will create light-switchable versions of CREB (photo-CREB) and a dominant negative form of CREB (photo-DN-CREB). Photo-CREB will allow activation of CREB with blue light irradiation. Photo-DN-CREB will allow inhibition of CREB with blue light irradiation. Both processes will be reversible (i.e., activation or inhibition will cease after the light source is removed). These proteins will be chimeras of the basic leucine zipper (bZIP) DNA binding region of CREB and the light sensitive protein PYP (photoactive yellow protein). These designed and evolved proteins will be analyzed and optimized in vitro using a range of biophysical techniques (X-ray, NMR, DNA binding assays). Experiments with cells in culture will assay kinetics of photoswitching and control of transcription using CREB reporter assays. Light-switchable CREB proteins will then be expressed in freely moving mice with optical stimulation delivered with an optical fiber directly to the mouse’s brain.

This approach using light-switchable proteins will permit the processes of memory formation to be probed on a time scale of seconds to minutes - an unprecedented degree of temporal resolution – and will therefore enable fundamental questions about the role of CREB in memory formation to be addressed for the first time.

The aim of the proposal is to determine how large scale search and taxis patterns of movement are produced by C. elegans at the motor output level and to quantify and predict the adaptive properties of the motor output in the presence of noisy sensory information in spatiotemporally structured environments. Understanding how movement patterns on the ecological scale are produced in response to (or absence of) sensory information is a fundamental, multi-scale, interdisciplinary problem of behavioral and evolutionary biology and biophysics. While most such work has been done by attempting to reduce behaviors to their molecular mechanisms, we propose to take a holistic, integrative approach. Our interdisciplinary team will comprehensively study and model sensory signalling and motor control underlying the adaptive movement strategies of a roundworm C. elegans, with particular emphasis on the thermal information processing system. We will bridge molecular, organismal, and ecological scales by investigating optimality of the behavioral patterns in experimentally controlled environmental contexts, and we will connect neuroscience, behavior, and sensory learning in these animals by characterizing the adaptive responses to the changes in the environment.

C. elegans is a "simple" organism with only 302 neurons, but with complex sensorimotor behavior, which includes nontrivial motion patterns, the ability to find food and to learn by conditionally associating temperature and food. These behaviors are adaptive and are influenced by the statistics of sensory cues in the environment. Our work will focus on answering questions such as: How do physical and statistical properties of the environment affect the worm behavior? Is the worm optimal in utilizing the available information? Can the worm's search patterns be predicted using optimization principles with appropriate physical constraints?

C. elegans is an ideal model for this research. Indeed, to date complete characterization of sensory motor behavior has been achieved, arguably, only for bacteria. However, physics on the microscale severely limits these behaviors, and actions performed by microbes are rather simple. On the other hand, C. elegans displays a wide range of behaviors, including associative learning, a hallmark of higher animals. Yet it is still simple enough to avail a large set of powerful genetic, molecular biology, and optical neurophysiology tools; the entire motor output can be efficiently represented within low dimensional computational model; and the worm environment can be precisely manipulated experimentally. In view of this, C. elegans will become a “hydrogen atom” of the integrated research that connects biophysics of sensory processing to motor behaviors and their adaptations and to ecological consequences of these behaviors. This proposal will be the first in the series developing this atom.

Malaria is one of mankind’s deadliest diseases. It is caused by a parasite that infects the circulating red blood cells, transmitted via the bite of an Anopheline mosquito. To survive and multiply in both man and mosquito, the parasite has evolved many novel biological adaptations to help it pass through the distinct tissues of the two hosts on its infectious journey. Perhaps most striking of all is the speed with which the parasite is able to move through tissues and enter host cells. To do this, it has evolved a molecular engine similar to that in our muscles, made up of two key proteins actin and myosin. However, in parasites these two proteins have diverged radically, making predictions about their function difficult. Actin, a protein that spontaneously comes together to form filaments, is crucial to many cell functions across life. Indeed, because of this property, cells, be they parasite, brain or cancer invest large resources in regulators to control the whereabouts and timing of filament formation. Once formed, these then provide a track for the myosins, the motor proteins of the cell, to run along, pulling cargo or even the entire cell as they go. In malaria parasites the actin-myosin engine is one of the secrets to this unique cells success. We want to understand how the engine works. How the filaments form and where these dynamic events occur inside the parasite cell. To this end, our three teams came together, comprising biologists, chemists and biophysicists, each with different interests in cell motility and parasite biology but united in using advanced technologies to see molecules in action either within parasites or in the Petri dish. However, in combining our efforts we aim to go one stage further. We want to generate fluorescent parasite lines that show us molecules in action in real time as well as mutants that show us how they move when actin regulation goes awry. We also want to build the entire parasite engine in vitro from its defined parts. This not only gives us a platform to test how the engine functions, but also allows us to test drugs for their capacity to inhibit engine parts and potentially halt parasites dead in their tracks. As it reveals key insights into how these deadly parasites move, this program will lay a foundation for new ways to treat malaria disease and rid mankind of one of its greatest scourges.

Understanding the specificity, affinity and dynamics of odor ligand-olfactory receptor interactions represents an outstanding problem of sensory neuroscience, structural and molecular biology. We recently identified a novel family of Drosophila olfactory receptors, the Ionotropic Receptors (IRs), which are homologous to ionotropic glutamate receptors (iGluRs), a conserved family of ligand-gated ion channels with a well-characterized “Venus flytrap” ligand-binding domain (LBD). The primary sequences of IR-LBDs are highly divergent and lack one or more key glutamate-binding residues, consistent with their implication in recognizing diverse small molecules. We propose the IRs as an excellent new model to define the structural basis of odor ligand-olfactory receptor interactions through a combination of Nuclear Magnetic Resonance, X-ray crystallography, molecular genetics and electrophysiology. Our results will provide insights into how a single protein fold can selectively accommodate myriad odor ligands with different affinities and how olfactory receptor repertoires evolve functional diversity. Moreover, the results will open the possibility to engineer IRs of novel ligand-binding specificity, with applications as genetically-encoded chemical sensors.

The discovery that four transcription factors – Oct4, Sox2, c-Myc and Klf4 – can convert somatic into pluripotent cells (iPS cells) has fueled the hope that individualized iPS cell-based therapies will be effective against prevalent diseases such as neurodegenerative disorders, diabetes and cancer. However, a quantitative foundation to the problem of induced pluripotency or reprogramming is missing. We will apply time-lapse microscopy and a powerful single-cell and single-molecule mRNA counting method developed in Dr. van Oudenaarden´s laboratory to state-of-the-art iPS models developed by Dr. Jaenisch´s group at M.I.T. With these tools in hand, we will measure the co-variation of multiple transcripts in thousands of cells exposed to various amounts and combinations of Oct4/Sox2/c-Myc/Klf4 and reconstruct the dynamics of pluripotency transcriptional networks as a function of the cell-cycle with single-cell resolution. In addition, we will count the same transcripts in iPS clones with various degrees of pluripotency to find which transcriptional network configuration associates with the highest potency. The proposed study will advance our knowledge on the molecular mechanisms of reprogramming and set a strong quantitative foundation for engineering large-scale iPS cells production for regenerative medicine.

CD8+ cytotoxic T cells are crucial to combat infectious pathogens. CD8+ T cells recognize peptides presented by MHC class I molecules on the surface of infected cells. The processing pathway for cytosolic proteins is relatively well understood but how exogenous material derived from vacuolar parasites is processed, remains unclear. This question is key for the understanding of host-pathogen interactions
Here I propose to study the molecular and cellular mechanisms leading to presentation of pathogen-derived material by MHC I. We will use the genetically tractable, obligate intravacuolar parasite Toxoplasma gondii (Toxo) as our main model. Toxo is an opportunistic pathogen which infects a large range of hosts including mice and humans.
During my HFSP fellowship, we have identified natural and protective antigens recognized by T cells in Toxo-infected mice. These tools are indispensable for the achievement of this project. We will examine the molecular steps leading to generation of the final epitope from the full-length parasite antigen (aim 1), study how parasite protein trafficking controls access to the MHC I pathway (aim 2) and identify host effectors that regulate organelle trafficking and may be involved in presentation of Toxo antigens (aim 3).
We will use a combination of gene targeting approaches in the parasite and host cell, advanced biochemical methods (mass spectrometry) and novel cellular assays (shRNA library). Our data should reveal new pathways facilitating or interfering with MHC I antigen processing and may ultimately suggest ways to facilitate antigenic generation and enhance parasite immunity.

As mechanical systems, cells control their shape via tight coordination of the dynamic self-assembly of cytoskeletal polymers (e.g. actin filaments) and modulation of intracellular tension exerted among these filaments by molecular motor proteins (e.g. myosins). Highly dynamic and contractile actomyosin-based structures, such as stress fibers of non-muscle cells and sarcomeres of muscle tissue, lie at the heart of fundamental cellular processes including morphogenesis, establishment of polarity and overall motility. Although it is well established that stress fibers play central roles in cell biology, the chemical and physical principles driving the formation of these highly organized contractile structures are largely unknown. The identification of the biochemical mechanisms and biophysical laws that govern the macroscopic organization of the cytoskeleton in cells is rendered difficult by the complexity of the systems. Simplified experimental systems in which actin dynamics can be reconstituted in the presence of regulatory factors and motor proteins are needed to understand the mechanism of stress fiber formation and the consequences of myosin-dependent force generation on the organization and stability of complex actin structures. We will use a multi-disciplinary approach to reconstitute the assembly, turnover, and contractility of stress fibers in vitro and determine how the mechanical and contractile properties of intermediate and assembled structures depend on the individual protein components. Our approach will integrate a new versatile micropatterning method capable of imposing spatial boundary condition to the growth of actin filaments, the use of diverse molecular motors with unique biochemical properties, and the use of AFM and laser trap nanometry to obtain reliable determination of the generated contractile fiber structures and mechanical properties. Theoretical biophysical modeling and computer simulation will complement these experimental works. General principles regarding cell mechanics and the hierarchy of molecular components will emerge from this multi-disciplinary work.

The epidermal growth factor (EGF) receptor family has been widely studied for its role in cancer. Recently, Her3 has attracted much attention because of its pivotal role as preferred dimerization partner for Her2 in Her2-positive breast cancer. It is thus of key importance to not only target Her2, but also Her3. To this purpose, structural and functional data of Her3 and in particular of the Her2/Her3 heterodimer are strongly needed. However, little is known about this heterodimer. Certainly no 3D structure is available: the heterodimer has so far shown to be too unstable to crystallize. Additionally, oligomerization studies would improve the understanding of receptor activation in situ.

X-ray crystallography and single molecule spectroscopy are methods of choice to answer these questions. Novel, alternative approaches will be applied, making use of the enormous potential of designed ankyrin repeat proteins (DARPins) targeting Her2 and Her3. For crystallization trials, the Her2 and Her3 ecotodomains will be produced with a C-terminal IgG-hinge to induce heterodimerization via disulfide bonds. DARPins can facilitate crystallization by acting as chaperones. Receptor oligomerization can be studied by applying Alexa-labelled DARPins in two colour coincidence detection microscopy (TCCD). Coincident fluorescence from single molecules labeled with two different fluorophores will be detected, as the molecules diffuse through a confocal volume. Finally, the potential biological effect of Her3-selective DARPins on tumor cells will be examined.

These data will extend our knowledge of tumorigenesis and aid in the development of new, much needed cancer therapeutics.

Many Nobel Prize winning advances in biology were driven by x-ray studies of crystallized biomolecular systems. A recent revolution in x-ray sources, the x-ray free electron laser (XFEL), eliminates the radiation damage problem that plagues studies using x-rays and electrons. XFELs open a new horizon of atomic resolution study of non-periodic biological structures. The Linac Coherent Light Source, LCLS, at the SLAC National Accelerator Laboratory recently became the first operational hard x-ray FEL. Its x-rays are a billion times more brilliant than any other source. One of the most promising scientific applications of XFELs is coherent diffraction imaging (CDI) of non-crystalline biological samples at sub-nanometer resolution. The basis of this "diffract-and-destroy" method is the recent observation that a sufficiently brief x-ray pulse provides a useful diffraction pattern before significant radiation damage commences. Our international team is comprised of the requisite expertise to uncover new scientific discoveries in complex non-periodic biological systems using XFEL experiments. We have i) extensive experience in algorithms for CDI, ii) over six years of experience in XFEL experiments, including LCLS, iii) the requisite biological expertise for virus preparation and data interpretation, and iv) unique experimental apparatus. Our primary objective is to achieve the highest resolution structure of algal (PBCV-1) viruses. Little is known structurally about the algal virus PBCV-1. 3D reconstructions of PBCV-1 at 2.2 nm resolution exist but many questions remain open. These include, for example, how the genome is packed and the structure of the unique vertex involved in infection. Since single particle CDI is so new, significant effort will be directed toward method development. At present, there are at least two computational methods in the literature that we will use for determining the relative orientations of diffraction patterns, permitting a 3D data set of diffraction intensity to be assembled for phasing and image reconstruction. XFEL-CDI methodologies are now on the verge of contributing to the biological questions posed for many viruses, including PBCV-1. We were awarded LCLS beamtime to begin addressing these questions and will use HFSP support to help stimulate new biological frontiers by the x-ray imaging of viruses.

Many proteins and enzymes frequently undergo structural rearrangement to execute their functions. Ligand/substrate entry and binding are often associated with some degree of localized disorder of both protein and solvent. Characterization of these states is vital to understand the mechanisms of general biological functions. In this agenda I will use human zinc dependent metalloproteinases, key physiological enzymes, as model system to study the correlation between protein and solvent motions with enzyme function. Here, I address the question of the enzyme-solvent-substrate interplay with special focus on the active site structural kinetics during the hydrolysis of peptide substrates and on the role of the omnipresent solvent: water. Specifically, I will combine stopped-flow x-ray and THz structural spectroscopic analyses to quantify both active site structural-kinetic transitions with coupled solvent-protein motions in real time and at atomic resolution. These studies will be complemented with solution NMR and computational analyses. Applying this multidisciplinary approach will allow exploring general and yet unknown aspects in enzymatic catalysis such as the effect of catalytic metal ions on solvent water dynamics, the correlation between localized active site chemical transitions with solvent dynamics and the influence of enzyme side chains in prearranging approaching substrate molecules to a preferential conformation for binding together with solvent molecules. Thus establishing a novel experimental procedure as well as revealing the detailed mechanism of metalloenzyme molecular recognition, onset of turnover and product release will be the ultimate goal of my agenda.

The primary virulence strategy utilized by many bacterial pathogens is the type III secretion system (T3SS), a molecular syringe through which effector proteins are injected into host cells. These effectors modify the host cell to promote bacterial infection, and as a result are critical determinants of pathogenesis. The study of effectors and their host targets is providing valuable insights into mechanisms of both pathogenesis and host immunity.

Despite the intensive investigation of effector targeting mechanisms, host targets for the majority of effectors remain unknown. This is largely due to the reliance upon target identification strategies, such as the yeast two hybrid assay and coimmunoprecipitation, which require isolation or over/ectopic-expression of the effectors and their proposed targets. The artificial nature of these approaches can prevent the observation of effector-target interactions.

Due to these limitations, I propose a novel strategy that employs a pathogen encoding effectors with photocrosslinking (PCL) unnatural amino acids for the in vivo detection of host targets. In this approach the genetically modified pathogen injects PCL-effectors into the host via natural infection. These PCL-effectors interact with cognate targets in their physiological environment. Ultraviolet light irradiation of the infected cells stimulates PCL in vivo, yielding covalently coupled effector-target complexes that are recovered and analyzed using affinity purification and mass spectroscopy. This in vivo tool will allow for the identification of the most relevant host targets and thus promises to advance our understanding of pathogen-host interactions.

The malaria parasite has developed an unusual mechanism to regulate gene expression, which seems largely dependent on epigenetic modifications such as chromatin remodeling. One of the main mechanisms involved in chromatin remodeling is DNA methylation. Although previous studies have been unable to detect DNA methylation in P. falciparum, recent data from the Le Roch lab obtained using the latest and most sensitive next-generation sequencing technology show that the genome of P. falciparum is in fact highly methylated. In the current project, I will investigate the functional role of DNA methylation in controlling malaria parasite development. First, I will determine DNA methylation patterns throughout the asexual blood stage and in sexually mature gametocytes, and relate these to changes in nucleosome positioning and mRNA transcript levels. Second, I will explore whether DNA methylation is involved in transcriptional regulation of the var genes that are responsible for parasite survival and immune escape. Third, I will characterize the cellular localization and biochemical activity of a putative DNA methyltransferase and determine whether this enzyme is essential for the development of the malaria parasite. In addition, I will perform pull-down experiments to identify proteins interacting with this putative MTase. Together, these studies will contribute toward a more comprehensive understanding of the complex process of transcriptional regulation in P. falciparum.

We propose to unravel a long-standing mystery in neuroscience: the origin and functionality of endogenous extracellular electric field in the brain. While neuroscientists have heavily relied on extracellular recordings to study mechanisms of brain processing very few have addressed their identity. We focus on the local field potential (LFP), the slow portion of the recorded extracellular voltage (Ve), thought to reflect neural circuit computations. We seek to quantitatively understand the origin of the LFP in neocortical and hippocampal structures and to manipulate such fields to characterize the extent to which these can directly affect the state and function of neurons and neural populations through field (ephaptic) effects. We do so in a highly interdisciplinary manner allowing us to study these effects in multiple scales. We will use novel multi-patch setups, silicon probe recordings, optical stimulation protocols to obtain intra- and extracellular recordings at multiple locations in- and outside neurons both in the hippocampus and the neocortex to understand the origin of extracellular fields induced by different neural types and the contribution of compartments such as the soma or dendrites. Concurrently, optical stimulation, transcranial electric stimulation and massively parallel recordings will be conducted to study Ve and the endogenous field in anesthetized and behaving rodents. In addition to the experiments, we plan to carry out biophysical simulations of individual neurons as well as neocortical and hippocampal circuits to understand the origin of Ve (and the LFP in particular) by examining various Ve-contributors such as postsynaptic currents from pyramidal neurons and other neural types. Finally, we seek to provide a theoretical framework to address the extent to which endogenous electric fields can alter or entrain the activity of a neural circuit via ephaptic coupling both in terms of subthreshold and spiking activity.

The kidneys are essential organs required for excretion of waste products and the regulation of fluid balance, blood pressure, blood volume, ion balance and bone density. The functional units of the kidney are the epithelial nephrons which can filter the entire blood volume every minute, creating 200 litres of filtrate each day and then recapturing 95% of this volume. All of the nephrons form during fetal development from two key stem/progenitor populations that interact with each other to form the intricately branched plumbing of the kidney as well as the filtering tubules. In humans, the number of nephrons per kidney varies 9 fold between individuals and there is an inverse correlation between nephron number and the likelihood of hypertension and kidney disease later in life. While much of kidney development has been described anatomically, due to a lack of quantitative and spatial modeling of this complex process across the fetal development, there is little understanding of how such variation in kidney size and nephron number can come about. The purpose of this application is to bring together mathematical modeling, advanced imaging and developmental genetics in order to develop a comprehensive model of the process of kidney development. Having created a framework that predicts the interactions between different cell types, we will test and refine this model by genetically altering the size of different cell populations to see how this affects development and whether the model holds. This will ultimately allow us to predict the outcome of perturbations in different cell compartments over space and time and potentially guide us with respect to understanding why nephron formation ends and what parameters might need to be altered to prolong the process.

The three-dimensional organization of the genome plays important and yet poorly understood roles in its regulation including control of gene expression and genome segregation. Control of gene expression by distantly located regulatory elements involves formation of chromatin loops driven by physical promoter-element interactions. Moreover, active and inactive segments of the genome appear spatially separated, which may contribute to their coordinated expression or silencing. Finally, formation of complex higher-order chromosome structures plays critical roles in chromosome condensation and segregation during mitosis. Thus, chromosomes assume multiple distinct conformations in relation to the expression status of resident genes, and undergo dramatic alterations in higher-order structure throughput the cell cycle. Detailed insights into chromosome conformation will greatly contribute to a more complete characterization of genome regulation.

Our aim is to combine light microscopy, cell and molecular biology with physics and statistics to determine a probabilistic map of the organization of chromatin domains in intact cells during cell cycle and in relation to gene expression. This project breaks important new ground by deriving high-resolution three-dimensional structures of higher-order chromatin organization and by discovering how these structures change dynamically during the cell cycle. This work will allow the characterization of the dynamics of higher-order chromatin folding and the identification of specific roles of genomic elements in these processes.

Lesions of the spinal cord or pathologies involving spinal motoneurons (such as amyotrophic lateral sclerosis, ALS) lead to partial or complete impairment of motor function in humans. Electrical stimulation of the spinal cord has been used both for diagnostic and rehabilitative purposes when spinal cord function has been compromised. It has also been an important basic research tool for exploring spinal cord function [1,2]. Nevertheless, electrical stimulation suffers from the disadvantage that current spreads in a highly non-specific manner, activating all nearby fibers of passage and making it difficult to isolate the effects of stimulation [3]. Optical stimulation of the spinal cord has the potential to overcome this drawback, as specific classes of spinal neurons can be focally targeted for excitation or inhibition. Given the considerable complexity of spinal circuits, the ability to focally activate specific sub-population of neurons is crucial to understanding overall spinal circuit function. Specifically, I begin with the hypothesis that the spinal cord is organized at the level of modules that control muscle activations in terms of functional groupings or “synergies” [4]. Where and how are these synergies constructed? Are they invariant or can they be modulated in a context-dependent fashion? I propose to investigate the hypothesis that synergies arise solely from the dynamics of inter-neuronal circuitry. This hypothesis can be rigorously tested by optogenetic manipulations of selectively targeted classes of neurons within the spinal cord.

The success of humans greatly relies on our ability to understand and manipulate our environment. The use of other life forms is no exception, though for ages this process resulted mostly from unconscious domestication. Spectacular advances in molecular biology now makes it theoretically possible to genetically modify existing organism, and particularly microbes, to endow them with new functionalities.
However impressive, our technical abilities to physically manipulate DNA largely exceed our understanding of the underlying genetic systems. The emerging field of synthetic biology aims at providing scientific tools to bridge this gap and make biology easier to engineer – be it for academic or technological applications.
A major roadblock is that the basic components specifying the expression of genetic informations are not described in a manner that support their rational use. Indeed, studies generally provide exquisitely refined descriptions of paradigmatic examples, analyzed in a single genetic context. As a result, most attempts to construct a particular genetic system with a defined output require a wasteful cycles of trial and errors.
The aim of this project is to analyze the activity of a large number of natural and engineered genetic components involved in gene expression, and to describe their interactions when composed with each others. The experimental data will be used to build predictive models linking sequences and activities. Beyond deepening our fundamental understanding of the central dogma, these results will be used to devise strategies to minimize interactions between components, thereby enabling their rational combination toward a predicted output.

The goal of this project is to understand how the HIV-1 virus that causes AIDS escapes from infected cells. Formation of retroviruses such as HIV-1 is an interplay between the multidomain viral protein Gag, plasma membrane lipids, and cellular proteins of the ESCRT machinery. Gag assembles into an ordered lattice on the inner face of the plasma membrane and recruits other components of the virus. Release of the nascent virus particle is not mediated by the viral proteins themselves, but by the cellular ESCRT complexes which are recruited by Gag to the budding site. In order to understand in detail how this happens, HIV-1 assembly and release will be reconstituted from the individual viral and cellular proteins and model membranes that mimic the plasma membrane. The Gag polyprotein and the individual human ESCRT complexes will be purified and fluorescently labeled. The proteins will be assembled in a sequential fashion on giant unilamellar vesicles (GUVs) and the ensuing formation of viral particles will be analyzed in real time by fluorescence microscopy. This will allow definition of the minimal protein machinery necessary for HIV formation, clarify what role specific lipids play in the membrane bending and scission, and possibly generate new strategies to inhibiting HIV release. The proposed reconstitution of HIV formation is a novel approach to study this process, and will provide a higher level of detail about the mechanism of enveloped virus assembly and release than has ever before been possible.