The use of antiretroviral therapy (ART) for HIV-1 treatment effectively suppresses viral replication if managed appropriately, but sometimes patients experience incomplete viral suppression with manifestation of either persistent low-level viremia (LLV) or discernible virological failure (VF). In the present study, potential risk factors associated with LLV and VF were investigated in a cohort of HIV-1 infected Kenyan children receiving ART. Drug resistant (DR) variants in children with or without LLV were examined using a next-generation sequencing-based HIV DR typing protocol. The potential association between HIV DR mutations (DRMs) and LLV and/or VF was then examined. To measure the potential impacts from other clinical and epidemiological confounding factors, a database comprising of epidemiological and clinical information from this patient cohort was established and sanitized for ensured accountability. Statistically significant correlations between the examined factors and LLV or VF were determined using chi-square test, Kaplan-Meier survival analysis, and Cox proportional hazard models. Of 293 examined patients, 20% had LLV and 22% of the selected patients progressed to VF with no significant association observed between LLV and VF. ART adherence during therapy, baseline CD4 counts, DRMs at LLV, WHO clinical stage, gender, ART therapy stage (1st/2nd line), ART drugs and co-morbidities were not significantly associated with LLV, whereas, the ART adherence, CD4 counts and co-infection with pneumonia was significantly associated with VF. This study highlights the factors predictive of VF, and the relevance of maintaining LLV in HIV-infected children. / February 2017

Le VIH-1 est un rétrovirus contenant deux copies de son génome ARN simple brin positif. Dans la cellule, son ARN est rétrotranscrit en ADN puis cet ADN est intégré dans le génome cellulaire. L'ADN viral est ensuite transcrit en ARN dont la moitié évite l'épissage. Dans le cytoplasme, cet ARN possède une double fonctionnalité : il sert d'ARNm pour la synthèse des protéines Gag et Gag-Pol et d'ARN génomique (ARNg) qui est encapsidé sous forme de dimère. Malgré sa faible abondance dans la cellule, l'ARNg est préférentiellement encapsidé. Cependant, il a été montré par nous et d'autres que d'autres ARN non-génomiques peuvent également être spécifiquement encapsidés, tels que des ARN viraux épissés et certains ARN cellulaires, bousculant ainsi le dogme d'une spécificité unique pour l'ARNg. Nous avons montré que le VIH-1 contrôlerait l'incorporation des ARN viraux, épissés et non épissés, par un mécanisme de compétition, impliquant des facteurs communs. En revanche, les ARN cellulaires 7SL et U6 semblent encapsidés par des mécanismes indépendants. Afin d'identifier les déterminants qui régissent la spécificité d'encapsidation, nous avons testé le rôle de la nucléocapside (NC) qui est fortement impliquée dans l'encapsidation. Nous montrons ici que de façon tou t à fait inattendue, des mutations de NC conduisent à l'encapsidation d'ADN et augmentent également l'encapsidation d'ARN viraux épissés. Nous avons ensuite cartographié les déterminants de la région 5' UTR de l'ARNg et déterminé que la tige-boucle polyA et la région PBS sont impliquées dans l'encapsidation des ARN viraux épissés. Ce travail apporte une meilleure compréhension de la spécificité d'encapsidation, primordiale pour la mise au point de thérapies géniques utilisant des vecteurs lentiviraux. / HIV-1 is a retrovirus containing two copies of its single stranded positive RNA genome. In the cell, its RNA is reverse transcribed into DNA and this DNA is integrated into the cellular genome. The viral DNA is then transcribed into RNA. One moiety of this RNA pool escapes splicing. Once in the cytoplasm, this RNA has a double function: it serves as mRNA for synthesis of Gag and Gag-Pol proteins and as genomic RNA (gRNA) that is packaged as a dimer. Despite its low abundance in the cell, the gRNA is preferentially encapsidated into virions. However, it has been shown by us and others that other non-genomic RNA can be specifically packaged, such as spliced viral RNA and some cellular RNA, thus shaking up the dogma of a unique specificity for the gRNA. We have shown that HIV-1 might control the incorporation of the spliced and unspliced RNA by a mechanism of competition, involving common factors. In contrast, cellular RNA 7SL and U6 seem encapsid ated through independent mechanisms. To identify the determinants that govern the specificity of encapsidation, we tested the role of the nucleocapsid (NC) which is strongly involved in the packaging. Here we show that unexpectedly, mutations of NC lead to the encapsidation of DNA and also increase the encapsidation of spliced viral RNA. We then mapped the determinants in the 5' UTR of the gRNA and determined that the polyA stem-loop and the PBS region are involved in the encapsidation of the spliced viral RNA. This work provides a better understanding of the specificity of encapsidation that is crucial for the development of gene therapy using lentiviral vectors.

Les virus sont des parasites intracellulaires obligatoires, qui détournent la quasi-totalité des voies cellulaires. La voie des miRNA et du RNAi ne font pas exception. D'abord, les miRNA peuvent reconnaître les ARN viraux, permettant le recrutement de la machinerie du RNAi, en particulier AGO2, sur les messagers viraux, ce qui peut moduler la réplication du virus. Pendant ma thèse, nous avons identifié un nouveau moyen de recruter AGO2, sur les messagers viraux, qui n'impliquent pas les miRNA, ni sa capacité à induire l'extinction des gènes. Nous avons montré qu'AGO2 interagit avec GAG et se fixe aux ARN viraux par les séquences d'encapsidation. Ensuite, les virus peuvent moduler le répertoire de miRNA cellulaires, de sorte à créer un contexte favorable à sa propre réplication. Ainsi, nous avions pour objectif, d'identifier de nouveaux partenaires cellulaires de VIH. Nous avons alors analysé des données transcriptomiques, obtenues à partir de cellules infectées par VIH-1 ou VIH-2, et reconstitué des réseaux de régulations impliquant les facteurs de transcription et les miRNA. Nous avons montré que les modulations de miRNA dépendent du mode d'entrée du virus, en particulier de l'utilisation des co-récepteurs. De plus, l'approche de Biologie Intégrative que nous avons suivie, nous a permis de caractériser une nouvelle protéine cellulaire, capable de réguler l'expression du VIH et de restreindre sa réplication. / Viruses are obligatory intracellular parasites that hijack many, if not all, cellular pathways. The RNA interference (RNAi) and the micro(mi)RNA pathways are no exceptions. First, cellular micro(mi)RNAs are able to recognize viral RNAs through imperfect micro-homologies. Similar to the miRNA-mediated repression of cellular translation, this recognition is thought to tether the RNAi machinery, in particular Argonaute(AGO)2, on viral messengers and eventually to modulate virus replication. During my PhD, we have unveiled another pathway by which AGO2 can interact with retroviral mRNAs without involving host miRNAs and translation repression. We have shown that AGO2 interacts with the retroviral GAG core proteins and preferentially binds unspliced retroviral RNAs through the RNA packaging sequences. The interaction between AGO2 and GAG, observed with both the Human Immunodeficiency Virus 1 (HIV-1) and the Primate Foamy Virus 1 (PFV-1), facilitates GAG multimerization and retroviral particle formation. Second, viruses modulate the miRNA repertoire presumably to create favorable conditions for viral replication. Hence, in order to identify novel cellular partners of HIV, we have analyzed transcriptomics data obtained from HIV1 and HIV-2-infected cells and reconstituted Transcription Factor- and miRNA-based regulation networks. Strikingly, we have noticed that the modulations of the transcriptome (coding and non-coding RNAs) depend on the mode of entry of the virus (i.e. co-receptor usage). Our in silico approach also helped us characterize a novel cellular protein able to regulate virus gene expression and

La protéine Nef du VIH-1 joue un rôle important dans la pathogenèse du VIH-1 en modulant les voies de signalisation de la cellule hôte. La signalisation par le TcR est essentielle à la sélection positive pour générer les cellules simples positives (SP) CD4+ et simples positives (SP) CD8+, processus largement dépendant de l’activité de la Src kinase Lck et de son habileté à lier la queue cytoplasmique des corécepteurs CD4 et CD8. Nous avons précédemment trouvé que l’expression de Nef dans le VIH ou VIS peut induire une sévère déplétion des thymocytes et une baisse d’expression du corécepteur CD4 à la membrane. Nous avons également montré que Nef bloque la génération des thymocytes doubles positifs (DP) CD4+ CD8+ en plus d’altérer la transition des cellules DP vers CD4+ SP. Par contre, ce phénotype est récupérable par plusieurs approches dont le croisement d’une souris transgéniques exprimant Nef avec une souris exprimant la forme constitutivement active de Lck Y505F. Les résultats indiquent que la maturation des cellules CD4+ est altérée par le dysfonctionnement de la signalisation CD4-Lck. Toutefois, les mécanismes moléculaires par lesquels Nef contribue au bloc de la génération des cellules CD4+ dans le thymus demeurent très imprécis. Dans cette étude, en utilisant des approches biochimiques et de microscopie confocale, nous avons trouvé que les thymocytes transgéniques Nef+ expriment plus de Lck que les thymocytes Nef-. Malgré cette augmentation, une partie significative de Lck est incapable d’atteindre la membrane plasmique. Cette fraction était significativement accumulée dans un compartiment intracellulaire des thymocytes transgéniques exprimant Nef. Également, en utilisant la technique d’essai kinase in vitro, nous avons trouvé que l’activité kinase de Lck est significativement augmentée dans les thymocytes transgéniques mais demeure stable suite à une stimulation par un α-CD3ε + α-CD4. Également, comparativement aux thymocytes Nef-, la kinase Lck dans les thymocytes transgéniques était résistante à la dégradation suite à une stimulation. En examinant le statut de c-Cbl, le principal régulateur négatif de Lck, nous avons montré que c-Cbl colocalise faiblement avec Lck, malgré son hyperphosphorylation constitutive. Ceci pourrait expliquer l’échec de la dégradation de Lck. En plus, nous avons trouvé que suite à une stimulation par un α-CD3ε + α-CD4, la phosphorylation de Zap-70 en tyrosine 493 par Lck est diminuée, résultant d’une importante baisse de l’activité kinase de Zap-70 et d’un bloc des premiers évènements de la voie de signalisation par le TcR. Ces données indiquent que la signalisation CD4-Lck est interrompue par la présence de Nef. / HIV-1 Nef protein plays an essential role in the HIV-1 pathogenesis by modulating the host signaling transduction pathways. TcR signalling is important for the thymic selection process to CD4 and CD8 single positive T cells and is greatly dependent on the activity of Src kinase Lck and its ability to bind to CD4 and CD8 cytoplasmic tail. We previously found that expression of HIV or SIV Nef can induce severe thymocytes depletion and downregulation of CD4 expression in Nef+ mice. We also recently showed that Nef blocks generation of double positive thymocytes and impairs DP to CD4+ SP T cells transition. The reversal of this phenotype was accomplished by several approaches, among them by crossing Nef+ mice with mice expressing constitutively active Lck Y505F. These results imply that the maturation of CD4+ T cells is disrupted due to impairment of Lck-mediated CD4 receptor signaling. However, the molecular mechanisms by which Nef contributes to the impairment in thymic CD4 generation remains largely unclear. In this study, using confocal microscopy and biochemical approaches, we found that Nef+ thymocytes express more Lck than the Nef- control. Despite of this increase, a significant portion of Lck molecules were unable to reach to the plasma membrane. It was significantly accumulated in the intracellular endosomal compartment of the Nef+ thymocytes. Moreover, using IVKA we found that the activity of Lck is significantly increased in Nef+ thymocytes but was not further increased upon stimulation by α-CD3ε, α-CD4 or α-CD3ε + α-CD4. Moreover, compared to Nef- controls, Lck kinase in Nef+ thymocytes was resistant to degradation upon stimulation. Examining the status of c-Cbl, the main negative regulator of Lck, showed that c-Cbl localized with Lck poorly, despite his constitutive hyperphosphorylation. This explains the failure of Lck degradation. In addition, we found that upon stimulation, Zap-70 phosphorylation at tyrosine 493 by Lck is decreased, resulting by a decrease of Zap-70 kinase activity and TcR proximal event block. These data indicate that CD4-Lck signaling was interrupted by the presence of Nef.

Human immunodeficiency virus (HIV) infected patients with a history of injection opiate abuse have higher incidences of acquired immunodeficiency syndrome (AIDS) and neurological dysfunction. The use of combined anti-retroviral therapy has significantly reduced the prevalence of mortality and progression to AIDS. Due to extended life expectancy, these patients are still at a great risk for HIV-associated neurological disorders and impairment in their later life. Neural progenitor cells (NPCs) play critical roles in brain growth and repair after injury and insult. Pediatric HIV patients whose glial populations are still developing are especially at risk for central nervous system (CNS) damage. Our previous reports suggest that HIV-1 transactivator of transcription (Tat) can directly cause pathology in neural progenitors and oligodendroglia (OLs) (Hauser et al. 2009). Thus, we have hypothesized that NPCs and/or glial progenitors may be targets of HIV proteins ± opiates drugs of abuse. To determine whether progenitors are targets of HIV-1, a multi-plex assay was performed to assess chemokine/cytokine expression after treatment with viral proteins Tat or glycoprotein 120 (gp120) with/without morphine. Murine striatal progenitors released increased amount of the beta-chemokines CCL5/regulated upon activation, normal T cell expressed and secreted (RANTES), CCL3/macrophage inflammatory protein-1α (MIP-1α), and CCL4/macrophage inflammatory protein-1β (MIP-1β) after 12 h exposure to HIV-1 Tat, but no to gp120. Secreted factors from Tat-treated progenitors were chemoattractive towards microglia, an effect blocked by 2D7 anti- C-C chemokine receptor type 5 (CCR5) antibody pre-treatment. Tat and opiates have interactive effects on astroglial chemokine secretion, but this interaction did not occur in progenitors. We also examined effects of Tat and morphine on proliferation and lineage progression of NPCs in vitro and in vivo. In vitro, Tat and morphine independently reduced the proliferation and population of Sox2+ and Olig2+ cells in the absence of cell death. The interactive effects of morphine and either Tat or supernatant from HIV-1SF162 infected monocytes varied depending on outcome measure and time of exposure, but interactive effects occurred primarily on proliferation. In rare instances, viable human progenitors were associated with p24 immunolabeling suggesting that progenitors may be infected, a concept that is still controversial. To investigate effects of Tat and morphine on NPCs in vivo, we used a mouse model in which HIV-1 Tat1-86 is conditionally expressed in astroglia. In vivo results in neonatal striata were similar to those in cell cultures. We extended the experiments into adult mice with inducing Tat expression for 3 month and the effect of sexes was examined in these animals. Intriguingly, males showed more Tat-induced impairment in behavioral tests (rotarod, grip strength, light-dark box) than females. Tat+ males also showed a greater reduction in the proportion of NeuN+ cells and NeuN immunoreactivity in the striatum, accompanied by greater microglial activation (3-nitrotyrosine+/Iba-1+). Unbiased stereological estimation in Nissl staining revealed that the depletion of NeuN immunoreactivity in these mice was not due to neuron cell death or loss, because the total neuron number in striatum and total striatal volume were not affected by long-term Tat induction. Tat exposure appears to selectively reduce levels of NeuN in living neurons, although the reason is not known. Therefore, both the enhanced microglial reactivity and depletion of NeuN levels in males may help to explain sex-specific behavioral outcomes. Sox2+ and Olig2+ cells showed equivalent reduction in their population in both sexes. Overall, our findings show that CNS progenitors, including both undifferentiated NPCs and glial progenitors, are vulnerable to individual or combined effects of HIV-1 or Tat and opiates. Changes in progenitor dynamics may alter the balance of cell populations in both the developing and adult CNS. We speculate that such changes may contribute to the behavioral abnormalities that we observed in Tat+ mice and which appear to model aspects of motor, cognitive and anxiety deficits in HIV-infected patients.

HIV-Associated Neurocognitive Disorders (HANDs) are becoming one of the largest problems in patients infected with HIV-1. The ability of infected cells such as monocytes and microglial cells to act as viral reservoirs causes extreme inflammation in the CNS and has led to several different types of neurocognitive problems. Specifically these HIV-1 infected monocytes are able to secrete inflammatory factors such as the regulatory protein Tat which acts as a chemoattractant for monocytes while also promoting the adhesion of leukocytes to the extracellular matrix (ECM). We have shown that one of the major features of the Tat protein is that it promotes cytoskeletal rearrangement resulting in increased adhesion. Specifically integrin and actin visualization was performed using confocal immunofluorescence while cytoskeletal morphology was shown with light and SEM. This microscopy work showed the Tat protein resulted in altered β1-integrin expression and distribution as well as changes in polymerized actin. These cytoskeletal changes resulted in increased adhesion to the ECM. Similarly we have also shown that these cytoskeletal changes of β1-integrin distribution and polymerized actin can be modulated through select cannabinoids THC and CP55940 that bind through the CB2 receptor which inhibits this adhesion as well as the morphological changes. The modulation of this reorganization is characteristic of a signal transduction pathway where a novel convergent point between extracellular Tat and the select cannabinoids THC and CP55940 exists. The aim of this project was to show the cytoskeletal reorganization using different microscopy techniques including light and scanning electron microscopy. This was followed by identifying and characterizing the convergent point along the signal transduction pathway linked to these changes. Different techniques were utilized in order to identify the putative convergent point in the signal transduction cascade including antibody arrays, RT-PCR, and western immunoblotting. The cytoskeletal rearrangements of β1-integrin and actin polymerization were shown successfully via light and scanning electron microscopy in the context of treatment with Tat in the presence and absence of select cannabinoids THC and CP55940. Several different pathways were identified as possibly linked to cannabinoid-mediated inhibition of signal transductional activation consequent of attachment to extracellular matrix proteins. However, the exact molecules implicated in specific signal transductional pathways as targets of cannabinoid-mediated action remain to be defined.

During development, oligodendrocytes (OLGs), the myelinating cells of the central nervous system (CNS), undergo a stepwise progression during which OLG progenitors, specified from neural stem/progenitor cells, differentiate into fully mature myelinating OLGs. This progression along the OLG lineage is characterized by well-synchronized changes in morphology and gene expression patterns. The studies presented in this dissertation identified the extracellular factor Autotaxin (ATX) as a novel upstream signal modulating HDAC1/2 activity and gene expression in cells of the OLG lineage. Using the zebrafish as an in vivo model system, as well as rodent primary OLG cultures, this functional property of ATX was found to be mediated by its lysoPLD activity, which has been well-characterized to generate the lipid signaling molecule lysophosphatidic acid (LPA). LPA binds to Gprotein-coupled LPA receptors (LPARs) on the surface of OLGs to initiate downstream signaling events. ATX’s lysoPLD activity was found to modulate HDAC1/2 regulated gene expression during a time window coinciding with the transition from OLG progenitor to early differentiating OLG. When looking further downstream of the ATX-LPA axis, down-regulation of LPA receptor 6 (LPA6) was found to reduce the expression of OLG differentiation genes as well as the overall process network area of OLGs. Thus, LPA6 plays a role in both the gene expression and morphology changes seen in OLG differentiation. These findings prove useful for future therapeutic targets needed for demyelinating diseases of the CNS such as Multiple Sclerosis (MS), in which OLGs fail to differentiate into mature OLGs, needed for remyelination. Additionally, white matter injury has been frequently reported in HIV+ patients. Previous studies showed that HIV-1 Tat (transactivator of transcription), a viral protein that is produced and secreted by HIV-infected cells, is a toxic factor to OLGs. We show here that Tat treatment reduces the expression of OLG differentiation genes and the overall process network area of OLGs. Additionally, Tat-treated OLGs have reduced ATX lysoPLD activity and there is a physical interaction between Tat and ATX. Together, these data strongly suggest functional implications of Tat blocking ATX’s lysoPLD activities and thus the ATX-LPA signaling axis proves to play a significant role in the development of OLGs.

Human immunodeficiency virus type 1 (HIV-1) is known to cause a spectrum of neurological, behavioral and motor deficits collectively termed as HIV-1 associated neurocognitive impairments (HAND). Opiates augment HIV-related CNS complications through both direct and indirect mechanisms that disrupt glial and neuronal function. All CNS macroglia and neurons derive from neural progenitor cells (NPCs) during development, and NPCs in the adult brain contribute to repair processes. Since disruptions in NPC function are known to impact CNS populations and brain function in a number of disease/injury conditions, we determined whether HIV ± opiate exposure affected the maturation and fate of human NPCs (hNPCs). As hNPC infection by HIV has occasionally been reported, we also reexamined this question, and parsed between effects due directly to hNPC infection by HIV, or to hNPC dysfunction caused by the infective milieu. Multiple approaches confirmed the infection of hNPCs by R5 tropic (CCR5 utilizing) HIVBaL, and demonstrated that active infection could be sequentially transferred to naïve hNPCs. Exposure to supernatant from HIVBaL-infected cells (HIVsup) reduced hNPC proliferation and led to premature differentiation into astrocytes and neurons. Morphine co-exposure prolonged hNPC infection and exacerbated functional effects of HIVsup. Neither purified virions nor UV-inactivated HIVsup altered proliferation, indicating that this effect did not require infection. Gene array analysis and RT-qPCR with immunoblot validation suggested that Cdk5 signaling was involved in HIV-morphine interactions. siRNA-mediated knockdown of Cdk5 expression attenuated the effect of HIV-1 and morphine on hNPC proliferation and MAP2 differentiation, but also increased hNPC death. Furthermore, in an attempt to understand the role of mu-opioid receptor (MOR) splice variants on the interactive effect of HIV-1 and morphine on hNPCs, we found that both MOR-1 and MOR-1K are differentially regulated by HIV-1 in these cells. This suggests that these splice variants may have differential actions in the response of hNPCs to HIV-1 and morphine co-exposure. Given the overlap of Cdk5 and MOR signaling, it is likely that MOR-1K and/or MOR-1 converge with Cdk5 in the mechanism underlying HIV-1 and morphine interaction in hNPCs. Overall, dysregulation of hNPC functions by the infectious environment may create cell population imbalances that contribute to CNS deficits in both adult and pediatric patients. Additionally, infected hNPCs may pass virus to their progeny, and serve as an unappreciated viral reservoir. The recent epidemic of opiate/heroin abuse highlights the clinical importance of HIV and opiate interactions.

L'autophagie est un mécanisme de dégradation lysosomale qui joue un rôle important dans l'immunité innée et adaptative. La relation entre le Virus de l'Immunodéficience Humaine de type 1 (VIH-1) et l'autophagie est complexe. En effet, l'équipe a montré que l'enveloppe virale du VIH-1 (Env), exprimée à la surface des cellules infectées, induit une autophagie massive dans les cellules T CD4 bystanders non infectées. Au contraire, lorsque ces cellules s'infectent de façon productive, l'autophagie est inhibée, suggérant qu'une ou plusieurs protéines virales soient capables de bloquer ce processus. L'objectif de ce travail de thèse a été de rechercher ces protéines virales et leur mécanisme d'action. Un crible double hybride en levure a permis de mettre en évidence que plusieurs protéines du VIH-1 sont capables d'interagir avec des protéines autophagiques (Atg), et plus spécifiquement que la protéine virale Vif (Virion infectivity factor) interagit directement avec la protéine LC3, protéine essentielle au processus autophagique. Cette interaction a été confirmée en système in vitro et in vivo (GST pull-down et immunoprécipitation). Plusieurs mutants des protéines Vif et LC3 ont été réalisés pour déterminer les domaines de liaison. La partie C-terminale de Vif ainsi que la glycine C-terminale de LC3, responsable de la conjugaison au phosphatidylethanolamine (PE), semblent être les domaines impliqués dans cette interaction. Une des principales fonctions de Vif connues est de dégrader, via le protéasome, les facteurs cellulaires APOBEC (Apolipoprotein B mRNA-editing, enzyme-catalytic, polypeptide-like) et en particulier la protéine APOBEC3G (A3G). Les résultats montrent que Vif est impliquée dans le blocage de l'autophagie induite par l'enveloppe indépendamment de son action sur A3G. / Autophagy is a lysosomal degradation pathway involved in the innate and adaptative immunity. The relationship between the Human Immunodeficiency Virus type 1 (HIV-1) and autophagy is complex. The team has demonstrated that autophagy is induced in uninfected CD4 T cells after contact with infected cells expressing HIV-1 envelope glycoproteins (Env), leading to apoptosis. In contrast, when these cells are productively infected, autophagy is repressed, suggesting that one or several viral proteins are able to block this process. The aim of the thesis was to search these viral proteins and to determine their mechanism of action. A two-hybrid screen has revealed that several HIV-1 viral proteins are able to interact with autophagic proteins (Atg). In particular, Vif (Virion infectivity factor) interacts directly with LC3, a protein involved in the formation of autophagosomes. This interaction has been confirmed in vitro and in vivo (GST pull-down and immunoprecipitation). Several mutants of Vif and LC3 have been done to analyze the binding domains. The C-terminal part of Vif and the C-terminal glycine of LC3, responsible for the conjugation to PE, seem to be involved in the interaction between Vif and LC3.Vif plays an important role during HIV-1 infection. One of its main functions is to degrade, by the ubiquitin-proteasome system, the antiviral factors called APOBECs (Apolipoprotein B mRNA-editing, enzyme-catalytic, polypeptide-like) and in particular APOBEC3G (A3G). Our results demonstrate that Vif is involved in the blockade of Env-mediated autophagy independently of its role on A3G.

Le cycle réplicatif du HIV-1 aboutit à la formation de virions qui s’assemblent dans des microdomaines spécifiques localisés à la membrane plasmique ou sur des compartiments intracellulaires particuliers, nommés VCC pour « Virus-Containing Compartments ». Selon les cas, ces virions sont ensuite relâchés par bourgeonnement ou exocytose. Ces étapes nécessitent un remodelage membranaire via le cytosquelette d’actine, ce qui est régulé par des voies de signalisation contrôlées par les RhoGTPases. Certains résultats suggèrent l’implication de ces protéines dans la biogénèse du HIV-1. Cependant, il reste à caractériser les mécanismes moléculaires spécifiquement impliqués dans la régulation cellulaire de l’assemblage viral.L’objectif de cette thèse consistait donc à identifier les RhoGTPases et les effecteurs des voies de signalisation spécifiquement requis durant la biogénèse virale. Cette étude a porté sur les GTPases Rac1, Cdc42 et RhoA car elles ont un rôle majeur dans la régulation du cytosquelette d’actine et de la dynamique membranaire. Elle a été réalisée sur les lymphocytes T (LT) Jurkat, cellules modèles pour l’infection HIV-1 où les virions s’assemblent à la membrane plasmique ; et les cellules adhérentes HeLa où les virions peuvent aussi s’assembler au niveau des VCC. Nos résultats ont révélé le rôle de la voie de signalisation Rac1-IRSp53-Wave2 dans l’assemblage de Gag à la membrane plasmique des LT Jurkat, et un rôle pour RhoA dans la régulation de l’assemblage viral suggéré au niveau des VCC. Ce travail améliore la compréhension des voies de signalisation cellulaires sollicitées lors de l’assemblage du HIV-1, en particulier dans les lymphocytes T, cibles du virus. / During the last steps of HIV-1 replication cycle, the Gag proteins come together in particular microdomains located at the plasma membrane or in some intracellular compartments, named “Virus-containing compartments”. Then, the viral particles are released by budding or exocytosis. All these steps involve membrane and actin cytoskeleton remodeling which is regulated by the RhoGTPases. In fact, some data suggest the implication of such proteins in HIV-1 biogenesis, but molecular mechanisms underlying this effect is not yet understood. During this thesis, our aim was to characterize the RhoGTPases and the effectors of cell signaling pathways which are specifically required during HIV-1 particle biogenesis. We focused our study on the GTPases Rac1, Cdc42 and RhoA because their influence on membrane and actin cytoskeleton was essential. Moreover, this work was accomplished on Jurkat T lymphocytes which are model cells to HIV-1 infection where the Gag proteins assemble at the plasma membrane, and on HeLa cells where the Gag proteins can also assemble on virus-containing compartments. Our results showed the requirement of the Rac1-IRSp53-Wave2 signaling pathway for HIV-1 Gag assembly at the plasma membrane of Jurkat T cells, and a role for RhoA GTPase in the regulation of viral particle assembly on virus-containing compartments in HeLa cells. This study improved understanding of cell signaling pathways required during the HIV-1 particle biogenesis and release, particularly in T cells which are the main host cell for HIV-1.