Vers la caractérisation structurelle du complexe ribonucléoprotéique du virus de l'hépatite D

Hepatitis D is an inflammatory liver disease caused by the hepatitis D virus, which relies on the hepatitis B virus for self-replication as it uses the HBV envelope proteins to form viral particles. About 5% of patients chronically infected with HBV are also infected with HDV, affecting between 15 and 20 million people worldwide. Although treatments are available, they do not completely eradicate the virus, highlighting the urgent need for further research.The Hepatitis D antigen, encoded by a genome of approximately 1700 nucleotides, plays a critical role in the replication of the virus. This antigen, which exists in both small and large forms, and the genome are packaged into virions as a ribonucleoprotein. However, the mechanism by which HDAg supports replication is still unclear. This work focuses on nuclear magnetic resonance studies of the different forms of HDAg produced by cell-free synthesis or in bacteria to study the RNP, providing detailed structural information at atomic resolution.First, we expressed the whole S-HDAg protein in a cell-free system. As this protein is completely insoluble, we were able to study it by using solid-state NMR spectroscopy. Cross-polarization experiments detected less than 60 rigid residues, whereas Insensitive nuclei enhanced by polarization transfer experiments revealed highly mobile residues, with most of them showing no overlap with CP signals. To deal with the complexity of S-HDAg, we implemented a 'divide-and-conquer' approach, dividing S-HDAg into essential subdomains. I studied separately the protein assembly domain, containing the first 60 residues, termed HDAg1-60, and the RNA binding domain, containing the following residues (61-195), termed S-HDAgΔ60. As the S-HDAgΔ60 protein is soluble, we were able to study it by solution NMR and determine its three-dimensional structure. S-HDAgΔ60 shows two intrinsically disordered protein segments separated by a structured helix-loop-helix motif, and is conserved in the full-length protein. Next, we determined the interaction between S-HDAgΔ60 and viral RNA1.7kb synthesised in vitro by individually assessing the binding affinity of the residues involved. Our results indicate specific regions within the helices that interact with RNA1.7kb, with affinities in the micromolar range, while disordered regions have little or no involvement in the interaction. These data allowed us to perform directed docking and to propose a structural model of the S-HDAgΔ60-RNA1.7kb complex.We then studied the HDAg1-60 assembly domain, which, like the whole protein, was insoluble and therefore studied by solid-state NMR. Using fast magic-angle spinning proton detection and selectively labelled samples, we were able to sequentially assign a large part of this domain. The CP spectra of HDAg1-60 showed a rigid portion similar to that of the whole S-HDAg.Finally, we looked at a form of S-HDAg truncated at the C-terminus by 35 amino acids, called HDAg1-160 (residues 1-160), which is the closest form to the full protein while remaining soluble. This protein can be studied by both solution NMR, as it is soluble, and solid-state NMR, as it contains the assembly domain that allows it to be sedimented in NMR rotors. In this thesis I have therefore used HDAg1-160 and viral RNA1.7kb to study the conformation of the HDV RNP. This analysis revealed changes in the NMR spectra upon binding of the RNA1.7kb to HDAg1-160. In addition, the formation of the complex was monitored by atomic force microscopy. Preliminary results suggest that the RNA1.7kb may rigidify specific protein conformations. My work has provided information on the structure and dynamics of the different domains of the HDV proteins and their interaction with nucleic acids using solution and solid-state NMR. These results will provide a better understanding of the formation of the RNP, the central component of the hepatitis D virus.