La SUMOylation est une modification post-traductionnelle qui gouverne divers processus cellulaires incluant immunité innée et défense antivirale. Des effecteurs de la synthèse d’IFN, de son signal de transduction ainsi que des facteurs de restriction sont modifiés par SUMO (Small Ubiquitin Modifier). Par ailleurs, certains virus exploitent la machinerie SUMO afin de contrecarrer les mécanismes de défense antivirale suggérant l’implication de SUMO dans l’interface virus et défense antivirale. A l’aide d’un modèle cellulaire exprimant les différents paralogues SUMO1, SUMO2 ou SUMO3 ou en diminuant l’expression de l’unique enzyme de conjugaison à SUMO, Ubc9, nous avons montré un effet différentiel de SUMO sur deux virus de la famille des Rhabdoviridae (virus de la stomatite vésiculaire (VSV) et le virus de la rage) et sur la réponse aux IFN alpha et IFN gamma. Le premier axe de recherche a permis de montrer que l’expression de SUMO inhibe la synthèse de l’IFN suite à l’infection par le VSV et le virus de la rage, rendant les cellules plus sensibles à l’infection par le virus de la rage. IRF3 est conjuguée à SUMO, ce qui corrèle avec l’inhibition de sa phosphorylation et l’inhibition de la synthèse d’IFN beta. En revanche, bien que la synthèse de l'IFN soit diminuée, l’expression de SUMO confère la résistance au VSV et inhibe sa transcription primaire. L’activité anti-VSV de SUMO est abolie par la déplétion de MxA. L’effet de SUMO est médié par sa capacité à augmenter l’oligomérisation et la stabilité de MxA. Par ailleurs, ce travail a permis d’identifier MxA comme nouvelle cible de la machinerie SUMO. MxA interagit avec SUMO de manière covalente sur la lysine K48 et de manière non covalente avec SUMO1. Le second axe de recherche a permis d’identifier SUMO comme un nouveau régulateur de la réponse aux IFN. La SUMOylation de STAT1 inhibe sa phosphorylation, régulant négativement le signal de transduction et par conséquent la transcription et la réponse biologique en réponse à l’IFN gamma. En revanche, l’expression de SUMO n’altère ni le signal de transduction ni la transcription en réponse à l’IFN alpha.Par ailleurs, dans les cellules exprimant SUMO3, l’IFN gamma et l’IFN alpha induisent la SUMOylation de PML par SUMO3 ce qui entraîne sa dégradation via le protéasome et inhibe les réponses biologiques médiées par PML. Ce travail a permis de montrer un rôle central de SUMO dans l’immunité intrinsèque et innée, médié par la SUMOylation de protéines cellulaires telles qu’IRF3, MxA, STAT1 ou PML. / SUMOylation modulates several cellular process including innate immunity and antiviral defense. Many key regulators involved in IFN induction, IFN signaling as well as various restriction factors are SUMOylated. Using cells stably expressing the different paralogs of SUMO; SUMO1, SUMO2 or SUMO3 and cells depleted of the only known SUMO conjugating enzyme, Ubc9, we show a differential effect on two viruses from Rhabdoviridae family (Vesicular Stomatitis Virus (VSV) and rabies virus) and on the response to IFN alpha and IFN gamma. First, we report that SUMO expression inhibits VSV- and rabies virus-induced IFN synthesis. Consequently, SUMO expression renders cells more sensitive to rabies virus infection. Overexpression of SUMO leads to IRF3 SUMOylation correlating rabies viral infection with both the inhibition of IRF3 activation and IFN beta synthesis. However, although SUMO inhibits VSV-induced IFN, SUMO confers resistance to VSV by inhibiting VSV primary transcription. Furthermore, the anti-VSV effect of SUMO is abolished in MxA depleted cells. Mechanistically, SUMO enhances MxA oligomerization resulting in the stabilization of the MxA protein. We also identified MxA as a new target of SUMO machinery. MxA interacts covalently with SUMO2/3 on lysine K48 and non-covalently with SUMO1. We then investigated the various roles of SUMO at different steps of the JAK/STAT pathway, including STAT activation, transcriptional response and IFN-induced biological effects, identifying SUMO as a new regulator of IFN response. The overexpression of SUMO leads to STAT1 SUMOylation and to a decrease in IFN-induced STAT1 phosphorylation resulting in an inhibition of IFN-gamma-induced transcription and biological responses. In contrast, SUMO expression does not alter IFN alpha signaling and transcriptional response. In addition, in SUMO3 expressing, IFN gamma;and IFN alpha induce SUMOylation of PML by SUMO3 inducing its degradation via the proteasome and inhibition of biological responses mediated by PML. Taken together our results show that SUMO plays a crucial role in innate and intrinsic immunity mediated by SUMOylated proteins such as IRF3, MxA, STAT1 or PML.

No description available.

Our innate immunity is our first line of defence against pathogens. We require this immunity to control the numerous viral infections we are challenged with during our lives. However, little is known about the exact molecular mechanisms of our innate immunity, particularly components that have specific antiviral potential. One potent mediator of this antiviral activity is the interferon system. Activation of the interferon system leads to the production of several interferon-induced proteins, which inhibit the multiplication of viruses by distinct mechanisms. A key example of one of these mediators is the human MxA protein. Human MxA has the capacity to inhibit many different viruses from diverse families. In many cases it is proposed that MxA interferes with key viral components, such as incoming or newly formed nucleocapsids. It is speculated that MxA traps and missorts these viral components so they are no longer available for virus production and virus dissemination is inhibited. West Nile virus belongs to Flaviviridae family of viruses and was involved in the outbreak of virus-associated encephalitis in New York City in 1999. In this thesis I show that West Nile virus is insensitive to antiviral activity of MxA and describe how West Nile virus has developed a replication strategy that avoids MxA recognition and activation. I show that virus-induced changes of cytoplasmic membranes provide a protective microenvironment for viral replication and the viral components essential for viral replication. This hypothesis was proven by preventing the formation of these membrane structures with the fungal chemical Brefeldin A. Under these conditions I observed that stable expression of MxA could partially restrict West Nile virus RNA replication. Subsequently, I showed that the assembly mechanism of West Nile virus prevents interaction between the MxA protein and the viral capsid proteins. This was achieved by the use of a trans-packaging cell line whereby the West Nile virus structural proteins are expressed stably in trans instead of in cis from the polyprotein. When this cell line was transfected with a West Nile virus replicon expressing the human MxA protein distinct co-localisation and redistribution of the MxA with West Nile virus capsid proteins into large tubular structures within the cytoplasm of transfected cells was observed. Strikingly, these tubular aggregates are visually analogous to structures observed during infection of MxA expressing cells infected with members of the Bunyaviridae, particularly La Crosse virus. Moreover, retargeting MxA to specific sites of the endoplasmic reticulum in cells transfected with the West Nile virus infectious clone resulted in co-localisation between MxA and the West Nile virus capsid proteins and significantly inhibited the production of infectious particles. These results suggest that the sequestering of viral capsids within cytoplasmic inclusions maybe a conserved mechanism for antiviral activity of the MxA protein across the viruses families and highlight the innate ability of such molecules to recognise key molecular patterns within pathogens. Finally, I sought to exploit the antiviral potential of MxA as a therapeutic agent against infection with Influenza A viruses; viruses that have a very high sensitivity for the antiviral activity of MxA. By expressing MxA from the West Nile virus replicon, infection with the highly pathogenic Influenza virus H5N1 strain could be inhibited in vitro. Furthermore, in vivo studies in Mx-negative mice indicated that intranasal inoculation with MxA expressed from the West Nile virus replicon can protect these mice against an otherwise lethal infection with a low pathogenic Influenza A virus. Taken all together, in this thesis I provide evidence that strongly supports the existence of an evolutionary working mechanism of a significant mediator of our immune system, the antiviral MxA protein. Furthermore, I show how an important human pathogen, such as West Nile virus has evolved a replication strategy to evade this antiviral protein. These results will open new pathways for the development of a new type of antiviral therapies that utilize the potent antiviral activity of the MxA protein.

Our innate immunity is our first line of defence against pathogens. We require this immunity to control the numerous viral infections we are challenged with during our lives. However, little is known about the exact molecular mechanisms of our innate immunity, particularly components that have specific antiviral potential. One potent mediator of this antiviral activity is the interferon system. Activation of the interferon system leads to the production of several interferon-induced proteins, which inhibit the multiplication of viruses by distinct mechanisms. A key example of one of these mediators is the human MxA protein. Human MxA has the capacity to inhibit many different viruses from diverse families. In many cases it is proposed that MxA interferes with key viral components, such as incoming or newly formed nucleocapsids. It is speculated that MxA traps and missorts these viral components so they are no longer available for virus production and virus dissemination is inhibited. West Nile virus belongs to Flaviviridae family of viruses and was involved in the outbreak of virus-associated encephalitis in New York City in 1999. In this thesis I show that West Nile virus is insensitive to antiviral activity of MxA and describe how West Nile virus has developed a replication strategy that avoids MxA recognition and activation. I show that virus-induced changes of cytoplasmic membranes provide a protective microenvironment for viral replication and the viral components essential for viral replication. This hypothesis was proven by preventing the formation of these membrane structures with the fungal chemical Brefeldin A. Under these conditions I observed that stable expression of MxA could partially restrict West Nile virus RNA replication. Subsequently, I showed that the assembly mechanism of West Nile virus prevents interaction between the MxA protein and the viral capsid proteins. This was achieved by the use of a trans-packaging cell line whereby the West Nile virus structural proteins are expressed stably in trans instead of in cis from the polyprotein. When this cell line was transfected with a West Nile virus replicon expressing the human MxA protein distinct co-localisation and redistribution of the MxA with West Nile virus capsid proteins into large tubular structures within the cytoplasm of transfected cells was observed. Strikingly, these tubular aggregates are visually analogous to structures observed during infection of MxA expressing cells infected with members of the Bunyaviridae, particularly La Crosse virus. Moreover, retargeting MxA to specific sites of the endoplasmic reticulum in cells transfected with the West Nile virus infectious clone resulted in co-localisation between MxA and the West Nile virus capsid proteins and significantly inhibited the production of infectious particles. These results suggest that the sequestering of viral capsids within cytoplasmic inclusions maybe a conserved mechanism for antiviral activity of the MxA protein across the viruses families and highlight the innate ability of such molecules to recognise key molecular patterns within pathogens. Finally, I sought to exploit the antiviral potential of MxA as a therapeutic agent against infection with Influenza A viruses; viruses that have a very high sensitivity for the antiviral activity of MxA. By expressing MxA from the West Nile virus replicon, infection with the highly pathogenic Influenza virus H5N1 strain could be inhibited in vitro. Furthermore, in vivo studies in Mx-negative mice indicated that intranasal inoculation with MxA expressed from the West Nile virus replicon can protect these mice against an otherwise lethal infection with a low pathogenic Influenza A virus. Taken all together, in this thesis I provide evidence that strongly supports the existence of an evolutionary working mechanism of a significant mediator of our immune system, the antiviral MxA protein. Furthermore, I show how an important human pathogen, such as West Nile virus has evolved a replication strategy to evade this antiviral protein. These results will open new pathways for the development of a new type of antiviral therapies that utilize the potent antiviral activity of the MxA protein.

Certains motifs moléculaires associés à la mort cellulaire semblent identifier les cancers prompts à répondre à une thérapie cytotoxique. Ceci en élaborant une réponse anti-tumorale basées sur une réponse T protectrice. Mon travail de thèse montre que le traitement par chimiothérapie immunogène active des voies moléculaires mimant une infection virale. Ceci conduit au niveau des cellules tumorales à une signalisation autocrine via l’IFNαβ / IFNAR1/2, initiée par la reconnaissance d’ARN double brin (dsRNA) endogène par les Récepteurs endosomaux de Reconnaissance des Motifs (PRRs). De façon plus détaillée, nous montrons que les axes TLR3/TRIF (senseurs endosomaux de dsRNA) et IFNAR1/2 (Récepteurs de l’IFN de Type I) doivent signaliser au niveau de la cellule tumorale pour que la chimiothérapie puisse aboutir à l’induction de l’axe CXCL10/CXCR3 et éliciter une réponse efficace in vivo. L’analyse du profil ARN de cellules tumorales Tlr3+/+ (mais pas Tlr3-/-) exposées aux anthracyclines a révélé une forte empreinte virale/IFN, indispensable à l’efficacité/activité anti-tumorale. Le fait d’affecter les axes TLR3 ou IFNAR1/2 au niveau tumorale soit à l’aide d’anticorps neutralisants, soit à l’aide de modèles KO abroge le relarguage de CXCL10 induit par la chimiothérapie, et ainsi la capacité à contrôler la pousse tumorale à moins que de l’IFNαβ ou du CXCL10 exogène soit co-administré aux anthracyclines. De plus la chimiorésistance des tumeurs traitées par des molécules n’induisant pas de signature virale peux être réversée par de l’IFN de Type I exogène. Enfin, la détection d’une signature IFN au niveau de biospies de cancers du sein humains permet de prédire la bonne réponse au traitement adjuvant par anthracyclines. D’un point de vue de l’évolution, alors que les tumeurs (comme les virus) ont élaboré des mécanismes pour échapper aux réponses IFN, la signature virale induite par la chimiothérapie devrait contribuer à contrecarrer cette immunoédition. / Distinct cell death-associated molecular patterns might define cancers proned to respond to a cytotoxic therapy by mounting a protective T cell-based anticancer immunity. My PhD Thesis work shows that immunogenic chemotherapy phenocopies viral infection leading to autocrine IFNαβ/IFNAR1/2 signalling in tumor cells initiated by recognition of self dsRNA by endosomal pattern recognition receptors (PRRs). In detail, TLR3/TRIF (endosomal dsRNA sensors) and IFNAR1/2 (Type I IFN receptors) must signal within the tumor cells so that chemotherapy can induce downstream CXCL10/CXCR3 axis and elicit therapeutic responsiveness in vivo. RNA profiling of Tlr3+/+ (but not Tlr3-/-) tumor cells exposed to anthracyclines revealed a strong IFN/viral fingerprint, indispensable for the tumoricidal activity. Neutralization by antibodies or genetic defects affecting tumor –associated TLR3 or IFNAR1/2 compromised chemotherapy-induced CXCL10 release and tumor control unless exogenous IFNαβ or CXCL10 are concomitantly supplied to anthracyclines. Moreover, chemoresistance of tumors treated by drugs failing to induce a viral signature can be reversed by exogenous Type I IFN. Finally, the IFN fingerprint of human breast cancers allowed to predict tumors proned to benefit from adjuvant anthracyclines. From an evolutionary viewpoint, while tumors (like viruses) have evolved mechanisms to evade an IFN response, chemotherapy-induced viral mimicry might contribute to bypass such as immunoediting.

The interferon (IFN) system forms an essential part of the innate immune response, up-regulating hundreds of IFN-stimulated genes (ISGs) in response to viral infection. A key protein in this response is the human myxovirus resistance protein MxA, an IFN-induced GTPase with broad-spectrum antiviral activity, capable of inhibiting many RNA and DNA viruses. One of the most studied antiviral effects of MxA is the inhibition of influenza A virus replication, yet the molecular mechanism of antiviral activity is still unknown. Influenza A viruses are inhibited by MxA at two distinct stages of viral replication; during viral entry and following primary transcription of viral mRNAs. The antiviral effects of MxA during viral entry are highly dependent on IFN, however activity exerted after primary transcription can occur in the absence of IFN. This study provides evidence that MxA exerts its antiviral activity at these two stages of viral replication through distinct mechanisms, and outlines a potential model of MxA antiviral activity following primary transcription. A potential third antiviral mechanism of MxA is proposed based on the findings that MxA is able to regulate cellular lipid metabolism, thereby potentially affecting virion composition. Mutational analysis of MxA highlights the significance of GTPase activity to the antiviral effects of MxA, while also demonstrating that natural single nucleotide polymorphisms in MxA have the potential to severely impair or prevent antiviral activity. Finally, this thesis shows for the first time that MxA exhibits antiviral activity against influenza B viruses. Overall this thesis provides new information illustrating how MxA provides potent antiviral activity against influenza viruses. Such information is vitally important as understanding the molecular basis of how proteins such as MxA function against many human pathogens is fundamentally important in our efforts to create better long-term treatment options for all viral diseases.

Hantaviren sind wichtige menschliche Krankheitserreger und können das Hämorrhagische Fieber mit renalem Syndrome (HFRS) auslösen, welches sich durch endotheliale Dysfunktion kennzeichnet. Pathogene und nicht-pathogene Hantaviren replizieren sich in Endothelzellen, ohne zytopathische Effekte auszulösen. Dies legt nahe, dass immunpathologische Mechanismen eine entscheidende Rolle in der Pathogenese spielen. Wir haben die antivirale Antwort nach Infektion mit dem pathogenen Hantaan Virus (HTNV) sowie mit dem weniger pathogenen Tula Virus (TULV) in humanen Endothelzellen (HUVEC) verglichen. Die mit HTNV und auch die mit TULV infizierten Zellen zeigten eine erhöhte Expression von Antigen-präsentierenden Molekülen. Hierbei induzierte TULV die Expression von HLA Klasse I-Molekülen noch effizienter. HTNV sorgte für die Induktion von Interferon (IFN)-???während dieses Zytokin im Überstand von TULV-infizierten HUVEC kaum nachzuweisen war. Trotzdem konnte die Hochregulation von HLA Klasse I-Molekülen auf HTNV- und TULV-infizierten Zellen durch anti-IFN-?-Antikörper blockiert werden. Interessanterweise wurde das antiviral wirksame MxA-Protein, welches die virale Replikation hemmt, bereits 16 Stunden nach einer Infektion mit TULV induziert. Im Gegensatz dazu war MxA in HTNV-infizierten Zellen erst nach 48 Stunden der Infektion nachzuweisen. Der Kinetik der MxA-Expression entsprechend, replizierte sich TULV nur sehr schwach in HUVEC, wohingegen HTNV-infizierte Zellen hohe Virustiter aufwiesen, die nach 48 Stunden der Infektion wieder zurückgingen. Beide Hantavirus-Spezies waren jedoch gleichermaßen effizient in der Lage, sich in Vero E6-Zellen zu replizieren, denen die IFN-induzierte MxA-Antwort fehlt. Die verzögerte Induktion des MxA nach einer Infektion der HUVEC mit HTNV, könnte die Virusausbreitung ermöglichen und mit zur Pathogenese des HFRS beitragen. Das Risiko, sich während der Arbeit im Forschungslabor versehentlich mit Hantaviren zu infizieren, macht spezielle Sicherheitsmaßnahmen zwingend erforderlich. Die Wirkung von chemischen oder physikalischen Inaktivierungsmethoden wurde an HTNV-infizierten Proben untersucht. Die beschriebenen Maßnahmen zur Virus-Desinfektion sind geeignet, eine sichere Handhabung der Proben zu gewährleisten. / Hantaviruses represent important human pathogens and can induce hemorrhagic fever with renal syndrome (HFRS), which is characterised by endothelial dysfunction. Both pathogenic and nonpathogenic hantaviruses replicate without causing any apparent cytopathic effect suggesting that immunopathological mechanisms play an important role in pathogenesis. We compared the antiviral response triggered by Hantaan virus (HTNV), a pathogenic hantavirus associated with HFRS, and Tula virus (TULV), a rather nonpathogenic hantavirus, in human umbilical vein endothelial cells (HUVEC). Both HTNV- and TULV-infected cells showed increased levels of molecules involved in antigen presentation. However, TULV-infected HUVEC more rapidly upregulated HLA class I molecules. Interestingly, HTNV clearly induced the production of interferon (IFN)-( whereas expression of this cytokine was barely detectable in the supernatant or in extracts from TULV-infected HUVEC. Nevertheless, upregulation of HLA class I on both TULV- and HTNV-infected cells could be blocked by neutralising anti-IFN-( antibodies. Most strikingly, antiviral MxA protein, which interferes with hantavirus replication, was induced already 16 h after infection with TULV. In contrast, HTNV-infected HUVEC showed no expression of MxA until 48 h postinfection. In accordance with the kinetics of MxA expression TULV only inefficiently replicated in HUVEC whereas HTNV-infected cells produced high titers of virus particles that decreased 48 h postinfection. Both hantavirus species, however, could replicate equally well in Vero E6 cells which lack an IFN-induced MxA response. Thus, a delayed induction of antiviral MxA in endothelial cells after infection with HTNV could allow viral dissemination and contribute to the pathogenesis leading to HFRS. The potential risk of accidental infection by hantaviruses in a clinical or research laboratory necessitates special precautionary measures. To study the elimination of hantavirus infectivity, the effects of different chemical and physical inactivation and depletion procedures were investigated on HTNV-containing materials. The virus inactivation and depletion methods described herein are suitable to prepare non-infectious samples for further use in immunological, virological and cell biological assays.

The Interferon (IFN) family of cytokines are produced in direct response to virus infection and they constitute the first line of defence against virus infection by inducing hundreds of interferon stimulated genes (ISGs) which act in concert to establish the so-called “antiviral state”. Influenza A viruses and parainfluenza virus type 5 (PIV5) are both small negative strand RNA viruses that must circumvent their hosts’ interferon (IFN) response for replication. However, the ways in which these viruses interact with the IFN system are very different. Although PIV5 replication is initially severely impaired in cells in a pre-existing IFN-induced antiviral state, it manages to overcome the antiviral state by targeting an essential component of type I IFN signalling, STAT1, for degradation. Thus the cells cannot maintain the antiviral state indefinitely without continuous signalling. Consequently, the virus resumes its normal replication pattern after 24-48 hours post-infection. In clear contrast, influenza virus fails to establish its replication in the majority of infected cells (90-95%) with a pre-existing IFN-induced antiviral state, although a few cells are still able to produce viral antigens. To further investigate how influenza virus interacts with cells in a pre-existing IFN-induced antiviral state, I have used in situ hybridization to follow the fate of input and progeny genomes in cells that have, or have not, been treated with IFN prior to infection. Here I show for the first time that IFN pre-treatment blocks the nuclear import of influenza A virus genome, which prevents the establishment of virus replication, but this can be overcome by increasing multiplicities of infection. Of those IFN-induced antiviral molecules, human MxA is an essential component of the IFN-induced antiviral state in blocking influenza virus genome import, as this block can be abolished by lentivirus-mediated knockdown of MxA. I also show that in cells constitutively expressing MxA the viral genome still manages to be transported into the nucleus, indicating that MxA might require an unidentified IFN-induced factor to block nuclear import of the influenza virus genome. These results reveal that IFN exerts its action at an early stage of virus infection by inducing MxA to interfere with the transport of viral genome into the nucleus, which is the factory for viral RNA production.

No description available.