Neointimal proliferation is a key process underlying many cardiovascular diseases such as atherosclerosis and angioplasty-induced restenosis. Vascular smooth muscle cells (SMC) are significant contributors to the development and stability of the neointimal lesion. This is due, in part, to their capacity to be phenotypically modulated, facilitating SMC proliferation in response to mechanical injury, their subsequent migration, and deposition of extracellular matrix. The aim of this thesis was to characterize the function of two genes identified in our laboratory to be upregulated shortly after mechanical injury of vascular SMC and their exposure to fibroblast growth factor (FGF)-2, an injury-induced cytokine. The first is activation transcription factor (ATF) 4, which is upregulated by FGF-2 and mechanical injury in vascular SMC in vitro, and by balloon-injury in the artery wall. The induction of ATF4 by FGF-2 was shown to be mediated through the PI3K pathway, and preceded by phoshorylation of eIF2alpha, a known upstream effector of ATF4 activation. Knock-down of ATF4 expression inhibited balloon-injury induced neointimal hyperplasia, suggesting that ATF4 is a key player in the SMC response to injury. Furthermore, microarray analysis identified several genes whose transcription in response to FGF-2 may be regulated by ATF4. In particular, this work demonstrates that ATF4 is necessary for VEGF-A upregulation in SMC in response to FGF-2 and mechanical injury in vitro and in the artery wall following balloon-injury. The second is a translation factor, YrdC203. Using confocal fluorescence microscopy, YrdC203 was found to localize partially to the ER, and with RPL12, a component of the 60S ribosomal subunit. Immunoprecipitation studies demonstrate that YrdC203 also interacts with an initiation factor, eIF5B. Mutation of an initiation factor’s signature on the exterior of YrdC203 perturbed its interaction with RPL12 and eIF5B, and inhibited the increase in protein synthesis observed with overexpression of YrdC203. This implicates YrdC203 as a translation factor responsible for ensuring protein synthesis in vascular SMC in response to injury. The present work provides evidence for new molecular mechanisms, transcriptional and translational, regulating the response of vascular SMC to injury. This would provide leads for future therapeutic targets.

Neointimal proliferation is a key process underlying many cardiovascular diseases such as atherosclerosis and angioplasty-induced restenosis. Vascular smooth muscle cells (SMC) are significant contributors to the development and stability of the neointimal lesion. This is due, in part, to their capacity to be phenotypically modulated, facilitating SMC proliferation in response to mechanical injury, their subsequent migration, and deposition of extracellular matrix. The aim of this thesis was to characterize the function of two genes identified in our laboratory to be upregulated shortly after mechanical injury of vascular SMC and their exposure to fibroblast growth factor (FGF)-2, an injury-induced cytokine. The first is activation transcription factor (ATF) 4, which is upregulated by FGF-2 and mechanical injury in vascular SMC in vitro, and by balloon-injury in the artery wall. The induction of ATF4 by FGF-2 was shown to be mediated through the PI3K pathway, and preceded by phoshorylation of eIF2alpha, a known upstream effector of ATF4 activation. Knock-down of ATF4 expression inhibited balloon-injury induced neointimal hyperplasia, suggesting that ATF4 is a key player in the SMC response to injury. Furthermore, microarray analysis identified several genes whose transcription in response to FGF-2 may be regulated by ATF4. In particular, this work demonstrates that ATF4 is necessary for VEGF-A upregulation in SMC in response to FGF-2 and mechanical injury in vitro and in the artery wall following balloon-injury. The second is a translation factor, YrdC203. Using confocal fluorescence microscopy, YrdC203 was found to localize partially to the ER, and with RPL12, a component of the 60S ribosomal subunit. Immunoprecipitation studies demonstrate that YrdC203 also interacts with an initiation factor, eIF5B. Mutation of an initiation factor’s signature on the exterior of YrdC203 perturbed its interaction with RPL12 and eIF5B, and inhibited the increase in protein synthesis observed with overexpression of YrdC203. This implicates YrdC203 as a translation factor responsible for ensuring protein synthesis in vascular SMC in response to injury. The present work provides evidence for new molecular mechanisms, transcriptional and translational, regulating the response of vascular SMC to injury. This would provide leads for future therapeutic targets.

Chez les mammifères, les carences nutritionnelles telles que les carences en acides aminés constituent un stress nutritionnel important. Pour faire face à ces situations, l'organisme dispose de processus adaptatifs tels que l'autophagie, régulés par de multiples voies de signalisation. Au niveau cellulaire, plusieurs voies de signalisation sont impliquées dans la régulation de ces processus adaptatifs qui permettent la survie cellulaire lors de différentes conditions de stress environnementaux y compris la carence en acides aminés. En particulie,r la voie eIF2α/ATF4 joue un rôle crucial dans l'adaptation des cellules à ces différents stress notamment en régulant la transcription de nombreux gènes cibles spécifiques. L'objectif de ce travail était donc de déterminer le rôle de la voie eIF2α/ATF4 dans la régulation de la transcription des gènes impliqués dans l'autophagie en réponse à carence en acides aminés. En utilisant p62 comme un modèle de travail, nous avons montré que la kinase GCN2 qui phosphoryle eIF2α et les facteurs de transcription ATF4 et CHOP jouent un rôle clé dans la régulation de la transcription d'un grand nombre de gènes impliqués dans le processus autophagique en réponse à une carence en acides aminés. Nous avons en particulier identifié 3 classes de gènes de l'autophagie selon leur dépendance à ATF4 et CHOP et la liaison de ces facteurs sur les éléments spécifiques de leurs promoteurs en fonction de l'intensité du stress. PLus généralement, nous avons démontré que ce mécanisme pouvait également être activé par la kinase PERK lors d'un stress du réticulum endoplasmique. Enfin, nous avons pu montrer que durant les 6 premières heures de la carence en acides aminés, la voie eIF2α/ATF4/CHOP n'est pas impliquée dans la diminition de la viabilité cellulaire. Cependant, lorsque la carence en acides aminés est prolongée (16-48h), CHOP joue un rôle clé dans la régulation de l'apoptose et dans la répression du processus autophagique en contrôlant la transcription des gènes cibles spécifiques. Ainsi, ce travail a permis de mettre en évidence qu'en cas de carence en acides aminés, la voie eIF2α/ATF4/CHOP joue un rôle clé dans le devenir de la cellule. En fonction de la durée et de l'intensité du stress, la régulation très coordonnée de ces mécanismes moléculaires va permettre successivement la survie de la cellule et ensuite l'apoptose. / In mammals nutritional deficiencies such as amino acid limitation are an important nutritional stress. To deal with these situations, the body has adaptive processes such as autophagy regulated by multiple signaling pathways. At the cellular level, several signaling pathways are involved in the regulation of these adaptive processes that allow cell survival in different conditions of environmental stress, including amino acid deficiency. In particular, the eIF2α/ATF4 pathways plays a crucial role in the adaptation of these cells to various stresses such as the transcriptional regulation of many specific target genes. The aim of this work was to identify the role of the eIF2α/ATF4 pathway in the stress-regulated transcription of mammalian autophagy genes. Using p62 as a working model, we have shown that the GCN2 eIF2α-kinase and ATF4 and CHOP transcription factors are required to increase transcription of a set of autophagy genes implicated in the formation, elongation and function of the autophagosome. We also identify 3 classes of autophagy genes according to their dependence on ATF4 and CHOP and the binding of these factors to specific promoter cis elements. Furthermore, different combinations of CHOP and ATF4 bindings to target promoters allow the trigger of a differential transcriptional response according to the stress intensity. Furthermore, we have demonstrated that the same mechanism can also be activated by ER stress through PERK eIF2α-kinase activation. We also show that during the first 6h of starvation, CHOP up-regulates a number of autophagy genes while cell viability is not affected. By contrast, when the amino acid starvation is prolonged (16-48h), we demonstrated that CHOP has a dual role in both limiting autophagy and inducing apoptosis through the transcriptional activation of specific target genes. Thus, this work establishes that following amino acid starvation, the eIF2α/ATF4 pathway plays a key role in the cell fate. Depnding on the duration and intensity of the stress, the highly coordinated regulation of these molecular mechanisms sequentially will allow the survival of the cell and subsequently apoptosis.

Le développement anarchique de tumeurs peut conduire à la formation de régions hypoxiques carencées en nutriments. A cause de l’instabilité génétique des cellules cancéreuses, certaines peuvent développer des résistances à ces conditions proapoptotiques. Parmi ces mécanismes de résistance, ceux impliqués dans la survie des cellules face à la carence prolongée en acides aminés n’ont pas été élucidés. La carence en acides aminés est particulièrement délétère pour les cellules de par leur importance dans le métabolisme cellulaire et l’impossibilité pour 9 d’entre eux d’être synthétisée de novo. L’objectif de ce travail était de déterminer les mécanismes permettant aux cellules cancéreuses de survivre face à la carence prolongée en acides aminés. Pour cela, nous avons développé un outil cellulaire par génétique fonctionnelle. Des fibroblastes embryonnaires de souris ont été soumis à une pression de sélection en les cultivant dans un milieu fortement carencé en acides aminés pendant plusieurs mois. Des clones capables de survivre et de proliférer dans ce milieu ont été générés. Le travail de thèse a d’abord consisté à caractériser la résistance de plusieurs clones à la carence en acides aminés et notamment, les voies de signalisation régulées par cette situation. Parmi ces dernières, nous avons pu observer que les clones par rapport aux cellules parentales présentaient une altération de la voie GCN2, et plus précisément, une sous-expression d’un des acteurs majeurs de cette voie, la protéine ATF4. Par des expériences de shRNA, nous avons pu mettre en évidence que cette sous-expression favorisait la survie des clones en milieu carencé en acides aminés. L’association entre survie et sous-expression d’ATF4 a également pu être faite dans une lignée de cancer du pancréas, BxPC-3. En rétablissant l’expression d’ATF4, les cellules résistantes redevenaient sensibles aux effets apoptotiques de la carence en acides aminés, confirmant le rôle apoptotique d’ATF4 lors de carence en acides aminés prolongée. Nous nous sommes également intéressés aux mécanismes que les cellules utilisaient pour se fournir en acides aminés à partir des protéines du milieu extracellulaire. Nous avons pu identifier qu’une des lignées de clones présentait un fort niveau de macropinocytose et que son inhibition lors de carence en acides aminés diminuait ses capacités de survie lors de carences en acides aminés. Notre approche a ainsi pu mettre en évidence des mécanismes de résistances des cellules à la carence en acides aminés. Ces mécanismes ont ensuite pu être observés dans des lignées cancéreuses, confirmant la pertinence de notre modèle pour l’identification de tels mécanismes. / Tumor development can be characterized by the formation of hypoxic area deprived in nutrients. Due to their genetic instability, tumor cells can become resistant to this apoptotic condition. Among the resistance mechanisms, those involved in cell survival against amino acids restriction is poorly known. Amino acids deprivation is particularly deleterious as nine of them cannot be synthetized de novo. The aim of this work was to identify the mechanisms by which tumor cells can survive to long-term amino acids deprivation. For that purpose, we exerted a selective pressure on mouse embryonic fibroblast by exposing them to amino acids deprivation for several months. We generated clones able to survive and proliferate in deprived amino acids medium. Firstly, we characterized the capacity of proliferation and survival of several resistant clones in amino acid deprivation condition. Secondly, we studied several signaling pathways which are regulated during this condition. We have found that every clones present an alteration of the GCN2 pathway, and notably, a low expression of its major actor, the protein ATF4, when compared to the parental cells. We have shown with shRNA experiment that this underexpression promotes cell survival during amino acid deprivation. The association between decreased expression of ATF4 and better survival in amino acid deprivation condition was also found in a pancreatic cancer cell line, BxPC-3. ATF4 overexpression in this resistant cell line restores the apoptotic effect of ATF4 during amino acid deprivation. We have also studied how cells can provide themselves in amino acids from the extracellular environment. We have observed that one of the clone cell line presents a high level of macropinocytosis and that, the inhibition of this mechanism decreases its survival during amino acid deprivation. Thanks to our approach, we were able to highlight mechanisms which confers to cell resistance to amino acid deprivation. By observing those mechanisms in cancer cell lines, we confirmed the relevance of our approach to identify such mechanisms.

Les leucémies aiguës myéloïdes (LAM) sont des hémopathies malignes caractérisées par une accumulation dans la moelle et le sang de progéniteurs hématopoïétiques bloqués dans un stade différenciation. La mutation FLT3-ITD, qui entraîne une activation constitutive du récepteur à activité tyrosine kinase FLT3, est retrouvée dans 20-25% des LAM et est associée à un mauvais pronostique. De nombreux inhibiteurs de FLT3 ont été développés et certains sont testés en clinique mais des études mettent en évidence l'apparition de résistance. Une meilleure compréhension des mécanismes oncogéniques de FLT3-ITD est donc nécessaire afin d'améliorer le traitement des LAM. Mes travaux de thèse ont été centrés sur l'analyse du processus autophagique qui correspond à l'un des mécanismes de résistance décrits dans les cellules cancéreuses en réponse aux traitements. Au cours de cette étude, nous avons constaté que l'expression de FLT3-ITD augmente l'autophagie basale des cellules de LAM, et que l'inhibition du récepteur réduit cette autophagie dans des échantillons primaires de LAM et dans des lignées cellulaires. Nous avons pu montrer que l'autophagie est requise pour la prolifération et la survie in vitro et in vivo des cellules de LAM et que sont ciblage permet de surmonté la résistance aux inhibiteurs de FLT3. De plus, nous avons identifié la protéine ATF4 comme un acteur essentiel au processus d'autophagie en aval de FLT3-ITD. Ces résultats suggèrent que le ciblage de l'autophagie ou d'ATF4 chez les patients exprimant les mutations de FLT3 peut représenter une stratégie thérapeutique prometteuse et innovatrice dans les LAM. / Acute myeloid leukemias (AMLs) are a family of hematological malignancies characterized by an accumulation in the marrow and blood of hematopoietic progenitors blocked in their differentiation process. The FLT3-ITD mutation is found in 20-25% of AMLs and is associated with a poor prognosis. Different FLT3 inhibitors have been developed and some of them are clinically tested but resistance to treatment has been observed in many patients. A better understanding of AML biology is necessary in order to improve the treatment of AMLs. My thesis project focused on the analysis of the autophagic process, which is one of the mechanisms described in the resistance of cancer cells. In this study, we found that the FLT3-ITD expression increases basal autophagy in AML cells, and that the receptor inhibition reduced this autophagy in primary patient samples and cell lines. We show that autophagy is required for proliferation and survival in vitro and in vivo of leukemic cells lines and inhibition of autophagy overcomes resistance to FLT3 inhibitors. In addition, we identified the ATF4 protein as a key actor of the autophagy process induced by the FLT3-ITD mutation. These results suggest that targeting autophagy or ATF4 may represent a promising and innovative therapeutic strategy for FLT3 mutated AMLs.

Skeletal muscle atrophy is a common, debilitating consequence of muscle disuse, malnutrition, critical illness, musculoskeletal conditions, neurological disease, cancer, and organ failure. Despite its prevalence, little is known about the molecular pathogenesis of this devastating condition due in large part to an incomplete understanding of the molecular mechanisms that drive the atrophy process. In previous studies, we identified the transcription factor ATF4 as a critical mediator of skeletal muscle atrophy. We found that ATF4 is necessary and sufficient for skeletal muscle atrophy during limb immobilization. However, ATF4 mKO mice were only partially protected from skeletal muscle atrophy during limb immobilization, indicating the existence of another pro-atrophy factor that acts independently of the ATF4 pathway. Using mouse models, we identify p53 as this ATF4-independent factor. We show that skeletal muscle atrophy increases p53 expression in skeletal muscle fibers. In addition, overexpression of p53 causes skeletal muscle atrophy. Further, p53 mKO mice are partially resistant to muscle atrophy during limb immobilization. Taken together, these data indicate that like ATF4, p53 is sufficient and required for skeletal muscle atrophy during limb immobilization. Importantly, overexpression of p53 induces muscle atrophy in the absence of ATF4, whereas ATF4-mediated muscle atrophy does not require p53. Furthermore, overexpression of p53 and ATF4 induces greater muscle atrophy than p53 or ATF4 alone. Moreover, skeletal muscle lacking both p53 and ATF4 is more resistant to skeletal muscle atrophy than muscle lacking either p53 or ATF4 alone. Taken together, these data indicate that p53 and ATF4 mediate distinct and additive mechanisms to skeletal muscle atrophy. However, the precise mechanism by which p53 and ATF4 cause skeletal muscle atrophy remained unclear. Using genome-wide expression arrays, we identify p21 as a skeletal muscle mRNA that is highly induced by p53 and ATF4 during limb immobilization. Further, overexpression of p21 causes skeletal muscle atrophy. In addition, p21 is required for muscle atrophy due to limb immobilization, p53, and ATF4. Collectively, these results identify p53 and ATF4 as critical and complementary mediators of skeletal muscle atrophy during limb immobilization, and discover p21 as an essential downstream mediator of the p53 and ATF4 pathways.

Skeletal muscle atrophy is a common and often debilitating complication of diverse stresses including muscle disuse, fasting, aging, critical illness and many chronic illnesses. However, the pathogenesis of muscle atrophy is still poorly understood. The thesis herein describes my studies investigating the molecular mechanisms of skeletal muscle atrophy. Using mouse skeletal muscle and cultured skeletal myotubes as experimental systems, I discovered a novel stress-induced pathway in skeletal muscle that causes muscle atrophy. The pathway begins with stress-induced expression of ATF4, a basic leucine zipper (bZIP) transcription factor with an evolutionarily ancient role in cellular stress responses. I found that diverse stresses including fasting and muscle disuse increase expression of ATF4 in skeletal muscle. ATF4 then activates the growth arrest and DNA damage-inducible 45a (Gadd45a) gene, leading to increased expression of Gadd45a protein, an essential and inducible subunit of DNA demethylase complexes. Gadd45a localizes to skeletal myonuclei where it interacts with and stimulates demethylation of a specific region in the promoter of the cyclin dependent kinase inhibitor 1a (Cdkn1a) gene. By demethylating the Cdkn1a promoter, Gadd45a activates the Cdkn1a gene, leading to increased expression of Cdkn1a protein, also known as p21WAF1/CIP1. Cdkn1a stimulates protein breakdown (a critical pro-atrophy process) and inhibits anabolic signaling, protein synthesis and PGC-1α expression (processes that maintain healthy skeletal muscle and protect against atrophy). As a result, Cdkn1a causes skeletal muscle fibers to undergo atrophy. Importantly, interventions that reduce any one component of this pathway (ATF4, Gadd45a or Cdkn1a) reduce skeletal muscle atrophy during fasting, muscle disuse, and perhaps other skeletal muscle stresses such as illness and aging. Conversely, forced expression of any one component of this pathway is sufficient to cause skeletal muscle fiber atrophy in the absence of upstream stress. These data suggest the ATF4/Gadd45a/Cdkn1a pathway as a potential therapeutic target. Collectively, my studies demonstrate that the sequential, stress-induced expression of ATF4, Gadd45a and Cdkn1a is a critical process in the pathogenesis of skeletal muscle atrophy. This significantly advances our understanding of how muscle atrophy occurs and it opens up new avenues of investigation into the causes and treatment of muscle atrophy.

An increasing body of evidence points to a key role of endoplasmic reticulum (ER) stress in chronic and acute neurodegenerative diseases. Indeed, markers of ER stress are common features of neurons destined to die in these conditions. In the present study we demonstrate that PUMA, a BH3-only member of the Bcl-2 family is essential for ER stress-induced cell death. PUMA is known to be a key transcriptional target of p53, however we have found that ER stress triggers PUMA induction and cell death through a p53-independent mechanism involving instead the ER stress inducible transcription factor ATF4. Specifically, we demonstrate that ectopic expression of ATF4 sensitizes neurons to ER stress induced apoptosis, and that ATF4-deficient neurons exhibit markedly reduced levels of PUMA expression and cell death. However, chromatin immunoprecipitation experiments suggest that ATF4 does not directly regulate the PUMA promoter. Rather, we found that ATF4 induces expression of the transcription factor CHOP, and that CHOP in turn directly activates PUMA induction. Specifically, we demonstrate that CHOP binds to the PUMA promoter during ER stress and that CHOP knockdown attenuates PUMA induction and neuronal apoptosis. In summary, we have identified a key signaling pathway in ER stress induced neuronal death involving ATF4-CHOP mediated transactivation of the pro-apoptotic Bcl-2 family member PUMA. Protein aggregates and markers of ER stress response have also been observed in dying neurons in several animal models of cerebral ischemia. Therefore, to decipher the significance of the ER stress apoptotic response, we investigate the role of ATF4-CHOP signaling pathway in ischemic neuronal injury. Ischemic stroke results from a transient or permanent reduction in cerebral blood flow in the brain. In spite of much research in trying to develop therapeutic strategies, most clinical trials have failed. These failures demonstrate that effective treatments require a more complete understanding of molecular signals that lead to neuronal death. However, stroke is a complex scenario since distinct mechanisms may involve in rapid and/or delayed neuronal death. The signaling pathways regulating these mechanisms however are not fully defined. Previous studies had suggested that ER stress playing a pivotal role in post-ischemic neuronal death. Yet, the relevance of ER stress signals was not fully known in ischemic neuronal injury. Accordingly, this thesis research attempts to explore the functional role of ER stress -inducible pathway, ATF4-CHOP axis, in different models of neuronal death (delayed and excitotoxic cell death) evoked by ischemia. The data indicates that ATF4 is essential in delayed type of death in vitro. In focal ischemia model (tMCAO) ATF4 also plays a role as a mediator of death signal in vivo. However, CHOP function looks more complex, and our data did not support the role of CHOP in ischemic neuronal death.

An increasing body of evidence points to a key role of endoplasmic reticulum (ER) stress in chronic and acute neurodegenerative diseases. Indeed, markers of ER stress are common features of neurons destined to die in these conditions. In the present study we demonstrate that PUMA, a BH3-only member of the Bcl-2 family is essential for ER stress-induced cell death. PUMA is known to be a key transcriptional target of p53, however we have found that ER stress triggers PUMA induction and cell death through a p53-independent mechanism involving instead the ER stress inducible transcription factor ATF4. Specifically, we demonstrate that ectopic expression of ATF4 sensitizes neurons to ER stress induced apoptosis, and that ATF4-deficient neurons exhibit markedly reduced levels of PUMA expression and cell death. However, chromatin immunoprecipitation experiments suggest that ATF4 does not directly regulate the PUMA promoter. Rather, we found that ATF4 induces expression of the transcription factor CHOP, and that CHOP in turn directly activates PUMA induction. Specifically, we demonstrate that CHOP binds to the PUMA promoter during ER stress and that CHOP knockdown attenuates PUMA induction and neuronal apoptosis. In summary, we have identified a key signaling pathway in ER stress induced neuronal death involving ATF4-CHOP mediated transactivation of the pro-apoptotic Bcl-2 family member PUMA. Protein aggregates and markers of ER stress response have also been observed in dying neurons in several animal models of cerebral ischemia. Therefore, to decipher the significance of the ER stress apoptotic response, we investigate the role of ATF4-CHOP signaling pathway in ischemic neuronal injury. Ischemic stroke results from a transient or permanent reduction in cerebral blood flow in the brain. In spite of much research in trying to develop therapeutic strategies, most clinical trials have failed. These failures demonstrate that effective treatments require a more complete understanding of molecular signals that lead to neuronal death. However, stroke is a complex scenario since distinct mechanisms may involve in rapid and/or delayed neuronal death. The signaling pathways regulating these mechanisms however are not fully defined. Previous studies had suggested that ER stress playing a pivotal role in post-ischemic neuronal death. Yet, the relevance of ER stress signals was not fully known in ischemic neuronal injury. Accordingly, this thesis research attempts to explore the functional role of ER stress -inducible pathway, ATF4-CHOP axis, in different models of neuronal death (delayed and excitotoxic cell death) evoked by ischemia. The data indicates that ATF4 is essential in delayed type of death in vitro. In focal ischemia model (tMCAO) ATF4 also plays a role as a mediator of death signal in vivo. However, CHOP function looks more complex, and our data did not support the role of CHOP in ischemic neuronal death.

Les leucémies aiguës myéloïdes (LAM) représentent un groupe d’hémopathies malignes agressives, de pronostic sombre en dépit des traitements intensifs actuellement proposés. Malgré une grande hétérogénéité clinique et moléculaire, les cellules de LAM sont caractérisées par l’activation de voies de signalisation essentielles à leur prolifération et leur survie, comme par exemple celle du complexe mTORC1 (mammalian target of rapamycin complex 1). Cependant, l’utilisation clinique d’inhibiteurs tels que la rapamycine ou des inhibiteurs catalytiques s’est avérée décevante, ce qui suggère qu’il n’y a pas d’addiction oncogénique à mTORC1 dans les LAM. Au cours de ce travail, nous avons démontré que l’activation de mTORC1 est au contraire une condition nécessaire à l’induction de la mort cellulaire en réponse à l’activation d’AMPK (AMP-activated protein kinase), établissant une relation de létalité synthétique entre ces deux voies. Pour cela, nous avons utilisé un nouveau composé activateur spécifique d’AMPK, le GSK621. En invalidant la sous-unité catalytique AMPKα1 par ARN interférence ou par le système CRISPR/Cas9, nous avons démontré que les effets antileucémiques de ce composé sont bien dépendants de l’activation d’AMPK. Nous avons observé que ce composé favorise l’autophagie, et que ce processus est impliqué dans la mort des cellules leucémiques puisque l’inhibition des protéines ATG5 ou ATG7 a un effet protecteur sur les cellules leucémiques. Les effets antileucémiques du composé GSK621 ont été confirmés sur des cellules primaires, ainsi que sur un panel de vingt lignées de LAM, et dans un modèle murin de xénogreffe. De façon intéressante, l’activation d’AMPK pourrait également compromettre la survie des cellules souches leucémiques, comme en atteste l’atténuation du potentiel clonogénique en méthylcellulose de cellules murines transformées par MLL-ENL ou FLT3-ITD. Nous avons observé que le composé GSK 621 n’avait pas de toxicité envers les progéniteurs hématopoïétiques normaux, ouvrant ainsi une fenêtre thérapeutique intéressante. Comme l’activation d’AMPK conduit dans de nombreux modèles cellulaires à l’inhibition de mTORC1, et comme l’activation de mTORC1 est observée dans les cellules de LAM mais pas dans les progéniteurs hématopoïétiques normaux, nous avons proposé l’hypothèse que le niveau d’activation de mTORC1 déterminait les effets de l’activateur d’AMPK. Pour cela, nous avons inhibé mTORC1 dans les cellules leucémiques d’une part, et activé mTORC1 dans les progéniteurs normaux d’autre part. De façon inattendue, mTORC1 échappe au contrôle d’AMPK dans les LAM, et nous avons observé que l’activation de mTORC1 est une condition nécessaire et suffisante pour que le composé GSK621 entraîne la mort des cellules. Le substrat moléculaire de cette létalité synthétique est le facteur de transcription proapoptotique ATF4, dont la transcription est favorisée par mTORC1, et la traduction par AMPK via la phosphorylation d’eIF2A. Ces travaux proposent donc que malgré l’absence d’addiction oncogénique, l’activation de mTORC1 dans les LAM représente une opportunité thérapeutique originale via une relation de létalité synthétique avec l’activation d’AMPK. Ils constituent un rationnel au développement clinique d’activateurs d’AMPK dans les LAM, voire dans d’autres cancers ayant une activation constitutive de mTORC1. / Acute myeloid leukemia (AML) is a heterogeneous disease with poor prognosis despite intensive treatments. Virtually all recurrent molecular alterations in AML functionally converge to cause signal transduction pathway dysregulation that drives cellular proliferation and survival. The mammalian target of rapamycin complex 1 (mTORC1) is a rapamycin-sensitive signaling node defined by the interaction between mTOR and raptor. Constitutive mTORC1 activity is nearly universal in AML. However, pharmacologic inhibition with rapamycin or second-generation mTOR kinase inhibitors has shown limited anti-leukemic activity in both preclinical models as well as in clinical trials, suggesting that addiction to this oncogene is not a recurrent event in AML. Here we report that sustained mTORC1 activity is nonetheless essential for the cytotoxicity induced by pharmacologic activation of AMP-activated protein kinase (AMPK) in AML. Our studies employed a novel AMPK activator called GSK621. Using CRISPR/Cas9 and shRNA-mediated silencing of the AMPKa1 catalytic subunit, we showed that AMPK activity was necessary for the anti-leukemic response induced by this agent. GSK621-induced AMPK activation precipitated autophagy, and blocking autophagy via shRNA-mediated knockdown of ATG5 or ATG7 protected AML cells from cytotoxicity resulting from treatment with GSK621, suggesting that autophagy promotes cell death in the context of active AMPK. GSK621 cytotoxicity was consistently observed across twenty different AML cell lines, primary AML patient samples and AML xenografts in vivo. GSK621-induced AMPK activation also impaired the self-renewal capacity of MLL-ENL- and FLT3-ITD-induced murine leukemias as measured by serial methylcellulose replating assays. Strikingly, GSK621 did not induce cytotoxicity in normal CD34+ hematopoietic progenitor cells. We hypothesized that the differential sensitivity to GSK621 could be due to the difference in amplitude of mTORC1 activation between AML and normal CD34+ cells. In contrast to most reported cellular models in which AMPK inhibits mTORC1, sustained mTORC1 activity was seen following GSK621-induced AMPK activation in AML. Inhibition of mTORC1 either pharmacologically (using rapamycin) or genetically (using shRNAs targeting raptor and mTOR) abrogated AMPK-induced cytotoxicity in AML cells, including primary AML patient samples. The same synthetic lethality could be recapitulated in normal CD34+ progenitors by constitutive activation of mTORC1 using a lentivirally-transduced myrAKT construct. We further observed that the level of ATF4 protein is under a transcriptionnal control by mTORC1 and a translational control by AMPK (through eIF2A), and explains the synthetic lethal relationship between AMPK and mTORC1. Taken together, these data show that the magnitude of mTORC1 activity determines the degree of cytotoxicity triggered by AMPK activation. Our results therefore support AMPK activation as a promising therapeutic strategy in AML and other mTORC1-active malignancies which warrants further investigations in clinical trials.