Epilepsy, a disease characterized by recurring spontaneous seizures, affects over 65 million people, 2% of the world’s population. Over 30% of patients are refractory to all current medical therapy, and for those that can be treated, many suffer from severe drug side-effects. Understanding the molecular basis of epilepsy is vital to the advancement of better therapeutic options and an eventual cure. Upregulation of brain-derived neurotrophic factor (BDNF) is highly associated with epileptogenesis in human patients, as well as animal models. Our laboratory discovered that BDNF induces the Janus Kinase/Signal Transducer and Activator of Transcription (JAK/STAT) pathway in neurons and that inhibition attenuates spontaneous seizures in a temporal lobe epilepsy model. The mechanism behind JAK/STAT signaling in neurons and its relationship to epilepsy still remains to be elucidated and is the subject of my thesis. Surprisingly, even though BDNF is such a major signaling molecule, its full genomic impact has never been assessed. We conducted a high-density RNA-sequencing analysis of the BDNF transcriptome in cortical neurons and probed such regulation with selective JAK inhibitors. Results suggest that 68% of BDNF-induced changes in gene expression implicated in epilepsy are regulated by JAK/STAT signaling. Eighty percent of BDNF-induced changes coding for proteins involved in synaptic neurotransmission (receptor subunits and ion channels) involve JAK/STATs. Additionally, these datasets include genes that have never been associated with BDNF regulation (such as Dopamine Receptor D5 and Galanin Receptor 1). Most interestingly, the datasets reveal that BDNF-induced JAK/STAT signaling in neurons is non-canonical, as STAT3 phosphorylation at tyrosine 705 is not required for action. To directly examine STAT3’s role in epileptogenesis, we studied the transcriptome of transgenic mice that express lower levels of STAT3 specifically in neurons. Using the intrahippocampal kainic-acid (KA) model of epilepsy, our datasets suggest that STAT3 knockdown in vivo, and selectively in neurons, protects mice from KA-induced dysregulation of the sphingolipid metabolism pathway that is associated with the trafficking, sorting, and stability of membrane-bound proteins, including neurotransmitter receptors and ion channels. Finally, we discuss a model for JAK/STAT signaling in neurons that includes structural aspects of an intracellular BDNF receptor (p75NTR) associated with JAK2. / 2021-06-14T00:00:00Z

Yes / The realisation that unregulated activation of the Janus kinase–signal transducer and activator of transcription (JAK–STAT) pathway is a key driver of a wide range of diseases has identified its components as targets for therapeutic intervention by small molecule inhibitors and biologicals. In this review, we discuss JAK-STAT signalling pathway inhibition by the inducible inhibitor “suppressor of cytokine signaling 3 (SOCS3), its role in diseases such as myeloproliferative disorders, and its function as part of a multi-subunit E3 ubiquitin ligase complex. In addition, we highlight potential applications of these insights into SOCS3-based therapeutic strategies for management of conditions such as vascular re-stenosis associated with acute vascular injury, where there is strong evidence that multiple processes involved in disease progression could be attenuated by localized potentiation of SOCS3 expression levels. / British Heart Foundation; Chief Scientist's Office; NHS Greater Glasgow and Clyde Research Endowment Fund; BBSRC

Hypothèse : Notre équipe étudie le rôle, mal connu, du trafic membranaire dans le contrôle de l’activation de la voie de signalisation JAK/STAT par les interférons (IFN), une voie clé du contrôle des processus cancéreux. La liaison de l’IFN-a à son récepteur IFNAR active les kinases JAK1 et TYK2 puis des transducteurs de signal comme STAT1, antiprolifératif, ou STAT3, qui a un pouvoir oncogénique. Le laboratoire a démontré récemment que le trafic membranaire d’IFNAR détermine la spécificité du signal des différents IFNs.L’objet de cette thèse est l’étude du rôle des cavéoles dans ce contrôle. Les cavéoles sont des invaginations membranaires enrichies en cholestérol et glycosphingolipides, formées par l’oligomérisation de la cavéoline1 (Cav1). Les cavéoles ou le gène CAV1 ont souvent été associés à la progression tumorale, notamment des cellules mammaires, mais ce rôle reste énigmatique et controversé. Le fait que IFNAR ait été détecté par biochimie dans des fractions de membrane enrichies en cholestérol et positives pour la cavéoline-1 chez la souris et le fait que l’expression du gène CAV1 ait été corrélée à l’action antitumorale de l’IFNa nous ont conduit à étudier le rôle des cavéoles dans l’action antitumorale des IFNs.Résultats: Le rôle putatif des cavéoles sur le contrôle de la voie JAK/STAT a été étudié dans des cellules murines MLEC n’exprimant pas Cav1 et dans des lignées humaines par interférence ARN contre Cav1. Nous avons pu démontrer que la présence de Cav1 régule de manière opposée deux étapes de la voie de signalisation de STAT3 activée par l’IFN-a. Par contre, ni l’activation de STAT1 par l’IFN-a ni celle de STAT3 par les autres IFNs ne nécessitent Cav1. Parallèlement, le laboratoire a montré que les cavéoles jouent un rôle capital dans la réponse cellulaire aux stress mécaniques en se dépliant lors d’un étirement membranaire, ce qui amortit la tension membranaire. Nous montrons qu’un tel stress mécanique par étirement module spécifiquement la signalisation de STAT3 par l’IFN-a d’une manière dépendante de Cav1 dans les cellules MLEC, suggérant pour la première fois un rôle de STAT3 et de l’IFN-a dans la mécanotransduction dépendante des cavéoles. Ce résultat permet aussi de relier les contraintes mécaniques présentes dans la masse tumorale et leur effet sur la progression tumorale. Perspectives : Les IFNs et la voie JAK/STAT sont bien caractérisés pour leur action antiproliférative, mais si l’IFN-a est utilisé en thérapeutique oncologique, les mécanismes de l’effet antitumoral sont mal connus. Nos résultats impliquent pour la première fois les cavéoles dans l’activation sélective du proto-oncogène STAT3 par l’IFN-a et proposent STAT3 comme un des nouveaux acteurs de la mécanotransduction par les cavéoles. Elucider les mécanismes moléculaires mis en jeu dans ces deux fonctions inédites des cavéoles devrait permettre d’identifier de nouvelles cibles thérapeutiques dans la progression tumorale. / Hypothesis: Our team studies the poorly investigated role of membrane trafficking in the control of the activation of the JAK / STAT signaling pathway by interferons (IFN), a key mechanism in the control of tumorigenesis. The binding of the IFN-a to its receptor IFNAR activates the kinases JAK1 and TYK2 and then, signal transducers and activators of transcription including the antiproliferative STAT1 or the oncogenic STAT3. The laboratory demonstrated recently that the trafficking of IFNAR at the plasma membrane determines the signal specificity of the various IFNs.The goal of this thesis was to study the role of caveolae in this control. Caveolae are specialized membrane invaginations enriched in cholesterol and glycosphingolipids, formed by the oligomerization of their main structural protein, caveolin-1 (Cav1). Caveolae or the CAV1 gene have often been associated with tumorigenesis, in particular in mammary cancer cells, but this role remains enigmatic and controversial. The fact that IFNAR was previously found in Cav1-positive lipid microdomains and the fact that the expression of the CAV1 gene had been functionally linked to the antitumoral function of IFN-a led us to investigate the role of caveolae in the antitumoral function of the IFNs.Results: The putative role of caveolae in the control of the JAK / STAT signaling pathway have been studied in murine lung endothelial MLEC cells that do not express Cav1 and in a human lineage by RNA interference against Cav1. We were able to demonstrate that the presence of Cav1 regulates in an opposite manner two stages of the signaling pathway of STAT3 activated by the IFN-a whereas the activation of STAT1 by IFN-a, or STAT3 by the other type I and II IFNs do not require Cav1.At the same time, the laboratory showed that caveolae play a major role in the cellular answer to mechanical stress by flattening during a membrane stretching, thus buffering the membrane tension. We show that mechanical stress by uniaxial cell stretching modulates specifically the signaling pathway of STAT3 activated by the IFN-a in a Cav1-dependant manner in MLEC cells. This result suggests for the first time a role of STAT3 and of IFN-a in caveolae-driven mechanotransduction. This result also allows us to link the mechanical constraints found in the tumoral mass to their effect on tumorigenesis.Prospects:The IFNs and the JAK / STAT signaling pathway protect the cells from tumorigenesis, but although IFN-a is used in oncology, the mechanisms of its antitumoral effect are poorly known. Our results involve for the first time caveolae in the selective activation of the proto-oncogenic STAT3 by the IFN-a and allow us to propose STAT3 and the IFN-a as new actors of the mechanotransduction by caveolae. Clarifying the molecular mechanisms involved in these two new functions of caveolae should allow us to identify new therapeutic targets in tumorigenesis.

Les néoplasmes myéloprolifératifs (NMP) sont hémopathies malignes aboutissant à la surproduction d'une ou plusieurs lignées myéloïdes. Elles sont dues à l'acquisition de mutations sur l'axe de signalisation MPL/JAK2 incluant des mutations de JAK2V617F, de MPL et plus récemment de la calréticuline (CALR), dont les deux principales sont CALRdel52 et CALRins5. Ces mutations de signalisations peuvent être accompagnées de mutations de l'épigénétique, les plus importantes étant des mutations dans TET2. Le but de cette thèse était d'étudier le rôle des mutations de TET2 et de la calrdel52 dans les NMP grâce à une technologie de cellules souches induites à la pluripotence (IPSC). Dans la première partie j'ai pu démontrer que TET2 joue un rôle dans le processus de reprogrammation, vraisemblablement de manière indépendante de son activité catalytique. Dans la seconde partie, j'ai démontré que CALRdel52 joue un rôle dans les MPN en provoquant une hypersensibilité et une pousse indépendante de la TPO des progéniteurs mégakaryocytaires ainsi qu'une hyperprolifération des mégacaryocytes, liées à l'activation constitutive de stat3 et de ERK. J'ai également démontré une pousse indépendante du GCSF des granulocytes. Ce travail a donc permis de mettre en lumière le rôle du facteur épigénétique TET2 dans le processus de reprogrammation ainsi que le rôle de CALRdel52 dans les MPN dans un contexte d'expression endogène. / Myeloproliferative neoplasms (NMP) are hematological malignancies that lead to an ovrproduction of one or more myeloid lineages. They are driving by mutations in MPLl/jak2 signaling pathway, mainly JAK2V617F, MPL, and more recently calreticulin (CARL), with two main mutations being calrdel52 and calrins5. These signaling mutations are sometimes associated with epigenetic mutations, the major one being in tet2. The objective of my thesis was to study the role of TET2 and CALRdel52 in MPN thanks to an induced pluripotent stem cells (IPSC) model. In the first part i demonstrated the role of TET2 in reprogramming process, probably independently of the catalytic domain. In the second part i demonstrated that CALRdel52 induced a TPO hypersensitivity and a TPO indenpendant growth of the megakaryocytic progenitors as well as a hyperproliferation of the megakaryocytes. This phenotype is associated with a constitutive activation of stat3 and ERK. A G-CSF independent growth of the granulocyte was also demonstrated. In conclusion this work underline the role of an epegenetic factor, TET2, in the reprogramming process and demonstrate the role of CALRdel52in MPN with an endogenous expression model.

Ligand-activated signal transduction is a process critical to cell survival and function as it serves as a means of communication between the cells and their environment. Endocytosis is generally thought to down-regulate incoming signals by reducing the surface availability of receptors. However, increasing evidence in many systems suggests a notion which is referred to as the „signalling endosome" hypothesis - that endocytosis can also actively contribute to signalling apart from clearance of activated receptors and thereby attenuation of signalling. The functional aspect of signalling endosomes has been well-characterized in several pathways including RTK and TGF-β signalling. There are, however, various other signalling pathways where the active mechanism of endocytotic regulation is yet to be understood. In this study, we probe this aspect in the cytokine signalling system, where the receptors are known to internalize but the significance of such internalization and precise mechanism is unclear. My thesis aims to elucidate the function and molecular details of internalization of cytokine receptor using interleukin-4 receptor (IL-4R) signalling as a model. IL-4 and IL-13 ligands can induce assembly of three distinct complexes: IL4 induced IL-4Rα – IL-2Rγ (type I), IL-4 induced IL-4Rα – IL-13Rα1 (type II) or the IL-13 induced IL-13Rα1-IL-4Rα (type II). The formation of any of these complexes triggers signalling through the JAK/STAT pathway. However, models of how the oligomerization of the transmembrane receptors and activation takes place are very diverse and lack a clear molecular and biophysical understanding of the underlying receptor dynamics. Previous results of the lab had shown that the affinities between subunits are low, precluding complex formation at the plasma membrane at physiological concentrations. In addition, IL-4R subunits localize in to endosomal structures adjacent to the plasma membrane. It had already been shown that the shared IL-4R subunit IL-2Rγ is internalized by a specific, actin dependent, Rac1/Pak1 regulated endocytosis route in the IL-2 context. We could show that pharmacological suppression of this endocytosis pathway also prevented IL-4 induced JAK/STAT signalling, placing endocytosis upstream of signalling. Here I show using immuno-EM techniques that these endosomal structures are multivesicular bodies. Importantly, I could show that receptor subunits are highly enriched in the limiting membrane of these endosomes relative to the adjacent plasma membrane. Using quantitative loading assays I could furthermore demonstrate that this enrichment is achieved by constitutive internalization of receptors from the cell surface into cortical endosomes. The trafficking kinetics of the receptor subunits is independent of ligand occupancy. Pharmacological inhibition shows that receptors and ligand traffic via the previously identified Rac1/Pak1 pathway. Finally, Vav2 was identified as a candidate Guanine Exchange Factor (GEF) that may regulate Rac1 activity and thereby control the actin polymerization cascade driving IL-4R endocytosis. Immunoprecipitations showed that Vav2 interacts both with the cytoplasmic tail region of the receptors and the receptor associated 2 kinase JAK3. Vav2 may thus couple the receptor/JAK complexes to the Rac1/Pak1 mediated endocytosis route. Taken together, our results suggests that stable „signalling endosomes‟ adjacent to the plasma membrane act as enrichment centres, where ligand and receptor concentrations are locally increased by constitutive trafficking. The confined environment of the endosome then compensates for the weak affinities between the ligand and receptor and facilitates ligand-mediated receptor dimerization. Importantly, overexpression of both type II IL-4R subunits renders signal transduction resistant to endocytosis inhibition, strongly suggesting that the critical factor effecting signalling is sufficient concentration, which the endosomes facilitate achieving. The endosomes are thus dispensable as signalling scaffolds when the receptors are in sufficient concentration, where activated receptors could interact with downstream pathway components. Endocytosis thus provides a crucial means for the signalling process to overcome the thermodynamic hurdles for receptor oligomerization. In conclusion, our data propose a novel, purely thermodynamic role of endosomes in regulating cytokine receptor signalling not seen in any other signalling pathway.

Submitted by PAULO CÉSAR JARK null (paulocjark@hotmail.com) on 2016-12-12T17:43:46Z No. of bitstreams: 1 TESE PAULO JARK IMPRESSÃO.pdf: 1837097 bytes, checksum: e5756c844b29f7062a50211bad6f5b0a (MD5) / Approved for entry into archive by Felipe Augusto Arakaki (arakaki@reitoria.unesp.br) on 2016-12-15T15:04:37Z (GMT) No. of bitstreams: 1 jark_pc_dr_jabo.pdf: 1837097 bytes, checksum: e5756c844b29f7062a50211bad6f5b0a (MD5) / Made available in DSpace on 2016-12-15T15:04:37Z (GMT). No. of bitstreams: 1 jark_pc_dr_jabo.pdf: 1837097 bytes, checksum: e5756c844b29f7062a50211bad6f5b0a (MD5) Previous issue date: 2016-11-18 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) / A via Janus Kinase (JAK) e do transdutor de sinal e ativador de transcrição (STAT) desempenham papéis importantes na patogênese de neoplasias hematopoiéticas. A ativação da via JAK2/STAT3 promove o crescimento e sobrevivência celular em uma variedade de linfomas humanos. Há uma necessidade de compreender a participação da via JAK2/STAT3 em linfomas caninos difusos de grandes células B e do potencial terapêutico dos inibidores de JAK no tratamento dessa doença. O objetivo do presente estudo foi avaliar a expressão de JAK2-STAT3 em linfomas difusos de grandes células B e o impacto do uso de inibidores de JAK2 como AZD1480 e CYT387 no crescimento in vitro dessa linhagem tumoral. Foi realizada técnica de imuno-histoquímica com os anticorpos anti-STAT3 e anti-STAT3 fosforilado (p-STAT3) em linfonodos acometidos por linfoma difuso de grandes células B e comparado à linfonodos normais e reativos. Para avaliação do efeito terapêutico dos inibidores de JAK2 (AZD1480 e CYT387) foi realizado ensaio de viabilidade celular pelo método de azul de tripan utilizando linhagens celulares de linfoma difuso de grandes células B (CLBL-1) e análise de apoptose por citometria de fluxo utilizando o sistema Annexin V. Houve aumento significativo na expressão de STAT3 e p-STAT3 em linfomas difusos de grandes células B em comparação com linfonodos normais. Ambos os fármacos inibiram o crescimento celular em proporções dependentes da dose administrada e houve um aumento significativo nas taxas de apoptose das células tratadas com inibidores de JAK-2 em comparação ao grupo controle tratado com DMSO. Este é o primeiro estudo a avaliar a via JAK2/STAT3 em linfomas difusos de grandes céluslas B canino e esses dados permitem compreender e explorar o potencial terapêutico dos inibidores de JAK permitindo estudos futuros da eficácia clínica desses fármacos na oncologia veterinária / The Janus Kinase (JAK) and signal transducer and activator of transcription (STAT) pathway play important roles in the pathogenesis of hematologic malignancies. Activated JAK2-STAT3 signaling pathway promotes the growth and survival of a variety of lymphomas in human. There is a great demand for understanding JAK-STAT pathway in canine diffuse large B cell lymphoma (DLBCLs) and evaluating the therapeutic potential of JAK inhibitors. Our study aims to evaluate the expression of JAK2-STAT3 pathway in canine DLBCLs and to assess the impact of AZD1480 and CYT387, two novel JAK inhibitors, on canine DLBCL cell growth. Immunohistochemistry was performed in canine DLBCLs, normal and reactive lymph nodes with primary antibodies against STAT3 and phosphorylated STAT3 (p-STAT3). To evaluate the therapeutic effect of novel JAK inhibitors, canine DLBCL cell line CLBL-1 was treated with either AZD1480 or CYT387 and trypan blue viability assay was performed post treatment. There was a significant increase in expression of STAT3 and pSTAT3 in canine DLBCLs compared with the normal lymph node. Both AZD1480 and CYT387 inhibited canine DLBCL cells in a dose dependent manner. This is the first study to evaluate the JAK2/STAT3 pathway in canine DLBCLs. The knowledge of JAK2-STAT3 activity in canine DLBCLs enables us to understand and explore the therapeutic potential of JAK inhibitors. The dose dependent cell growth inhibition by novel JAK inhibitors in this study will lead into the future studies of the underlying mechanism

Submitted by Fabio Sobreira Campos da Costa (fabio.sobreira@ufpe.br) on 2017-07-12T15:39:14Z No. of bitstreams: 2 license_rdf: 811 bytes, checksum: e39d27027a6cc9cb039ad269a5db8e34 (MD5) Dissertação Juan Luiz Coelho da Silva.pdf: 2693101 bytes, checksum: b946d507d9f21698d6349e8ecf91e259 (MD5) / Made available in DSpace on 2017-07-12T15:39:14Z (GMT). No. of bitstreams: 2 license_rdf: 811 bytes, checksum: e39d27027a6cc9cb039ad269a5db8e34 (MD5) Dissertação Juan Luiz Coelho da Silva.pdf: 2693101 bytes, checksum: b946d507d9f21698d6349e8ecf91e259 (MD5) Previous issue date: 2016-03-11 / FACEPE / Algumas evidências destacam mutações no gene ASXL1 como um evento importante na evolução clínica de pacientes com neoplasias hematológicas, particularmente em leucemias mieloides agudas e síndrome mielodisplásicas. Contudo, seu impacto prognóstico em neoplasias mieloproliferativas (NMP) ainda é pouco explorado. Aqui, nós caracterizamos 208 pacientes com NMP cromossomo Filadélfia (Ph) negativo (policitemia vera, PV; trombocitemia essencial, TE; mielofibrose primária, MFP), de acordo com mutações no gene ASXL1, e correlacionamos esses achados com características clinico-laboratoriais desses pacientes. A pesquisa das mutações foi realizada por sequenciamento sanger, em que polimorfismos germinativos e mutações sinonímias foram excluídas das análises. Mutações no ASXL1 foram detectadas em 22/208 pacientes (10%), das quais quatro foram observadas em pacientes com PV (4/54; 7%), onze em pacientes com TE (11/123; 9%) e sete com MFP (7/31; 22%). As características clínicas e laboratoriais foram similares entre pacientes com ASXL1 mutado e não mutado. Quando as entidades foram avaliadas individualmente (PV, TE e MFP), observou-se associação entre mutações no ASXL1 e idade mais avançada em pacientes com TE (P = 0,049) e desenvolvimento de esplenomegalia em pacientes com MFP (P = 0,026). Com uma mediana de seguimento de 5,1 anos (IC95%: 4,5 a 7,3 anos), 136 pacientes (65%) desenvolveram algum tipo de manifestação clínica, sendo o desenvolvimento de complicações vasculares o mais frequente (n=54; 26%), seguido por esplenomegalia (n=47; 22%), eventos hemorrágicos (n=30; 14%) e trombose (n=21; 10%). Mutações no gene ASXL1 não foram associadas com o desenvolvimento das referidas manifestações. Dentro deste seguimento, apenas dois pacientes evoluíram para síndrome mielodisplásica e um para leucemia mieloide aguda, todos sem mutações no gene ASXL1. / Accumulating evidences report mutation in ASXL1 as an important predictor to clinical outcomes of patients with hematological malignancies, particularly acute myeloid leukemia and myelodysplastic syndrome. However, the prognostic impact in myeloproliferative neoplasm (MPN) remains underexplored. Here, we evaluated clinical and laboratory features of 208 Philadelphia negative MPN patients (polycythemia vera, PV; essential thrombocythemia, ET; primary myelofibrosis, PMF), according to mutations in ASXL1. Screening for ASXL1 mutations were performedby Sanger sequencing. Germline variations were excluded. ASXL1 mutations were detected in 22/208 patients (10%), of which four in PV patients (4/54-7%), 11 in ET patients (11/123-9%) and seven in PMF (7/31-22%). Baseline features were similar between ASXL1-mutated and non-mutated patients. Evaluated individually (PV, ET, PMF), we observed that ET patients harboring ASXL1 mutations were older (P = 0,049) than ASXL1 non-mutated patients. Similarly, PMF patients presented higher frequency of splenomegaly in ASXL1mutated group (P = 0,026). No other features were associated with ASXL1mutations. The median follow-up was 5,1 years (CI95%: 4,5-7,3 years). One hundred and thirty six patients (65%) developed some of the clinical common manifestations, which the most frequent was vascular complications (n=54; 26%), followed by splenomegaly (n=47; 22%), bleeding (n=30;14%) and thrombosis (n=21;10%). ASXL1 mutations were not associated with development of such events. In our cohort, only two patients have evolved for myelodysplastic syndrome and one for acute myeloid leukemia, all of them without mutations in ASXL1.

Ligand-activated signal transduction is a process critical to cell survival and function as it serves as a means of communication between the cells and their environment. Endocytosis is generally thought to down-regulate incoming signals by reducing the surface availability of receptors. However, increasing evidence in many systems suggests a notion which is referred to as the „signalling endosome" hypothesis - that endocytosis can also actively contribute to signalling apart from clearance of activated receptors and thereby attenuation of signalling. The functional aspect of signalling endosomes has been well-characterized in several pathways including RTK and TGF-β signalling. There are, however, various other signalling pathways where the active mechanism of endocytotic regulation is yet to be understood. In this study, we probe this aspect in the cytokine signalling system, where the receptors are known to internalize but the significance of such internalization and precise mechanism is unclear. My thesis aims to elucidate the function and molecular details of internalization of cytokine receptor using interleukin-4 receptor (IL-4R) signalling as a model. IL-4 and IL-13 ligands can induce assembly of three distinct complexes: IL4 induced IL-4Rα – IL-2Rγ (type I), IL-4 induced IL-4Rα – IL-13Rα1 (type II) or the IL-13 induced IL-13Rα1-IL-4Rα (type II). The formation of any of these complexes triggers signalling through the JAK/STAT pathway. However, models of how the oligomerization of the transmembrane receptors and activation takes place are very diverse and lack a clear molecular and biophysical understanding of the underlying receptor dynamics. Previous results of the lab had shown that the affinities between subunits are low, precluding complex formation at the plasma membrane at physiological concentrations. In addition, IL-4R subunits localize in to endosomal structures adjacent to the plasma membrane. It had already been shown that the shared IL-4R subunit IL-2Rγ is internalized by a specific, actin dependent, Rac1/Pak1 regulated endocytosis route in the IL-2 context. We could show that pharmacological suppression of this endocytosis pathway also prevented IL-4 induced JAK/STAT signalling, placing endocytosis upstream of signalling. Here I show using immuno-EM techniques that these endosomal structures are multivesicular bodies. Importantly, I could show that receptor subunits are highly enriched in the limiting membrane of these endosomes relative to the adjacent plasma membrane. Using quantitative loading assays I could furthermore demonstrate that this enrichment is achieved by constitutive internalization of receptors from the cell surface into cortical endosomes. The trafficking kinetics of the receptor subunits is independent of ligand occupancy. Pharmacological inhibition shows that receptors and ligand traffic via the previously identified Rac1/Pak1 pathway. Finally, Vav2 was identified as a candidate Guanine Exchange Factor (GEF) that may regulate Rac1 activity and thereby control the actin polymerization cascade driving IL-4R endocytosis. Immunoprecipitations showed that Vav2 interacts both with the cytoplasmic tail region of the receptors and the receptor associated 2 kinase JAK3. Vav2 may thus couple the receptor/JAK complexes to the Rac1/Pak1 mediated endocytosis route. Taken together, our results suggests that stable „signalling endosomes‟ adjacent to the plasma membrane act as enrichment centres, where ligand and receptor concentrations are locally increased by constitutive trafficking. The confined environment of the endosome then compensates for the weak affinities between the ligand and receptor and facilitates ligand-mediated receptor dimerization. Importantly, overexpression of both type II IL-4R subunits renders signal transduction resistant to endocytosis inhibition, strongly suggesting that the critical factor effecting signalling is sufficient concentration, which the endosomes facilitate achieving. The endosomes are thus dispensable as signalling scaffolds when the receptors are in sufficient concentration, where activated receptors could interact with downstream pathway components. Endocytosis thus provides a crucial means for the signalling process to overcome the thermodynamic hurdles for receptor oligomerization. In conclusion, our data propose a novel, purely thermodynamic role of endosomes in regulating cytokine receptor signalling not seen in any other signalling pathway.

The JAK-STAT pathway is an intracellular signaling pathway that is found to have crucial roles in hematopoiesis, immune response and the development of many other tissues in mammals. The pathway is conserved in Drosophila melanogaster, and is much simpler: there is only one Drosophila JAK (Hopscotch, Hop) and STAT (STAT92E) respectively, while there are at least 4 JAKs and 7 STATs in mammals. The pathway has been intensively studied in Drosophila, and has been implicated in many tissue development and cellular processes. In this work, I present several roles of JAK signaling in oogenesis.First, JAK signaling is required for cell differentiation within a specific lineage of follicle cells – stalk cells and polar cells. Unpaired (upd), which encodes the known ligand for the pathway, is expressed specifically in the polar cells in the developing egg. Reduced function of Upd or Hop results in fusions of egg chambers, which is primarily caused by improper formation of stalk cells, while general activation of the pathway in the egg chamber produces an extra number of stalk cells and sometimes eliminates polarfollicle cells. Based on the known function of the Notch pathway in oogenesis, we propose a model that Notch signaling determines a pool of precursors for the polar and stalk cells while JAK activity determines their specific fates within that pool.Second, JAK signaling is also involved in epithelial follicle cell differentiation. Consistent with the expression pattern of upd in the ovary, there is a gradient of JAK activity expanding from the poles, and this JAK activation gradient is both required and sufficient to suppress the main body follicle cell fate. Also, different levels of JAK activity are required and sufficient to determine both anterior and posterior terminal follicle cell fates. Consistent with these data is a model that a gradient of JAK activity triggered by Upd from the poles pre-patterns the epithelium into three domains and pre-determines sub-populations of terminal follicle cell fates prior to the EGFR activation, and cooperates with EGFR activity later to define posterior terminal follicle cell fates. This provides the first evidence for a morphogenic function of the JAK-STAT pathway in any organism.

The JAK/STAT pathway is a non-receptor tyrosine kinase signaling pathway that is well conserved and highly re-utilized in many mammalian and Drosophila developmental processes. Compared to dozens of ligands and receptors in mammalian JAK/STAT, Drosophila JAK/STAT pathway is simpler with one receptor and three ligands, Upd, Upd2 and Upd3, which have similar amino acid sequences. Previous literature shows that upd and upd2 exhibit the same dynamic striped expression pattern in embryos and have semi-redundant functions during embryogenesis. Do Upd and Upd3 also have redundant functions? To answer this question, the functions of Upd3 in Drosophila development were investigated in this dissertation. In addition, the coordinate expression mechanism of upd and upd3 in eye discs was also analyzed. To study the functions of Upd3 in development, the expression pattern of upd3 was examined and detected in larval eye discs, wing discs, haltere discs, lymph glands and adult ovaries with in situ hybridization to upd3 mRNA and an upd3 reporter line. Consistent with the expression pattern, the loss of function mutants of upd3 exhibit small eyes, outstretched wings, downward extended halteres and reduced circulating blood cell concentration, demonstrating the roles of Upd3 in these tissues’ development. However, functions of Upd3 in other aspects of immune response were not detected. To investigate the mechanism of the coordinate expression of upd and upd3, the genetic and molecular relationship of upd, upd3 and os was dissected. The os alleles, oso, oss and os1, are a group of classical alleles which display outstretched wings, small eyes, or both, respectively. The genetic complementation tests of upd, upd3 and os showed that both upd and upd3 failed to complement os while upd complemented upd3, suggesting functions of both upd and upd3 are affected in os alleles. Consistent with the genetic tests, the expression of upd and upd3 in eye discs is lost in os allele. Molecularly, putative enhancer regions are deleted at the 5’ end of upd3 in os alleles. Hence, a transcriptional co-regulation model of upd and upd3 is proposed in which upd and upd3 share a common cis-regulatory region, lesions of which cause the os phenotype.