Remodelage membranaire par des protéines dans le trafic intracellulaire : une approche par systèmes modèles des rôles des golgines dans la structuration du Golgi et des SNAREs dans la fusion synaptique

Biological membranes play a crucial role in cells. Over the past few decades, they have been recognized not only as barriers that define cellular compartments, but also as transducers, reactors, and sensors. These membranes are highly complex two-dimensional structures. Their composition, with lipids, proteins, and sugars, is variable and depends on their specific functions. They are also very dynamic, which is particularly evident during intracellular trafficking. This tightly regulated process ensures the transport of proteins, lipids, and other molecules between organelles, notably through membrane-enveloped cargos called vesicles. The formation and fusion of these sacs constantly cause deformation of organelles membrane, as well as import and export of membrane material. Surprisingly enough, the system remains globally stable despite these constant fluxes. Although many components of the molecular machineries involved in this lively yet stable ballet have been identified, precise operational molecular mechanisms remain unclear. Traditional approaches in cellulo to studying protein players in intracellular trafficking are limited by the interdependence of multiple pathways, making it hard to untangle complex interactions. A complementary approach, used in our lab, is bottom-up reconstruction. By using minimal and controlled systems, we aim ta model these phenomena from scratch and understand the contributions of individual molecular effectors. This thesis focuses on two distinct intracellular trafficking processes. First, we studied the organization of the Golgi apparatus, a key site for sorting and modifying proteins alter synthesis. The unique structure of the Golgi, resembling a stack of pancakes, is remarkably resilient. lt disassembles and reforms during cell division and remains stable during interphase, despite continuous material exchange. We hypothesized that certain proteins, such as GM130, actas scaffolds to maintain this structure. Using giant unilamellar vesicles (GUVs) as a model, we studied GM130 and showed that its phase separation on membrane is driven by a 200-aminoacid segment near its N-terminus. This segment is predicted to form a coiled-coil structure. This is surprising because liquid-liquid phase separation generally involves intrinsically disordered domains. We also explored the mechanical properties that the presence of the protein could give to the membrane. Although experimental challenges limited our ability ta fully characterize these properties, we observed that protein crowding, for densities similar to what can be found on Golgi membranes and regardless of protein type, can lead ta membrane plasticity in our model system. The second process studied is synaptic fusion. SNARE proteins are the main drivers of vesicle fusion with target membranes. Using simplified models and rule-of-thumb calculations, we identified bottlenecks in SNARE-mediated fusion and highlighted the role of the additional protein effectors known to catalyze the process in vivo. Furthermore, we used a suspended membrane set up and a bulk assay to characterize mutant SNAREs. Our goal was to test a more complex theoretical model of fusion. We found that some results of the model needed to be revisited due to a misinterpretation of the experimental data it was based on. Hence the experiments were discontinued. However, this work provided valuable insights into the model, thereby highlighting the importance of close collaboration between theoretical and experimental approaches. Overall, this thesis deepens our understanding of Iwo proteins involved in different intracellular trafficking pathways. It also highlights the value of in vitro assays, which we improved during the course of this work, and paves the way for future research. On a personal note, it was also a humble harvest of a few epistemological take-aways.