<p>Apoptosis is a conserved mode of cell death executed by a group of proteases named caspases, which collectively ensure tissue homeostasis in multicellular organisms by triggering a program of cellular "suicide" in response to developmental cues or cellular damage. </p><p>Accumulating evidence suggests that cellular metabolism impinges directly upon the decision to initiate cell death. Several links between apoptosis and metabolism have been biochemically characterized. Using <italic>Xenopus</italic> oocyte extracts, our laboratory previously discovered that caspase-2 is suppressed by NADPH metabolism through an inhibitory phosphorylation at S164. However, the physiological relevance of these findings has not been investigated at the whole organism level. Studies presented in this dissertation utilize both Schneider's <italic>Drosophila</italic> S2 (S2) cells and transgenic animals to untangle the influence of metabolic status on fly apoptosis.</p><p>We first demonstrate a novel link between <italic>Drosophila</italic> apoptosis and metabolism by showing that cellular NADPH levels modulate the fly initiator caspase Dronc through its phosphorylation at S130. Biochemically and genetically blocking NADPH production removed this inhibitory phosphorylation, resulting in the activation of Dronc and the subsequent apoptotic cascade in cultured S2 cells and specific neuronal cells in transgenic animals. Similarly, non-phosphorylatable Dronc was found to be more potent than wild-type in triggering neuronal apoptosis. Moreover, upregulation of NADPH prevented Dronc-mediated apoptosis upon abrogation of <italic>Drosophila</italic> Inhibitor of Apoptosis (IAP) protein 1 (DIAP1) by double-stranded RNA (dsRNA) or cycloheximide (CHX) treatment, revealing a novel mechanism of DIAP1-independent apoptotic regulation in <italic>Drosophila</italic>. Mechanistically, the CaMKII-mediated phosphorylation of Dronc hindered its activation, but not its catalytic activity. As NADPH levels have been implicated in the regulation of oocyte death, we demonstrate here that a conserved regulatory circuit also coordinates somatic apoptosis and NADPH levels in <italic>Drosophila</italic>.</p><p>Given the regulatory role of NADPH in the activation of Dronc in <italic>Drosophila</italic> and caspase-2 in vertebrates, we then attempted to further elucidate the underlying signaling pathways. By tracking the catabolic fate of NADPH, we revealed that fatty acid synthase (FASN) activity was required for the metabolic suppression of Dronc, as both the chemical inhibitor orlistat and FASN dsRNA abrogated NADPH-mediated protection against CHX-induced apoptosis in S2 cells. Interestingly, it has been previously demonstrated that blocking FASN induces cell death in numerous cancers, including ovarian cancer; however, the mechanism is still obscure. As our results predict that suppression of FASN activity may prevent the inhibitory phosphorylation of Dronc and caspase 2 (at S130 and S164 respectively), we examined the contribution of caspase-2 to cell death induced by orlistat using ovarian cancer cells. Indeed, caspase-2 S164 was dephosphorylated upon orlistat treatment, initiating the cleavage and activation of caspase-2 and its downstream target, Bid. Knockdown of caspase-2 significantly alleviated orlistat-induced cell death, further illustrating its involvement.</p><p>Lastly, we developed an assay based on bimolecular fluorescence complementation (BiFC) to monitor the oligomerization of Dronc in S2 cells, a crucial step in its activation. The sensitivity of this assay has been validated with several apoptotic stimuli. A future whole-genome screen employing this assay is planned to provide new insights into this complex apoptotic regulatory network by unbiasedly identifying novel apoptotic regulators.</p> / Dissertation

Le métabolisme central de la souche E. coli MG1655 Δpgi ΔudhA Δedd Δqor a été rationnellement modifié afin de produire deux moles de NADPH et deux moles de NADH lors de l’oxydation du glucose en acétyl-CoA, alors qu’une souche sauvage produit quatre moles de NADH. La conséquence de cette modification est une forte diminution de son taux de croissance sur milieu minimum et glucose. Afin d’évaluer les aptitudes de cette souche à s’adapter à un tel stress métabolique, son évolution in vivo a été forcée lors de cultures par repiquages successifs sur glucose. Ainsi, après quatre cultures d’évolution un clone pur a été réisolé et caractérisé : un taux de croissance multiplié par six par rapport à la souche non évoluée a été mesuré. L’analyse par CGS (Séquençage par comparaison de génomes) a permis de corréler l’augmentation du taux de croissance à l’apparition d’une mutation, NuoF*(E183A), dans la sous-unité NuoF du complexe respiratoire I, complexe NADH-dépendant. Des études biochimiques et physiologiques de l’impact de cette mutation ont permis de démontrer que le complexe I évolué peut oxyder à la fois le NADPH et le NADH, créant ainsi une nouvelle voie d’oxydation du NADPH dans la cellule. L’évolution in vivo a ensuite été poursuivie au cours de onze repiquages et un nouveau clone pur a été réisolé et caractérisé : un taux de croissance proche de la souche sauvage et onze fois supérieur à celui de la souche non évoluée a alors été mesuré. L’analyse par CGS a permis cette fois de corréler l’augmentation du taux de croissance à l’apparition de deux mutations : NuoF*(E183A) et d’une deuxième dans la sous-unité α de l’ARN polymérase, rpoA*. Enfin, une deuxième souche E. coli MG1655 ΔpfkAB ΔudhA Δedd Δqor a été construite afin de détourner son métabolisme pour produire cette fois trois moles de NADPH et une mole de NADH lors de l’oxydation du glucose en acétyl-CoA. Cette souche étant incapable de se développer en milieu liquide et glucose, une étape de criblage en milieu solide et glucose a permis de sélectionner des clones capables de croître sur glucose. Tous ces clones possédaient soit la mutation NuoF*(E183A), soit une nouvelle mutation NuoF*(E183G), dont la caractérisation biochimique a montré que les deux enzymes évoluées permettent l’oxydation du NADPH par la chaîne respiratoire. Le phénomène d’évolution in vivo a conduit à la création d’une nouvelle fonction pour le NADPH qui n’est plus seulement impliqué dans les réactions de synthèse anabolique mais qui peut être utilisé pour produire directement de l’énergie catabolique. La compréhension globale du phénomène d’évolution a finalement permis la conception de nouvelles souches adaptées pour la production NADPH-dépendante de composés chimiques d’intérêt / Bacterial metabolism is characterized by robustness and plasticity that allow it to adjust too many metabolic perturbations. This present work demonstrates Escherichia coli abilities of evolution and adaptation under stress of NADPH accumulation. We constructed the E. coli MG1655 Δpgi::FRT ΔudhA::FRT Δedd::FRT Δqor::FRT strain where central metabolism has been rationally engineered to produce two mol of NADPH and two mol of NADH during the oxidation of glucose to acetyl-CoA, while a wild-type strain produces 4 mol of NADH per mole of glucose. Consequently, this strain presents a weak growth on glucose mineral medium. So as to evaluate bacterial abilities to overcome such metabolic stress, in vivo evolution of this strain has been forced in laboratory by serial transfer subcultures. After four evolution subcultures, an individual clone has been characterized by a six fold increased growth rate compared to non-evolved strain. CGS (Comparative Genome Sequencing) analysis allowed us to correlate growth improvement with one mutation apparition in respiratory complex: NuoF*(E183A) in NuoF subunit from the NADH dependant complex I. Further biochemical and physiological studies demonstrated that the evolved respiratory complex is able to oxidize both NADH and NADPH, resulting in a new NADPH reoxydation pathway in the cell. In vivo evolution experiments were then continued until eleven subcultures, where a new individual clone has been characterized by an eleven fold increased growth rate compared to non-evolved strain. Additional CGS analysis allowed us to correlate growth improvement with apparition of two mutations: NuoF*(E183A) and another mutation within the RNA polymerase alpha subunit, rpoA*. Thus, a second E. coli MG1655 ΔpfKA::FRT ΔpfKB::FRT ΔudhA::FRT Δedd::FRT Δqor::FRT strain has been rationally engineered to produce three mol of NADPH and one mole of NADH per mole of glucose oxidized to acetyl-coA. As this train was unable to growth in liquid glucose mineral medium, we performed a solid-state screening on glucose mineral medium that led to two different types of NuoF mutations in strains having recovered growth capacity. In addition to the previously seen E183A mutation other clones showed an E183G mutation, both having NADH and NADPH oxidizing ability. This result highlights need of this new NADPH reoxydation pathway for NADPH accumulating cells. This solution creates a new function for NADPH that is no longer restricted to anabolic synthesis reactions but can now be also used to directly produce catabolic energy. Finally, global understanding of evolution process allowed conception of new engineered strains, well designed for NADPH dependant production of chemicals of interest