La protéine Promyelocytic Leukemia (PML) est impliquée dans de nombreux processus cellulaires, et identifiée comme un suppresseur de tumeur. Cette protéine est le composant structural des Corps Nucléaires PML (CNs-PML) dont l'intégrité, compromise dans certaines leucémies, dépend strictement de sa SUMOylation. Ce projet de thèse visait à identifier de nouveaux régulateurs des CNs-PML, et par extension de la voie SUMO, en utilisant comme 'read-out' l'anatomie des CNs-PML, laquelle est extrêmement sensible au niveau de SUMOylation cellulaire globale. Ces travaux sont basés sur un criblage siARNs à grande échelle qui a conduit à l'identification de deux candidats SKP1et RBX1, tous deux faisant partie intégrante d'un complexe d'ubiquitine E3 ligase appelé " SKP1-CUL1-F-Box containing complex " (SCF). Nous avons pu démontrer l'implication de SKP1 et RBX1 dans la stabilité de la protéine PML avec des expériences de gain et perte de fonction. Nous avons également identifié FBXO9 comme la protéine F-Box capable de reconnaitre spécifiquement PML, causant son ubiquitination par le complexe SCFFBXO9, suivie de sa dégradation par le protéasome. En revanche, le site d'interaction de FBXO9 sur PML -tout comme la kinase impliquée dans ce processus de reconnaissance- restent encore à identifier. PML étant dégradé dans de nombreux cancers, il apparait essentiel d'avoir une meilleure compréhension des mécanismes post-traductionnels menant à sa dégradation. Ces travaux devraient à long terme permettre de révéler de nouveaux régulateurs des CNs-PML, et potentiellement permettre le développement de nouvelles stratégies thérapeutiques, visant à moduler les CNs-PML dans la cellule tumorale. / ProMyelocytic Leukemia (PML) protein is implicated in a number of key cellular processes, and was identified as a tumor suppressor. This protein is one of the main structural components forming the PML Nuclear Bodies (PML-NBs) whose integrity -compromised in some leukemias- is strictly dependent on PML SUMOylation. The goal of this thesis project was to identify new regulators of PML Nuclear Bodies, and by extension of the SUMO pathway, using PML-NBs, which are extremely sensitive to global cellular SUMOylation level, as a read out. This work is based on a high throughput siRNA screen, which led to the identification of two proteins, SKP1a and RBX1, which are both part of an Ubiquitin E3 ligase complex called SKP-Cullin-F-Box containing complex (SCF). We were able to show the involvement of SKP1 and RBX1 in PML protein stability through gain and loss of function experiments. We also identified FBXO9 as the F-Box capable of specifically recognizing PML, causing its ubiquitination by SCFFBXO9 complex and subsequent degradation by the proteasome. However, FBXO9 site of interaction on PML and the identity of the kinase implicated in this recognition processes are yet to be discovered. PML being degraded in numerous cancers, it is essential to acquire a better understanding of post-translational mechanism leading to the degradation of this tumour suppressor. In the long term, this work should, allow the discovery of new PML Nuclear Body regulators and potentially allow the development of new strategies aiming to modulate PML Nuclear Bodies in tumoral cells.

The dissertation deals with a cell response to genotoxic stress, specifically to anti-cancer treatments with a genotoxic mechanism of action. In principle, cells can respond to these perturbing stimuli in several ways: in case of severe DNA damage, they usually undergo apoptosis or enter senescence. In case of minor DNA damage, or upon defective checkpoint mechanisms, they may continue the cell cycle, either with successfully repaired DNA or with mutations of various kind. Thanks to selection pressure, the mutations that provide cells with a certain growth advantage under conditions of continuing genotoxic stress, gradually accumulate and render the tumor treatment-resistant. In my thesis, I focus on several aspects of this whole process. First, I participated in a characterization of a radioresistant and anoikis-resistant population of prostate cancer cells. This population was generated by irradiating cells 35 times by 2 Gy, a regime used in clinics. After this treatment, a population of low-adherent cells emerged that demonstrated increased expression of EMT- and stem cell markers. The low-adherent state of these cells was maintained by Snail signaling and their anoikis resistance by ERK1/2 signaling. Interestingly, after a protracted period of time, these cells were able to re-adhere and...