The majority of our physiological and metabolic processes are coordinated by an internal clock, which has evolved as an adaptive response to the daily light-dark cycles. Thus, several physiological and behavioral activities display an oscillatory rhythmic period of 24 hours. This highly conserved molecular mechanism is achieved through a specific program of gene expression, characterized by a complex interaction between clock-core proteins, chromatin remodelers and epigenetic events associated with the oscillatory nature of circadian transcriptional activity in the genome. Clock disruption leads to a wide spectrum of severe health problems including chronic metabolic disorders, muscle waste and cardiopathies. Previous studies revealed that each cell and organ possess an intrinsic clock and that coordination between central versus peripheral clocks is key for health. Furthermore, it has been found that under nutritional challenge such as High Fat Diet (HFD), the circadian transcriptome and metabolome are rapidly remodeled in the mouse model. Surprisingly, metabolome and gene expression analysis on various tissues revealed that skeletal muscle is the most affected under HFD. Mechanisms that regulate circadian cycle and stress induced rapid adaptation and in particular metabolic stress at the chromatin level are largely unknown. In this study, we investigated the role of Polycomb proteins group (PcG) mediate cell memory system by maintaining transcriptional gene silencing, in particular the PRC2-Ezh1. We hypothesized that Ezh1 could play an important role in circadian clock regulation in post-mitotic skeletal muscle, and this pathway has never been explored in this context. We explored the circadian role of PRC2-Ezh1 in the mouse skeletal muscle. Intriguingly, we found that the oscillatory profile of a novel isoform of Ezh1 (Ezh1beta), localized specifically in the cytoplasm and controlling stress induced nuclear PRC2 activity, was completely disrupted under HFD. More interestingly, the circadian pattern of core clock components was impaired in Ezh1 depleted cells. Our data unveils an interesting physiological role of the PcG memory system, from cytoplasm to chromatin, which could indicate a new link between the chromatin remodeler Polycomb proteins and the endogenous clock in adaptation mechanism in skeletal muscle.

Human epidermis relies on a population of adult stem cells to maintain its homeostasis. Stem cells transit from a dormant to an active state and undergo a tightly regulated process of differentiation that replenishes the tissue according to its needs. This process either replaces cells that get shed away, or contributes to tissue healing upon injuries, such as wounding. Distinct molecular mechanisms are required to keep human epidermal stem cells localized in their niche and for their active proliferation and mobilization, while others regulate their differentiation status. However, little is known about the proper global chromatin modifications that ensure the correct transition between these stem cell states. This work shows that Cbx4, a Polycomb Repressive Complex-1 (PRC1)-associated protein, maintains human epidermal stem cells slow-cycling and undifferentiated, while protecting them from senescence. Interestingly, abrogating the polycomb activity of Cbx4 impairs its anti-senescent function without affecting stem cell differentiation, indicating that differentiation and senescence are independent processes in human epidermis. Conversely, Cbx4 inhibits stem cell activation and differentiation through its SUMO ligase activity. Global transcriptome and chromatin occupancy analyses indicate that Cbx4 regulates modulators of epidermal homeostasis and represses factors, such as Ezh2, Dnmt1, and Bmi1, to prevent the active stem cell state. Interestingly, Cbx4 also represses genes required for neuronal fate repression, suggesting that it might have a role in ectoderm patterning during development. Cbx proteins are differently expressed during epidermal differentiation and the activity of Cbx4 towards promoting human epidermal stem cell quiescence is unique among the Cbx proteins. This suggests that different Polycomb complexes are assembled, based on the availability of its core member, and balance epidermal stem cell dormancy and activation, while continually preventing senescence and differentiation. / La homeostasis de la epidermis humana depende de una población de células troncales adultas (CTAs). Las CTAs alternan ciclos de quiescencia y actividad, seguidos por una regulación estricta de su diferenciación, según las necesidades celulares del tejido. Este proceso es esencial para repoblar el tejido de células envejecidas o dañadas. Cada estadío por el que transita una CTA está regulado por procesos moleculares específicos. Sin embargo, aún sabemos poco sobre los procesos que regulan la reorganización de la cromatina necesarios para mediar dichas transiciones en la población de las CTAs. Estos resultados demuestran que la proteina Cbx4, pertenciente al complejo Polycomb Repressive Complex-1 (PRC1), es necesaria para mantener a las CTAs de la epidermis humana quiescentes, indiferenciadas, y protegidas de la senescencia. A nivel molecular, la actividad polycomb de Cbx4 es únicamente necesaria para su función antisenescente, pero es dispensable para la regulación de la proliferación y diferenciación de las CTAs. La inhibición de la proliferación y diferenciación celular sin embargo depende de la activdad E3 SUMO ligasa de Cbx4. Analisis del transcriptoma global y de unión a la cromatina (ChIP), demuestran que Cbx4 regula la expresión de moduladores esenciales de la homeostasis de la epidermis, y reprime la expresión de factores necesarios para la activación de las CTAs, tales como Ezh2, Dnmt1 y Bmi1. Cabe destacar que Cbx4 también reprime la expresión de genes que determinam el linage neuronal, lo que sugiere que Cbx4 pueda ser importante para separar el neuroectodermo entre ectodermo y neuronas, durante el desarrollo embrionario. Cbx4 es la única proteina Cbx capaz de inducir entrada en quiescencia de las CTAs, y el resto de proteinas Cbx se expresa de forma diferente durante la diferenciación en la epidermis. Por lo tanto, nuestros estudios sugieren que la actividad de distintos complejos Polycomb actúa en los sucesivos estadíos de quiescencia, proliferación y diferenciación de las CTAs, a la vez que impiden su senescencia de forma constante.

Pavel Jůda - Abstract The cell nucleus is a complex cellular organelle. The nucleus and nuclear processes are organized into functionally and morphologically separated nuclear subcompartments. This thesis is particularly concerned with the three following nuclear subcompartments: sites of DNA replication, Polycomb bodies and nuclear inclusions constituted of inosine monophosphate dehydrogenase 2 (IMPDH2). First, we examined the relationship between MCM proteins and DNA replication. Using immunofluorescent labeling of cells extracted prior fixation and applying cross-correlation function analysis, we showed that MCM proteins are present at the sites of active DNA synthesis. Our results contributed to the solving of the first part of so-called MCM paradox. Second, we studied the structural basis of the Polycomb bodies. Based on fluorescence microscopy studies, Polycomb bodies have been considered to be the nuclear subcompartments formed by the accumulation of Polycomb proteins in the interchromatin compartment. In our work, using correlative light electron microscopy and experimental changes in macromolecular crowding, we clearly showed that a Polycomb body is a chromosomal domain formed by an accumulation of heterochromatin structures, rather than a typical nucleoplasmic body. Third, we were interested in...

Pavel Jůda - Abstract The cell nucleus is a complex cellular organelle. The nucleus and nuclear processes are organized into functionally and morphologically separated nuclear subcompartments. This thesis is particularly concerned with the three following nuclear subcompartments: sites of DNA replication, Polycomb bodies and nuclear inclusions constituted of inosine monophosphate dehydrogenase 2 (IMPDH2). First, we examined the relationship between MCM proteins and DNA replication. Using immunofluorescent labeling of cells extracted prior fixation and applying cross-correlation function analysis, we showed that MCM proteins are present at the sites of active DNA synthesis. Our results contributed to the solving of the first part of so-called MCM paradox. Second, we studied the structural basis of the Polycomb bodies. Based on fluorescence microscopy studies, Polycomb bodies have been considered to be the nuclear subcompartments formed by the accumulation of Polycomb proteins in the interchromatin compartment. In our work, using correlative light electron microscopy and experimental changes in macromolecular crowding, we clearly showed that a Polycomb body is a chromosomal domain formed by an accumulation of heterochromatin structures, rather than a typical nucleoplasmic body. Third, we were interested in...

L’ubiquitination, une modification post-traductionnelle importante pour le contrôle de nombreux processus cellulaires, est une réaction réversible. La réaction inverse, nommée déubiquitination est catalysée par les déubiquitinases (DUB). Nous nous sommes intéressés dans nos travaux à étudier l’ubiquitination de l’histone H2A (H2Aub), au niveau des résidus lysines 118 et 119 (K118/K119), une marque épigénétique impliquée dans la régulation de la prolifération cellulaire et la réparation de l’ADN. Le régulateur transcriptionnel BAP1, une déubiquitinase nucléaire, a été initialement identifié pour sa capacité à promouvoir la fonction suppressive de tumeurs de BRCA1. BAP1 forme un complexe multi-protéique avec plusieurs facteurs transcriptionnels et sa fonction principale est la déubiquitination de H2Aub. Plusieurs études ont démontré que BAP1 est un gène suppresseur de tumeurs majeur et qu’il est largement muté et inactivé dans une multitude de cancers. En effet, BAP1 émerge comme étant la DUB la plus mutée au niveau des cancers. Cependant, le ou les mécanismes d’action et de régulation du complexe BAP1 restent très peu connus. Dans cette étude nous nous sommes intéressés à la caractérisation moléculaire et fonctionnelle des partenaires protéiques de BAP1. De manière significative nous avons caractérisé un mécanisme unique de régulation entre deux composants majeurs du complexe BAP1 à savoir, HCF-1 et OGT. En effet, nous avons démontré que HCF-1 est requis pour maintenir le niveau protéique de OGT et que cette dernière est indispensable pour la maturation protéolytique de HCF-1 en promouvant son clivage par O-GlcNAcylation, une signalisation cellulaire nécessaire au bon fonctionnement de HCF-1. Également, nous avons découvert un nouveau mécanisme de régulation de BAP1 par l’ubiquitine ligase atypique UBE2O. En effet, UBE2O agit comme un régulateur négatif de BAP1 puisque l’ubiquitination de ce dernier induit sa séquestration dans le cytoplasme et l’inhibition de sa fonction suppressive de tumeurs. D’autre part nous nous sommes penchés sur la caractérisation de l’association de BAP1 avec deux facteurs de la famille des protéines Polycombes nommés ASXL1 et ASXL2 (ASXL1/2). Nous avons investigué le rôle de BAP1/ASXL1/2, particulièrement dans les mécanismes de déubiquitination et suppression de tumeurs. Nous avons démontré que BAP1 interagit directement iii via son domaine C-terminale avec le même domaine ASXM de ASXL1/2 formant ainsi deux complexes mutuellement exclusifs indispensables pour induire l’activité déubiquitinase de BAP1. De manière significative, ASXM s’associe avec BAP1 pour créer un nouveau domaine composite de liaison à l’ubiquitine. Ces interactions BAP1/ASXL1/2 régulent la progression harmonieuse du cycle cellulaire. De plus, la surexpression de BAP1 et de ASXL2 au niveau des fibroblastes induit la sénescence de manière dépendante de leurs interactions. D’autre part, nous avons identifié des mutations de cancers au niveau de BAP1 le rendant incapable de lier ASXL1/2, d’exercer sa fonction d’autodéubiquitination et de ce fait d’agir comme suppresseur de tumeurs. Ainsi nous avons révélé un lien étroit entre le gène suppresseur de tumeurs BAP1, son activité déubiquitinase et le contrôle de la prolifération cellulaire. / The reverse reaction of ubiquitination, a crucial post-translational modification, is catalyzed by deubiquitinases (DUBs). BAP1 is an ubiquitously expressed nuclear DUB that recently emerged as an important tumor suppressor highly mutated and inactivated in an increasing number of cancers of diverse origins. Both somatic and germline mutations with loss of heterozygosity were observed in tumors, making BAP1 the most mutated DUB in human malignancies. We previously reported that BAP1 is a component of a large multi-protein complex that includes several transcription regulators. The Drosophila homologue of BAP1, Calypso, forms the Polycomb-repressive DUB (PR-DUB) complex with Additional Sex Comb, ASX. This complex catalyzes the deubiquitination of histone H2A, an essential chromatin modification that regulates gene expression. Despite the ever increasing number of findings describing the occurrence of BAP1 mutations in cancers, few studies investigated the mechanisms of action of this DUB as a tumor suppressor. Therefore, the biological function and the mechanism of action and regulation of BAP1 remains largely uncharacterized. In the work described in this thesis, we investigated the roles of BAP1 partners in modulating its catalytic activity and tumor suppressor function. More specifically we discovered a unique mechanism of regulation between two major components of BAP1 complexes, namely HCF-1 and OGT. Indeed, HCF-1 is important for the maintenance of the cellular levels of OGT. OGT, in turn, is required for the proper proteolytic maturation of HCF-1 by promoting its O-GlcNAcylation. This signaling event is required for HCF-1 function as a cell cycle regulator. On the other hand, we deciphered an intricate mechanism of regulation of BAP1 by the atypical E2/E3 ligase, UBE2O. UBE2O, promote the multi-monoubiquitination of BAP1 on its NLS mediating its cytoplasmic sequestration and thus inhibition of its tumor suppressor function. Another aspect of modulation of BAP1 H2Aub catalysis is provided by the association of BAP1 with ASXL1 and ASXL2 (ASXL1/ASXL2), two orthologs of ASX. We investigated the role of BAP1/ASXL1/2, particularly in the mechanisms of deubiquitination and tumor suppression. We have demonstrated that BAP1 interacts directly via its C-terminal domain with the ASXM domain of ASXL1/2, thus forming two mutually exclusive complexes. Significantly, ASXM promote, through assembly with BAP1, the generation of a composite ubiquitin binding domain (CUBI), indispensable for inducing the deubiquitinase activity of BAP1 towards H2Aub. The interactions between BAP1 and ASXL1/2 regulate cell cycle progression. In addition, overexpression of BAP1 or ASXL2 in fibroblasts induces senescence in CTD- and ASXM-dependent manner. We also identified cancer-derived mutation of BAP1 that selectively abolish its interaction with ASXL1 and ASXL2 as well as its H2A deubiquitinase activity. Significantly, this mutant suppressed senescence induced by BAP1 overexpression. Thus we provided a link between the tumor suppressor BAP1, its deubiquitinase activity and the control of cell proliferation.

L’ubiquitination est une modification post-traductionnelle des protéines qui consiste à attacher, d’une manière covalente, le groupement ubiquitine sur un résidu lysine de la protéine cible. Cette modification peut avoir un impact considérable sur la fonction, la localisation et la stabilité de ces cibles. Une fois établie par des enzymes appelées E3 ligases, l’ubiquitination peut être enlevée par des enzymes spécifiques appelées déubiquitinases, modulant ainsi les effets causés par cette modification. BAP1 (BRCA1-Associated Protein 1) est une déubiquitinase de la famille des UCH (Ubiquitin C-terminal Hydrolases) qui a été initialement identifiée comme partenaire du suppresseur de tumeurs BRCA1 (BReast Cancer Associated gene 1). De nombreux groupes de recherche, incluant le nôtre, ont montré que BAP1 est associée avec d’autres cofacteurs formant un large complexe multiprotéique. Ce dernier est impliqué dans plusieurs processus cellulaires comme la transcription des gènes, la régulation de la chromatine, la coordination du cycle cellulaire et la réponse aux dommages à l’ADN. La cible majeure de BAP1 est l’histone H2A ubiquitinée sur la lysine 119, une marque d’histone qui a été souvent associée avec une conformation répressive de la chromatine. Quels sont les mécanismes régulant le complexe BAP1 lui permettant d’exécuter ces fonctions biologiques? Cela implique-t-il des modifications post-traductionnelles touchant les partenaires de BAP1 ? Ces questions restent encore sans réponse définitive. Ainsi, les objectifs de cette thèse sont de caractériser le mécanisme et la fonction du complexe BAP1 en étudiant les modifications post-traductionnelles de ses partenaires. Pour répondre à ces questions nous avons étudié les modifications post-traductionnelles touchant BAP1 et ses cofacteurs mutuellement exclusifs ASXL1 et ASXL2 (Additional Sex Comb-like 1,2). Nous avons démontré qu’ASXL1 et ASXL2 sont monoubiquitinés uniquement lorsqu’ils sont associés à BAP1. Sachant que les complexes BAP1/ASXLs sont conservés au cours de l’évolution, nous avons aussi démontré que la monoubiquitination des ASXLs est conservée chez la Drosophile. En utilisant des méthodes de déplétion de protéines par siARN et CRISPR/Cas9 ainsi que des mutants de perte de fonction de BAP1 et ASXL2, nous avons identifié les enzymes responsables de la monoubiquitination des ASXLs ainsi que leur effet sur l’activité catalytique de BAP1. D’autre part, nous avons étudié le développement chez la Drosophile ainsi que le cycle cellulaire des cellules humaines pour identifier la fonction biologique de la monoubiquitination de ASXL2. Nos résultats démontrent que la monoubiquitination d’ASXL2 sur la lysine 370 en présence de BAP1 est une modification post-traductionnelle conservée et catalysée directement par la famille UBE2Es des enzymes de conjugaison de l’ubiquitine (UBE2E1,2,3 chez les mammifères et UbcD2 chez la Drosophile). Cette monoubiquitination stimule l’activité catalytique de BAP1 chez les mammifères et de son orthologue Calypso chez la Drosophile envers H2Aub. Le blocage de la monoubiquitination des ASXLs par des mutations ciblant la lysine K370 induit une inhibition de l’activité de BAP1, ce qui cause une dérégulation du cycle cellulaire chez les cellules mammifères et une transformation homéotique haltère-aile chez la Drosophile. De plus, il nous a été possible de constater l’importance de cette monoubiquitination dans le cancer en démontrant la forte corrélation d’expression de BAP1/ASXL2 et les UBE2Es au niveau du mésotheliome, un cancer connu pour la dérégulation de BAP1. Nos résultats indiquent l’importance des modifications post-traductionnelles, dont la monoubiquitination, dans la régulation de la fonction et la stabilité du complexe BAP1. De plus, nous décrivons un nouveau mécanisme d’activation d’une deubiquitinase par la monoubiquitination de son cofacteur. D’autres études seront nécessaires afin de comprendre le lien entre l’activation de BAP1/ASXL2 par monoubiquitination et la fonction suppresseur de tumeurs de BAP1 via la deubiquitination d’H2Aub. D’autre part, nous avons fait l’observation que la déplétion de la deubiquitinase associée à la particule régulatrice du protéasome, PSMD14, induit non seulement une réduction drastique d’H2Aub dans la cellule, mais aussi une mort cellulaire rapide. Ceci nous a poussé initialement à investiguer l’implication de l’activité catalytique du protéasome dans la régulation d’H2Aub en lien avec la mort cellulaire. Malgré le fait que nous n’ayons pas trouvé un lien direct entre PSMD14 et la deubiquitination d’H2Aub, nous avons identifié plusieurs candidats (DUBs et E2s) impliqués dans l’induction de la mort cellulaire tout en surmontant une résistance acquise contre des inhibiteurs ciblant l’activité catalytique du protéasome. Ces candidats pourraient représenter des cibles intéressantes pour développer des inhibiteurs spécifiques afin de contrecarrer la résistance aux inhibiteurs du protéasome. / Ubiquitination is a post-translational modification of proteins that involves covalently attaching the ubiquitin moiety to the lysine residues of the target protein. This modification has been reported to have a significant impact on the function, localization and stability of these targets. Once established by enzymes called E3 ligases, ubiquitination can be removed by specific enzymes called deubiquitinases, thus modulating the effects caused by this modification. BAP1 (or BRCA1-Associated Protein1) is a deubiquitinase, from the UCH (Ubiquitin C-terminal Hydrolases) family, that was originally identified as a partner of the BRCA1 (BReast Cancer Associated gene 1) tumor suppressor. We and other research groups have shown that BAP1 is associated with other co-factors forming a multi-protein complex involved in several cellular processes such as gene transcription, chromatin regulation, cell cycle regulation and DNA damage response. The major target of BAP1 is ubiquitinated histone H2A, a histone mark that has been frequently associated with a repressive chromatin conformation. What are the mechanisms regulating the BAP1 complex allowing it to perform its biological functions? Does this involve post-translational modifications affecting BAP1 partners? These questions are still incompletely answered. Thus, the objectives of our studies are to characterize the mechanism and the function of the BAP1 complex by studying the post-translational modifications that could affect its obligate partners including ASXLs. To address these questions, we studied the post-translational modifications affecting BAP1 and its two mutually exclusive co-factors ASXL1 and ASXL2 (Additional Sex Comb-like 1,2). We demonstrated that ASXL1 and ASXL2 are mono-ubiquitinated only when associated with BAP1. Taking into account that the BAP1/ASXLs complexes are highly conserved during evolution, we also demonstrated that the mono-ubiquitination of ASXLs is important for Drosophila development. Using RNAi and CRISPR/Cas9 gene depletion methods and loss-of-function mutants of BAP1 and ASXL2, we identified the precise site of ASXLs ubiquitination, the enzymes responsible for establishing this mono-ubiquitination as well as its effect on catalytic activity of BAP1. On the other hand, we investigated Drosophila development as well as human cell cycle progression to identify the biological function of ASXLs mono-ubiquitination. Our results indicate that the mono-ubiquitination of ASXL2 on lysine 370 in the presence of BAP1 is a conserved post-translational modification catalyzed directly by the UBE2E family of ubiquitin-conjugating enzymes (UBE2E1, 2, 3 in mammals and UbcD2 in Drosophila). This mono-ubiquitination event stimulates the catalytic activity of BAP1 in mammals and its Drosophila ortholog Calypso towards H2Aub in vivo and in vitro. Blocking the mono-ubiquitination of ASXLs, by mutations targeting lysine K370, induces an inhibition of BAP1 catalytic activity causing a deregulation of human cell cycle progression and a haltere-to-wing homeotic transformation in Drosophila. In addition, we were able to assess the importance of ASXLs mono-ubiquitination in cancer using the mesothelioma tumor model, demonstrating a strong correlation between the expression of BAP1/ASXL2 and UBE2Es. Our results indicate the importance of post-translational modifications, including mono-ubiquitination, in the regulation of the function and stability of the BAP1 complex. Moreover, we describe a novel mechanism of activation of a deubiquitinase by the mono-ubiquitination of its co-factor. Further studies will be needed to shed more light on the link between BAP1/ASXLs activation by mono-ubiquitination and the tumor suppressor function of BAP1 via H2Aub deubiquitination. On the other hand, we have noticed that the depletion of PSMD14, a deubiquitinase associated with the proteasome regulatory particle, induces not only a drastic reduction of H2Aub in the cell, but also rapid cell death. This prompted us initially to investigate the involvement of the catalytic activity of the proteasome in the regulation of H2Aub in connection with cell death. Although we did not find a direct link between PSMD14 and H2Aub deubiquitination, we identified several candidates (DUBs and E2s) involved in the induction of cell death while overcoming acquired resistance against proteasome catalytic inhibitors. These candidates may represent attractive targets for developing specific inhibitors to counteract resistance to proteasome inhibitors.