Afin d’initier la transcription par l’ARN Polymérase II, la chromatine est modifiée par des coactivateurs, dont certains catalysent des modifications post-traductionnelles des queues des histones. La protéine GCN5 est une enzyme qui possède une activité histone acétyltransférase (HAT). Elle fait partie du complexe coactivateur SAGA, qui acétyle les histones H3. Or, il existe un second complexe HAT contenant GCN5 : le complexe ATAC, mis en évidence chez la drosophile. Chez l’homme en revanche, l’existence d’un tel complexe n’avait pas encore été démontrée au début de ma thèse.L’objectif de ma thèse a consisté tout d’abord en la purification et la caractérisation du complexe HAT ATAC chez l’homme. Puis, j’ai cherché à comprendre le fonctionnement et la spécificité d’action du complexe ATAC, par rapport au complexe SAGA.Dans une première partie, j’ai ainsi pu montrer que GCN5 fait partie d’un second complexe chez l’homme, le complexe ATAC. La composition en sous-unités du complexe ATAC a été déterminée et l’activité de ce dernier sur les histones étudiée. Nous avons pu démontrer que, comme hSAGA, hATAC acétyle les histones in vitro et in vivo, et préférentiellement la lysine 14 de l’histone H3. Chez les vertébrés, un paralogue de GCN5, PCAF peut se substituer à GCN5 dans les complexes ATAC ou SAGA.Par la suite, j’ai poursuivi la caractérisation de ces complexes HAT afin de comprendre le rôle des enzymes au sein des complexes et leurs fonctions. Pour cela, j’ai voulu comprendre le rôle des sous-unités, comment elles influencent l’activité de l’enzyme, et ainsi identifier les protéines qui permettent la spécificité de hATAC par rapport à hSAGA. / In order to initiate the transcription by the RNA polymerase II, chromatin needs to be modified by coactivators. Some of these coactivators are histone post-translational modifying complexes. GCN5 is a histone acetyltransferase enzyme (HAT), which can acetylate the histones. This enzyme is found in a multiproteic complex named SAGA. Recently, a second HAT complex containing GCN5 was discovered: ATAC, in drosophila. At the beginning of my thesis, the existence of such complex in human was not shown.My thesis objectives were then to identify and characterize an ATAC complex in human cells. In a first part, we purified and identified the composition in subunits of human ATAC. Then we studied the activity and specificity of ATAC on histone substrates, compared to SAGA. Next, we were wondering how the subunits of the two HAT complexes could play a role on the regulation of the activity of the enzyme GCN5, in order to understand the histone specificity of ATAC and SAGA.

In higher eukaryotes, DNA is packaged with histones and other proteins into chromatin. While this is important in the control of unwanted gene expression, chromatin also serves as a barrier to many vital functions in the cell. Therefore, cells have evolved many different types of chromatin remodeling enzymes to contend with this inhibitory structure and enable gene expression and other functions. The Saccharomyces cerevisiae PHO5 gene is triggered in response to phosphate starvation. In this study, I evaluate the chromatin remodeling requirements of this gene with respect to the multisubunit complexes SWI/SNF and SAGA. I show, for the first time, physical recruitment of SWI/SNF to the PHO5 promoter. I also demonstrate the role of promoter occupancy in influencing requirements for chromatin remodeling enzymes. Further, I describe various interactions between these two complexes at the PHO5 promoter. This study presents evidence for the first instance of excess recruitment of an ATP-dependent remodeler potentially compensating for the lack of a histone acetyltransferase.

Acetylation was previously thought to occur exclusively on histones, but recent high-throughput screens have identified thousands of non-histone substrates. Despite the identification of these sites, little is known about how these acetyltransferase enzymes target their substrates. Gcn5 is the catalytic acetyltransferase found within the highly conserved SAGA complex. Recently, a member of this complex, Ada2, was found to impact Gcn5 substrate selection. In the yeast model organism Saccharomyces cerevisiae, a synthetic substrate developed from a proposed Gcn5-specific consensus sequence is used to identify regulators of Gcn5 substrate selection. This work is the first to demonstrate that addition of a consensus sequence is enough to confer acetylation of a non-substrate. With this method, Ada3 was identified as a key regulator, and acetylome profiling identified novel targets for Gcn5 dependent acetylation specifically regulated by Ada3. This system could be adapted for other acetyltransferases to identify regulators of substrate selection.

Indiana University-Purdue University Indianapolis (IUPUI) / The obligate intracellular protozoan parasite Toxoplasma gondii is a medically relevant pathogen that has infected a third of the world’s population. Toxoplasma is the causative agent of toxoplasmosis, which can have severe repercussions such as encephalitis and even death in immunocompromised patients. Current treatments for toxoplasmosis only target acute infection and can be toxic to patients, resulting in complications including allergy and bone marrow suppression. Thus, the identification of novel drug targets and therapies for toxoplasmosis is vital. Epigenetic modulators of lysine acetylation, including ‘writers,’ ‘erasers,’ and ‘readers,’ have been identified as promising drug targets for protozoan parasites. The lysine acetyltransferase (KAT) GCN5b appears to be an essential gene for Toxoplasma viability. The KAT domain of GCN5b is essential to GCN5b function and is targetable by small molecule inhibitors. While the acetyltransferase activity of this gene is well-characterized, the functionality of its C-terminal bromodomain (BRD) remains to be understood. Bromodomains are readers of lysine acetylation, and recently, bromodomain inhibitors have shown promise in a number of human diseases, as well as in protozoan parasites. We hypothesized that the GCN5b bromodomain is critical for Toxoplasma viability. The data reported herein suggest that the GCN5b bromodomain is important for tachyzoite viability and may serve as a novel therapeutic target in Toxoplasma.

<P> The first histone acetyltransferase to be described was GCN5, from the yeast species Saccharomyces cerevisiae. To date, the GCN5-related N-acetyltransferases (GNATs) comprise one of the largest enzyme superfamilies with over 10,000 identified members in sequenced genomes. This protein is known to acetylate specific lysine residues on the amino-terminal tails of nucleosomal histones, thereby loosening their contact with the tightly packed DNA and facilitating transcription. </p> <p> In this study, I determined that GCN5 is able to shuttle between the nucleus and the cytoplasm using fluorescence recovery after photobleaching (FRAP). Mutational studies revealed that its nuclear import is regulated by a classical bipartite nuclear localization signal (NLS) that is dependent on the transporters importin a and f3. In contrast, we found that GCN5 lacks a CRM1-dependent nuclear export signal (NES), as demonstrated by mutational and leptomycin B (LMB) studies; instead, IKB, a previouslydescribed transcription inhibitor with a CRMl-dependent NES, was found to modulate the export of GCN5 from the nucleus. This was initially discovered while performing the LMB assays, for which IKB served as a positive control, and was subsequently confirmed by mutational studies and protein complementation assays (PCAs). Furthermore, while the PCAs demonstrated a physical interaction between these two proteins in vivo, GST pull-down experiments were employed to confirm their interaction in vitro. </p> <p> Furthermore, this study also revealed that over-expression of GCN5-e YFP in NIH 3T3 cells causes -10% of the transfected cells to exhibit nuclear GCN5-eYFP-associated filaments; these structures were confirmed to be F -actin filaments comprised of f3-actin through co-localization studies with both TRITC-phalloidin and a mRFP-f-actin construct. GCN5's acetyltransferase activity was shown to be responsible for the formation ofthese filaments through mutation of its catalytic residue. Moreover, a protein complementation assay (PCA) demonstrated an in vivo interaction between GCN5 and f-actin, while FRAP analysis of a single filament showed that GCN5-e YFP molecules rapidly and randomly associate with these filaments along their entire length. Together these results suggest that GCN5's acetyltransferase activity is responsible for the structural maintenance of these filaments. Finally, GCN5-eYFP-associated filaments were found to be spatially separate from both lamin A (a nuclear envelope structural protein) and DNA; however, this does not exclude the possibility of an indirect interaction between these cellular constituents, as treatment of a live cell with Hoechst DNA stain, which disrupts the structure of DNA, was shown to disturb the structural integrity of these filaments. </p> / Thesis / Master of Science (MSc)

Les cellules souches de gliomes (CSG), une sous-population de cellules tumorales, seraient en partie responsables de l’échec des traitements des gliomes de par leur résistance et leur capacité régénérative. Le mécanisme alternatif (ALT) de maintenance des télomères, basé sur la recombinaison homologue et non pas sur la télomérase, est détecté dans environ 30% des gliomes humains suggérant que des stratégies thérapeutiques spécifiquement dirigées contre ALT pourraient avoir un intérêt thérapeutique. Dans ce travail, nous avons poursuivi la caractérisation du premier exemple de CSG humaines ayant un phénotype ALT, les cellules TG20. Nous avons montré que malgré leur très fort taux de recombinaison, les télomères de ces cellules continuaient à assurer leur fonction de protection des chromosomes. Nous avons vérifié que les cellules TG20 conservaient leur capacité à générer des tumeurs intracérébrales après des transplantations sériées chez les souris immunodéprimées tout en gardant un phénotype ALT. Ces résultats confirment à la fois les propriétés de cellules souches cancéreuses des cellules TG20 et la capacité de ALT à assurer la maintenance des télomères nécessaire à l’autorenouvellement et au fort taux de prolifération des CSG in vivo. La greffe intracérébrale de cellules TG20 chez des souris immunodéprimées représente donc un bon modèle d’étude préclinique des gliomes ALT. Nous avons ainsi montré qu’un traitement précoce par un ligand des G-quadruplexes télomériques, 360B, juste après la greffe de cellules TG20, était capable d’inhiber le développement tumoral suggérant l’intérêt de l’utilisation de ligands des G-quadruplexes pour cibler spécifiquement les CSG ALT. Une étude des profils d'expression transcriptomique des cellules TG20 et de plusieurs lignées de CSG humaines exprimant la télomérase, nous a conduit à nous intéresser aux rôles de deux lysine acétyl transférases homologues, PCAF (P300/CBP Associated Factor) et GCN5 (General Control Nonderepressible 5) dans la régulation de la recombinaison télomérique des cellules ALT. Nous avons montré que les inhibitions de ces deux protéines ont des effets opposés sur le mécanisme ALT. Nous proposons qu’une balance d’expression de PCAF et GCN5 régule la maintenance des télomères dans les cellules ALT via le contrôle du turnover de TRF1 ce qui pourrait constituer la base d’une nouvelle stratégie thérapeutique vis-à-vis des gliomes ayant un phénotype ALT. / Glioma stem cells (GSC), a subpopulation of tumor cells, are partly responsible for the failure of treatment of gliomas because of their resistance and regenerative capacity. The mechanism of alternative lengthening of telomere (ALT), based on homologous recombination, is detected in approximately 30 % of human gliomas. Therefore, therapeutic strategies directed specifically against ALT may have a therapeutic value. In this work, we further characterized the first model of human ALT GSC, the TG20 cells. We showed that despite their very high rate of recombination, the telomeres were still capable of fulfilling their protective function of chromosomes. We verified that the TG20 cells retained their ability to generate intracerebral tumors after serial transplantations in immunocompromised mice, while preserving an ALT phenotype. These results confirm the cancer stem properties of TG20 cells and the ability of ALT to ensure telomeres maintenance, which is required for the self-renewal and the high proliferation rate of GSC in vivo. Intracerebral grafts of TG20 cells in immunocompromised mice represent thus a good preclinical model for studying ALT gliomas. We have shown that treatment with a ligand of telomeric G-quadruplexes, the 360B, at an early stage of TG20 tumor engraftment, was able to inhibit tumor growth, showing the interest of the use of G-quadruplex ligands to specifically target ALT GSC. Transcriptomic profiling of TG20 cells and several other GSC telomerase-positive lines, incited us to study the roles of two homologous lysine acetyl transferases, PCAF (p300/CBP Associated Factor) and GCN5 (General Control Nonderepressible 5), in the regulation of telomeric recombination in ALT cells. We showed that the inhibition of these two proteins has opposite effects on the ALT mechanism. We propose that a balance of expression of PCAF and GCN5 regulates the telomere maintenance in ALT cells by controlling the turnover of TRF1. This model could serve for the development of new therapeutic strategies targeting ALT gliomas.

La condensation mitotique des chromosomes est l'un des mécanismes assurant la transmission fidèle de l'information génétique. Les complexes condensines et leur association à la chromatine sont nécessaires à cette condensation. Cependant, les mécanismes par lesquels ces complexes s'associent aux chromosomes et contribuent à leur condensation sont mal compris. L'objectif de ma thèse était d'identifier et de caractériser des facteurs de condensation encore inconnus collaborant avec le complexe condensine présent chez S. pombe. Par un crible génétique, nous avons recherché des mutants viables lorsque le complexe condensine est complètement fonctionnel mais morts lorsque ce complexe est partiellement défectif. Nous avons ainsi identifié 7 protéines jusqu'alors jamais impliquées dans la condensation mitotique. Parmi ces dernières, nous avons identifié des protéines impliquées dans le remodelage de la chromatine et des facteurs de transcription comme Gcn5, une HAT très conservée, connue pour son rôle de coactivateur de la transcription ; suggérant un lien entre la condensation et la machinerie transcriptionnelle. Gcn5 s'associe à la chromatine au niveau des promoteurs des gènes où elle acétyle principalement H3K9, H3K14 et H3K18. Sa présence au niveau des promoteurs est directement corrélée avec le niveau de transcription des gènes correspondants. Bien que la majorité de la chromatine soit dé-acétylée et que la présence de Gcn5 soit réduite au niveau des chromosomes en mitose, des traces de H3K9 acétylée persistent au niveau de certains promoteurs. Nos résultats suggèrent que cette acétylation persistante pourrait être liée au recrutement du complexe condensine à la chromatine / From yeasts to human, Condensin is essential for mitotic chromosome condensation. However, how Condensin binds to chromatin and, in this context, shapes mitotic chromosome remain poorly understood. Mappings performed from yeasts to mouse have revealed that condensin is enriched near highly expressed genes along chromosome arms, suggesting that as yet identified features associated with transcription take part in condensin binding to chromatin. To identify factors that collaborate with Condensin we performed a synthetically lethal genetic screen in fission yeast. We searched for mutants that are alive when Condensin is fully functional but dead when Condensin is partly defective. We identified 7 proteins never known for their roles in the mitotic condensation, such as some chromatin remodelling and some transcription factors. All these results were consistent with a link between condensation and transcription. Among theses 7 proteins, we found Gcn5, which encodes a conserved HAT, well known for the role it plays as a transcriptional co-activator. Gcn5 binds to gene promoters where it acetylates mainly H3K9, K14 and K18, and its occupancy correlates with transcription rates. Remarkably, although the bulk of chromatin is de-acetylated and Gcn5 reduced from chromatin upon mitosis entry, traces of Gcn5 dependant H3K9 acetylated persist at condensin binding sites. Here, we provide evidence that Gcn5-mediated histone H3 K9 acetylation could assist the binding of Condensin to chromatin

The CCAAT/Enhancer Binding Protein Beta (C/EBPβ) is part of the leucine zipper family of transcription factors and is involved in a myriad of processes including cellular proliferation and differentiation. C/EBPβ is expressed as three isoforms (LAP*, LAP, LIP), translated from a single mRNA by a leaky ribosomal scanning mechanism. While LAP* and LAP have activating functions, LIP is recognized as being a repressor of transcription due to its lack of activation domains. Numerous studies have shown that C/EBPβ acetylation state modulates its activity in a promoter-specific manner. For instance, the acetyltransferases GCN5/PCAF and the deacetylase complex mSin3A/HDAC1 regulate C/EBPβ activity on the C/EBPa promoter. GCN5/PCAF-mediated acetylation of C/EBPβ was shown to positively affect its transcriptional activity in a steroid-dependent mechanism via the glucocorticoid receptor (GR). GR relieves HDAC1 association from C/EBPβ by targeting the deacetylase for proteasomal degradation, hence favouring GCN5-mediated acetylation of C/EBPβ and allowing maximum activation capacity to be reached. In order to further elucidate C/EBPβ activation, I sought to characterize the interplay between GCN5 and HDAC1 in regulating C/EBPβ LAP/LIP activity during murine adipogenesis by identifying their binding domain in C/EBPβ. I identified a minimal domain located within regulatory domain 1 (RD1) of C/EBPβ that is required for both GCN5 and HDAC1 binding. Furthermore, the loss of the identified domain in C/EBPβ appears to partially mimic the GR effect, thus giving C/EBPβ a higher basal transcriptional activity that accelerates NIH 3T3 and 3T3 L1 adipogenesis. Moreover, I also showed that the LIP isoform inhibitory mode of action is partially mediated through the mSin3A/HDAC1 repressor complex, which gives LIP an active repressor function. In addition to LIP inhibitory function, I also showed that a cysteine residue located in LAP* negatively regulates its transactivating function during murine adipogenesis. Although RD1 of C/EBPβ has been suggested to act as a negative regulatory domain, I showed that only five residues are responsible for most of its inhibitory effect. Hence, in an attempt to further define sub-domains within RD1, I characterized a new positive regulatory domain at its N-terminal region, which seems to be required for C/EBPβ activity in a promoter-specific manner. In conclusion, this study not only supports previously hypothesized mechanisms by which C/EBPβ is regulated, but it also redefines the contribution of LAP*, LAP and LIP in regulating transcription. Most importantly, the results emphasize the countless possibilities by which C/EBPβ transactivation potential could be modulated during cellular differentiation.

The CCAAT/Enhancer Binding Protein Beta (C/EBPβ) is part of the leucine zipper family of transcription factors and is involved in a myriad of processes including cellular proliferation and differentiation. C/EBPβ is expressed as three isoforms (LAP*, LAP, LIP), translated from a single mRNA by a leaky ribosomal scanning mechanism. While LAP* and LAP have activating functions, LIP is recognized as being a repressor of transcription due to its lack of activation domains. Numerous studies have shown that C/EBPβ acetylation state modulates its activity in a promoter-specific manner. For instance, the acetyltransferases GCN5/PCAF and the deacetylase complex mSin3A/HDAC1 regulate C/EBPβ activity on the C/EBPa promoter. GCN5/PCAF-mediated acetylation of C/EBPβ was shown to positively affect its transcriptional activity in a steroid-dependent mechanism via the glucocorticoid receptor (GR). GR relieves HDAC1 association from C/EBPβ by targeting the deacetylase for proteasomal degradation, hence favouring GCN5-mediated acetylation of C/EBPβ and allowing maximum activation capacity to be reached. In order to further elucidate C/EBPβ activation, I sought to characterize the interplay between GCN5 and HDAC1 in regulating C/EBPβ LAP/LIP activity during murine adipogenesis by identifying their binding domain in C/EBPβ. I identified a minimal domain located within regulatory domain 1 (RD1) of C/EBPβ that is required for both GCN5 and HDAC1 binding. Furthermore, the loss of the identified domain in C/EBPβ appears to partially mimic the GR effect, thus giving C/EBPβ a higher basal transcriptional activity that accelerates NIH 3T3 and 3T3 L1 adipogenesis. Moreover, I also showed that the LIP isoform inhibitory mode of action is partially mediated through the mSin3A/HDAC1 repressor complex, which gives LIP an active repressor function. In addition to LIP inhibitory function, I also showed that a cysteine residue located in LAP* negatively regulates its transactivating function during murine adipogenesis. Although RD1 of C/EBPβ has been suggested to act as a negative regulatory domain, I showed that only five residues are responsible for most of its inhibitory effect. Hence, in an attempt to further define sub-domains within RD1, I characterized a new positive regulatory domain at its N-terminal region, which seems to be required for C/EBPβ activity in a promoter-specific manner. In conclusion, this study not only supports previously hypothesized mechanisms by which C/EBPβ is regulated, but it also redefines the contribution of LAP*, LAP and LIP in regulating transcription. Most importantly, the results emphasize the countless possibilities by which C/EBPβ transactivation potential could be modulated during cellular differentiation.

Engineered cellular systems are a promising avenue for production of a wide range of useful products including renewable fuels, commodity and specialty chemicals, industrial enzymes, and pharmaceuticals. Achieving this breadth of biological products is facilitated by the diversity of organisms found in nature. Using biological and engineering principles, this diversity can be harnessed to make efficient and renewable bio-based products. Such advancements rely upon our ability to modify host genetics and metabolism. This work focuses on the development of new biotechnological tools which enable cellular engineering, and the implementation of these tools in eukaryotic systems. Mammalian cell engineering has important implications in protein therapeutics and gene therapy. One major limitation, however, is the ability to predictably control gene expression. We address this challenge by examining critical aspects of gene expression in human cells. First, we evaluate the impact of selection markers, a common mammalian expression element, on cell line development. In doing so, we determine that Zeocin is the best selection agent for human cells. Next, we identify loci across the genome that support high level expression of recombinant DNA and demonstrate their advantage for stable integration. Finally, we optimize a Cre recombinase based methodology that enables efficient retargeting of genomic loci. Collectively, this work augments the current genetic toolbox for human cell lines. Beyond basic gene expression, there is interest in understanding global interactions within the cell and how they relate to phenomena including gene regulation, expression and disease states. Although our tools are not yet sufficient to study these phenomena in many hosts, methods can be developed in lower eukaryotes and then adapted for more complex hosts later. We demonstrated two methods in S. cerevisiae that utilize a systems-level approach to understand complex phenotypes. First, we developed condition-specific codon optimization that utilizes systems biology information to optimize gene sequence in a condition-specific manner. Additionally, we developed a Graded Dominant Mutant Approach which can be used to dissect multifunctional proteins, understand epigenetic factors, and quantitatively determine protein-DNA interactions. Both can be implemented in many cellular hosts and expand our ability to engineer complex phenotypes in eukaryotic cell systems.