Influenza A virus (IAV) is a major threat to human health with seasonal epidemics, occasional pandemics and emergence of new highly pathogenic strains from the animal reservoir. Our laboratory has shown that the human Annexin A6 (AnxA6) interacts with the IAV M2 proton channel and limits production of progeny IAV from infected cells. We have found that overexpression of AnxA6 impairs morphogenesis and release of progeny viruses. These findings are supported by another study showing that AnxA6 has a critical role in the late endosomal cholesterol balance and affects IAV replication and propagation in AnxA6-overexpressing cells. However, the molecular mechanism responsible for restriction of IAV morphogenesis by AnxA6 is still unclear. AnxA6 is a calcium-dependent phospholipid-binding protein which plays a major role in cellular events such as regulation of cholesterol homeostasis and membrane organisation or repair. AnxA6 is also implicated in the regulation of intracellular signalling pathways required for IAV infection. In this study, we used a combination of virology, cellular biology and biochemistry approaches to decipher the restriction mechanism of IAV by human AnxA6. We found that AnxA6 down-regulates M2 viral protein expression and impairs viral morphogenesis and budding. We also found that AnxA6 regulates chemokines and cytokines expression during viral infection, suggesting that AnxA6 triggers an innate immune response to IAV by modulating signalling pathways required for viral replication. Finally, we observed that IAV down-regulates AnxA6 expression at mRNA level during early stages of infection and at protein level during late infection, suggesting that IAV has developed a strategy to respond to AnxA6 restriction mechanism during viral infection. We conclude that it is essential to better understand the interaction between human AnxA6 and IAV to elucidate the potential of AnxA6 as an antiviral candidate.

Bacterial genomes provide records of the molecular processes associated with emergence and evolution of different bacterial lifestyles. This thesis is based on whole-genome comparisons within the genus Bartonella, an excellent model system for studies of host- and vector-specificity and infection outcome in animal-associated bacteria. The louse-borne human specialist and trench fever agent Bartonella quintana was contrasted to the flea-borne generalist relatives Bartonella henselae and Bartonella grahamii, which cause asymptomatic infection in cat and mouse respectively. While B. henselae is commonly isolated from humans, and causes cat scratch disease, there is only one reported case of B. grahamii human infection. The gene complements of the three species are nested like Russian dolls with the smaller genome (B. quintana) being entirely contained in the medium sized (B. henselae), which in turned is contained in the largest (B. grahamii). Size differences reflect differences in the horizontally and vertically acquired gene content, and in the number of genus- and species- specific genes, owing to differential impact of bacteriophages and plasmids, and to different degrees of genome decay. These processes can be attributed to the three distinct lifestyles. Comparisons with other alpha-proteobacteria suggest that the Bartonella genus as a whole evolved from plant-associated species, and that horizontal transfer, in particular of genes involved in interaction with the host, played a key role in the transition to animal intracellular lifestyle. The long-term genome decay associated with this lifestyle is most advanced in the host-restricted B. quintana. The broad host-range species B. grahamii has the largest genome and the largest proportion of auxiliary DNA of the three, probably because it has access to a larger gene pool. In encodes all the known pathogenicity determinants found in the genomes of B. henselae and B. quintana, suggesting that these genes primarily evolved to facilitate colonization in the reservoir host.

HIV-2 differs from HIV-1 in that many infected people experience normal survival, whilst only 20% progress rapidly to AIDS. Understanding mechanisms of delayed HIV-2 disease progression could provide new insights into HIV control. The Caio Community Cohort was established in Guinea-Bissau in the setting of high HIV-2 prevalence. This thesis investigates the role of polymorphic host restriction factors of the TRIM family in HIV-2 outcome. TRIM proteins are a family of E3 ubiquitin-ligases, where closely-related TRIM5α and TRIM22 are thought to inhibit HIV-1 transcription, uncoating and budding. There was an association between TRIM5α amino acid substitution R136Q and reduced HIV-2 viral load/prolonged survival. Conversely, P479L was enriched among HIV-2 infected participants and progressors with CD4+ T cell decline. TRIM22 was highly polymorphic in this cohort, revealing three novel coding variants. Although most substitutions were located in the putative virus-interacting PRYSPRY domain, two in the coiled-coil, D155N and R242T, showed significant and divergent associations with survival. R242T was enriched in HIV-2 infected participants, who progressed to death at twice the rate of wild-type controls. In silico studies predicted D282, D360, and R321 of TRIM22 to be highly conserved, exposed residues, for which polymorphisms would be deleterious. When aligned with sequences from the potent HIV-1 restriction factor, rhesus macaque TRIM5α, TRIM22 substitutions R321K, T415I, and D360Y were spatially relevant to residues involved in HIV-1 restriction. The role of TRIM22 in HIV restriction was supported by in vitro pilot studies showing that TRIM22 was upregulated by HIV-1 infection in a lymphoid cell line and co-localised with the HIV-1 capsid protein p24. Overexpression of TRIM22 resulted in the restriction of VSV-G pseudotyped HIV-1 and SIVmac. The R242T substitution diminished TRIM22's restriction of HIV-1 and SIVmac: protein analysis suggested that this may be due to the inability of the R242T mutant to fully dimerise.

L’infection par le virus de l’hépatite B (HBV) est problème majeur de santé publique mondial car, en dépit d’un vaccin efficace, les traitements curatifs actuels ne permettent pas l’élimination complète du virus. Comprendre les mécanismes de réplication du virus et son rôle dans la survenue du cancer hépatocellulaire (CHC) reste un enjeu majeur.Le rôle de la protéine HBx dans l’infection par HBV et l’oncogenèse viro-induite, reste mal connu, malgré un grand nombre de publications, car les fonctions décrites jusqu'alors sont limitées à des contextes d’études particuliers, souvent loin des conditions physiologiques.Mes travaux de thèse reposent sur l’utilisation de modèles d’études proches de la physiologie naturelle d’une infection par HBV, notamment des cellules primaires infectables in vitro. J’ai pu démontrer lors de mon étude que HBx est indispensable à la réplication de HBV, et qu’il agit essentiellement via son interaction avec DDB1 pour contrer la restriction du virus due au complexe SMC5/6, en induisant sa dégradation. Ce facteur de restriction permet de bloquer la transcription de l’ADN viral au niveau épigénétique. Ce nouveau rôle inattendu de SMC5/6 ouvre de nombreux axes de recherche, notamment sur les mécanismes de restriction des virus à ADN épisomal. SMC5/6 est connu pour son implication dans les voies de réparation de l’ADN : la dernière partie de ce manuscrit montre que sa dégradation dans les cellules infectées, altère ces mécanismes et sensibilise les cellules aux dommages à l’ADN, induits notamment par la radiothérapie. La présence de HBx dans les CHC pourrait ainsi s’avérer un atout pour le traitement du CHC / Hepatitis B virus (HBV) infection is a major health problem worldwide as (1) despite an effective preventive vaccine over 240 million individuals are chronically infected and (2) the actual viral suppressive treatments available do not eliminate viral DNA from cells. Thus, infected individuals are at a high risk of developing hepatocellular carcinoma (HCC) and understanding viral replication mechanisms and how it impacts on hepatocarcinogenesis is a major challenge.The role of the HBx protein, notably in viral replication and oncogenic processes, is the subject of many publications. However, many studies have often used non-physiological infection conditions. My thesis project has addressed these limitations by using cellular models, including primary human hepatocytes which can be infected by HBV, to investigate HBx’s role in these processes. I have shown that HBx is indispensable for HBV replication and that HBx associates with the infected cell’s DDB1/ E3 ubiquitin complex to target its Smc5/6 complex for degradation via the proteasome. These studies have identified that the Smc5/6 complex is a novel viral restriction factor that acts at an epigenetic level to block viral replication. This unexpected role of SMC5/6 has led to new research into the evolutionary conservation of restriction factors for episomal DNA viruses. As SMC5/6 is implicated in DNA Damage Repair (DDR), the last section of my thesis reports how SMC5/6 degradation in infected cells can sensitise cells to the cell killing effects of DNA damaging agents such as ionizing radiation and hydroxyurea. These results open-up possibilities for HCC treatment where HBx expression may be of therapeutic benefit