Les interférons (IFNs) de type I sont des cytokines qui jouent un rôle capital dans les défenses immunes, antivirales et antiprolifératives de l'organisme. En se liant à leur récepteur de surface, composé des deux sous-unités IFNAR1 et IFNAR2, ils induisent la cascade de signalisation JAK/STAT qui aboutit à leur effets biologiques. L'objectif de ma thèse était d'identifier des mécanismes de régulation de la signalisation des IFNs. Dans ce contexte, nous nous sommes intéressés à la palmitoylation du récepteur de l'IFN de type I, une modification lipidique souvent impliquée dans le trafic et la signalisation des protéines. Par marquage métabolique au palmitate tritié, nous avons montré qu'IFNAR1 et IFNAR2 sont palmitoylées. Le domaine cytoplasmique d'IFNAR1 contient deux cystéines, Cys463 et Cys502, qui sont des sites potentiels de palmitoylation. A l'aide de mutants sur chacune de ces cystéines, nous avons montré qu'IFNAR1 est palmitoylée uniquement sur sa cystéine la plus proche de la membrane plasmique, la Cys463. Un mutant non palmitoylé dans lequel cette cystéine a été remplacée par une alanine nous a permis de constater que la palmitoylation d'IFNAR1 n'est pas impliquée dans son trafic intracellulaire, dans son endocytose ni dans sa stabilité, mais qu'en revanche elle joue un rôle crucial dans l'activation de la voie de signalisation JAK/STAT. De façon concordante, la palmitoylation d'IFNAR1 est requise pour l'activation transcriptionnelle des gènes induits spécifiquement par l'IFN-a. Par contre, un défaut de palmitoylation n'influence nullement l'activité antiproliférative de l'IFN-a, en dépit du rôle de cette modification dans la signalisation.

Mammalian Scribble is a target for proteins encoded by human papilloma virus, retro- and flaviviruses. Tick-borne encephalitis virus (TBEV) is a flavivirus that have evolved distinct strategies to escape antiviral responses. Information of how flaviviruses intrude on cell integrity comes from understanding of the roles that host-factors play when they interfere with viruses. The first part of this thesis describes a novel interaction between the TBEVNS5 protein and Scribble. The importance of the interaction was demonstrated by RNAi-mediated depletion of Scribble, which prevented suppression of JAK-STAT signaling by NS5. Together, these results define Scribble as a novel target for NS5. TBEV is known to cause central nervous system disease TBE in humans that can lead to cognitive dysfunction. A unifying theme in CNS related diseases are defects in neuronal extensions. We therefore addressed the effects of TBEV expression in PC12 cell differentiation, which is characterized by extensive neurite growth. Our data show that TBEVNS5 suppresses neurite outgrowth through the Rho GTPase Rac1. These findings provide evidence that Rac1 is an indirect target of NS5 in neurite inhibition. Scribble was recently implicated in spine morphogenesis. Thus, we tested the role of Scribble in neurite elongation. Depletion of Scribble in PC12 cells, reduced neurite density but increased length of those remaining. Moreover, Scribble bound components in the Ras/ERK cascade in a growth factor dependent manner. Together, these results demonstrate that Scribble controls neurite elongation by scaffolding MAPK components. Moreover, as loss of dendritic spines, actin-rich protrusions on neurons, is a feature in cognitive dysfunction we speculate that cognitive dysfunction in TBE might involve disturbed Scribble expression by NS5. We also investigated the binding between NS1 of Influenza A virus and Scribble. The PDZ domains of Scribble are usually selective for specific C-terminal motifs in proteins. Because NS1 has a canonical PDZ motif we tested if binding to Scribble depends on this motif. We found that Scribble binds NS1; the association is dependent on the NS1 C-terminus that is recognized by PDZ3-4 of Scribble. Together, these results suggest that Scribble is a target for the H5N1 NS1 protein / At the time of the doctoral defense, the following papers were unpublished and had a status as follows: Paper 2: In press. Paper 3: Manuscript. Paper 4: Manuscript.

The JAK/STAT pathway is but one of the signal transduction cascades responsible for proper development and homeostasis. Gain-of-function mutations of pathway components are causative agents of several leukemias, highlighting the necessity for proper regulation of signal transduction. Drosophila presents an attractive model to study JAK/STAT signaling because mutations in the pathway behave in an analogous manner. Furthermore, the Drosophila cascade is much simpler as only one of each component required for activation has been characterized; whereas in mammals, there are many ligands, receptors, 4 JAKs and 7 STATs.Suppressors of Cytokine Signaling (SOCS) are one family of molecules which regulate JAK/STAT signaling via a negative feedback loop. All SOCS share a distinct modular domain architecture, which we exploited to locate three putative SOCS homologues within the Drosophila genome. I present the identification and initial characterization of one of these homologues, Socs44A. I show that Socs44A is not responsive to or dependent on JAK activity. However, I demonstrate that Socs44A is capable of downregulating JAK/STAT signaling in the developing wing but not inoogenesis, indicating that its ability to regulate the pathway is tissue specific, a phenomenon observed in the mammalian model.Signal transduction pathways are integrated at multiple levels. This interplay allows for combinatorial signaling, resulting in a higher order of complexity in the signals that can be received and interpreted by a cell. Well documented are the interactions between the JAK/STAT and the EGFR/MAPK pathways. In this work, I show that Socs44A can genetically interact with, and upregulate, the EGFR/MAPK pathway, analogous to a recent report involving SOCS-3.Starting with the Drosophila genome sequence, I initiated a reverse genetic approach to studying the function of the Socs44A locus. During the course of this investigation, I designed and implemented a novel post-processor of the BLAST algorithm, called Multi-BLAST, which facilitates retrieval of multiple domain sequences from public databases. In what would have been the ultimate achievement of this study, I attempted two mutagenesis screens designed to isolate Socs44A loss-of-function alleles. Progress on these screens is reported.

Equine herpesvirus-1 (EHV-1) is one of the major viral pathogens causing respiratory disease, abortion, perinatal mortality and neurologic disease among horses resulting in significant economic losses to the equine industry. The virus can also remain latent in the horses and recrudesce at any time. Type-I interferons (IFNs) act as a first line of defense against many viral infections. In this study we investigated the type-I IFN response against the neuropathogenic T953 strain of EHV-1 in equine endothelial cells (EECs). The results showed that after a transient induction of IFN-β mRNA as well as protein at an early time (3h) post infection (p.i.), T953 strain of EHV-1 suppressed further induction of IFN-β at later times (12h onwards). Studies were done to confirm that the suppression of type-I IFN induction at later time points was not due to the normal IFN-β induction kinetics, it was rather because of the active interference by the virus. Investigation of the mechanisms by which T953 interferes with IFN-β production revealed that the virus degraded the endogenous level of the transcription factor, interferon regulatory factor 3 (IRF-3) and also down-regulated the activation of IRF-3 followed by its accumulation in the nucleus. However, T953 infection caused degradation of nuclear factor κB (NF-κB) inhibitory protein IκBα and also induced p50 subunit to translocate into nucleus from cytoplasm suggesting activation of NF-κB signaling. This also indicated that inhibition in the type-I IFN production was probably not due to the inhibition of NF-κB. The results of these studies also indicated that T953 virus was resistant to the biological effect of the recombinant equine IFN-α in vitro. Investigation of the reason of this resistance showed that T953 virus interfered with the cellular JAK-STAT signaling mechanism by which type-I IFN exerts its antiviral effect. Moreover, the studies revealed that downstream of the JAK-STAT signaling, T953 virus also inhibited the expression of cellular antiviral proteins including interferon stimulated gene 56 (ISG56) and viperin. Altogether, these data indicate that the T953 strain of EHV-1 interfered with the host cell innate immune responses by modulating type-I IFN mediated immune responses at multiple levels in vitro.

In the Drosophila testis, sperm are derived from germline stem cells (GSCs) which undergo a stereotyped pattern of divisions and differentiation. The somatic cells at the tip of the testis form the hub, which is the niche for both the somatic cyst stem cells (CySCs) and GSCs. The hub expresses Upd, a ligand for the JAK/STAT pathway that has roles in the maintenance of CySCs and GSCs. Male mutants of upd3, another ligand of the JAK/STAT pathway, become sterile much earlier than the wild-type, leading to the hypothesis that similar to upd, upd3 also promotes the self-renewal of stem cells in testis. It was found here that upd3 is also expressed in the hub, and that mutants of upd3 have fewer CySCs and GSCs. Using a GFP reporter of the JAK/STAT pathway, it was found that the JAK/STAT pathway is not only activated in the stem cells, consistent with its known function in the maintenance of stem cells, but is also activated in the elongated cyst cells that encapsulate late stage differentiating spermatids. The reduction of JAK/STAT activity in the somatic cyst cells led to impaired spermatid individualization, a late stage of spermatogenesis during which the syncytial spermatids are separated. The impairment of individualization was shown by the loss of three characteristic structures: individualization complexes (ICs), cystic bulges (CBs), and waste bags (WBs). The failure of IC formation implies STAT activity is required for the initiation of individualization, and the loss of CBs and WBs suggests STAT activity is required for the progression of individualization. Activation of caspases in elongated spermatids is known to be required for individualization. The reduction of JAK/STAT activity in cyst cells almost completely eliminated the activation of two effector caspases: drICE and DCP-1. It was concluded that JAK/STAT activity in somatic cyst cells promotes individualization by stimulating caspase activity in spermatids. The JAK/STAT pathway is not only required for the maintenance of stem cells at the tip, but also required for individualization away from the tip during late differentiation, thus is reutilized in Drosophila spermatogenesis.

The duration and intensity of the JAK/stat signaling must be tightly regulated to prevent excessive transcriptional response and to reset the pathway to receive additional signals. Socs are the largest class of these regulators in mammals. Eight Socs genes have been found in mammals. CIS, and SOCS1-3, the canonical Socs, are transcriptionally activated by and down-regulate the JAK signaling. Socs4-7, the non-canonical Socs, are less studied and their relationship with the JAK/STAT pathway has not been well established. The Drosophila genome encodes three non-canonical Socs homologues, Socs16D, Socs36E, and Socs44A. Expression of Socs36E is controlled by the JAK pathway and misexpression causes phenotypes similar to that from reduction of JAK in both ovary and wing, which may make it functionally more similar to the canonical Socs. Expression of Socs44A is not controlled by the JAK pathway and misexpression causes JAK mutant phenotypes in wing but not in ovary. Imprecise excision mutants of the three Socs genes have been generated by us and have no visible phenotypes. The mutants of Socs36E and Socs44A significantly enhance the tumor formation in hopTum-l mutant, a gain-of-function mutation of the JAK/STAT pathway. The function of Drosophila Socs will be further studied with different strategies.

L’adaptateur LNK est un régulateur négatif des voies de signalisation, dont la voie JAK/STAT,essentielle au développement du système hématopoïétique. Son implication dans les hémopathies chroniques, notamment les Néoplasmes Myéloprolifératifs (NMP), a été mise en évidence par l’analyse de souris invalidées pour cet adaptateur et l’identification de mutations de LNK chez les patients atteints de ces pathologies. Toutefois, le mécanisme permettant la régulation de ses partenaires, dont la kinase JAK2, et l’implication fonctionnelle des mutations de LNK dans les NMP, restent à définir. Ainsi, mon projet de thèse a porté sur l’analyse structurale et fonctionnelle des complexes de signalisation LNK/JAK2 et sur le développement d’une stratégie moléculaire pour l’utilisation thérapeutique de LNK dans les NMP. Nos résultats ont montré pour la première fois, la fonction inhibitrice de la région N-terminale incluant le domaine d’homologie à la Pleckstrine deLNK sur JAK2 normale et de manière plus importante, sur la forme mutée JAK2-V617F, retrouvée chez les patients atteints de NMP. De plus, nos études sur les mutations de LNK localisées dans cette région régulatrice, ont permis de comprendre leur contribution dans le développement de ces hémopathies et de proposer un mécanisme d’inhibition de l’activation de JAK2 par LNK. Nos résultats permettent d’utiliser le ciblage de la région N-terminale de LNK comme stratégie moléculaire inhibant spécifiquement la forme oncogénique JAK2-V617F à l’aide de peptides pénétrants (CPP). A long terme, cette approche pourrait être utilisée comme outil thérapeutique dans le traitement de patients atteints de NMP positifs pour JAK2-V617F. / The LNK adaptor protein is a key negative regulator of signalling pathways, such as JAK/STAT, important in the development of the hematopoietic system. Its implication in chronic blood diseases, such as Myeloproliferative Neoplasms (MPN) has been confirmed by studies on Lnk-deficient mice, as well as the identification of LNK mutations in MPN patients. However, the LNK mechanism of regulation on its partners and the functional implication of LNK mutations in MPN pathogenesis, are still unclear. Therefore, my PhD project covers the structural and functional analysis of theLNK/JAK2 signalling complex and the development of a molecular strategy to use LNK as a therapeutic tool for the treatment of MPN patients. Our study showed, for the first time, the inhibitory function of the N-terminal region and the pleckstrin homology domain of LNK on JAK2 activity, which occurs more importantly on JAK-V617F than JAK2 wild type form. Moreover, our study provided evidence on how LNK mutations located in this LNK region could contribute to these haematological diseases and has allowed us to propose a model for LNK regulatory function on JAK2activity. Furthermore, we developed a cell penetrating peptide-based strategy to deliver this regulatory region of LNK in hematopoietic cells to specifically inhibit JAK2-V617F oncogenic form. The finalaim is to use this region as a therapeutic molecule to treat JAK2-V617F-positive MPN patients.

Submitted by Renata Lopes (renatasil82@gmail.com) on 2016-02-15T17:10:47Z No. of bitstreams: 1 renatamendesdefreitas.pdf: 1082430 bytes, checksum: 17933c41d6c615bede876aefdcce85ef (MD5) / Approved for entry into archive by Adriana Oliveira (adriana.oliveira@ufjf.edu.br) on 2016-02-26T12:27:10Z (GMT) No. of bitstreams: 1 renatamendesdefreitas.pdf: 1082430 bytes, checksum: 17933c41d6c615bede876aefdcce85ef (MD5) / Made available in DSpace on 2016-02-26T12:27:10Z (GMT). No. of bitstreams: 1 renatamendesdefreitas.pdf: 1082430 bytes, checksum: 17933c41d6c615bede876aefdcce85ef (MD5) Previous issue date: 2014-02-21 / CAPES - Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / As Neoplasias Mieloproliferativas são originadas por uma proliferação clonal de um progenitor hematopoético. Descrita inicialmente em 1951 como “Doenças Mieloproliferativas” e reavaliada pela Organização Mundial da Saúde em 2011, as Neoplasias Mieloproliferativas agrupam a Policitemia Vera, Trombocitemia Essencial e Miefibrose Primária em um subgrupo chamado de BCR-ABL negativo. De acordo com a revisão dos critérios adotados para o diagnóstico das Neoplasias Mieloproliferativas a presença da mutação JAK2 V617F é considerada critério de maior importância para o diagnóstico do subgrupo BCR-ABL negativo representando um marcador clonal. A Policitemia Vera é principalmente diagnosticada por um aumento na massa eritrocitária independente de fator estimulante, aumento de leucócitos no sangue periférico, esplenomegalia e trombocitose. Ocorre, geralmente, em homens e mulheres com faixa etária de 60 anos, com incidência de 0,7 a 2,6/100.000 habitantes/ano. A mutação V617F no gene JAK2 produz uma proteína alterada que ativa constitutivamente a via JAK-STAT e outras vias downstream de modo que, as proteínas de ativação transcricional e transdutoras de sinal (STAT do inglês Signal Transducerand Activator of Transcription) são fosforiladas posteriormente impactando na expressão de genes envolvidos na regulação da apoptose e proteínas regulatórias, além de alterar a taxa de proliferação das células hematopoéticas. Neste trabalho foram selecionados 26 pacientes com Policitemia Vera, os quais foram atendidos na cidade de Juiz de Fora, Minas Gerais. Foram realizadas análises de mutações no gene JAK2 a partir do material genético isolado do sangue periférico. Os dados clínicos de cada paciente foram relacionados entre si e correlacionados com os dados moleculares. / Myeloproliferative neoplasms (MPN) are caused by a clonal proliferation of a hematopoietic progenitor. First described in 1951 as “Myeloproliferative Diseases” and reevaluated by the World Health Organization in 2011 (WHO, 2011), the Myeloproliferative Neoplasms gather the Polycythemia Vera, Essential Thrombocythemia and Primary Mielofibrose in a subgroup called negative BCR-ABL. According to WHO the revised diagnosis criteria for myeloproliferative neoplasms, the presence of JAK2 V617F mutation, is considered the most important criterion for the diagnosis of negative BCR-ABL subgroup representing a clonal marker. The Polycythemia Vera is primarily diagnosed by an independent stimulating factor in red cells mass increasing, followed by an increase of leukocytes in peripheral blood, spleen and thrombocytosis. It usually occurs in men and women of all age groups with higher incidence in 60-year-old patients ranging from 0.7 to 2.6/ 100.000 hab/year. The V617F mutation in the JAK2 gene produces an altered protein that activates constitutively the JAK-STAT pathway and other pathways downstream as a result the protein transcriptional activation and signal transduction (STAT) are subsequently phosphorylated causing impact in the expression of genes involved in regulation of apoptosis and regulatory proteins, as well as altering the rate of proliferation of hematopoietic cells. In this study 26 patients with Polycythemia Vera were selected in Juiz de Fora, Minas Gerais. The peripheral blood was used for genetic material isolation (JAK2 mutation). The clinical data of each patient were related to each other and correlated with the molecular data.

<p>Functioning as a “powerhouse”, the liver adapts to the metabolic needs of the body by maintaining a homeostatic balance. Prolactin receptor (PRLR) has been found to have a copious existence in the liver. Having established a well-defined role in both reproductive and endocrine systems, the role of this transmembrane protein in hepatocytes is yet to be elucidated. Due to its abundant nature, we hypothesized that PRLR is required for maintaining hepatic homeostasis and plays a role in liver diseases. To test this hypothesis, we defined two specific aims. The first was to explore whether PRLR loss-of-function affects liver structure and function in physiological conditions. The second was to determine whether PRLR is associated with liver pathology. We deleted the <i>Prlr</i> gene specifically in hepatocytes using a virus-based approach and evaluated liver function, transcriptome, and activities of downstream signaling molecules. Due to the absence of PRLR, we found that the urea cycle was disrupted, concomitant with excessive accumulation of urea in the blood; 133 genes exhibited differential expression, largely associated with hepatocyte structure, metabolism, and inflammation; and the activities of STAT3 and 5 were reduced. The results signify that PRLR indeed plays a homeostatic role in the liver. We also used <i>Prlr</i>^+/- mice to assess whether the loss of one allele of the <i>Prlr</i> gene alters maternal hepatic adaptations to pregnancy. As a result, in the pre-pregnancy state and during the first half of gestation, the expression of maternal hepatic PRLR protein was reduced approximately by half owing to <i>Prlr</i> insufficiency. However, during the second half of pregnancy, we observed compensatory upregulation of this molecule, leading to minimal interference in STAT 3 and 5 signaling and liver size. Contrary to a previous study in the breast and ovary, our results suggest that one allele of <i>Prlr</i> may be sufficient for the maternal liver to respond to this physiological stimulus (pregnancy). Furthermore, we examined the expression and activity of PRLR in fatty as well as cholestatic livers. Using a high fat diet, we induced non-alcoholic fatty liver disease (NAFLD). Strikingly and for the first time, we discovered that the short isoform of PRLR (PRLR-S) was completely inactivated in response to NAFLD, whereas the long isoform remained unchanged. This finding strongly suggests the involvement of PRLR-S in lipid metabolism. We also postulate that PRLR-L may be the major regulator of STAT signaling in the liver, consistent with other reports. Lastly, we induced extrahepatic cholestasis via bile duct ligation (BDL) in mice. As this liver disease progressed, the expression of both isoforms of PRLR generally declined and was surprisingly accompanied by increased STAT 3 and 5 activity. The data suggests that PRLR participates in this disease progression, with a disconnection between PRLR signaling and STAT proteins. Collectively, our preliminary studies suggest that PRLR signaling is required to maintain liver homeostasis and more prominently, is involved in liver diseases, especially NAFLD. These findings lay a foundation for our future studies.</p>

Indiana University-Purdue University Indianapolis (IUPUI) / Functioning as a “powerhouse”, the liver adapts to the metabolic needs of the body by maintaining a homeostatic balance. Prolactin receptor (PRLR) has been found to have a copious existence in the liver. Having established a well-defined role in both reproductive and endocrine systems, the role of this transmembrane protein in hepatocytes is yet to be elucidated. Due to its abundant nature, we hypothesized that PRLR is required for maintaining hepatic homeostasis and plays a role in liver diseases. To test this hypothesis, we defined two specific aims. The first was to explore whether PRLR loss-of-function affects liver structure and function in physiological conditions. The second was to determine whether PRLR is associated with liver pathology. We deleted the Prlr gene specifically in hepatocytes using a virus-based approach and evaluated liver function, transcriptome, and activities of downstream signaling molecules. Due to the absence of PRLR, we found that the urea cycle was disrupted, concomitant with excessive accumulation of urea in the blood; 133 genes exhibited differential expression, largely associated with hepatocyte structure, metabolism, and inflammation; and the activities of STAT3 and 5 were reduced. The results signify that PRLR indeed plays a homeostatic role in the liver. We also used Prlr+/- mice to assess whether the loss of one allele of the Prlr gene alters maternal hepatic adaptations to pregnancy. As a result, in the pre-pregnancy state and during the first half of gestation, the expression of maternal hepatic PRLR protein was reduced approximately by half owing to Prlr insufficiency. However, during the second half of pregnancy, we observed compensatory upregulation of this molecule, leading to minimal interference in STAT 3 and 5 signaling and liver size. Contrary to a previous study in the breast and ovary, our results suggest that one allele of Prlr may be sufficient for the maternal liver to respond to this physiological stimulus (pregnancy). Furthermore, we examined the expression and activity of PRLR in fatty as well as cholestatic livers. Using a high fat diet, we induced non-alcoholic fatty liver disease (NAFLD). Strikingly and for the first time, we discovered that the short isoform of PRLR (PRLR-S) was completely inactivated in response to NAFLD, whereas the long isoform remained unchanged. This finding strongly suggests the involvement of PRLR-S in lipid metabolism. We also postulate that PRLR-L may be the major regulator of STAT signaling in the liver, consistent with other reports. Lastly, we induced extrahepatic cholestasis via bile duct ligation (BDL) in mice. As this liver disease progressed, the expression of both isoforms of PRLR generally declined and was surprisingly accompanied by increased STAT 3 and 5 activity. The data suggests that PRLR participates in this disease progression, with a disconnection between PRLR signaling and STAT proteins. Collectively, our preliminary studies suggest that PRLR signaling is required to maintain liver homeostasis and more prominently, is involved in liver diseases, especially NAFLD. These findings lay a foundation for our future studies.