The regulation of mRNA translation is crucially important in determining which cellular proteins are produced in response to intracellular and extracellular cues. The resulting collection of functional proteins determines which physiological processes a given cell will carry out; therefore, the deregulation of protein production is strongly implicated in diseases, such as cancer, in which cells fail to appropriately respond to stimuli. The mammalian target of rapamycin(mTOR) signaling pathway, which links amino acid, growth factor, and energy availability to mRNA translation resulting in cellular growth and proliferation, is frequently deregulated in cancer and is an active target for drug discovery. Among the effectors of mTOR signaling are the eukaryotic translation initiation factor 4E (eIF4E) binding proteins (4E-BPs), which when phosphorylated bymTOR release the mRNA 5'-cap protein eIF4E to promote translation of progrowth/proliferation mRNAs. While previous biochemical fractionation experiments have suggested that 4E-BP1 is exclusively cytoplasmic, we show using immunoassays that this protein is also present in the nuclei of mammalian cells, where it sequesters eIF4E upon mTOR inhibition. This nuclear eIF4E accumulation is useful as a biomarker for mTOR signaling, as we use it as the read-out for a chemical genetic screen for novel mTOR pathway inhibitors. This discovery of a nuclear eIF4E:4E-BP1 complex opens the door to the potential for4E-BPs to impact nuclear RNA processing events. Evidence for and against iv such a nucleus-specific function for 4E-BPs is evaluated and the nuclear function of 4E-BPs is assessed experimentally. / La régulation de la traduction des ARN messagers (ARNm) est d'une importance cruciale afin de contrôler quelles protéines sont produites en réponseaux signaux intra- et extracellulaires. La collection de protéines fonctionnelles quien résulte détermine quels processus physiologiques seront effectués par la cellule. Conséquemment, la dérégulation du contrôle traductionnel est fortement impliquée dans plusieurs pathologies, incluant le cancer, ceci dû au fait que les cellules ne répondent pas de manière appropriée aux stimuli qu'elles reçoivent. Une voie de signalisation impliquée dans la croissance et la prolifération cellulaire qui est souvent dérégulée dans les cancers, la voie de la cible mammifère de la rapamycine (mTOR), et qui intègre la disponibilité en acides aminés, facteurs de croissance et énergie avec la traduction des ARNm, est une cible pharmacologique préférentielle. Parmi les effecteurs de la voie mTOR, on retrouve les protéines s'associant au facteur d'initiation de la traduction eIF4E, les 4E-BP, qui lorsqu'elles sont phosphorylées relâchent la protéine liant la coiffe5' des ARNm, eIF4E, promouvant ainsi la traduction des ARNm encodant des protéines impliquées dans la croissance et la prolifération. En contraste avec les résultats de fractionnement subcellulaires reportés précédemment dans la littérature suggérant que 4E-BP1 est une protéine exclusivement cytoplasmique, nous montrons ici, par essais immunologique, que cette protéine est également v résidente du noyau des cellules mammifères où elle séquestre eIF4E suivant l'inhibition de mTOR. Cette accumulation nucléaire de eIF4E est un biomarqueur de choix que nous avons utilisé comme lecture du niveau d'activité de la voie mTOR lors d'un criblage chimio-génétique entrepris dans le but de trouver de nouveaux inhibiteurs de la voie mTOR. Cette découverte d'un complexe nucléaire eIF4E :4E-BP1 à ouvert la porte à une possible fonction des 4E-BP dans certains processus nucléaires. Les évidences en faveur et en opposition àun tel rôle nucléaire spécifique des 4E-BPs sont évaluées et la fonction nucléaire des 4E-BPs est testée expérimentalement.

Angiogenesis is the formation of new blood vessels by sprouting from existing vessels. In order for new blood vessels to invade the surrounding tissue, the extracellular matrix must be degraded and subsequently remodeled. Fibulin (fbln)-5 (DANCE, EVEC) is an elastin and integrin binding extracellular matrix protein that is expressed strongly throughout the cardiovascular system during development and is re-expressed during vascular injury. Recently, an increase in the number of cutaneous blood vessels was observed on the posterior surface of the abdominal wall of the fbln5-/- mouse. In the present study, the role of fibulin-5 as an anti-angiogenic factor was investigated in vivo. The increase in vascular sprouting in the fbln5 -/- skin was confirmed to be significant by quantitative analysis. Using a well-established angiogenic model, polyvinyl alcohol (PVA) sponges were implanted subcutaneously in wild-type and fbln5-/- mice to determine the effect of the absence of fibulin-5 on fibrovascular invasion. Results showed no difference in fibroblast migration into the sponges from fbln5-/- and wild-type mice; however, PECAM-1 staining revealed a significant increase in microvascular invasion in sponges from fbln5 -/- mice. Sponges from fbln5-/- mice showed increased expression levels of VEGF and the angiopoietins relative to wild-type counterparts by quantitative RT-PCR. In addition, fibulin-5 inhibited bFGF induced neovascular invasion in Matrigel. In an ex vivo system, a lack of fibulin-5 stimulated vascular sprouting, and this response was blocked by the addition of exogenous fibulin-5. Taken together, these results indicate that fibulin-5 functions as an inhibitor of physiological angiogenesis and the absence of fibulin-5 results in an increase in vascularization.

Fibrillins constitute the major backbone of multi-functional microfibrils in elastic and non-elastic extracellular matrices. Proper expression, assembly and homeostasis mechanisms are central to the formation and function of these microfibrils. These properties are often compromised in pathological situations such as Marfan syndrome and other fibrillinopathies caused by mutations in fibrillins.While the developmental expression of fibrillin-1 and -2 has been intensely studied, fibrillin-3 is much less characterized in this regard. In this thesis, we analyze the developmental expression of human fibrillin-3 and compare it to that of other fibrillin isoforms. Fibrillin-3 is widely expressed in connective tissues of many organs as well as in close proximity to basement membranes of developing epithelia and endothelia. Generally, fibrillins are expressed in the same organs with a number of differences on the tissue level suggesting that fibrillin-3 fulfills specific functions in human development, both overlapping and distinct compared to the other fibrillin isoforms.In the present work, we identify fibronectin as a novel binding ligand for all fibrillins and demonstrate the requirement of a preassembled fibronectin network for fibrillin-1 network assembly and homeostasis. We show that multimerized C-terminal halves of all fibrillins and the N-terminal half of fibrillin-1 interact with high affinity with the collagen/gelatin binding region of fibronectin as well as with some lower affinity sites outside this region. In this study, we further characterize the role of heparin/heparan sulfate in microfibril assembly and its interaction with fibrillins. We demonstrate that fibrillin-2 and -3 interact with heparin/heparan sulfate and that fibrillin multimerization increases the avidity for heparin/heparan sulfate. We also show that heparin/heparan sulfate acts as a regulator of fibrillin homo- and heterotypic interactions, which are critical for microfibril assembly. Our results refine the sequence of events leading to microfibril assembly, consolidated in a new model. / Fibrilline forme l'ossature des microfibrilles multifonctionnelles présentes dans les matrices élastiques et non-élastiques. Un bon fonctionnement des mécanismes d'expression, d'assemblé et d'homéostasie est essentiel à la formation de microfibrilles fonctionnelles. Ces mécanismes sont souvent altérés par des mutations dans les fibrillines lors de pathologies telles que le syndrome de Marfan et autres fibrillinopathies. Bien que l'expression développementale des fibrillines-1 et -2 eut été intensément étudiée, peu d'informations sur l'expression de fibrilline-3 sont connues. Dans cette thèse, nous analysons l'expression développementale de la protéine humaine fibrilline-3 et la comparons à l'expression des autres fibrillines. Fibrilline-3 est exprimée dans le tissue conjonctif de nombreux organes ainsi que proches des membranes basales épithéliales et endothéliales. Les trois fibrillines sont généralement exprimées dans les mêmes organes avec quelques différences au niveau tissulaire. Ces résultats suggèrent que fibrilline-3 ait des fonctions précises dans le développement humain, ces fonctions pouvant être identiques ou différentes des autres fibrillines.Dans cette étude, nous définissons fibronectine comme interagissant avec toutes les fibrillines et démontrons qu'un réseau préétabli de fibronectine est essentiel à la formation et l'homéostasie du réseau de fibrilline-1. Nous montrons que la moitié carboxylique multimérique de toutes les fibrillines ainsi que la moitié aminique de fibrilline-1 interagissent avec grande affinité avec la région d'interaction de collagène et gélatine de fibronectine ainsi qu'avec d'autres régions. Dans le travail présent, nous caractérisons aussi le rôle joué par héparine-sulfate d'héparane dans l'assemblé des microfibrilles ainsi que son intéraction avec les fibrillines. Nous démontrons que fibrilline-2 et -3 intéragissent avec héparine-sulfate d'héparane et que la multimerization des fibrillines augmente l'avidé pour l'héparine-sulfate d'héparane. Nous démontrons aussi que l'héparine-sulfate d'héparane agit en tant que régulateur des intéractions homo- et hétérotopiques entre fibrillines, ce qui est crucial pour l'assemblé des microfibrilles. La séquence d'évènements participant à la formation de microfibrilles est redéfinie par nos résultats et est présentée dans un nouveau modèle.

A major limitation to the success of islet cell transplantation as a therapy for type I diabetes is the cell loss induced by the islet isolation procedure. The aim of this thesis was to elucidate signal transduction events and related intracellular activities that are implicated in islet cell survival/death following islet isolation in order to develop therapeutic interventions to promote islet survival. / The isolation of pancreatic islets imposes considerable stress on these cells, resulting in significant levels of cell death following isolation which was associated with activation of the stress-activated c-jun NH2-terminal kinase (JNK). However, within 24 hours in culture, JNK activation was greatly reduced concomitant with an increase in AKT activation. Inhibition of phosphatidylinositol 3-kinase (PI3K)/AKT signalling resulted in sustained JNK phosphorylation, while activators of AKT suppressed JNK phosphorylation, indicating that the rise in AKT activity during islet culture suppresses JNK. One of the stimulus of the PI3K/AKT pathway was found to be insulin secreted by the islets themselves, acting in an autocrine manner. The result of this autocrine activation of the prosurvival AKT pathway, and subsequent suppression of JNK, was a decrease in the appearance of apoptotic cells in islets after 72 hours in culture. Caspase inhibition alone was unable to prevent cell death in isolated islets. In addition, the amount of mitochondrial depolarization occurring in isolated islets was unaffected by caspase inhibition, leading to the notion that the commitment to islet cell death could be occurring at the level of, or upstream of, mitochondrial dysfunction. Indeed, inhibition of BAX translocation to the mitochondria, a critical event mediating mitochondrial permeabilization, prevented islet cell death. Inhibition of JNK also prevented mitochondrial permeabilization and cell death. / The current results demonstrate that insulin can act as an autocrine survival signal in isolated human islets. These findings also reveal the interdependence of necrosis and apoptosis in isolated islets, suggesting therapeutic strategies which target early events in cell death signalling in order to prevent multiple forms of islet cell death.

The field of subcellular proteomics aims to describe and analyze all the proteins present in a precise subcellular compartment at a given time. In contrast to whole-cell or whole-organism proteomics, the analysis of individual organelles has provided simpler proteomes from which relevant biological information could be more easily derived. To date, the protein complement of several subcellular structures, including the mitochondria, lysosome, peroxysome, phagosome and nucleus has been described. The powerful and rapidly evolving instrumentation as well as the development of biochemical and bioinformatics tools now allow scientists to derive whole organelle models based on the proteomic data generated. This field however still faces numerous challenges. Among those is the analysis of membrane-associated proteins, whose large and hydrophobic character complicate their extraction, subsequent separation and analysis by mass spectrometry. Another emerging limitation in subcellular proteomic resides in the difficulty in collecting enough material of a very pure preparation of the organelle of interest, which still depend on lengthy and labor-intensive density-based centrifugation as the method of choice for subcellular fractionation. The work presented in this thesis describes the development of new methods for subcellular proteomics that address the above-mentioned limitations, and their application to relevant biological models. In chapter 2, we present the design of a non-discriminatory investigative approach to study membrane proteins. Relying on detergent-free and gel-free procedures, this strategy allowed identification of hundreds of cell surface-exposed proteins from freshly ejaculated bovine spermatozoa, as presented in chapter 3. Diverging from traditional cell fractionation protocols, we have refined in chapter 4 a fluorescence-assisted organelle sorting method and have employed it to acquire the proteome of corticotropes-derived secretory granules. Our proc / Le domaine de la protéomique subcellulaire vise à décrire et analyser toutes les protéines présentes dans un compartiment subcellulaire précis à un temps donné. Contrastant avec la protéomique des cellules ou d'organismes complets, l'analyse d'organelles individuelles a généré des protéomes plus simples desquels l'information biologique pertinente peut être plus facilement extraite. À ce jour, le complément de protéines de plusieurs structures subcellulaires, incluant la mitochondrie, le lysosome, le peroxysome, le phagosome et le noyau a été décrit. L'évolution rapide et la puissance de l'instrumentation disponible couplées au développement d'outils biochimiques et bioinformatiques permettent maintenant aux scientifiques de générer des modèles d'organelles complets basés sur les données générées par la protéomique. Ce domaine, cependant, fait toujours face à plusieurs défis. Parmi ceux-ci, on doit mentionner l'analyse des protéines associées à la membrane dont la taille et l'hydrophobicité compliquent l'extraction, la séparation subséquente et l'analyse par spectrométrie de masse. Une autre limitation émergeante en protéomique subcellulaire est l'obtention d'une préparation très pure d'organelles d'intérêt et ce, en quantité suffisante, qui dépend toujours de longues et laborieuses centrifugations basées sur la densité comme méthode de choix pour la fractionnement subcellulaire. Le travail présenté dans cette thèse décrit le développement de nouvelles méthodes en protéomique subcellulaire qui s'adressent aux défis mentionnés précédemment, et leur application à des modèles biologiques pertinents. Dans le chapitre 2, nous présentons l'élaboration et la mise au point d'une approche investigatrice non discriminatoire pour étudier les protéines de membrane. Basée essentiellement sur des procédures sans détergent et sans gel de séparation, cette stratégie a permis l'identification de centaines$

Netrins are a small family of secreted proteins that guide growing axons during neural development by binding to the receptor DCC (Deleted in colorectal cancer). Our lab and others have previously shown that the activity of the Rho family GTPases Rac1 and Cdc42 are essential for DCC- mediated neurite outgrowth. Rac1 and Cdc42 act as molecular switches, mediating cytoskeleton remodelling when they are active and bound to GTP. Rac1 and Cdc42 are regulated positively by GEFs (guanine exchange factors) and negatively by GAPs (GTPase-activating proteins). Since DCC does not interact directly with Rac1, there should be an indirect link between DCC and Rac1. Trio is a GEF that activates Rac1 and RhoA. The orthologs of Trio in C.elegans (unc-73) and in D. melanogaster have been shown to play important roles in axon guidance, suggesting that mammalian Trio may link DCC to Rac1 activation. Here, we investigated how netrin-1 and its respective guidance receptor DCC are linked to Rac1 through studying the role of Trio in this signaling pathway. We found that Trio, Nck1, PAK1, and DCC are present in the same signaling complex, and that netrin-1-induced Rac1 activation is impaired in the absence of Trio. Trio -/- cortical neurons fail to extend neurites in response to netrin-1, while they are able to respond to glutamate. Accordingly, netrin-1-induced commissural axon outgrowth is severely impaired in Trio -/- spinal cord explants and commissural axon projections are defective in Trio -/- embryos. In addition to defects in spinal cord development, the anterior commissure is absent in Trio-null embryos, and netrin-1/DCC-dependent axonal projections that form the internal capsule and the corpus callosum are also defective in Trio -/- embryos. Thus, Trio through its ability to activate Rac1 mediates netrin-1 signaling in axon growth and guidance. / Le facteur de guidage chémotropique nétrine-1 favorise la croissance axonale à travers son récepteur DCC (Deleted in Colorectal Cancer) via l'activation de Rac1. Cependant, le facteur d'échange nucléotidique (GEF) qui lie nétrin-1/DCC à Rac1 n'a pas encore été identifié. Nous démontrons que Trio est la protéine GEF impliquée dans ce phénomène. Nous avons trouvé que Trio, Nck1, PAK1, et DCC sont présents dans le même complexe de signalisation et que l'activation de Rac1 induite par nétrine-1 est inhibée en absence de Trio. Les neurones corticaux Trio-/- échouent dans l'extension de neurites en réponse à la nétrine-1 alors qu'ils répondent à la stimulation par le glutamate. Par conséquent, l'induction de la croissance des axones commissuraux est sérieusement entravée dans les explants de moelle épinière des embryons Trio -/-. En plus du défaut dans le développement de la moelle épinière, les commissures antérieures sont absentes dans les embryons Trio-/-, et les projections axonales, qui forment la capsule interne et le corpus callosum, sont aussi affectées dans les embryons Trio -/-. Donc, par sa capacité d'activer Rac1, Trio favorise la signalisation par la nétrine-1 dans la croissance et le guidage axonale.

Telomeres are nucleoprotein structures that protect the ends of chromosomes from recombination and fusion. Telomeres are prefentially maintained by the enzyme telomerase. Telomerase is a reverse transcriptase, minimally composed of two core components, a catalytic subunit, TElomerase Reverse Transcriptase (TERT) and an RNA subunit, Telomerase RNA (TR) that carries the template for the replenishment of the telomeres. Telomerase is present throughout evolution, from ciliates and yeasts to vertebrates. Likely because it performs reverse transcription to maintain telomeres in these organisms, telomerases are remarkably conserved in terms of structure as well as in term of function. Nevertheless, some species-specific differences exist between telomerases from different model organisms. Studying these differences can deepen our knowledge of the telomerase enzyme, critical for human diseases such as cancer. The conservation and distinction between telomerases was explored by studying the functional conservation of a pseudoknot structure between Tetrahymena and human TRs and the response of mouse and human cells to the mutations of the subunits of telomerase. The hypothesis that the pseudoknot domains of the Tetrahymena and human TRs (tTR and hTR, respectively) may be functionally interchangeable was tested by constructing a chimeric TR (htTR) where the hTR pseudoknot domain was exchanged for tTR pseudoknot domain. htTR exhibited a weak, non processive telomerase activity in vitro, and was defective in telomere elongation, demonstrating that the Tetrahymena and human pseudoknots are not fully functionally interchangeable. Then, the importance of telomerase for cellular proliferation in mouse compared to human cells was investigated by studying the cellular consequences of mutating each core subunit of the mouse telomerase. First, the TERT-DN, a dominant-negative catalytically-inactive mutant of mTERT, was stably introduced into a mouse cell line (CB17) whose telomere / Les télomères sont des structures nucléoprotéiques qui protègent les extrémités des chromosomes contre la recombinaison et la fusion. Les télomères sont maintenus de manière préférentielle par l'enzyme télomérase. La télomérase est une transcriptase inverse, composée minimalement de deux constituants essentiels, la sous-unité catalytique TElomerase Reverse Transcriptase (TERT) et la sous-unité ARN, Telomerase RNA (TR) qui comprend la matrice pour la génération des télomères. La télomérase est présente à travers l'évolution, des ciliés au vertébrés en passant par les levures. Vraisemblablement parce que l'enzyme utilise un mécanisme de transcription inverse pour maintenir les télomères dans ces organismes, l'enzyme est remarquablement conservée, tant en terme de structure qu'en terme de fonction. Néanmoins, quelques différences propres aux espèces existent entre les télomérases d'organismes modèles. Étudier ces différences peut approfondir notre connaissance de la télomérase, une enzyme cruciale pour des maladies humaines tel que le cancer. La conservation et la distinction entre les télomérases furent explorées en étudiant la conservation fonctionnelle d'une structure de pseudonœud entre les sous-unités TRs de Tétrahyména et de l'humain, et en étudiant la réponse des cellules murines et humaines à des mutations dans les deux sous-unités essentielles de la télomérase. L'hypothèse que les domaines pseudonoeuds de l'ARN de la télomérase de Tétrahyména et de l'humain (tTR et hTR, respectivement) soient fonctionnellement interchangeables a été testée en construisant un TR chimérique (htTR) où le domaine pseudonoeud de hTR a été échangé avec le domaine pseudonoeud de tTR. htTR a démontré une faible activité non processive in vitro et était défectif en terme d'élongation des télomères, démontrant que les pseudonoeuds de Tétrahyména et de l'humain ne sont pas complètement fonctionnel

Gap junctions are membrane channels between closely apposed cells that allow for intercellular exchange of small molecules. These channels are formed from a family homologous proteins known as connexins (Cx). In the present study we studied connexin trafficking within the cell with particular emphasis on 1/ the mechanisms of gap junction formation and removal and on 2/ the early events associated with connexin biosynthesis. The mechanisms of gap junction formation and removal were studied by treating Normal Rat Kidney and BICR-MlR$ sb{ rm k}$ rat mammary tumour cells with ilimaquinone (IQ), a novel and reversible inhibitor of protein secretion. The effects of IQ treatment on Cx43 maturation and gap junction assembly were then examined. The events associated with biosynthesis were examined through localization of Cx32 messenger RNA to either free or membrane-bound polyribosomes in rat liver hepatocytes through Northern blot analysis. In these studies we observed that IQ inhibited gap junction formation without inhibiting trafficking of Cx43 to the cell surface. Upon removal of the drug, gap junction plaques reformed. We concluded that additional factors other than phosphorylation are necessary for Cx43 assembly into gap junctions and this process is independent of microtubules. Preliminary studies revealed Cx32 mRNA in populations of free and membrane-bound polyribosomes suggesting that Cx32 is inserted into the endoplasmic reticulum membrane both post- and co-translationally.

The present study examines the endocytosis of testicular sulfated glycoprotein-1 (SGP-1) by the nonciliated cells of the efferent ducts. SGP-1 is a 70 KDa protein secreted by the Sertoli cells. Once secreted in the seminiferous lumen, the protein binds to the tail of spermatozoa. In the efferent ducts, it is endocytosed by the nonciliated cells, presumably via a receptor-mediated process. Because the initial steps of receptor-mediated endocytosis result from the binding of a ligand's terminal oligosaccharide to a receptor on the ceil surface, several monosaccharides were injected into the lumen of the rete testis to study their effect on the endocytosis of SGP-1 in the efferent duct. The labeling density of various endocytic compartments was estimated and compared in untreated and treated animals with various sugars. The following sugars were tested: glucose, galactose, mannose, mannose 6-phosphate, N-acetylglucosamine, N-acetylgalactosamine, and sialic acid. The findings suggest that, via its sialic acid binding domain, SGP-1 may bind to glycolipids on the tail of spermatozoa, remove them from the membrane forming a lipo-protein complex. The complex may then be endocytosed by the nonciliated cells, via a receptor that would recognize SGP-1's terminal sialic acid residues, and be delivered to the lysosomes to be degraded. (Abstract shortened by UMI.)

Regulators of G-protein Signaling (RGSs) are negative regulators of G-protein Coupled Receptor (GPCR) mediated signaling that function to limit the lifetime of receptor-activated Galpha proteins. Heterologously expressed mammalian RGSs can functionally complement a yeast mutant lacking its RGS containing gene SST2. Here we show that four mammalian RGSs differentially inhibit the activation of a FUS1-LacZ reporter gene by the STE2 encoded GPCR in yeast with the apparent rank order potency: RGS1 > RGS16 > RGS2 > RGS5. In order to examine the role of the GPCR in modulating RGS function, we functionally expressed the human somatostatin receptor 5 (SSTR5) in yeast. / The ability of RGSs to inhibit SSTR5 signaling was further assessed in cells expressing modified Gpa1 proteins. / Yeast have also been shown to be a useful model organism for the study of the localization of mammalian RGS proteins. We have constructed a series of vectors that allow us to express proteins fused to a Green Fluorescent Protein (GFP). (Abstract shortened by UMI.)