archives@tulane.edu / A-to-I RNA editing, catalyzed by a family of enzymes called adenosine deaminases acting on RNA (ADARs), brings broad significance in various biological processes. To date, the roles of ADARs and its associated RNA editing in Epstein-Barr virus (EBV)’s life cycle and pathogenesis are still largely unknown. To fill this significant knowledge gap, we utilized our well-established next-generation RNA sequencing-based computational approaches and traditional molecular biology methodologies to elucidate the triangle relationship between ADARs, RNA-editing, and EBV infection. The expression of ADARs was first evaluated in a cohort of EBV-associated lymphoma cells. A constitutive expression of ADAR1, the predominant form of ADARs, was observed in the examined cells. In synchronous EBV reactivation cell models, we found that EBV reactivation led to a decreased expression of ADAR1 as well as a global suppression of A-to-I RNA editing. Further, we found that expression of the key viral trans-activator Zta inhibited ADAR1 expression in EBV-associated lymphoma cells. Analyses of the ADARs-mediated RNA editing events revealed novel editing sites on viral lytic transcripts. Knockdown of ADAR1 led to a global suppression of RNA-editing accompanied by a more robust EBV reactivation. Meanwhile, the enhanced expression of ADAR1 inhibited Zta’s expression and transactivation function. Together, our findings reveal a novel mechanism controlling the balance of EBV life cycle, in which ADAR1 and associated RNA editing events help maintain the viral latency by silencing Zta; whereas a bona fide lytic signal leads to high-level Zta expression by inhibiting ADAR1 and ADARs-mediated RNA editing. / 1 / Yi Yu

Les cellules de le crête neurale sont une population de cellules précurseurs multipotentes qui émergent aux frontières du tube neural et de l’ectoderme non neural, migrent de manière extensive dans tout l'embryon et se différencient en une variété de types de cellules, notamment les cellules pigmentaires de la peau (mélanocytes), la glie du système nerveux périphérique (y compris les cellules de Schwann formant la myéline) et les neurones et les cellules gliales du système nerveux entérique. Chacune des étapes du développement de ces cellules est sous le contrôle de stimuli externes et de facteurs de transcription étroitement régulés formant un réseau complexe. Le rôle des modifications épigénétiques et post-transcriptionnelles a également été mis en évidence, mais leurs contributions aux troubles associés sont encore mal décrites. L'objectif de ma thèse était d'étudier le rôle de l'une des modifications post-transcriptionnelles les plus rependues : le RNA editing, dans le développement normal et pathologique de la crête neurale. Nous rapportons ici que ADAR1, enzyme responsable de la modification adénosine-inosine de l'ARN, est nécessaire au développement de trois dérivés de la crête neurale : les mélanocytes, les cellules de Schwann et le système nerveux entérique. L'invalidation conditionnelle spécifique d'Adar1 dans la crête neurale chez la souris provoque une dépigmentation généralisée et l'absence de myéline des nerfs périphériques résultant d'altérations affectant la survie des mélanocytes et de la différenciation des cellules de Schwann, respectivement. Des défauts de la glie entérique ont également été mis en évidence. Ces défauts sont tous trois précédés par l’activation d’une réponse immunitaire innée médiée par l’IFN. L’invalidation concomitante de MDA5, un senseur clé de détection d'ARNs non édités, corrige les défauts de myélinisation et de pigmentation observés chez les mutants adar1, suggérant qu'ADAR1, via son activité d’editing, protège ces dérivés de la crête neurale d'une production aberrante d’IFN délétère à leur survie ou différenciation. L’ensemble de ces résultats étendent le spectre d’action d’Adar1 au développement de la crête neurale normal et pathologique. / Neural crest cells are a population of multipotent precursor cells that emerge at the borders of the neural tube, migrate extensively throughout the embryo, and differentiate into a variety of cell types including the skin pigment cells (melanocytes), glia of the peripheral nervous system (including Schwann cells that form myelin) and the neurons and glia of the enteric nervous system. Each steps of the development of these cells is under the control of external stimuli and tightly regulated transcription factors that form a complex network. The role of epigenetic and post-transcriptional modifications was also highlighted, but their contributions to related disorders are still poorly described. The aim of my PhD project was to investigate the role of one of the most common post-transcriptional modification: the Adenosine to Inosine (A to I) RNA editing, in normal and pathologic NC development. Here, we report that adenosine deaminase acting on RNA (ADAR1), responsible for A to I editing of RNA, is required for regulating the development of three neural-crest derivatives: melanocytes, Schwann cells and enteric nervous system. Neural-crest specific conditional invalidation of Adar1 in mice led to global depigmentation and absence of myelin from peripheral nerves, resulting from alterations in melanocyte survival and late differentiation of Schwann cells respectively. Defects of enteric glia is also evidenced. These defects were preceded by upregulation of an innate immune inflammatory response. Simultaneous extinction of MDA5, a key sensor for the detection of unedited RNA, rescued the pigmentation and myelin defects of Adar1 mutants, suggesting that ADAR1 safeguards a subset of neural-crest derivatives from aberrant MDA5-mediated interferon production. We thus extend the landscape of ADAR1 function to the fields of neural-crest development and disease.

Disabled-2 (Dab2) is a mitogen-responsive adaptor protein playing a key role in multifaceted cellular functions, such as endocytosis, epithelial-mesenchymal transition (EMT), immune function, stem cell differentiation, oncogenesis, cell signaling, and inflammatory responses. The adenosine deaminase RNA-specific binding protein (ADAR1) is a multifunctional RNA-editing enzyme that can convert adenosine to inosine, which can modulate gene expression and cellular functions in multiple pathways, such as mRNA translation by changing codons and the subsequent protein sequence, pre-mRNA splicing by changing splice site sequences, RNA stability by altering sequence for nuclease recognition, and RNA structure-dependent functions by altering RNA-protein interactions. ADAR1 has displayed a largely pro-tumorigenesis role, especially its immunosuppressive function in cancer cells, which attributes ADAR1 as a potential novel immune checkpoint for cancer treatment. In our lab, we employed an F9 mouse teratocarcinoma stem cell differentiating model and confirmed that Dab2 is an indispensable element for retinoic acid (RA)-induced F9 cell differentiation. Interestingly, our new findings indicated that during the process of RA-induced F9 cell differentiation, both the protein levels of Dab2 and ADAR1 are significantly upregulated, and siRNA-mediated Dab2 silence results in the silence of ADAR1. In addition, results from EMT models and statistical analysis from the human TCGA database further indicated that there is a positive correlation between the expression of Dab2 and ADAR1. Our results imply that Dab2 and ADAR1 may cooperate with each other to modulate cellular functions, which will present a novel mechanism for the mechanistic study of Dab2 in tumorigenesis.

Le virus de la maladie de Marek (GaHV-2), est un "-herpèsvirus induisant des lymphomes T chez le poulet. L’étude des pri-miARN des régions RL à partir des miR-M4, -M11, M31 et -M1 a montré des initiations de transcription dispersées et internes aux longs pri-miARN initialement décrits. Les tests de fonctionnalité des régions en amont ont permis de conclure quant à l’absence d’activité promotrice suggérant une régulation par le prmiR-M9M4, caractérisé dans cette étude, ou par le prMeq. Le gène de 7,5 kpb d’un ARNlnc (ERL, Edited Repeat Long) épissé et antisens à ces transcrits a été caractérisé. Une hyper-édition A-en-G des transcrits ERL par ADAR1 a été mis en évidence. Principalement lié au cycle lytique, l’édition indique une répression fonctionnelle durant cette phase. Les régulations de l’expression d’ADAR1 chez l’humain, positives par la voie de réponse aux interférons et négatives par SOCS1, ont été validées chez le poulet. L’inhibition de SOCS1 par le gga-miR-155 et son orthologue viral mdv1-miR-M4-5p a été montré fonctionnelle chez le poulet comme chez l’humain et mène à une surexpression d’ADAR1 lors de l’activation des voies de réponse aux interférons. / Marek’s disease virus (GaHV-2) is an "-herpesvirus that induces T-cell lymphoma in chickens. In this study, we have shown that some pri-miRNAs, which are specific of miR-M4, -M11, M31 et -M1, initiated at dispersed transcription start sites. They are located in internal positions from previously described pri-miRNAs but upstream sequences lack promoter activity, indicated a potential regulation by the upstream prmiR-M9M4, characterised during this study, or the prMeq. The 7.5 kbp gene of the ERL (Edited Repeat Long) lncRNA, which is alternatively spliced et antisense of these pri-miRNAs was defined. An extensive A-to-G hyperediting of it sequence was observed strongly linked to the lytic phase, indicated a functional repression during this phase. We showed that, like the human one, the chicken ADAR1 expression was positively controlled by the IFN response pathway et negatively by the suppressor of cytokine signaling 1 (SOCS1). Like the human et murine miR-155-5p, the chicken gga-miR-155-5p et the GaHV-2 analog mdv1-miR-M4-5p deregulate this pathway by targeting et repressing expression of SOCS1, leading to the upregulation of ADAR1.

The adenosine deaminase acting on RNA (ADAR) enzymes catalyse the hydrolytic deamination of adenosine (A) to inosine (I) in double stranded (ds) RNA. Mutations in ADAR1 underlie the autoimmune disorder Aicardi Goutiѐres syndrome (AGS). Patients with AGS display heightened levels of type I interferon (IFN) and IFN stimulated genes (ISGs). The first aim of my thesis was to determine whether the mutations found in the human ADAR1 gene affected RNA editing. I found that the ADAR1 mutants identified in the AGS patients have reduced editing activity. Interestingly, the mutations have a greater effect on the IFN-inducible cytoplasmic isoform, ADAR1p150 than on the constitutive ADAR1p110 isoform. These results imply that A-to-I editing plays a role in regulating the type I IFN response. The Adar1 null mouse dies by E12.5 with a type I IFN signature similar to that observed in the AGS patients. The second aim of my thesis was to characterize the immune signalling pathway aberrantly activated in the absence of Adar1. A colleague in our research group rescued the Adar1 null mouse to birth by blocking the cellular response to cytoplasmic dsRNA by generating a double mutant with the mitochondrial antiviral signalling adaptor, Mavs. In the Adar1-/-; Mavs-/- mutant I found that the aberrant immune response is rescued at E11.5. This indicates that MAVS is the downstream adaptor in the aberrant immune response that underlies the embryonic lethality in the Adar1-/- mouse. The third aim of my thesis was to determine if the lack of inosine modification within cellular RNA was triggering the aberrant immune response in the Adar1-/- mouse. To study this, Adar1-/-; p53 -/- mouse embryonic fibroblasts (MEFs) were generated. By reintroducing various ADAR isoforms into the Adar1-/-; p53 -/- MEFs I found that to rescue the aberrant immune response requires both catalytic activity and the location of an ADAR protein within the cytoplasm. Moreover, I demonstrated that transfecting inosine-containing dsRNA oligonucleotides into Adar1-/-; p53 -/- MEFs suppresses the aberrant immune response. Overall my results suggest that A-to-I editing by ADAR1 is an essential RNA modification that is required by the cell to distinguish between ‘self’ and ‘non-self’ RNA. Editing of cellular RNAs prevents an autoimmune response whereas editing of viral RNA may act to suppress a heightened antiviral immune response and prevent long-term damage to the cell.

Afin de se répliquer et de se propager efficacement, les virus ont développé de multiples stratégies leur permettant d’échapper au système de défense innée : le système IFN de type I. Ce travail de thèse a alors consisté à étudier les interactions entre protéines virales et protéines de ce système de défense afin de mieux comprendre les mécanismes de subversion virale et d’identifier d’éventuelles cibles cellulaires thérapeutiques. La reconstruction d’un réseau d’interactions entre ces protéines nous a permis d’identifier des stratégies différentielles de subversion pour 4 familles virales et de montrer un ciblage massif et significatif des protéines du système IFN de type I par les virus. Les protéines en interaction directe avec ces protéines sont également fortement touchées par les virus et sont de potentiels modulateurs du système IFN de type I. Parmi ces modulateurs, le processus biologique sur-représenté est le transport nucléocytoplasmique et la protéine KPNA1 impliquée dans ce processus a retenu notre attention. L’étude fonctionnelle de l’interaction entre la protéine KPNA1 et la protéine NS3 du VHC a montré que la protéine NS3 associée à son cofacteur NS4A inhibe partiellement la réponse IFN de type I en empêchant l’import nucléaire de STAT1. Ce phénotype pourrait résulter de la dégradation de KPNA1 par NS3/4A. Par ailleurs, l’identification de nouveaux inter-acteurs de la protéine NS1 du virus influenza par criblage double-hybride levure a révélé la protéine induite par les IFN de type I, ADAR1, comme partenaire de la protéine NS1 de multiples souches virales et nous avons montré qu'ADAR1 est un facteur pro-viral dont la fonction editing est activée par NS1 / To replicate and propagate efficiently, viruses have developed multiple strategies allowing them to escape the innatedefense system: the type I IFN system, This work of thesis then consisted in studying the interactions between viralproteins and proteins of this defence system in order to understand better the mechanisms of viral subversion andidentifY possible therapeutic cellular tatgets. The reconstruction of a network of interacting proteins involved in the typeI IFN system allowed us to identifY differentiai subversion strategies for 4 viral families and to show a massive andsignificant targeting of proteins of the type I IFN system by viruses. Proteins directly interacting with the type Iinterferon system network are also strongly targeted by viruses and are potential modulators of the type I IFN system.Among these modulators, the most tatgeted function conesponds to the transport of NLS-bearing substrates to thenucleus and the KPNAI protein involved in this process held our attention. The functional study of the interactionbetween KPNA1 and NS3 protein of the HCV showed that NS3 protein associated with its cofactor NS4A inhibitsprutially the type I IFN response by preventing the nuclear translocation of ST A Tl. This phenotype could result fromthe degradation of KPNAI by NS3/4A. Besides, the identification of new cellular prutners ofNS 1 prote in of influenzavirus by yeast two-hybrid screens revealed ADARI, an interferon-stimulated prote in, as partner of NS 1 of ali testedvirus strains and we showed that ADARI is an essential host factor for viral replication and its editing function isactivated by NS 1 protein