La protéine Rnd3 est un membre atypique de la famille des Rho GTPases. Dénuée d’activité GTPasique, elle est ainsi constitutivement sous forme active et liée au GTP. La régulation de cette protéine ne passe donc pas par le cycle classique des Rho GTPases mais par d’autres mécanismes transcriptionnels, post-transcriptionnels ou encore traductionnels. Dans le carcinome hépatocellulaire (CHC), Rnd3 est significativement sous-exprimée, et cette sous-expression procure un avantage invasif aux hépatocytes. Ce projet de thèse avait pour objectif de déterminer plus précisément les mécanismes à la base de la régulation de Rnd3 dans les cellules tumorales, et notamment les cellules dérivées de carcinome hépatocellulaire. Les travaux de cette thèse ont été divisés en deux axes principaux. Dans une première partie, la régulation de Rnd3 par la β-caténine a été étudiée. En effet, la β-caténine est retrouvée mutée dans plus d’un tiers des CHC, et la présence de mutations activatrices de la β-caténine corrèle avec un faible niveau d’expression de Rnd3 dans les CHC. L’établissement d’un modèle original dans les cellules de CHC, HepG2, a permis d’étudier indépendamment l’implication de la β-caténine sauvage et la β-caténine mutée dans la régulation d’expression de Rnd3. Ce modèle a permis de mettre en évidence une régulation différentielle de Rnd3 par les deux formes de la β-caténine, la forme sauvage régulant Rnd3 au niveau transcriptionnel, et la forme mutée régulant Rnd3 au niveau post-transcriptionnel. La deuxième partie de ce travail, qui constitue la partie principale du projet, s’est intéressée à la régulation de Rnd3 par la voie de mécanotransduction MRTF/SRF. L’activation de cette voie de signalisation est très intimement reliée à l’organisation du cytosquelette d’actine, et cette voie régule en retour l’expression de nombreux gènes impliqués dans la dynamique de l’actine. Les résultats obtenus ont permis de déterminer Rnd3 comme une nouvelle cible directe de la voie MRTF/SRF dans les cellules tumorales, et placent Rnd3 au centre d’une boucle de régulation de cette voie de mécanotransduction. L’ensemble des résultats obtenus au cours de ce projet de thèse ont permis de mieux caractériser la régulation de l’expression de Rnd3 dans les cellules tumorales. / Rnd3 protein is an atypical member of the Rho GTPase family, devoid of GTPase activity and constitutively active and bound to GTP. Rnd3 regulation does not occur through the classical GTPase cycle but is achieved at transcriptional, posttranscriptional or translational level. Rnd3 is underexpressed in hepatocellular carcinoma (HCC), and this down-regulation increases HCC cell invasion and is linked to HCC progression. The aim of this thesis project was to better decipher the mechanisms involved in Rnd3 expression in tumor cells, and particularly in HCC cells. In a first part, Rnd3 regulation by β-catenin was studied. β-catenin is found mutated in one third of HCC, and activating β-catenin mutations in human HCC correlates with the lowest levels of Rnd3. An original model established in HepG2 cells allowed the study of the involvement of WT β-catenin versus mutated β-catenin in the regulation of Rnd3 expression.Interestingly, our results demonstrated that both forms of ß-catenin independently regulate Rnd3 mRNA expression. The WT β-catenin regulates Rnd3 at the transcriptional level, whereas the mutated β-catenin acts through the 3’UTR of Rnd3 mRNA. The second and main part of this thesis project was the study of the regulation of Rnd3 expression by the mechanotransduction pathway MRTF/SRF. The activation of this signaling pathway is tightly regulated by actin cytoskeleton, and the MRTF/SRF pathway directs in return the expression of a huge number of genes involved in actin dynamics. Our results uncovered Rnd3 as a new direct target of MRTF/SRF pathway in tumor cells. Indeed, upon actin dynamics changes, MRTF/SRF is able to bind Rnd3 promoter in order to favor its expression. As Rnd3 also acts as a regulator of the actin cytoskeleton, our results highlight Rnd3 at the center of a feedback loop of the MRTF/SRF mechanotransduction pathway. Taken together, all of the results obtained helped to better decipher the mechanisms of Rnd3 regulation in tumor cells.

Membrane protein mechanotransduction is the altered function of an integral membrane protein in response to mechanical force. Such mechanosensors are found in all kingdoms of life, and increasing numbers of membrane proteins have been found to exhibit mechanosensitivity. How they mechanotransduce is an active research area and the topic of this thesis. The methodology employed is classical molecular dynamics (MD) simulations. MD systems are complex, and two programs were developed to reduce this apparent complexity in terms of both visual abstraction and statistical analysis. Bendix detects and visualises helices as cylinders that follow the helix axis, and quantifies helix distortion. The functionality of Bendix is demonstrated on the symporter Mhp1, where a state is identified that had hitherto only been proposed. InterQuant tracks, categorises and orders proximity between parts of an MD system. Results from multiple systems are statistically interrogated for reproducibility and significant differences at the resolution of protein chains, residues or atoms. Using these tools, the interaction between membrane and the Escherichia coli mechanosensitive channel of small conductance, MscS, is investigated. Results are presented for crystal structures captured in different states, one of which features electron density proposed to be lipid. MD results supports this hypothesis, and identify differential lipid interaction between closed and open states. It is concluded that propensity for lipid to leave for membrane bulk drives MscS state stability. In a subsequent study, MscS is opened by membrane surface tension for the first time in an MD setup. The gating mechanism of MscS is explored in terms of both membrane and protein deformation in response to membrane stretch. Using novel tension methodology and the longest MD simulations of MscS performed to date, a molecular basis for the Dashpot gating mechanism is proposed. Lipid emerges as an active structural element with the capacity to augment protein structure in the protein structure-function paradigm.