It has been demonstrated in some cases that not all cancer cells are the same. In some cancers, only a small proportion of cells is primarily responsible for the growth of the cancer. These tumorigenic cells have become known as ‘cancer stem cells’, are distinguishable by cell surface marker expression and able to give rise to other cancer stem cells as well as non-tumorigenic cells within the cancer. The existence of cancer stem cells has profound implications for the way that research into the causes and treatment of cancer is conducted, as long term control over this disease will likely require their specific targeting. My research proposal aims primarily to determine whether cancer stem cells exist in human melanoma. In order to do this, I will take fresh specimens removed from patients with melanoma and evaluate them for heterogeneous expression of cell surface markers. Phenotypically defined melanoma cell populations will be purified and evaluated for tumorigenicity by transplantation into immunocompromised mice. If tumorigenic and non-tumorigenic cancer cells co-exist, I will also test whether the properties of tumorigenic melanoma cells change during disease progression. In addition, I will also perform gene-expression profiling on tumorigenic and non-tumorigenic melanoma cell subpopulations to identify genes that correlate with tumorigenicity. Together, these approaches have the potential to reveal important new insights into the biology of melanoma that could enhance the management of this disease.

Mutations in Pkd1 and Pkd2 genes, encoding polycystin-1 (PC-1, a membrane receptor) and polycystin-2 (PC2, an ion channel) proteins respectively, are implicated in the development of skeletal diseases. The PC1 and PC2 complex is believed to form a mechanosensitive receptor-ion channel complex that is involved in flow detection in kidney epithelial cells and is essential to normal kidney physiology. Although recent evidence suggests the presence of PC1 and PC2 in osteocytes (the mechaonsensor in bone) may be important for their physiology, their role in the mechanosensitivity of these cells remains unknown. To fully address the functional role of the ciliary PCs in the mechanosensitivity of osteocyes, our research plan consists of three main steps, divided as follows: 1). Mechanosensitivity of the primary cilium in OB/OC cells. 2). Role of PCs in the mechanosensitivity of OB/OC cells. 3) Link between PC-stimulation and cellular activation of OB/OC cells. We believe these approaches will not only lead to a better understanding of the molecular mechanisms underlying the mechanosensitivity of bone cells, but will also shed light on mechanisms involved in the pathogenesis of skeletal diseases.

Cell to cell variability in gene expression and protein levels has recently become a subject of intense study. However, it is still unclear whether such variability has any bearing on cell fate in mammalian cells. This question is especially relevant to cell death, which has a central role in tumor suppression, the immune response, and host-pathogen interactions. Is cell death a deterministic process, in which genetically identical cells in a homogenous environment behave the same? Or is there an element of chance, where cells under the same conditions behave differently? To try to answer this question, I propose to challenge macrophage cells with diverse pro-apoptotic factors, alone and in combination with TNF, a pro-survival factor in macrophages. I will also examine variability in cell death after infection with macrophage tropic HIV. I will use time-lapse microscopy to determine whether the probability of cell death during a given time interval is the same or different between cells, and whether the extent of variability depends on the pathway involved. If significant variability in cell fate does exist, I will investigate how different cell fates are correlated to different temporal activation dynamics of the NF-kappaB pathway, a central pro-survival pathway in many cell types. Understanding variability in cell death may lead to better understanding of processes where cell death plays a key role, such as HIV infection, cancer therapy, and the immune response. This systems biology based understanding of immune cell physiology will be a new direction for me, but my background in systems biology and live cell imaging gives me the tools to carry out this project.

While considerable research has examined the neurophysiology of a single cognitive operation, much less is known on how we chain these basic operations into complex tasks. Understanding this important aspect of neuroscience is a main goal in my career and the research proposal which accompanies my CDA application constitutes a major step in this direction. I propose an empirical and theoretical investigation to understand how a cognitive task can be parsed in independent processing stages, and the factors that determine whether these stages can be carried out in parallel (simultaneously with other concurrent processes) or rather establish a serial processing bottleneck. A particular strength of this proposal is that it is guided by a theoretical framework -developed during the course of my HFSP postdoctoral fellowship- which guides neurophysiological experiments establishing very concrete predictions. I expect these investigations to feed back to the models which at this stage are simple mathematical schemes and remain to be determined at the neurophysiological level. Thus, the two principal goals of my CDA are 1) to parse, using human imaging and single-cell recordings in monkeys, the stream of activations of a cognitive task in discrete states, understand the temporal ordering of these states and their rearrangement when the main task is carried out with a concurrent interfering task and 2) to develop a biophysical (biologically plausible) model capable of explaining the key observations of different behavioral interference experiments, as a first step to develop realistic models of complex task execution.

Human embryonic stem cells (hESCs) offer a unique opportunity for studying early human development and at the same time, form an exciting potential source of cells for regenerative medicine. Central to both goals is to map out the hierarchy of emerging precursors, and understand the underlying mechanisms of lineage specification and the regulation of large-scale organization across an extended embryonic tissue. Accordingly, my goal is to study the spatio-temporal regulation of early human morphogenesis using models of hESC differentiation. My approach consists of 3 interconnected research directions: (i) Multi-layer analysis of the kinetics of differentiation in response to morphogenetic signals, (ii) identification and isolation of early hESC-derived precursors followed by analysis of their differentiation program, their developmental potential, and the interactions between different precursors, and (iii) Studying the large-scale organization of morphogenesis using 3d culture models allowing juxtaposition of distinct, hESC-derived precursors and their exposure to gradients of morphogenetic signals. Preliminary kinetic profiles of mRNA, miRNA, and ribosome density are already beginning to reveal interesting insights about the early developmental program of differentiating hESCs. A large screen of cell surface markers in collaboration with the Weissman lab (Stanford University) further identified distinct precursors that are either present in the undifferentiated culture or emerge during early stages of differentiation. Altogether these studies should lead to a significant progress in defining and characterizing unmarked, therapeutically relevant, hESC-derived cells.

The control of translation initiation is an important mechanism for the regulation of gene expression in eukaryotes, and is known to be involved in cell growth, cancer, and stress responses such as starvation, hypoxia, ER-stress, and viral infection. At least 12 essential initiation factors (eIFs) play central roles in facilitating initiation as well as in translational regulation. Although distinct functions of each eIF are so far been characterized genetically and biochemically, it remains unclear how they interact with one another on ribosome to exert the integrated functions. As most eIFs form intermediate complexes with ribosome, initiator Met-tRNAi, or mRNA at specific stage of initiation, it is important to analyze the structures of the initiation intermediates to understand basis of functional association of eIFs. The purpose of the research proposed here is to investigate the initiation muti-factor complex assembly and its relation with individual functions. The study is also focused on structure and function of the largest and yet most uncharacterized initiation factor eIF3 consisting of 13 different subunits. The proposal will potentially shed new light into the comprehensive understanding of initiation factor-driven initiation and its regulation. The focus of my PhD work was to reveal the structure of archaeal homologs of translational apparatuses by X-ray crystallography. Studying the eukaryotic initiation complex extends my scope into the larger and complicated system as well as expanding my repertoire of techniques to preparation of multi-subunit factors from eukaryotic cells, and structural analysis with cryo-electron microscopy reconstitution.

Ubiquitination is the posttranslational modification of selected proteins by the covalent attachment of one or more ubiquitin molecules. The biological importance of protein ubiquitination is reflected in the fact that disruption of the conjugation pathway leads to developmental aberrations, abnormal responses to stimuli and defects in cell division and growth. At the center of the conjugation cascade is the substrate recruiting E3 ligase, which largely governs specificity. Although the role of E3 ligases in selectively controlling protein abundance, localization and activity is well studied, many aspects of E3 ligase regulation needs to be elucidated. The aim of this proposal is to understand the mechanism of E3 ligase activity regulation in response to external stimuli via analysis of KEG, a novel Arabidopsis RING E3. KEG is active as a E3 ligase and kinase, capable of post-translationally modifying one or more proteins, including itself, by ubiquitination and/or phosphorylation. The precise roles that these modifications play in the regulation of KEG function will be investigated. KEG is a negative regulator of ABA signaling, a major phytohormone which plays a key role in plant development and survival under stressful conditions. A combination of in vivo and in vitro approaches will be used to define the mechanism by which ABA signaling modulates KEG E3 ligase and/or kinase activity to control protein ubiquitination and subsequent degradation by the 26S proteasome.

In some non-human primate species, such as the pig-tailed (PT) macaque, infection by simian immunodeficiency virus (SIV) is pathogenic, resulting in high viral loads, chronic T cell activation, inflammation and destruction of hematolymphoid microenvironments. These manifestations of immunopathology are not observed in SIV-infected African green monkeys (AGM), even in the face of high plasma viral loads. A consistently observed phenotype distinguishing these two outcomes is the absence of an antibody response to the p27 Gag capsid protein in the setting of non-pathogenic infection. Low or undetectable anti-p27 Gag antibody responses in SIV-infected AGM suggests a state of relative tolerance against p27 Gag epitopes. My proposal aims to determine how such selective tolerance is induced, with the hypothesis that the absence of anti-p27 Gag antibodies in AGM is due to the expression of endogenous retroviral sequences encoding a homologous Gag transcript. I will screen tissues from AGM and PT macaques by RT-PCR, in situ hybridisation and microarrays for expression of Gag transcripts, and identify the cell subpopulations in which they reside. I will also determine the frequency and functional capacity of p27 Gag-specific T regulatory cells in AGM and PT macaques, before and after infection with SIV. Experiments to induce tolerance by expressing Gag epitopes in PT macaques will then address the causal nature of antibody responses to p27 Gag in SIV disease pathogenesis. These studies may promote future consideration of HIV vaccines that selectively suppress the induction of anti-p27 Gag antibodies and will help to elucidate the mechanisms resulting in nonpathogenic SIV infection.

My aim is to contribute to our understanding of how perception is achieved. In the past I worked on the electrophysiology of rat tactile perception. Now I would like to approach the problem of perception from a new perspective by studying reading capacity in humans and learn a new technique, fMRI. Writing was invented only ~5400 years ago. Our ability to recognize words might thus result from the recycling of already existing systems for object recognition. Areas of ventral occipitotemporal cortex activated by objects (faces, tools...) lie next to the Visual Word Form Area (VWFA), an area activated preferentially by written words. I plan to investigate the universal features that have predisposed the VWFA to become specialized in word recognition. Letters are composed of elements analogous to viewpoint-invariant object junctions present in natural environments (e.g. T, L, X). We propose that the area that becomes the VWFA has an intrinsic predisposition for such junctions relatively to other object features. To test this hypothesis I will measure fMRI activations for objects presented either as photographs or drawings with lines and junctions selectively removed. To determine what function does the VWFA serve before one learns to read I will extend my experiments to special subject populations: adult illiterates entering literacy class and children entering school. In parallel, I will use the unique opportunity offered by patients with drug-resistant epilepsies to carry out single-neuron human recordings which will probe the neural code underlying word recognition. These experiments should provide an important insight into the distinctly human gift of written language.

Protein misfolding has been implicated in the pathogenesis of many neurodegenerative disorders such as Alzheimer’s disease (AD), Parkinson’s disease (PD), and prion diseases. Despite distinctive symptoms, they are all characterized by deposition of amyloid fibers in or around the affected neurons. Amyloid deposits are often present in many cell types, yet they affect only a subset of cells, specific to the disease. Amyloid formation is therefore not toxic to all cells, and conversely, not all amyloids are toxic. Clearly, there must be both intrinsic and extrinsic factors that render a misfolded protein toxic to a given cell type. I propose to study the mechanism of amyloid toxicity in budding yeast by combining chemical biology and genomics in a high-throughput manner. I will systematically screen for yeast genes that suppress or enhance the toxicity of Rnq1, a yeast prion. As a complementary approach, I will use a novel method, SICLOPPS, to select for cyclic peptides that suppress Rnq1 toxicity. The two-pronged approach will elucidate both genes and protein domains mediating amyloid toxicity, and also facilitate cross-platform verification of hits. The positive hits will be further studied in vivo and in vitro with methods pioneered in the host laboratory. The results should increase our understanding of cellular factors contributing to amyloid toxicity and provide new tools to study amyloid biology.

Zinc is an essential micronutrient for organisms but is also toxic when present in excess. The yeast Zrt1, a zinc uptake transporter, is a member of ZIP family of transporters that are involved in metal uptake in bacteria, fungi, plants, and mammals. During the transition from zinc-limiting to zinc-replete growth conditions, the Zrt1 transporter is inactivated by endocytosis for preventing zinc accumulation to a toxic level. Zrt1 is ubiquitinated shortly after zinc treatment and prior to endocytosis. A region of Zrt1 required for ubiquitination and endocytosis in response to zinc was mapped and was designated as a potential ‘‘metal response domain’’ (MRD). The aim of this study is to understand how cells sense zinc concentration and transduce the signal to ubiquitinate Zrt1. I propose three hypotheses regarding the molecular mechanism: Zrt1 senses zinc as it’s being transported Zinc binds to MRD and its binding status controls ubiquitination An intra-or extra-cellular zinc binding protein (zinc sensor) transduces the signal directly or indirectly to MRD by protein-protein interaction. I will test these hypotheses by directed approaches and also by identification and characterization of novel proteins required for the mechanism. This study will contribute for understanding of the molecular mechanisms of substrate-dependent endocytosis of various transporters, important processes for nutrient homeostasis in various organisms.

Negative feedback is found all over biology and is crucial to suppress fluctuations and keep systems away from undesired states. However, because it always acts indirectly, either after a delay or through an intermediate molecular species, negative feedback could also increase fluctuations. It is in fact mathematically impossible for any control loop to completely overcome the problems associated with indirectness. This places general lower bounds on variances and creates frustration trade-offs where suppressing one noise source instead amplifies another, whether in transcriptional autorepression or the control of body temperature. The aim of my proposal is to achieve a more quantitative understanding of negative feedback acting under biological constraints, illustrated by replication control of bacterial plasmids. Bacterial plasmids are simple, tractable, and well characterized, and use negative feedback to suppress fluctuations and reduce extinction rates. I will develop methods to accurately count plasmids copy number in single cells, and to separate between different origins of fluctuations. I will then test and illustrate the mathematically predicted trade-offs and limits, and finally go back to refine the theory. The project and host laboratory were chosen to make my transition from synthetic organic chemistry to systems biology as interesting as possible.

My research interest is to determine how cell polarity mechanisms influence the fate of dividing progenitors and the morphology of their differentiating progeny during mammalian neurogenesis. To approach both questions, I propose to exploit the simplicity and the accessibility of the neural retina as a model system. By coordinating cell polarity with mitotic spindle alignment, the Inscuteable (Insc) adapter protein is central to the mechanism of asymmetric cell division. In Drosophila, this process ensures unequal inheritance of polarized determinants during mitosis, hence implementing distinct fates in the two daughter cells. We hypothesize that during retina development, Insc is a key regulator of the asymmetric divisions that generate a dividing progenitor and a post-mitotic neuron, or two neurons of different types. In addition, we will address a potential implication of Insc in neuronal morphology by testing the functional relevance of its polarized accumulation in the axons of retinal ganglion cells, the projection neurons which transmit the visual input to the cortex. To gain insights into subcellular mechanisms and their impact on whole retina architecture, the experimental approach will combine cell biology and live microscopy with mouse transgenesis. Questionings in cellular neurobiology and all analytical tools involved in this project are new to me. Previously, I applied a strictly genetic approach where analysis of mutant phenotypes lacked the cellular resolution important for addressing many biological questions today. This new training will thus provide me with the multidisciplinary skills important for succeeding as an independent investigator.

In vertebrates and many other metazoans, Hox genes specify segments along the developing body axis (and along other structures such as limbs) by following a spatial and temporal colinear activation which reflects their genomic order. By using a genetic approach in mice, our lab has brought the main studies throwing light on the cis-acting control mechanisms of Hox genes’ colinearity during limb development. They have shown that intergenic regulatory sequences are globally sufficient for the spatial and temporal colinear activation (with occasionally tissue-specific functions). Nevertheless, some specific distant regulatory elements are necessary for limb, external genitalia and intestinal hernia development. These overall data suggest that each developing structure shares common and specific regulation mechanisms of Hoxd genes’ colinearity. Here, I propose to address one of the major questions of vertebrate Developmental Biology since the discovery of the colinear Hox expression in 1988 : what are the mechanistic bases of the Hox colinear activation along the developing body axis ? In order to modify the topography of this cluster, I shall use series of mice carrying simple/combined deletions or duplications of Hoxd genes that have been generated in the lab. In each genetic configuration, the expression analysis of genes neighboring the mutation will be informative about how colinearity is implemented during the developing body axis. We hope that few of these configurations will reveal those unexpected features that are often the keys to understand complex processes.

Integral membrane proteins (IMPs) constitute up to 30% of cellular proteomes, and they represent targets for the majority of currently employed drugs. Thus, detailed knowledge of their structure, function and biogenesis is of major importance for biology and medicine. Membrane insertion of most IMPs occurs co-translationally by ribosomes bound to the protein conducting channel (PCC) of the Sec machinery. In order to obtain mechanistic insights into IMP biogenesis we propose to analyze the exact biochemical composition and the three dimensional structure of membrane protein insertion intermediates. Rather than using an in vitro system containing purified components, we propose to generate the insertion intermediates in vivo by making use of the well characterized SecM translation arrest sequence C-terminally fused to IMPs at specific positions. Subsequent affinity purification via tags on the arrested IMP and/or components of the PCC in combination with established protocols for isolation of cytoplasmic membranes and ribosomes should allow us to purify amounts that are sufficient for biochemical studies and cryo-EM. The components that remain attached to the insertion intermediate, such as cytoplasmic chaperones, targeting factors or the membrane protein integrase YidC, will be identified by mass spectrometry and immunodetection. Single particle analysis in combination with 3D reconstruction should result in structures at sufficiently high resolution to be interpreted on a molecular level. This novel approach will provide a highly versatile method that can be used to study the insertion mechanism of membrane proteins of various lengths and topologies.