Transcription elongation regulator 1 (TCERG1) has previously been demonstrated to be an inhibitor of the transactivation and growth arrest activities of CCAAT/Enhancer binding protein alpha (C/EBPα). Furthermore, TCERG1 had been demonstrated to become relocalized from nuclear speckles to the pericentromeric regions where C/EBPα resides when both proteins are co-expressed in the cell. This thesis demonstrates that the deletion of a unique, imperfect series of 38 glutamine-alanine (QA) repeats near the amino terminus of TCERG1 is able to abrogate the ability of TCERG1 to inhibit C/EBPα-mediated growth arrest, the physical interaction between TCERG1 and C/EBPα, and the relocalization of TCERG1 from nuclear speckles when C/EBPα is co-expressed in the cell. The deletion of the QA domain demonstrated that there was a threshold amount of QA repeats required in TCERG1 for the relocalization and growth arrest inhibitory activities between TCERG1 and C/EBPα. It was demonstrated that between 11 and 20 QA repeats were required in TCERG1 to produce the relocalization from nuclear speckles or to be able to inhibit C/EBPα-mediated growth arrest. The physical interaction of TCERG1 and C/EBPα as examined by co-immunoprecipitation was also found to be QA dependent, with a diminishing interaction observed as the number of QA repeats in TCERG1 were reduced. However, experiments examining the isolated QA domain indicated that it was insufficient to relocalize an mCherry fluorescent protein fusion to either the nucleus or to pericentromeric regions where C/EBPα is concentrated. This inability to produce relocalization suggests that the QA domain requires another domain or domains from TCERG1 to mediate the relocalization activity. When expressed with the WT TCERG1, ΔQA TCERG1 was able to act in a dominant negative manner, preventing the relocalization of the WT TCERG1 protein to pericentromeric domains. Interestingly, the transactivation inhibitory activities of TCERG1 on C/EBPα do not appear to require the QA domain, but rather are localized to the carboxy half of TCERG1, somewhere within amino acids 612-1098. The data obtained provides the first report of a role for this unique QA repeat domain.

TDP-43 est une protéine de liaison aux acides nucléiques qui joue un rôle essentiel dans le métabolisme de l'ARN. À l'état physiologique, un contrôle strict des niveaux d’expression de cette protéine est critique pour la fonction et la survie cellulaire. Une boucle d'autorégulation négative est à la base de ce contrôle du taux intracellulaire de TDP-43. Laquelle a été identifiée comme le constituant principal des inclusions observées chez une majorité des patients atteints de Sclérose Latérale Amyotrophique (SLA) ou de Dégénérescence Lobaire Fronto-Temporale (DLFT). A ce jour, plus de 50 mutations faux-sensdu gène TARDBP/TDP-43 ont été décrites chez des patients DLFT/SLA, démontrant le rôle clé de TDP-43 dans ces pathologies neurodégénératives. Notons cependant que les conséquences fonctionnelles de ces mutations ne sont pas complètement déterminées. Plusieurs études suggèrent qu’une élévation des niveaux d’accumulation de TDP-43 pourraitparticiper aux mécanismes physiopathologiques. La modulation du cycle de production de TDP-43 pourrait donc constituer une nouvelle stratégie thérapeutique. Ce travail de recherche avait donc pour principal objectif d’identifier des modulateurs génétiques de la production de TDP-43 en utilisant un nouveau modèle de drosophile transgénique mimant les principales étapes d’autorégulation de TDP-43. Nous avons ainsi pu montrer que la modulation des niveaux d’expression de la protéine TCERG1 et de plusieurs facteurs d'épissage, parmi lesquels SRSF1, SRSF3 et SF3B1, influe sur les niveaux de production deTDP-43. Nous avons également montré que la présence des mutations DLFT/SLA n’altère pas la capacité de la protéine à s’autoréguler. / TDP-43 is a DNA/RNA binding protein that plays an important role in RNA metabolism. In the physiological state, strict control of its expression levels is critical for cell function and survival. TDP-43 expression is tightly regulated through an autoregulatory negative feedback loop. This protein has been identified as the principal component of the inclusions observed in a majority of patients with Amyotrophic Lateral Sclerosis (ALS) or FrontoTemporal Lobar Degeneration (FTLD). To date, more than 50 missense mutations of the TARDBP / TDP-43 gene have been described in FTLD / ALS patients, demonstrating the key role of TDP-43 in these neurodegenerative pathologies. However, the functional consequences of TDP-43 mutations are not completely determined. Several studies suggest that high accumulation levels of TDP-43 may participate in pathophysiological mechanisms. The modulation of the production cycle of TDP-43 may therefore provide a new therapeutic strategy. The main goal of this research project was to identify genetic modulators of TDP-43 production by using a novel transgenic Drosophila model mimicking main steps of TDP-43 the autoregulatory mechanism. We identified several splicing factors, including SF2, Rbp1 and Sf3b1, as genetic modulators of TDP-43 production. We have also shown that modulation of TCERG1 expression levels affect TDP-43 production levels in flies. Finally, we found that FTLD/ALSlinked TDP-43 mutations do not alter TDP-43’s ability to self-regulate its expression and consequently of the homeostasis of TDP-43 protein levels.