Les lymphocytes T CD4+ auxiliaires ou T helper en anglais sont capables de soutenir une grande diversité de fonctions grâce à leur capacité à se différencier en différents sous-types effecteurs en fonction de l’antigène rencontré et de l’environnement cytokinique dans lequel ils se trouvent. Les connaissances actuelles sur la différenciation des cellules T helper mettent en avant l’existence de réseaux transcriptionnels particulièrement complexes et spécifiques à chaque sous-ensemble T helper. En 2008, les cellules T CD4 sécrétrices d’IL-9 (Th9) sont identifiées comme un nouveau sous-type de cellules T helper. Différenciées en présence d’IL-4 et TGF-β, les cellules Th9 sécrètent de l’IL-9 et de l’IL-21, et contribuent au développement de maladies auto-immunes et allergiques. Les lymphocytes Th9 présentent également des propriétés anti-tumorales particulièrement intéressantes.Le réseau transcriptionnel des cellules Th9 résulte d’un équilibre entre les voies de signalisation induites par les différentes cytokines nécessaires à sa polarisation. L’IL-4 permet l’activation de STAT6 et l’expression de GATA3 et IRF4, tandis que le TGF-β conduit à l’activation de la voie des Smad et l’expression du facteur PU.1. Le module transcriptionnel IRF4/BATF ainsi que le facteur PU.1 sont des messagers indispensables au développement des cellules Th9 et à la sécrétion d’IL-9.IRF8 est un facteur de transcription critique pour le développement des cellules myéloïdes et des lymphocytes B. Récemment, il est apparu qu’IRF8 était impliqué dans la polarisation de sous-ensembles T helper. En effet, IRF8 limite la sécrétion d’IL-17 par les cellules Th17, de même qu’il réprime l’expression de l’Il4 et l’Il17 dans les cellules Treg. Structurellement proche d’IRF4, IRF8 interagit avec des cofacteurs tels que PU.1 ou BATF afin de réguler l’activité transcriptionnelle.Ce travail présenté ici révèle que le facteur IRF8 participe à la polarisation des cellules Th9 in vitro et in vivo. Le TGF-β nécessaire à la différenciation des cellules Th9 régule directement l’expression d’Irf8 grâce à l’activation de Smad3. Comme dans d’autres types cellulaires, la fonction transcriptionnelle d’IRF8 est dépendante de ces partenaires d’interaction. Nous montrons qu’en présence des facteurs PU.1, IRF4 et BATF, IRF8 participe à un complexe multiprotéique nécessaire à l’induction des cytokines caractéristiques des cellules Th9, notamment l’Il9 et l’Il21. Nous démontrons également qu’en présence de la protéine ETV6, IRF8 est capable de former un complexe initiant la répression de l’activité transcriptionnelle de l’Il4. Nous soulignons ainsi le rôle bivalent joué par IRF8 dans le développement des cellules Th9 dépendamment de ses partenaires. Pour finir, l’expression d’Irf8 est nécessaire aux cellules Th9 pour exercer leurs fonctions anti-tumorales. / CD4 helper T cells support a wide range of functions due to their ability to differentiate into different effector subsets depending on the antigen encountered and the cytokine environment in which they are. Current knowledge on the differentiation of helper T cells highlights the existence of complex transcriptional networks specific to each T helper subset. In 2008, IL-9 secreting CD4 T cells (Th9) are identified as a new helper T cell subtype. Differentiated in the presence of IL-4 and TGF-β, Th9 cells secrete IL-9 and IL-21, and contribute to the development of autoimmune and allergic diseases. Th9 lymphocytes also exhibit strong anti-tumor properties.The transcriptional network of the Th9 cells results from a balance between the signaling pathways induced by the different cytokines required for its polarization. IL-4 allows activation of STAT6 and expression of GATA3 and IRF4, whereas TGF-β leads to activation of the Smad pathway and expression of the transcription factor PU.1. The IRF4 / BATF transcriptional module and the PU.1 factor are essential messengers for the development of Th9 cells and IL-9 secretion.IRF8 is a crucial transcription factor for the development of myeloid cells and B lymphocytes. Recently it appeared that IRF8 was involved in helper T subset polarization. Indeed, IRF8 limits the secretion of IL-17 by Th17 cells, as well as repressing the expression of Il4 and Il17 in Treg cells. Structurally close to IRF4, IRF8 interacts with cofactors such as PU.1 or BATF in order to regulate transcriptional activity.This work reveals that the IRF8 transcription factor contributes to the polarization of Th9 cells in vitro and in vivo. The TGF-β needed for Th9 cell differentiation activate Smad3 pathway which directly modulates the Irf8 expression. As in many cellular subtypes, the transcriptional function of IRF8 is dependent on these interaction partners. We show that in the presence of the transcription factors PU.1, IRF4 and BATF, IRF8 participates in a multiprotein complex essential for the induction of the Th9 cytokines, Il9 and Il21. We also demonstrate that in the presence of the ETV6 protein, IRF8 is able to form a complex responsible for the repression of Il4 expression. We underline the bivalent role played by IRF8 in the development of Th9 cells depending on its partners. Finally, expression of Irf8 is crucial for Th9 cells to exercise their antitumor functions.

<p>Epigenetic modifications were shown to play an essential role in tumorigenesis. Epigenetic mechanisms can alter transcription in several ways, through DNA methylation and/or through histone modification. DNA methylation at the TSS (transcriptional start site) has been implicated in tumor development and gene silencing. However, several examples of atypical methylation were shown. In Paper I we present the ICSBP/IRF8 gene that belongs to the IRF family and has characteristics of a tumor suppressor gene. The ICSBP/IRF8 is fully methylated in the promoter and TSS regions in U-937 and despite high expression of the gene. Presence of positive histone marks suggests that methylated DNA can be overridden by histone modification.</p><p>In Paper II a panel of 13 MM (multiple myeloma) cell lines and 9 primary patient tumors were analysed for methylation status of the ICSBP/IRF8 gene. In most cell lines (8/13) the gene was partially or fully methylated and partial methylation was also observed in 1/9 primary tumors. In vitro methylation analysis and treatment with 5-aza-2’deoxycytidine (DAC) proved that the ICSBP/IRF8 gene is silenced by methylation and may be associated with the malignant phenotype.</p><p>In Paper III and IV the NFκB signalling pathway was analysed and the role of ATRA and TNFα induction. In Paper III the data shows that activation of the NFκB pathway is essential in ATRA-induced terminal differentiation in the U-937 cell line and IκBα (S32A/S36A) inhibits ATRA-induced differentiation and G1 cell cycle arrest. This was accompanied by delayed down-regulation of several cyclins (A and E) and up-regulation of p21^WAF1/CIP1 (CDKN1A) and p27^KIP1 (CDKN1B).</p><p>TNFα alone did not induce expression of RA-induced genes analysed in Paper IV. However, ATRA in combination with TNFα showed enhanced activation of RA-induced genes. TNFα triggers demethylation of H3K9me3/H3K9me2 and H3K4me3 at RAR/RXR target genes, which were not accompanied by changes in the level of H3K9-ac. This decrease in H3 methylation by TNFα may pave way for the later ATRA-induced gene transcription.</p>

Epigenetic modifications were shown to play an essential role in tumorigenesis. Epigenetic mechanisms can alter transcription in several ways, through DNA methylation and/or through histone modification. DNA methylation at the TSS (transcriptional start site) has been implicated in tumor development and gene silencing. However, several examples of atypical methylation were shown. In Paper I we present the ICSBP/IRF8 gene that belongs to the IRF family and has characteristics of a tumor suppressor gene. The ICSBP/IRF8 is fully methylated in the promoter and TSS regions in U-937 and despite high expression of the gene. Presence of positive histone marks suggests that methylated DNA can be overridden by histone modification. In Paper II a panel of 13 MM (multiple myeloma) cell lines and 9 primary patient tumors were analysed for methylation status of the ICSBP/IRF8 gene. In most cell lines (8/13) the gene was partially or fully methylated and partial methylation was also observed in 1/9 primary tumors. In vitro methylation analysis and treatment with 5-aza-2’deoxycytidine (DAC) proved that the ICSBP/IRF8 gene is silenced by methylation and may be associated with the malignant phenotype. In Paper III and IV the NFκB signalling pathway was analysed and the role of ATRA and TNFα induction. In Paper III the data shows that activation of the NFκB pathway is essential in ATRA-induced terminal differentiation in the U-937 cell line and IκBα (S32A/S36A) inhibits ATRA-induced differentiation and G1 cell cycle arrest. This was accompanied by delayed down-regulation of several cyclins (A and E) and up-regulation of p21WAF1/CIP1 (CDKN1A) and p27KIP1 (CDKN1B). TNFα alone did not induce expression of RA-induced genes analysed in Paper IV. However, ATRA in combination with TNFα showed enhanced activation of RA-induced genes. TNFα triggers demethylation of H3K9me3/H3K9me2 and H3K4me3 at RAR/RXR target genes, which were not accompanied by changes in the level of H3K9-ac. This decrease in H3 methylation by TNFα may pave way for the later ATRA-induced gene transcription.

Microglia are the resident macrophage in the central nervous system (CNS) that actively survey their environment and participate in shaping neuronal circuits. Among the transcription factors necessary for microglia development, interferon regulatory factor 8 (IRF8) is a known risk gene for multiple sclerosis and lupus and it has recently been shown to be downregulated in schizophrenia. These studies suggest that lack of microglial IRF8 can subsequently impact neuronal function in disease, but the mechanisms underlying these effects remain unknown. While most studies have focused on IRF8-dependent regulation of immune cell function, little is known about how it impacts neural circuits. To interrogate the impact of disrupted microglial IRF8 signaling on brain circuits, I first show by RNAseq that several genes known to regulate neuronal function are dysregulated basally in Irf8-/- brains. I then found that these molecular changes are reflected in heightened neural excitability and a profound increase in susceptibility to chemically-induced lethal seizures in Irf8-/- mice. Importantly, I also show that developmental synaptic pruning, a key function for microglia, proceeds normally in Irf8-/-mice. Finally, I identified that these IRF8-dependent effects on circuits are due to elevated TNF-α in the CNS as genetic or acute pharmacological blockade of TNF-α in the Irf8-/- CNS rescued the seizure phenotype. These results provide important insights into the consequences of IRF8 signaling and TNF-α on neural circuits. The next steps are to use cell-specific genetic approaches to manipulate this signaling, which I have further developed over the course of this project.

Systemic lupus erythematosus (SLE) is an autoimmune, inflammatory disease characterized by autoantibody production and an activated type I interferon system. Cardiovascular disease (CVD) is as a major cause of morbidity and mortality. The aim of this thesis was to identify genetic risk factors for CVD in SLE. The role of T cells in regulation of the interferon-α (IFNα) production by plasmacytoid dendritic cells (pDCs) was also investigated.    In paper I, a thicker intima, thinner media and increased intima/media ratio was found in young premenopausal women with SLE compared to healthy controls indicating increased cardiovascular risk. As traditional ultrasound assessment of the common carotid intima-media thickness (CCA-IMT) in SLE has given conflicting results separate measurement of the intima and media can be a useful tool to identify SLE patients at increased risk of CVD.    In paper II, an association was demonstrated in SLE between a STAT4 risk allele and ischemic cerebrovascular disease and presence of anti-phospholipid antibodies (aPL). The association remained after adjustment for traditional CVD risk factors. A possible mechanism for this association is that the risk allele leads to increased production of aPL, which promotes thromboembolism.    In paper III, a genetic locus in IRF8 was identified to be associated to coronary heart disease (CHD) in SLE. The association remained after adjustment of other CHD risk factors.  Patients with the IRF8 risk variant had increased CCA-IMT, more carotid plaques and reduced frequency of circulating B cells. Weaker binding of nuclear protein to the risk allele was demonstrated, suggesting a regulatory function of the IRF8 risk variant.    In paper IV, activated T cells were found to strongly enhance the IFNα production by pDC stimulated with RNA-containing immune complexes via GM-CSF and IL-3. Activated SLE T cells enhanced the IFNα production to the same extent as T cells from healthy controls. This finding together with previous observations in SLE of increased levels of GM-CSF and IL-3 suggests that T cells contribute to the activated type I interferon system in SLE.    In conclusion, this thesis demonstrates that genetic predisposition is important for CVD in SLE and describes a new role for T cells in the pathogenesis of SLE.

The dichotomy between type 1 and 2 conventional DCs under steady-state conditions is well defined. Bosteels et al. demonstrate that, upon inflammation, cDC2s acquire a hybrid inf-cDC2 phenotype, sharing phenotype, gene expression, and function with cDC1s and monocyte-derived cells, to optimally boost CD4 and CD8 immunity via Fc receptors.