Plasmacytoid dendritic cells (pDC) are a specialized group of circulating dendritic cells that respond to viral nucleic acids with Type I IFN production as well as other cytokine and chemokines. These pDC responses lead to the production of antiviral molecules and recruitment of defense cells. During zoonotic B virus infection, a simplex virus of the subfamily Alphaherpesviridae, our lab has observed that infected individuals who succumb to infection have little-to-no-antibody or cell-mediated defenses. To identify whether this was partly due to failure of pDCs to produce antiviral interferon responses or produce chemokine and cytokines, we tested the hypothesis that B virus modulates the IFN response during zoonotic infection by blocking pDC activation and subsequent IFN signaling pathways to circumvent host defenses, while these pathways remain intact in the macaque hosts. We showed that human pDCs respond to B virus through the production of IFN-a, IL-1a, IL-6, TNF-a, MIP-1a/b and IP-10. Human pDCs co-cultured with B virus infected fibroblasts produced fewer cytokines and at lower levels. The macaque response to B virus was measured using PBMCs, as there are no specific reagents available to enrich macaque pDCs. Human and macaque PBMCs produced IFN-a when exposed directly to B virus infected lysates. Co-cultures of PBMCs with B virus infected fibroblasts from both hosts failed to produce any significant amounts of IFN-a. To quantify the antiviral effects of PBMC induced IFN-a, we measured B virus titers after exposure to supernatants from B virus exposed PBMCs, PBMC co-cultures with infected fibroblasts and exogenous recombinant Type I IFN. Our data further suggest that B virus resistance was not due to virus specific blockade of the Type I IFN signaling pathway because STAT-1 was activated in infected fibroblasts when treated with Type I IFNs. These data demonstrate for the first time that B virus replication is unimpeded in the presence of any source of IFN-a in either host cell type. In conclusion, this dissertation shows that the IFN-a production by both hosts in response to B virus is similar and that IFN-a treatment of B virus infected fibroblasts did not reduce B virus replication.

PhD (Microbiology) / Department of Microbiology / Background Conversion of Cholesterol to 25-HydroxyCholesterol (25HC) by Cholesterol 25-hydroxylase (CH25H) has been shown to exert broad antiviral properties. Given its antiviral activities, CH25H is part of an increasingly appreciated connection between type I interferon (IFN-I) and lipid metabolism. Moreover, the details of this connection appear to differ in mouse and human cells. Nevertheless, the molecular basis for the induction of CH25H in humans is not known. Objective Elucidation of signaling and transcriptional events for induction of CH25H expression is critical to design therapeutic antiviral agents. Materials and methods: Wildtype THP-1 monocytic cell-line or THP-1 MyD88 Knockout cell-line were treated with PMA for 72 hours for differentiation into macrophages. Differentiated macrophages or Microglial cells were stimulated with either TLR-agonists, pro-inflammatory cytokine, or interferons, and CH25H mRNAs expression levels were measured by qPCR. Results In this study, we show that CH25H is induced by Zika virus infection or TLR stimulation. Interestingly, CH25H is induced by pro-inflammatory cytokines including 1L- 1, TNF-, and IL-6, and this induction depends on STAT-1 transcription factor. Additionally, we have observed that ATF3 weakly binds to the CH25H promoter, suggesting co-operation with STAT-1. However, ZIKV induced CH25H was independent of type I interferon. Conclusion This study has demonstrated for the first time that pro-inflammatory cytokines such as 1L-1, TNF-, and IL-6 induce CH25H expression. Moreover, this provides further understanding to the connection between innate immunity and sterol metabolism and encourages the exploration of cytokines in antiviral immunity. / NRF

No description available.

The production of Nitric Oxide (NO) under physiological conditions has beneficial roles in acting as a key signaling component of many biological processes as well as having an anti-microbial effect. However its effects following excess production by the inducible NO pathway is potentially detrimental in the pathogenesis of chronic inflammation including sepsis and several other inflammatory diseases. Understanding the mechanisms that regulate the expression of the inducible nitric oxide synthase (iNOS) responsible for producing the excessive amounts of NO in disease states is therefore critical. In this regards, experiments were carried out to identify the signaling pathways that may mediate this process, focusing specifically on the JAK/STAT cascade. The reason for selecting the latter is because our research group, amongst others, has carried out extensive work investigating other signaling pathways, including the mitogen activated kinases (MAPK). Moreover, studies have also been carried out in an attempt to identify the critical role of JAK/STAT signaling for iNOS induction. These studies however failed to conclusively demonstrate whether, as with the MAPKs, the JAK/STATs may also play an essential role. Furthermore there is indeed controversy in the literature with researchers unable to agree whether expression of iNOS does require JAK/STAT activation. Thus, the aim of the project described in this thesis was to establish unequivocally whether activation of the JAK/STATs preceeds induction of iNOS. The studies were extended to L-arginine transport as well because the latter is widely reported to be induced in parallel with iNOS and substrate supply to iNOS may be critical for sustained NO production. Changes in transporter activity as well as their expression profiles were assessed. All experiments were carried out in either rat aortic smooth muscle cells (RASMCs) or in the J774 macrophage cell line. These cell types were selected because RASMCs are one of the prime targets for induced NO production in vascular inflammation and the macrophages are involved in host defence, acting in part through NO production. To establish the role of JAK/STATs, pharmacological and molecular approaches were used. Pharmacologically, two inhibitors were used and these were AG490 and JAK inhibitor I. The former is reported to be a selective JAK2 inhibitor and the other blocks all known JAK proteins. The potential of the GTPases to regulate the induction of iNOS was also examined using selective inhibitor known to regulate these proteins. In addition to these drugs, siRNA targeting JAK2 was also exploited and western blotting was extensively used to detect expression of various proteins including iNOS, native and phosphorylated JAK2 and TYK2. Changes in iNOS activity was monitored by determining nitrite production using the Griess assay and L-arginine transport was monitored using tritiated arginine (L-[3H]arginine). RASMCs were treated with a combination of LPS (100 µg/ml) and IFN- (100 U/ml) and the macrophages with LPS (1 µg/ml) to induce iNOS and transporter activity. Consistent with previous reports, the above treatment of both cell types resulted in the expression of iNOS, production of NO and enhanced transport of L-arginine. These effects were not affected by AG490 but blocked by JAK inhibitor I. Furthermore, although both cell types expressed the key JAKs (JAK2 and TYK2), neither of these proteins were phosphorylated under conditions of induced NO production. Moreover, siRNA experiments showed that JAK2 expression could be abolished without any significant change in NO production, confirming that at least JAK2 may not be required for this process. Whether TYK2 is involved still remains to be resolved as the phosphor-protein could not be detected. However the conclusive siRNA knockdown studies could not be carried out due to time and cost constraints. Apart from iNOS and NO production, changes in induced L-arginine transport were also not significantly affected under the experimental conditions described above suggesting that like with iNOS, induction of L-arginine transport is independent of at least JAK2. Interestingly however, STAT-1 was phosphorylated and this was blocked by JAK inhibitor I but not AG490. Thus, STAT-1 activation may be essential but its activation may be independent of the JAKs. One possible alternate upstream activator of STAT-1 may be the GTPases. Indeed these proteins have been indicated to phosphorylate STAT-1 independent of the JAKs. However, in this project, inhibition of the GTPase pathway enhanced NO production and L-arginine transport suggesting that the GTPases downregulate these processes. In conclusion, the studies carried out in this thesis have shown that induction of iNOS, NO production and L-arginine transport in both RASMCs and J774 macrophages are independent of JAK2 but require STAT-1 activation which may be phosphorylated independently of the JAKs. The role of other JAKs such as TYK2 although unlikely, will need to be resolved using a more specific approach such as siRNA.

Plus de 70% des tumeurs mammaires expriment le récepteur des œstrogènes alpha (ERα), un facteur de transcription dépendant de ses ligands, les œstrogènes. Deux types d’anti-œstrogènes (AE) peuvent être utilisés en clinique pour traiter ces tumeurs : les SERM (Selective Estrogen Receptor Modulators) qui sont des agonistes partiels, tels que le tamoxifène, le SERM le plus communément utilisé en première ligne de traitement. Les SERD (Selective Estrogen Receptor Degraders) sont des AE purs et induisent la dégradation de ERα. Le fulvestrant a été le premier SERD utilisé en clinique, en seconde ligne de traitement après rechute ou dans un contexte métastatique. Cependant, il est très peu biodisponible oralement. ERα est ubiquitiné en présence de SERD, ce qui conduit à sa dégradation par la voie du protéasome. De plus, nous avons montré qu’ERα est également SUMOylé en présence de fulvestrant et d’autres AE purs. Cette SUMOylation contribue à supprimer ses propriétés transcriptionnelles. La SUMOylation et l’ubiquitination sont deux modifications post-traductionnelles similaires et consistent en la conjugaison d’une petite protéine (SUMO1/2/3 ou ubiquitine) sur des résidus d’une protéine cible lors d’une cascade de réactions impliquant trois enzymes : une enzyme activatrice E1, une enzyme de conjugaison E2 et une ligase E3. Ces processus peuvent être réversibles à l’aide de déSUMOylases (SENP) ou déubiquitinases (DUB). Dans un premier temps, nous avons étudié les déterminants structuraux de ERα nécessaires pour sa SUMOylation. Nos analyses Western et nos tests BRET (Bioluminescence Resonance Energy Transfer), dans lesquels nous avons mesuré le transfert d’énergie entre un donneur luminescent (ERα ou ERβ fusionnés à la luciférase) et un accepteur fluorescent (SUMO1 ou SUMO3 fusionnée à la YFP), ont montré que ERα est SUMOylé en présence de fulvestrant, mais pas son paralogue ERβ. Nous avons ensuite créé des chimères dans lesquelles nous avons échangé les domaines de ERα et de ERβ et avons étudié le profil de SUMOylation de ces chimères à l’aide des mêmes essais. Nous avons ainsi mis en évidence que les séquences spécifiques à ERα dans les hélices H3-H4 du domaine de liaison au ligand sont nécessaires et suffisantes pour sa SUMOylation en présence de fulvestrant. Dans cette région, les acides aminés spécifiques à ERα ne sont pas de potentiels substrats de modification. De plus, ces hélices font partie du sillon de recrutement de cofacteurs de ERα, ce qui suggère qu’il y a un recrutement différentiel de la machinerie de SUMOylation par ERα et ERβ en présence de fulvestrant. Nous avons montré que la surexpression de PIAS1 et PIAS2 peut augmenter le signal de SUMOylation. Nous avons ensuite étudié le lien entre la SUMOylation et l’ubiquitination de ERα induites par les AE. Nous avons adapté nos tests BRET pour mesurer l’ubiquitination de ERα, en réalisant des essais Ubi-BRET (Ubiquitine fusionnée à la YFP). Nous avons ainsi pu observer que la SUMOylation et l’ubiquitination de ERα sont affectées par les mêmes mutations en présence d’AE, suggérant un lien entre ces deux voies de modification de ERα induites par les AEs purs. Des essais en cinétique BRET ont montré que ces modifications se font en parallèle et que l’utilisation d’inhibiteurs de SUMOylation chimiques (ML-792 et TAK-981) ou protéiques (déSUMOylase SENP1/2) abrogent la SUMOylation de ERα. En revanche, l’ubiquitination de ERα par différents AE est partiellement diminuée lorsque ces inhibiteurs sont utilisés. La surexpression des protéines RNF4 et RNF111, qui sont des enzymes E3 ubiquitine ligases qui vont ubiquitiner spécifiquement les protéines SUMOylées (STUbL), augmente l’ubiquitination de ERα en présence de fulvestrant et d’autres SERD, de manière dépendante de la SUMOylation. Cependant, l’inhibition de la SUMOylation par les inhibiteurs ML-792 et TAK-781 n’a pas eu d’effet sur la dégradation de ERα par la voie du protéasome dans les cellules ER-positives MCF-7 et T47D. Ces résultats suggèrent que la SUMOylation et l’ubiquitination de ERα ont lieu en parallèle et nécessitent la même conformation de ERα induite par les AE. Enfin, ces deux modifications peuvent être reliées, grâce aux protéines STUbL de manière spécifique des cellules. En résumé, ces projets nous ont permis de mieux comprendre les mécanismes d’action des anti-œstrogènes purs utilisés dans le cancer du sein ER-positif. Nos découvertes pourront contribuer à anticiper des mécanismes possibles de résistance à ces molécules et permettre d’optimiser le développement de nouveaux anti-œstrogènes plus efficaces dans l’induction de la SUMOylation. / More than 70% of mammary tumors are positives for the expression of the estrogen receptor alpha (ERα), a ligand-dependent transcription factor, activated by estrogens. Two types of antiestrogens (AE) are used in the clinic to treat these tumors: the SERMs (Selective Estrogen Receptor Modulators), which are partial agonists, with the tamoxifen, the most common SERM used in first line of treatment. The SERDs (Selective Estrogen Receptor Degraders) are pure antiestrogens that induced ERα degradation. Fulvestrant has been the first SERD used in the clinic, in second line of treatment after relapse or in a metastatic setting. However, fulvestrant is poorly orally bioavailable. ERα is ubiquitinated in the presence of SERDs and this induce its degradation via the proteasome pathway. In our lab, we have showed that ERα is also SUMOylated in the presence of fulvestrant and other pure antiestrogens. This SUMOylation suppresses its transcriptional properties. SUMOylation and ubiquitination are two similar post-translational modifications that consist in the conjugation of a small protein (SUMO1/2/3 or ubiquitin) on residues of a target protein during an enzymatic cascade implicating three enzymes: an E1 activating, an E2 conjugating and an E3 ligase. These processes are reversible thanks to deSUMOylases (SENPs) and deubiquitinases (DUBs). First, we have studied the structural determinants of ERα required for its SUMOylation. Our Western and BRET (Bioluminescence Resonance Energy Transfer) analyses, where we measured the energy transfer between a luminescent donor (ERα or ERβ fused to the luciferase) to a fluorescent acceptor (SUMO1 or SUMO3 fused to the YFP), have showed that ERα is SUMOylated in the presence of fulvestrant but not its paralog ERβ. We have created chimeras by exchanging domains between ERα and ERβ et we have studied the SUMOylation profile of these chimeras with these assays. We showed that the specific sequence to ERα in the H3-H4 helices of the ligand binding domain are necessary and sufficient to induce its SUMOylation in the presence of fulvestrant. In this region, the amino acids that are specifics to ERα are not potential substrates of modification. Moreover, these helices are part of the cofactor binding groove of ERα, suggesting that there is a differential recruitment of the SUMOylation machinery by ERα compared to ERβ in the presence of fulvestrant. We have also showed that the overexpression of the E3 SUMO ligases PIAS1 and PIAS2 can increase the SUMOylation signal. We then studied the parallel between SUMOylation and ubiquitination of ERα induced by AE. We adapted our BRET assays to measure the ubiquitination of ERα, by performing Ubi-BRET assays (ubiquitin fused to the YFP). We observed that SUMOylation and ubiquitination of ERα are affected by the same mutations in the presence of AE, suggesting that there is a crosstalk between these two modification pathways of ERα induced by AE. BRET kinetic assays showed that these modifications happened in parallel and that the use of SUMOylation inhibitors (chemicals ML-792 and TAK-981; or deSUMOylases SENP1/2) abrogated the SUMOylation of ERα. However, the ubiquitination of ERα by different AE is partially decreased when these inhibitors are used. The overexpression of RNF4 and RNF111, which are SUMO-targeted ubiquitin ligases (STUbLs), that will specifically ubiquitinate SUMOylated target proteins, increased the ubiquitination of ERα in the presence of fulvestrant and other SERDs, in a SUMO-dependent manner. However, the inhibition of SUMOylation by ML-792 and TAK-981 inhibitors did not impact the degradation of ERα induced by the proteasome in ER-positive breast cancer cells MCF7 and T47D. These results suggest that SUMOylation and ubiquitination of ERα happened in parallel and required the same conformation of ERα induced by AE. Finally, these two modifications can be linked thanks to STUbLs in a cell-specific manner. These projects led us to better understand the mechanisms of action of pure AE used in the treatment of ER-positive breast cancer. Our findings will contribute to anticipate possible mechanisms of resistance to AE et will help for the optimization and the development of new AE, more efficient in inducing ERα SUMOylation.

Tumor-induced immunosuppression can occur by multiple mechanisms, each posing a significant obstacle to immunotherapy. Evidence presented in this dissertation suggests that aberrant cytokine signaling, as a result of altered metabolism of Chronic Lymphocytic Leukemia (CLL) cells, confers a selective advantage for tumor survival and growth. Cells from CLL patients with aggressive disease (as indicated by high-risk cytogenetics) were found to exhibit prolongation in Interferon (IFN)-induced STAT3 phosphorylation, and increased levels of reactive oxygen species (ROS) in these cells reflected these signaling processes. Changes in the relative balance of phospho-STAT3 and phospho-STAT1 levels, in response to combinations of IL-2 + Toll-like receptor (TLR)-7 agonist + phorbol esters, as well as IFN, were associated with the immunosuppressive and immunogenic states of CLL cells. In addition, immunosuppressive leukemic cells were found to express high levels of proteins with O-linked N-acetylglucosamine (O-GlcNAc) modifications, due to increased metabolic activity through the Hexosamine Biosynthetic Pathway (HBP), which caused impaired intracellular signaling responses and affected disease progression. A conclusion of the studies presented here is that the intrinsic immunosuppressive properties of leukemic cells may be overcome by agents such as Resveratrol that target metabolic pathways of these cells.

Tumor-induced immunosuppression can occur by multiple mechanisms, each posing a significant obstacle to immunotherapy. Evidence presented in this dissertation suggests that aberrant cytokine signaling, as a result of altered metabolism of Chronic Lymphocytic Leukemia (CLL) cells, confers a selective advantage for tumor survival and growth. Cells from CLL patients with aggressive disease (as indicated by high-risk cytogenetics) were found to exhibit prolongation in Interferon (IFN)-induced STAT3 phosphorylation, and increased levels of reactive oxygen species (ROS) in these cells reflected these signaling processes. Changes in the relative balance of phospho-STAT3 and phospho-STAT1 levels, in response to combinations of IL-2 + Toll-like receptor (TLR)-7 agonist + phorbol esters, as well as IFN, were associated with the immunosuppressive and immunogenic states of CLL cells. In addition, immunosuppressive leukemic cells were found to express high levels of proteins with O-linked N-acetylglucosamine (O-GlcNAc) modifications, due to increased metabolic activity through the Hexosamine Biosynthetic Pathway (HBP), which caused impaired intracellular signaling responses and affected disease progression. A conclusion of the studies presented here is that the intrinsic immunosuppressive properties of leukemic cells may be overcome by agents such as Resveratrol that target metabolic pathways of these cells.