Several functions have been attributed to the cellular prion protein, PrPc, amongst which its anti-oxidant role has rapidly been gaining interest in the recent years. We and others have previously shown, that PrPc expressing cells, of neuroblastoma or epithelial origin, seem to exhibit a higher overall viability towards paraquat toxicity than cells expressing basal or low levels of the protein. Although several studies propose a protective mechanism that involves PrPc dependent activation of the superoxide dismutase (SOD) enzymatic machinery or an activation of its own intrinsic antioxidant function, others argue against this SOD-like role. Our objective was to investigate, at a biological and proteomic level, by which potential mechanism PrPc could protect neuroblastoma cells against paraquat induced oxidative damage. Using a biological aproach, we firstly evaluated the status of the Cu/Zn-SOD enzyme in Human neuroblastoma cells expressing different forms of PrPc following their exposure to paraquat. Next, we performed a proteomic study to investigate by which other potential mechanism(s), PrPc could protect the cell against paraquat induced oxidative stress. Our proteomic approach made use of an optimised two-dimensional liquid chromatography system, the ProteomeLab PF-2D, and reverse phase chromatography coupled with lava purple stained SDS-PAGE, both interfaced with tandem mass spectrometry. An interesting aspect of our study has been the development of an original immunoproteomic technique called immuno-PF2D-MS/MS, coupling classical immunological methods to a two-dimensional liquid chromatography proteomic tool interfaced with tandem mass spectrometry. We have proposed this technique for antigenic and serological characterization that have important implications in the study of biomarkers. Another important aspect of our study has been the detection of several candidates that could participate in PrPc-mediated protection against paraquat induced oxidative stress. Although, it was out of our scope to investigate each of these candidates in the present study, it presents an interesting perspective for future studies. We have, however, shown the implication of one such candidate: PARP-1. Complimentary tests will be necessary in the future to confirm the actual interaction of this candidate with PrPc.

La sclérose latérale amyotrophique (SLA) et la démence fronto-temporale (DFT) sont des maladies neurodégénératives fatales, actuellement sans traitement. Ces maladies entrainent la dégénérescence des neurones moteurs et corticaux, engendrant des troubles moteurs et cognitifs et ultimement menant à la mort des patients souvent par détresse respiratoire trois à cinq ans après l’apparition des premiers symptômes. À l’échelle d’une vie, le risque de développer ces pathologies est de 1 pour 300-400 pour la SLA et 1 pour 742 pour la DFT, faisant de ces pathologies un risque majeur. Avec le vieillissement de la population que nous connaissons actuellement, il est évident que l’incidence de ces maladies deviendra de plus en plus élevée. Ainsi il est essentiel de comprendre les mécanismes moléculaires sous-jacents à ces pathologies dans le but de développer des thérapies effectives et prévenir l’impact de ces pathologies dans notre société. À ce jour, l’étiologie de la SLA-DFT est encore débattue, cependant la communauté scientifique s’accorde sur le fait que l’interaction entre la génétique et l’environnement joue un rôle essentiel dans le développement de ces maladies. La caractéristique moléculaire principale de ces pathologies est la localisation cytoplasmique de la protéine, normalement, nucléaire TDP-43. TDP-43 est un régulateur clef de l’homéostasie des ARNs. Parmi ces nombreuses fonctions, TDP-43 régule la formation des granules de stress, en régulant leur protéine régulatrice G3BP1. Ces granules formés d’ARN et de protéines se forment pour protéger les cellules durant une période de stress. Récemment, ces granules ont fait l’objet de nombreuses études et leurs dysfonctions ont été associées à la SLA-DFT. Dans cette thèse, nous avons approfondi l’étude de la régulation de TDP-43 sur G3BP1. Nous avons défini que TDP-43 stabilise les transcrits de G3BP1 de par une liaison forte à une séquence conservée à travers l’évolution se situant dans le 3’UTR de G3BP1. La perte de localisation nucléaire, la présence de mutations ou de TDP-35, une isoforme pathologique de TDP-43, sont associées à une diminution des niveaux de G3BP1. Également, d’un point de vue histopathologique, dans le cortex orbitofrontal des patients atteints de SLA-DFT, les neurones présentant une localisation cytoplasmique de TDP-43 ont une perte des niveaux transcriptionnels de G3BP1, associant alors directement G3BP1 à la maladie. Par la suite, nous avons défini que la perte de fonction en tant que stabilisateur, permet la liaison de microARNs sur les transcrits de G3BP1, engendrant leur dégradation. Le blocage de la liaison de microARNs sur G3BP1 empêche la dégradation des transcrits et restaure les fonctions de la protéine. Ainsi, nous avons déterminé un moyen de contrer la perte de fonction de TDP-43 sur G3BP1. De façon intéressante, en plus de la formation des granules de stress, G3BP1 est essentielle pour l’homéostasie neuronale et la survie neuronale post-stress. Par conséquent, la restauration de la protéine est potentiellement une avenue thérapeutique multi-approche pour le traitement de ces maladies. / Amyotrophic lateral sclerosis (ALS) and frontotemporal dementia (FTD) are two fatal neurodegenerative diseases, currently without cure. These diseases lead to the degeneration of motor and cortical neurons, causing motor and cognitive disorders and ultimately leading to death, often from respiratory distress three to five years after the onset. Over a lifetime, the risk of developing these conditions is 1 in 300-400 for ALS and 1 in 742 for FTD, making these conditions a major risk. With the current aging of the population, it is evident that the incidence of these diseases will become increasingly high. It is therefore essential to understand the molecular mechanisms underlying these pathologies in order to develop effective therapies. To this day, the etiology of ALS-FTD is still debated. However, the scientific community agrees that the interaction between genetics and the environment play an essential role in the development of these diseases. The main molecular characteristic of these pathologies is the cytoplasmic localization of the normally nuclear protein TDP-43. TDP-43 is a key regulator of RNA homoeostasis. Among these many functions, TDP-43 regulates the formation of stress granules, by regulating their nucleator protein G3BP1. These granules of RNA and protein form to protect cells during times of stress. Recently these granules have been the subject of several studies and their dysfunction has been associated with ALS-FTD. In this thesis, we have deepened the study of the regulation of TDP-43 on G3BP1. We have defined that TDP-43 stabilizes G3BP1 transcripts by strong binding to a sequence conserved through evolution located in the 3'UTR of G3BP1. Loss of nuclear localization, the presence of mutations or of TDP-35, a pathological isoform of TDP-43, are associated with decreased levels of G3BP1. Also, histopathologically, in the orbitofrontal cortex of patients with ALS-DFT, neurons with cytoplasmic localization of TDP-43 have a loss of transcriptional levels of G3BP1, directly associating G3BP1 with the disease. Subsequently, we defined that TDP-43 loss of function as a stabilizer allows the binding of two microRNAs on the G3BP1 transcripts, causing their degradation. Blocking the binding of these microRNAs to G3BP1 prevents the degradation of the transcripts and restores the functions of the protein. Thus, we have determined a way to counter the loss of function of TDP-43 on G3BP1. Interestingly, in addition to the formation of stress granules, G3BP1 is essential for neuronal homoeostasis and post-stress neuronal survival. Therefore, the restoration of the protein is potentially a multi-approach therapeutic avenue for the treatment of these diseases.