Objectifs: La cytokine TRAIL (TNF Related Apoptosis Inducing Ligand) suscite un intérêt majeur en thérapie anti-cancéreuse grâce à sa capacité à induire l’apoptose des cellules cancéreuses tout en épargnant les cellules saines. L’association du récepteur Fas et de l’actine via l’ezrine, une protéine de la famille ERM (Ezrin, Moesin, Radixin), régule les premières étapes de l’induction de l’apoptose par FasL. Au cours de mon projet de thèse, nous avons voulu déterminer le rôle que pouvait jouer l’ezrine au cours de l’apoptose induite par TRAIL, dans des lymphomes B ou des cellules cancéreuses adhérentes (HeLa, HCT116 et SW480). Matériel et Méthodes: Des approches biochimiques et moléculaires nous ont permis d’étudier et de déterminer l’implication de l’ezrine et sa phosphorylation dans la régulation de la mort induite par TRAIL. Résultats: Ce travail démontre que l’ezrine peut réguler de manière négative l’apoptose induite par TRAIL et FasL. Cette activité inhibitrice est régulée par la phosphorylation/déphosphorylation sur la serine 66 ainsi que sur la thréonine 353. Néanmoins cette régulation n’affecte ni la formation, ni l’activation du DISC (Death Inducing Signalling Complex). Des mutations de ces résidus par une alanine (S66A) ou un acide aspartique (Y353D) augmente sélectivement la capacité de TRAIL à induire l’apoptose. Au contraire, des mutations ponctuelles de ces résidus permettant de mimer la phosphorylation de l’ezrine sur la serine 66 (S66D) ou l’expression d’un variant non phosphorylable sur la thréonine 353 (Y353F) protègent les cellules cancéreuses de l’apoptose induite par TRAIL. De manière concordante, l’utilisation du H89, un inhibiteur de PKA, kinase responsable de la phosphorylation de la serine 66 augmente la sensibilité des cellules cancéreuses à TRAIL, alors qu’au contraire, un activateur de PKA (8bromocyclic AMP) rend ces mêmes cellules plus résistantes à TRAIL. Enfin, l’association de TRAIL et du cisplatine permet de dépasser l’inhibition de l’apoptose par l’ezrine. / Background and Aim: TRAIL has sparked a growing interest in oncology due to its ability to selectively trigger cancer cell death while sparing normal cells. The Fas/actin association through ezrin, a member of the ERM protein family, has been reported to regulate early steps of Fas-mediated apoptosis. In this project, we addressed the role of ezrin regarding TRAIL-induced cell death in B lymphoma cell lines, or adherent cancer cell lines (HeLa WT, HCT116, SW480). Methods: Molecular and biochemical approaches were employed to study the relevance of ezrin and its phosphorylation status in TRAIL signaling. Results: We found that ezrin displays a negative function towards TRAIL- and Fas-mediated apoptosis and that the ezrin-mediated TRAIL-induced cell death inhibition led to ezrin activation through phosphorylation/dephosphorylation events at serine 66 and tyrosine 353, but is mainly independent of TRAIL DISC (Death Inducing Signalling Complex) formation or activation. Mutations of these residues to alanine (S66A) or aspartic acid (Y353D) selectively enhanced TRAIL-induced cell death, whereas point mutations mimicking ezrin phosphorylation on S66 (S66D) or a nonphosphorylable variant on Y353 (Y353F) strongly protected cancer cells from apoptosis induced by TRAIL. Moreover, inhibition of the ezrin serine 66 PKA target site, using H89, increased cancer cell sensitivity to TRAIL, while treatment with 8bromocyclic AMP, a PKA activator, decreased TRAIL-induced cell death. In addition, combined TRAIL/cisplatin treatments abrogated ezrin-mediated inhibition of TRAIL-induced apoptosis.

Les ATPases-P4 sont des transporteurs membranaires couplant l'hydrolyse de l'ATP au transport de lipides dans les membranes cellulaires eucaryotes. Avec leurs partenaires, les protéines CDC50, les ATPases-P4 transportent les phospholipides, en particulier la phosphatidylsérine (PS) et la phosphatidyléthanolamine (PE), du feuillet exoplasmique au feuillet cytosolique des membranes, assurant ainsi le maintien de l'asymétrie membranaire.Drs2p est l'une des cinq ATPases-P4 de la levure Saccharomyces cerevisiae. Elle est localisée dans les membranes du trans-Golgi (TGN), et elle a comme partenaire la protéine Cdc50p, qui est nécessaire à l'adressage correct et probablement au transport catalysé par Drs2p. Drs2p est principalement responsable du transport de la phosphatidylsérine (PS) dans les membranes du TGN et son activité est essentielle pour le maintien de la PS dans le feuillet cytosolique de ces membranes. En raison du rôle crucial de la PS dans de nombreuses voies de signalisation, aussi bien à l’extérieur (au cours de l’apoptose par exemple) qu’à l’intérieur de la cellule (par le recrutement de protéines impliquées dans des processus cellulaires essentiels), il est important de comprendre le mécanisme par lequel l’asymétrie de la PS est établie.Afin de progresser dans la compréhension du mécanisme moléculaire du transport de lipides, nous avons mis au point une procédure qui nous a permis de co-exprimer Drs2p et Cdc50p dans Saccharomyces cerevisiae. La purification de Drs2p par chromatographie d'affinité sur résine streptavidine a permis d'obtenir une fraction purifiée contenant très majoritairement Drs2p et Cdc50p, à raison de 1-2 mg/L de culture. Les deux protéines sont sous forme de complexe avec une stœchiométrie d'association de 1:1. Le complexe purifié est fonctionnel, et présente une activité d’hydrolyse de l’ATP stimulée par son substrat, la PS. Cette stimulation n’est cependant possible qu’en présence de PI4P, un phosphoinositide impliqué dans la régulation du trafic membranaire.De par leur rôle crucial dans le maintien de l'asymétrie membranaire, les ATPases-P4 ne peuvent qu'être régulées. Comme de nombreuses ATPases de type P sont soumises à une auto-régulation de leur activité, nous avons examiné la possibilité d’une telle auto-régulation dans le cas des ATPases P4. Pour ce faire, une approche par mutagenèse dirigée et protéolyse ménagée associée à l’identification par spectrométrie de masse des peptides ont été effectuées. La protéolyse ménagée du complexe purifié Drs2p/Cdc50p montre une activité ATPasique dépendante au PI4P de 30-50 fois plus importante. La protéolyse par la thrombine engendre un Drs2p dépourvu d'une partie N-terminale (R104) et d'une partie C-terminale (R1290) qui reste toujours associé à Cdc50p. Ce résultat montre qu'une coupure appropriée au niveau des extrémités terminales de Drs2p peut augmenter de façon significative, en présence du PI4P, l'activité ATPasique du complexe, nous amenant ainsi à identifier un rôle auto-inhibiteur des extrémités N- et/ou C-terminales de Drs2p.Ce travail ouvre des perspectives quant à la caractérisation structurale et fonctionnelle du mécanisme de transport de lipides par le complexe. Par ailleurs, il laisse entrevoir la possibilité d’étudier les bases moléculaires des pathologies associées aux mutations de certaines ATPases P4 humaines. / Maintenance of phospholipid asymmetry in eukaryotic cell membranes is essential for cellular integrity and function. P4-ATPases, from the P-type ATPases family, are energy-dependent transporters, together with their CDC50 accessory subunits couple ATP hydrolysis to lipid transport from the exoplasmic to cytoplasmic leaflet to maintain membrane asymmetry.Drs2p is one of these P4-ATPases in the yeast Saccharomyces cerevisiae. Drs2p is localised in trans-Golgi (TGN) membranes in association with its binding partner Cdc50p, which contributes to the correct addressing of Drs2p and probably in the catalyzed transport by Drs2p. Drs2p transport principally phosphatidylserine (PS) in TGN membranes. The PS is important for a several signalling pathways, for example, in apoptosis and recruitment of the proteins implied in various essential cellular process, so, it's very important to understand the mechanism that establishes this asymmetry.To gain in comprehension of molecular mechanism of lipid transport, robust protocols for expression and purification are required. In this work, we present a procedure for high-yield co-expression of Drs2p and Cdc50p. The purification of Drs2p and Cdc50p is achieved in a single step by affinity chromatography on streptavidin beads, yielding, 1-2 mg purified Drs2p/Cdc50p per liter of culture. This procedure allows purification of the complex Drs2p/Cdc50p with stoichiometry to 1:1. Our complex is functional, overal ATP hydrolysis by the complex is dependent of PS, favourite substrate of Drs2p. This hydrolyze is critically dependent on the presence of PI4P, a phosphoinositide involved in regulation of membrane trafficking.Like many P-type ATPases auto-regulate their activity, we examined the possibility that P4-ATPases are auto-regulated. In this work, we use directed mutagenesis and limited proteolysis associated with mass spectrometry for identify peptides. We show that limited proteolysis of a purified complex Drs2p/Cdc50p resulted in up to a 30-50 fold increase of it ATPase activity, which however remained dependent on PI4P. Using thrombin as the protease, Cdc50p remained intact and in complex with Drs2p, which was cleaved at two positions, namely after R104 and after R1290. Our results therefore reveal that trimming off appropriate regions of the terminal extensions of Drs2p can increase its ATPase activity in the presence of PI4P by an enormous factor, thereby identifying a role of N and/or C-terminal extensions in auto-inhibition of Drs2p.Our results open perspectives on the structural and the functional characterization of the lipid transport mechanism by the complex Drs2p/Cdc50p. Furthermore, our procedures open up the possibility of studying the molecular bases of the pathologies associated with the mutations of human P4-ATPases.

La chromatine chez les eucaryotes, porte des informations génétiques et épigénétiques. Les mécanismes garantissant le maintien de ces informations lors de la division cellulaire ou la réparation de l’ADN sont encore mal connus et ils constituent l’enjeu principal du projet de thèse. Plus particulièrement, l’objectif du projet de thèse est de chercher à comprendre comment les chaperons d’histones coordonnent leur action avec des partenaires associés à la fourche de réplication pour conserver les marques épigénétiques portées par les histones parentales et les reporter sur les histones nouvellement synthétisées. Cette thèse décrit précisément comment ASF1 (Anti Silencing Function 1) coopère avec le complexe CAF-1 (Chromatin Assembly Factor 1) et la sous-unité de l’hélicase réplicative MCM2 (Mini Chromosome Maintenance 2), pour la prise en charge des H3-H4 dans la réplication et la réparation de l’ADN.La thèse s’intéresse également à la régulation de l’activité de ces chaperons d’histones par des kinases activées suite à des stress réplicatifs ou des dommages de l’ADN. En particulier nous avons cherché à mieux comprendre comment l’ajout de groupements phosphate sur ASF1 par une enzyme appelée TLK (Tousled Like Kinase) module son activité au cours du cycle cellulaire et en réponse aux dommages de l’ADN. La caractérisation de l'importance des sites phosphorylés sur les propriétés de liaison du chaperon, permet de mieux comprendre le rôle joué par différent forme d’ASF1 dans l’assemblage des histones sur l’ADN et le maintien des informations épigénétiques. Le travail de thèse contient d’analyses biochimiques et structurales par une combinaison de techniques (SEC-MALS, AUC, ITC, RMN, cristallographie des rayons X) et d’analyses fonctionnelles sur des modèles cellulaires. / In eukaryotes, chromatin carries both, the genetic and epigenetic information. Mechanisms implicated in maintenance of these information during cell division or DNA repair remain poorly understood and they constitute the main issue of this thesis project. More specifically, the goal of the project is to understand how histone chaperones coordinate their action with partners associated with the replication fork to recognize and preserve the epigenetic marks carried by parental histones and to copy on the newly synthesized histones. The work unravels how ASF1 (Anti-Silencing Function 1) cooperates with the CAF-1 complex (Chromatin Assembly Factor 1) and with the replicative helicase subunit MCM2 (Mini Chromosome Maintenance 2), for the management of H3-H4 histones in DNA replication and repair.Moreover, this thesis investigates the regulation of histone chaperones activities by kinases activated after a replicative stress or DNA damage. In particular, we analyzed the consequences of ASF1 phosphorylation by the enzyme called TLK (Tousled like kinase). The activity of TLK is modulated during the cell cycle and after DNA damage. Characterization of the importance of phosphorylated sites on the chaperone binding properties, allows a better understanding of the role played by different forms of ASF1 in the assembly of histones on DNA and maintenance of epigenetic information. The thesis work included biochemical and structural analysis with a combination of different techniques (SEC-MALS, AUC, ITC, NMR, X-ray crystallography) and functional analysis in cellular models.

La chimiothérapie constitue encore le traitement de référence pour la majorité des cancers. Or certains agents chimiothérapeutiques sont capables de déclencher des signaux de stress pre-mortem permettant d’activer une réponse immunitaire antitumorale et confèrent ainsi une protection à long terme. A l'aide d'un modèle construit par intelligence artificielle, nous avons identifié, parmi une librairie comprenant 50 000 composés, des agents anti-cancéreux qui, d'après leurs propriétés physico-chimiques, pourraient induire une mort cellulaire immunogène (ICD, de l'anglais "immunogenic cell death"). Cet algorithme nous a permis d'identifier la dactinomycine, qui, en effet, active les mécanismes sous-jacents à l'activation des cellules dendritiques in vitro et a un effet anti-cancéreux dépendant du système immunitaire in vivo. La dactinomycine, utilisée en clinique pour le traitement de sarcomes pédiatriques, est connue pour sa capacité à inhiber la transcription. Nous nous sommes donc demandé si d'autres inducteurs de l'ICD partageaient cette propriété. Différentes chimiothérapies immunogènes induisent en effet une inhibition de la synthèse d’ARN, qui est suivie d'une inhibition de la traduction et s’accompagne de l’activation des différentes voies de l’ICD. De plus, une étude rétrospective in silico révèle que les agents classés comme inhibiteurs de la synthèse d’ARN ou de protéines sont prédits comme étant immunogènes. Ces résultats montrent que l’inhibition de la transcription est un évènement précurseur essentiel à l’activation d’une mort cellulaire immunogène. / Chemotherapy still constitutes the standard treatment for most cancers. Yet, some chemotherapeutics are able to trigger pre-mortem stress signals which activate an antitumor immune response and thereby confer long term protection. We used an established model built on artificial intelligence to identify, among a library of 50,000 compounds, anticancer agents that, based on their physicochemical characteristics, were predicted to induce immunogenic cell death (ICD). This algorithm led us to the identification of dactinomycin, which indeed activates the mechanisms preceding dendritic cell activation in vitro and demonstrates immune-dependent anticancer effects in vivo. Dactinomycin, mainly used to treat pediatric sarcomas, is known as able to inhibit transcription. We therefore investigated whether other ICD inducers would share this characteristic. Different immunogenic chemotherapeutics indeed inhibited RNA synthesis and secondarily translation, accompanied by an activation of ICD-related signaling. A retrospective in silico study revealed that agents annotated as inhibitors of RNA or protein synthesis are predicted as immunogenic. These results establish the inhibition of RNA synthesis as a major initial event for ICD induction.

Pseudomonas aeruginosa PA14 est une bactérie pathogène très résistante aux antibiotiques et impliquée dans de nombreuses infections nosocomiales. Toutefois, la disponibilité d'agents antibactériens efficaces contre cette bactérie manque cruellement à ce jour. Explorer la physiologie de P. aeruginosa au niveau des modifications post-traductionnelles (PTMs) pourrait fortement contribuer au développement de nouveaux agents thérapeutiques. En effet, il a été montré certaines corrélations entre les PTMs et la virulence, l’adaptation et la résistance bactérienne. De plus, les progrès récents en protéomique ont permis d’accéder à un nombre croissant des protéines modifiées. Pourtant, leur description reste un véritable challenge.Dans une première partie, nous avons étudié l'impact des kinases et des phosphatases sur la physiologie de P. aeruginosa PA14. Cependant, aucune différence de phénotype n'a été observée entre les 8 mutants de ces enzymes et la souche sauvage.Dans une deuxième partie, nous avons caractérisé le succinylome et l'acétylome de la lysine chez P. aeruginosa PA14 dans quatre sources de carbone (glucose, citrate, succinate et glutamate) par enrichissement par anticorps couplé à la spectrométrie de masse. Ainsi, 1 530 sites succinylés (617 protéines) et 1 109 sites acétylés (526 protéines) ont été identifiés. De façon intéressante, 622 sites (312 protéines) ont été observés acétylés ou succinylés sur la même lysine, révélant ainsi l'existence de protéoformes pour une même protéine. Les protéines modifiées sont impliquées dans tous les processus biologiques. Toutefois, certaines d'entre elles ont des fonctions dans la résistance aux antibiotiques, le chimiotactisme et la virulence.Nous avons également quantifié les peptides succinylés et/ou acétylés dans les 4 sources de carbone. Les peptides succinylés étaient principalement sur-exprimés en citrate, mais aucune différence significative n’a été observée pour les peptides acétylées.Dans une troisième partie, nous avons étudié par immunoprécipitation le succinylome et l’acétylome de la lysine des protéines extracellulaire de P. aeruginosa. Nous avons montré que certaines lysines des protéines LasB et CbpD, deux facteurs de virulence, sont modifiées par 9 PTMs différentes. Une approche d’électrophorèse bi-dimensionnelle (2D) a permis de révéler et de quantifier les protéoformes des protéines extracellulaires et plus spécifiquement de ces facteurs de virulence.Dans une quatrième partie, une approche quantitative « label-free » a permis de mettre en avant 581 protéines qui varient différemment selon la source de carbone. Parmi ces protéines, 67 biomarqueurs ont été identifiés par approche statistique.Ces travaux constituent un point de départ prometteur pour de futures études sur le rôle de la succinylation de la lysine et d'autres PTMs chez P. aeruginosa. / Pseudomonas aeruginosa PA14 is a multi-drug resistant human pathogen largely involved in nosocomial infections. Unfortunately, today, effective antibacterial agents lacked. Explore its physiology at the post-translational modification (PTMs) level may contribute to the renewal of combat tactics. Indeed, some correlations between PTMs and the bacterial virulence, adaptation and resistance have been shown. The recent improvements in proteomics have increased the number of modified proteins. However, their characterization believes a real challenge.In the first part, we focused on the impact of kinases and phosphatases on bacterial physiology of P. aeruginosa PA14. For this purpose, we compared different phenotypes of 8 mutants of kinase and phosphatase with the WT strain. Unfortunately, no difference was observed.In the second part, we characterized the lysine succinylome and acetylome in P. aeruginosa PA14 in 4 carbon sources (glucose, citrate, succinate and glutamate) by mass spectrometry. Overall, a total of 1 530 succinylated sites (617 proteins) and 1 109 acetylated sites (526 proteins) were identified. Interestingly, we noticed that 622 sites (312 proteins) can be either acetylated or succinylated on the same lysine. This reveals the existence of proteoforms for a same protein. As expected, many modified proteins are involved in a wide range of biological processes but some of these proteins have interesting functions like antibiotic resistance, chemotaxis and virulence.We also tried to quantify succinylated and/or acetylated peptides in the 4 carbon sources. Succinylated peptides were mainly over-represented in the citrate condition whereas no significant difference was observed for the acetylated forms.In the third part, we investigated the lysine succinylome and acetylome of P. aeruginosa in extracellular compartment by immunoprecipitation. We showed that some lysines of two virulence factors, LasB and CbpD, were modified by 9 different PTMs. We also used a 2-dimensional gel approach to reveal and quantify proteoforms of extracellular proteins and more specifically virulence factors. In the fourth part, we did a label-free quantitative approach to obtain protein abundance in each carbon source. In total, 581 proteins vary differently depending on the carbon source. Among these proteins, 67 biomarkers were identified by statistical approach.This work is a promising starting point for further investigations on the biological role of lysine succinylation, and others PTMs, in P. aeruginosa.

Le facteur neurotrophique dérivé du cerveau (BDNF, Brain-Derived Neurotrophic Factor) est une protéine qui joue un rôle essentiel dans la survie et la différenciation des neurones, ainsi que dans l'induction et l'expression de la plasticité synaptique (Deinhardt and Chao, 2014; Lu et al., 2005). Le BDNF est très exprimé à l'âge adulte et la réduction de son expression est impliquée dans de nombreuses maladies neurodégénératives et troubles psychologiques (Anastasia and Hempstead, 2014). Son action sur la plasticité synaptique est majeure pour la mise en place de fonctions cognitives et pour la mise en place de la mémoire (Bekinschtein et al., 2008; Egan et al., 2003).Le BDNF existant sous deux formes ayant des fonctions opposées, on peut parler d'effet Yin et Yang du BDNF (Lu et al., 2005). En effet, le BDNF est synthétisé en tant que molécule précurseur, le proBDNF, qui a des effets négatifs " Yin ". Par sa liaison au récepteur p75NTR, le proBDNF favorise l'apoptose, la rétraction dendritique et la dépression à long-terme. Au contraire, la forme clivée, le BDNF mature (mBDNF), se lie préférentiellement au récepteur TrkB qui active les voies de signalisation de survie cellulaire, de différentiation et de potentialisation à long terme. On parle alors d'effets positifs ou " Yang ". Le clivage de BDNF joue donc un rôle capital dans la régulation de sa balance fonctionnelle vers l'une ou l'autre voie. Il est généralement admis que le proBDNF est clivé en mBDNF par la furine dans le Golgi ou par la pro-protéine convertase 1/3 dans les vésicules de sécrétion (Mowla et al., 2001; Seidah et al., 1996). Le proBDNF restant n'ayant pas été clivé dans la cellule peut être sécrété et être rapidement clivé par la plasmine ou les métalloprotéases de matrice (MMP7) dans l'espace extracellulaire (Lee et al., 2001). Cependant, si les enzymes de conversion du BDNF sont bien connues, les mécanismes de régulation du clivage ne sont pas encore compris.Nous montrons ici un nouveau mécanisme de régulation de la maturation de BDNF via une phosphorylation qui impacte directement la balance fonctionnelle. En effet, nous montrons que la phosphorylation du résidu S130, localisé à l'interface entre le pro-domaine et le domaine mature, diminue l'efficacité du clivage du BDNF par la furine, régulant ainsi l'équilibre entre les formes immature et mature. Cette phosphorylation au site S130 est catalysée par les ectokinases FJX1 et FAM69B qui sont localisées dans l'appareil de Golgi avec BDNF. De plus, grâce à l'utilisation de souris transgéniques knock-in phospho-mutantes, nous montrons que la phosphorylation de BDNF, en favorisant la forme proBDNF, inhibe la potentialisation à long-terme et diminue la dynamique de plasticité des épines dendritiques après stimulation neuronale. Ces résultats suggèrent ainsi une nouvelle voie de régulation de la balance fonctionnelle de BDNF et suggèrent un rôle critique de la phosphorylation S130 dans les processus d'apprentissage et de la mémoire.En parallèle, la recherche de kinases potentielles de BDNF nous a amenés à identifier une exoPKA, localisée dans le Golgi et qui interagit directement avec BDNF. Cette exoPKA atypique phosphoryle BDNF au niveau de la sérine S130 et régule sa maturation et donc l'équilibre entre les formes pro et matures de BDNF. Cette exoPKA est différente de la PKA cytosolique ce qui suggère l'existence d'un nouveau mécanisme de régulation de la plasticité par PKA via BDNF. Enfin, nous sommes en train de tester le rôle de l'exoPKA sur la régulation négative de la plasticité synaptique via la régulation du clivage de BDNF. Ces résultats permettront de déterminer l'existence d'une dichotomie d'action de PKA sur la plasticité en fonction de la forme activée, cytosolique ou golgienne/sécrétée. / Brain-Derived Neurotrophic Factor (BDNF) is a protein that plays an essential role in the survival and differentiation of neurons, as well as in the induction and expression of synaptic plasticity (Deinhardt and Chao, 2014; Lu et al., 2005). BDNF is highly expressed during adulthood and the reduction of its expression is implicated in many neurodegenerative diseases and psychological disorders (Anastasia and Hempstead, 2014). Its action on synaptic plasticity is critical for the establishment of cognitive functions and for the establishment of memory (Bekinschtein et al., 2008; Egan et al., 2003).BDNF exists as two forms with opposite functions, or Yin and Yang effects (Lu et al., 2005). Indeed, BDNF is synthesized as a precursor molecule, proBDNF, which has negative effects "Yin". By binding to the p75NTR receptor, proBDNF promotes apoptosis, dendritic retraction and long-term depression. In contrast, the cleaved form, mature BDNF (mBDNF), binds preferentially to the TrkB receptor that activates the signaling pathways promoting cell survival, differentiation, and long-term potentiation. Thus, mBDNF has positive effects or "Yang". The cleavage of BDNF therefore plays a key role in regulating its functional balance towards one or the other pathway. It is believed that proBDNF is cleaved into mBDNF by furin in the Golgi network or by pro-protein convertase 1/3 in secretory vesicles (Mowla et al., 2001; Seidah et al., 1996). The remaining proBDNF that has not been cleaved in the cell can be secreted and rapidly cleaved by plasmin or matrix metalloproteases (MMP7) into the extracellular space (Lee et al., 2001).Having diametrically opposite roles, together, mature BDNF and proBDNF allow fine regulation of neuronal survival and differentiation and of their activity-dependent synaptic plasticity (Lu et al., 2014; Yang et al., 2014). Since BDNF has critical roles in neuronal functions, it is not surprising that its expression and action are extremely regulated in time and space. However, even if the enzymes processing BDNF are well known, the mechanisms regulating BDNF cleavage are not yet understood.Here we show a new mechanism for the regulation of BDNF maturation via phosphorylation which directly impacts the functional balance. Indeed, we show that phosphorylation of the S130 residue, located at the interface between the pro- and the mature domain, decreases the efficiency of BDNF cleavage by furin, thus regulating the balance between immature and mature forms. This phosphorylation at the S130 site is catalyzed by ectokinases FJX1 and FAM69B which are localized in the Golgi apparatus with BDNF. Furthermore, by using phospho-mutant knock-in transgenic mice, we show that the phosphorylation of BDNF, by promoting the proBDNF form, inhibits the long-term potentiation and decreases the plasticity dynamics of the dendritic spines after neuronal stimulation. These results suggest a novel regulatory pathway for BDNF functional balance and suggest a critical role for S130 phosphorylation in learning and memory processes.In parallel, the search for potential kinases of BDNF led us to identify an exoPKA, located in the Golgi and interacting directly with BDNF. This atypical exoPKA phosphorylates BDNF at serine S130 and regulates its maturation and therefore the balance between the pro- and mature forms of BDNF. This exoPKA is different from the cytosolic PKA which suggests the existence of a new mechanism of regulation of plasticity by PKA via BDNF. Finally, we are testing the role of exoPKA on the downregulation of synaptic plasticity via the regulation of BDNF cleavage. These results will allow to determine the existence of a dichotomous action of PKA on plasticity whether cytosolic or Golgi/secreted PKA is activated.

Indiana University-Purdue University Indianapolis (IUPUI) / Growth and division are tightly coordinated with available nutrient conditions. Cells of the budding yeast, Saccharomyces cerevisiae, grow to a larger size prior to budding and DNA replication when preferred carbon sources such as glucose, as opposed to less preferred sources like ethanol and acetate, are available. A culture’s doubling time is also significantly reduced when the available carbon and nitrogen sources are more favorable. These physiological phenomena are well documented but the precise molecular mechanisms relaying nutrient conditions to the growth and division machinery are not well defined. I demonstrate here that Cdc34, the ubiquitin conjugating enzyme that promotes S phase entry, is phosphorylated upon a highly conserved serine residue which is part of a motif that defines the family of Cdc34/Ubc7 ubiquitin conjugating enzymes. This phosphorylation is regulated by multiple, nutrient sensing kinases including Protein Kinase A, Sch9 and TOR. Furthermore, this phosphorylation event is regulated through the cell cycle with the sole induction occurring in the G1 phase which is when nutrients are sensed and cells commit to another round of division. This phosphorylation likely activates Cdc34 and in turn propagates a signal to the cell division cycle machinery that nutrient conditions are favorable for commitment to a new round of division. This phosphorylation is critical for normal cell cycle progression but must be carefully controlled when cells are deprived of nutrients. Crippling the activity of Protein Kinase A, SCH9 or TOR increases the proportion of cells that survive stationary phase conditions, which because of the metabolic conditions that must be maintained and the similarity to post-mitotic mammalian cells, is referred to as a yeast culture’s chronological lifespan. Yeast cells expressing Cdc34 mutants that are no longer subject to this regulation by phosphorylation have a reduced chronological lifespan. A precise molecular mechanism describing the change in Cdc34 activity after phosphorylation of this serine residue is discussed.

No description available.

A fosforilação de proteínas é o tipo de modificação pós-traducional mais recorrente nas vias de sinalização, desempenhando papel central numa vasta gama de eventos celulares. Um completo entendimento das circunstâncias que coordenam o evento de fosforilação permanece como um desafio para a ciência, a despeito do crescente número de abordagens e estudos realizados no assunto. Um mecanismo largamente descrito e aceito como essencial para coordenar a fosforilação de proteínas é a existência de sequências de aminoácidos que facilitam a fosforilação, conhecidos como consensos de fosforilação. Nesse modelo, cada proteína quinase reconhece sítios de fosforilação se os mesmos estiverem inseridos em uma sequência específica de resíduos na estrutura primária do substrato. Porém, com o crescente volume de dados sobre fosforilação, é possível notar a existência de sítios que são validados experimentalmente como fosforilados por uma determinada proteína quinase, que não apresentam o consenso de fosforilação. Neste trabalho, foi testada e comprovada a hipótese de que estes sítios de fosforilação sem consenso sequencial apresentam resíduos localizados em regiões da estrutura terciária adjacentes ao sítio de fosforilação, cuja as características estereoquímicas mimetizam um peptídeo substrato contendo o consenso de fosforilação. Para essa avaliação, utilizando substratos da PKA, foi constatado que mais de 90% dos sítios de fosforilação que não apresentam o consenso na estrutura primária, apresentam essa disposição na estrutura terciária. Resíduos distantes na estrutura primária se apresentam próximos espacialmente na estrutura tridimensional, em uma conformação semelhante a de um sítio com o consenso de fosforilação. Com isso nós propomos a existência de sítios conformacionais de fosforilação. Para confirmar que esses sítios conformacionais poderiam ser cruciais no reconhecimento do substrato, foram construídos modelos da interação da proteína quinase com os substratos, visando demonstrar a viabilidade da interação dos resíduos formadores do consenso conformacional com a proteína quinase de maneira análoga a de um substrato com o consenso de fosforilação. Para a comprovação experimental do fenômeno, foi utilizado o modelo de fosforilação da -Tubulina, no qual foi constatada uma fosforilação no resíduo T253 que depende da atuação dos resíduos K163 e K164 para a interação com a proteína quinase, confirmando a coerência do modelo proposto. Diante da novidade da proposta, dos estudos computacionais feitos e da validação conseguida, torna-se clara a relevância de se estudar a estrutura tridimensional dos substratos de fosforilação, não só como uma forma de aprofundar os conhecimentos gerais na área de fosforilação, mas também como uma alternativa com potencial de ser explorada no desenvolvimento de novas tecnologias / Protein phosphorylation is the most frequent type of post-translational modification in signaling pathway, developing a key role in a wide range of cell events. The full understanding of the circumstances that coordinate the phosphorylation event remains a challenge for science, despite the growing number of approaches and studies on the subject. A broadly described and accepted mechanism as essential for the coordination of protein phosphorylation is the existence of amino acids sequences that contribute to phosphorylation occurrence, known as phosphorylation consensus. In this model, each protein kinase is able to recognize phosphorylation sites inserted in a specific sequence on the primary structure. However, as the data about phosphorylation sites increases, it is possible to notice that there are sites that are validated experimentally as phosphorylated by a particular protein kinase, which do not have the consensus phosphorylation. In this work, it was tested and proved that phosphorylation sites without the sequence consensus presents anchors residues, that are close to the phosphorylation site on the tertiary structure, creating a structural conformation that mimics the stereochemical features of a substrate peptide containing the phosphorylation consensus. For this evaluation, using substrates of PKA, it was found that more than 90% of phosphorylation sites that have no consensus on the primary structure, presented this kind of disposition on the tertiary structure. Distant residues in the primary structure are spatially close on the three-dimensional structure, in a conformation similar to a phosphorylation site containing the consensus. Thus we proposed the existence of conformational phosphorylation sites. To confirm that these conformational sites could be crucial in substrate recognition, it was built kinase-substrate models, aiming to demonstrate the feasibility of residues forming the conformational consensus on the substrate to interact with the kinase analogously to a substrate with consensus phosphorylation. For experimental verification of this phenomenon, we used the phosphorylation model of -Tubulin, in which we observed a phosphorylation at residue T253 that depends of residues K163 and K164 to interact with the protein kinase, confirming the consistency of proposed model. Faced with the novelty of the proposal, the computational data and the experimental validation, it becomes clear the importance of studying the three dimensional structure of phosphorylation sites, not only as a way of achieving deeper knowledge on phosphorylation field, but also as a potential prospect to be explored on the development of new technologies

A fosforilação de proteínas é o tipo de modificação pós-traducional mais recorrente nas vias de sinalização, desempenhando papel central numa vasta gama de eventos celulares. Um completo entendimento das circunstâncias que coordenam o evento de fosforilação permanece como um desafio para a ciência, a despeito do crescente número de abordagens e estudos realizados no assunto. Um mecanismo largamente descrito e aceito como essencial para coordenar a fosforilação de proteínas é a existência de sequências de aminoácidos que facilitam a fosforilação, conhecidos como consensos de fosforilação. Nesse modelo, cada proteína quinase reconhece sítios de fosforilação se os mesmos estiverem inseridos em uma sequência específica de resíduos na estrutura primária do substrato. Porém, com o crescente volume de dados sobre fosforilação, é possível notar a existência de sítios que são validados experimentalmente como fosforilados por uma determinada proteína quinase, que não apresentam o consenso de fosforilação. Neste trabalho, foi testada e comprovada a hipótese de que estes sítios de fosforilação sem consenso sequencial apresentam resíduos localizados em regiões da estrutura terciária adjacentes ao sítio de fosforilação, cuja as características estereoquímicas mimetizam um peptídeo substrato contendo o consenso de fosforilação. Para essa avaliação, utilizando substratos da PKA, foi constatado que mais de 90% dos sítios de fosforilação que não apresentam o consenso na estrutura primária, apresentam essa disposição na estrutura terciária. Resíduos distantes na estrutura primária se apresentam próximos espacialmente na estrutura tridimensional, em uma conformação semelhante a de um sítio com o consenso de fosforilação. Com isso nós propomos a existência de sítios conformacionais de fosforilação. Para confirmar que esses sítios conformacionais poderiam ser cruciais no reconhecimento do substrato, foram construídos modelos da interação da proteína quinase com os substratos, visando demonstrar a viabilidade da interação dos resíduos formadores do consenso conformacional com a proteína quinase de maneira análoga a de um substrato com o consenso de fosforilação. Para a comprovação experimental do fenômeno, foi utilizado o modelo de fosforilação da -Tubulina, no qual foi constatada uma fosforilação no resíduo T253 que depende da atuação dos resíduos K163 e K164 para a interação com a proteína quinase, confirmando a coerência do modelo proposto. Diante da novidade da proposta, dos estudos computacionais feitos e da validação conseguida, torna-se clara a relevância de se estudar a estrutura tridimensional dos substratos de fosforilação, não só como uma forma de aprofundar os conhecimentos gerais na área de fosforilação, mas também como uma alternativa com potencial de ser explorada no desenvolvimento de novas tecnologias / Protein phosphorylation is the most frequent type of post-translational modification in signaling pathway, developing a key role in a wide range of cell events. The full understanding of the circumstances that coordinate the phosphorylation event remains a challenge for science, despite the growing number of approaches and studies on the subject. A broadly described and accepted mechanism as essential for the coordination of protein phosphorylation is the existence of amino acids sequences that contribute to phosphorylation occurrence, known as phosphorylation consensus. In this model, each protein kinase is able to recognize phosphorylation sites inserted in a specific sequence on the primary structure. However, as the data about phosphorylation sites increases, it is possible to notice that there are sites that are validated experimentally as phosphorylated by a particular protein kinase, which do not have the consensus phosphorylation. In this work, it was tested and proved that phosphorylation sites without the sequence consensus presents anchors residues, that are close to the phosphorylation site on the tertiary structure, creating a structural conformation that mimics the stereochemical features of a substrate peptide containing the phosphorylation consensus. For this evaluation, using substrates of PKA, it was found that more than 90% of phosphorylation sites that have no consensus on the primary structure, presented this kind of disposition on the tertiary structure. Distant residues in the primary structure are spatially close on the three-dimensional structure, in a conformation similar to a phosphorylation site containing the consensus. Thus we proposed the existence of conformational phosphorylation sites. To confirm that these conformational sites could be crucial in substrate recognition, it was built kinase-substrate models, aiming to demonstrate the feasibility of residues forming the conformational consensus on the substrate to interact with the kinase analogously to a substrate with consensus phosphorylation. For experimental verification of this phenomenon, we used the phosphorylation model of -Tubulin, in which we observed a phosphorylation at residue T253 that depends of residues K163 and K164 to interact with the protein kinase, confirming the consistency of proposed model. Faced with the novelty of the proposal, the computational data and the experimental validation, it becomes clear the importance of studying the three dimensional structure of phosphorylation sites, not only as a way of achieving deeper knowledge on phosphorylation field, but also as a potential prospect to be explored on the development of new technologies