Immune activation is increasingly recognized as a critical element of HIV infection and pathogenesis, causing expansion of virus founder populations at the mucosal port of entry and eventual exhaustion of cellular immune effectors. A cohort of HIV-resistant (HIV-R) commercial sex workers (CSW) in Nairobi, Kenya, have increased levels of anti- inflammatory factors in vaginal secretions and reduced peripheral immune activation ("immune quiescence"). The mucosal immune micro-environment underlying HIV susceptibility is well-known to be influenced by concurrent sexually transmitted infections, however the role of commensal microbiota is poorly characterized. Bacterial vaginosis (BV), characterized by a shift from Lactobacillus to Gardnerella and Prevotella as dominant members of vaginal microbiota, is a risk factor for HIV acquisition in studies worldwide. However, the etiology and ecological dynamics of BV remain enigmatic, and the mechanisms by which BV increases HIV susceptibility are not fully defined. Protective functional characteristics of Lactobacillus microbiota, including acid and hydrogen peroxide (H2O2) production, may reinforce physicochemical defences of vaginal mucus, stimulate innate epithelial defences and/or modulate activation status of HIV target cells. Therefore, the goal of this study was to determine if reduced BV and increased Lactobacillus colonization are the basis for resistance to HIV in this cohort. Vaginal specimens from a group of 242 CSW were examined, including microscopic diagnosis of BV, culture-based functional analyses and phylogenetic profiling by ultra-deep sequencing. HIV-R individuals were just as likely to have BV compared to other HIV- negative (HIV-N) individuals, and no more likely to be colonized with acid- or H2O2- ii producing bacteria, however two BV-related phylotypes identified by deep sequencing were significantly more likely to be observed in HIV-N individuals (p=0.0002 and p=0.006). HIV+ individuals were significantly more likely than HIV– individuals to have E. coli detected by deep sequencing (p<0.0001) and less likely to have Lactobacillus crispatus (p=0.0006). A coherent set of differences in culture-based and culture- independent characteristics were observed in individuals with BV diagnoses compared to BV– individuals. This study has generated an unprecedented amount of information regarding the composition, structure and function of the vaginal microbiota in African CSW, fundamentally defining many aspects of BV microbiology. Elucidation of the relationship between complex microbial communities and protective mucosal responses against HIV infection should be a priority for future research.

Master of Science / Entomology / Ming-Shun Chen / Robert "Jeff" J. Whitworth / MicroRNA (miRNA) plays a role in nearly all the biological pathways and therefore may provide opportunities to develop new means to combat the Hessian fly, Mayetiola destructor, a destructive pest of wheat. This study presents a comprehensive analysis of miRNA species via deep-sequencing samples from Hessian fly second instar larvae, pupae and adults. A total of 921 unique miRNA species were identified from approximately 30 million sequence reads. Among the 921 miRNA species, only 22 were conserved among Hessian fly and other insect species, and 242 miRNA species were unique to Hessian fly, the remaining 657 share certain sequence similarities with pre-miRNA genes identified from various insect species. The abundance of the 921 miRNA species based on sequence reads varies greatly among the three analyzed stages, with 20 exclusively expressed in adults, two exclusively expressed in pupae and two exclusively expressed in second instar larvae. For miRNA species expressed in all stages, 722 were with reads lower than 10. The abundance of the remaining 199 miRNA species varied from zero to more than eight-fold differences among different stages. Putative miRNA-encoding genes were analyzed for each miRNA species. A single putative gene was identified for 594 miRNA species. Two putative genes were identified for 138 miRNA species. Three or more putative genes were identified for 86 miRNA species. The three largest families had 14, 23 and 34 putative coding genes, respectively. No gene was identified for the remaining 103 miRNA species. In addition, 1516 putative target genes were identified for 490 miRNA species based on known criteria for miRNA targets. The putative target genes are involved in a wide range of processes from nutrient metabolism to encoding effector proteins. Analysis of the expression patterns of miRNA and pre-miRNA for the miRNA family PC-5p-67443, which contains 91 genes, revealed that miRNA and pre-miRNA were expressed differently in different developmental stages, suggesting that different isogenes are regulated by different mechanisms, or pre-miRNAs had other functions in addition to as an intermediate for miRNA biogenesis. The large set of miRNA species identified here provides a foundation for future research on miRNA functions in Hessian fly and for comparative studies in other species. The differential expression patterns between a pre-miRNA and its encoded mature miRNA in a multigene family is an initial step toward understanding the functional significance of isogenes in dramatically expanded miRNA families.

The unveiling of small RNA (sRNA)-mediated gene regulatory pathways has profoundly shaped our understanding of the complexity of gene regulation. In eukaryotes, sRNAs have been found to control cellular metabolism, growth and differentiation, to maintain genome integrity, and to combat viruses and mobile genetic elements. To gain insight into the roles of small RNAs in apple and peach, we conducted sRNA-seq, computational analysis and molecular experiments to genome-widely characterize their microRNAs (miRNAs) and trans-acting siRNA (tasiRNA) pathways. We identified totally 75 miRNAs or families, including 23 conserved, 10 less-conserved and 42 apple-specific ones, and 118 miRNA target genes in apple. Two classical trans-acting siRNA (tasiRNA) pathways, miR390-TAS3 and miR828-TAS4, were characterized with similar but unique tasiRNA biogenesis profiles and target specificities. Importantly, miR159, miR828 and miR858 can collectively target up to 81 MYB genes potentially involved in diverse aspects of plant growth and development. In contrast to the location of the miR159 target site in a sequence-divergent region, the target sites of miR828 and miR858 are located in the region encoding the conserved R3 repeat domain of MYB proteins. 10 out of the 19 miR828-targeted MYBs undergo the biogenesis of various phased siRNA (phasiRNA), which potentially regulate diverse genes outside the MYB family. In peach, totally 94 miRNAs or families and 80 target genes were identified. Similar pathways of tasiRNA (miR828-TAS4 and miR390-TAS3) or phasiRNA (miR828-MYB-siRNA) processing were also characterized in peach. Taking advantage of reverse computation and public available deep-sequencing data, we demonstrated that the miRNA-TAS-PPR-siRNA pathway is a highly dynamic and widespread feature of eudicots. Nine eudicot plants, representing six different plant families, have evolved similar tasiRNA pathways to instigate phasiRNA production from PPR �genes, which are triggered by different 22-nt miRNAs, including miR7122, miR1509, and fve-PPRtri1/2 and through distinct mechanistic strategies, like miRNA direct-targeting or indirect-targeting through TAS-like genes, one-hit or two-hit, or even two layers of tasiRNA-TAS interactions. We found that the MIRNA genes of these miRNA triggers show great identity with the Arabidopsis MIR173, implying a common origin of this group of miRNAs (super-miR7122). Combined results from phylogenetic analyses and conservation extent profiling revealed that the super-miR7122 was potentially evolved from another miRNA superfamily (super-miR4376), which probably originated from the miR390. Additionally, the miR482/2118-NB-LRR-siRNA pathway was found to be conserved, but evolved with distinct features, in apple and peach. Taken together, widespread and complex miRNA and tasiRNA regulatory networks have been adapted in apple and peach. They add another crucial layer of regulation on gene activity and stability, and must exert essential functions in all aspects of plant life. / Ph. D.

京都大学 / 0048 / 新制・論文博士 / 博士(医学) / 乙第12843号 / 論医博第2083号 / 新制||医||1006(附属図書館) / 31426 / (主査)教授 朝長 啓造, 教授 小柳 義夫, 教授 小川 誠司 / 学位規則第4条第2項該当 / Doctor of Medical Science / Kyoto University / DFAM

Les lymphocytes T (LT) sont des acteurs-clés du système immunitaire, un système complexe et dynamique évoluant au cours de la vie de l'organisme. On appelle " répertoire lymphocytaire ", une collection de lymphocytes partageant un même phénotype, une même fonction ou tout autres critères, chacun caractérisé par un récepteur membranaire unique, appelé TCR, lui permettant de reconnaitre de manière spécifique les antigènes. Les TCR sont caractérisés par des régions variables, produites par une série de réarrangements somatiques ayant lieu pendant la différenciation thymique, et qui assurent la diversité de reconnaissance des LT. On parle de répertoire TCR lorsque l'on s'attache à définir les caractéristiques clonales des populations lymphocytaires T sur la base de la diversité des TCR exprimés à l'échelle de la population. Le séquençage à haut débit des chaînes TCR permet désormais de décrire cette diversité avec une précision sans précédent. Cette approche requiert néanmoins des outils adaptés pour permettre une caractérisation pertinente de la structure des répertoires analysés. Un axe de recherche de L'unité I3 est l'analyse du répertoire TR de plusieurs populations lymphocytaires T en situation d'auto-immunité ou d'inflammation. Dans ce contexte, les objectifs de ma thèse ont été de : i) approfondir le concept de diversité du répertoire lymphocytaire, ii) mettre au point une méthodologie adaptée permettant d'exploiter les données de séquençage de manière optimale en prenant en compte les limites de cette technologie, et iii) développer un outil permettant aux immunologistes une caractérisation approfondie et facilement interprétable des répertoires qu'ils étudient. / T lymphocytes (LT) are key players in the immune system, a complex and dynamic system evolving over the organism’s life. The concept of "lymphocyte repertoire" designates a collection of lymphocytes sharing the same phenotype, the same function or any other criteria. Each LT is characterized by a unique membrane receptor, called TCR, allowing it to recognize specifically antigens. TCRs are characterized by variable regions produced by a series of somatic rearrangements that occur during the thymic differentiation; these regions engage LT recognition diversity. The “TCR repertoire” approach focuses the clonal characterisation of LT populations on the diversity of the TCR expressed on the scale of the population. The high-throughput sequencing of TCR chains (RepSeq) describes this diversity with unprecedented precision. However, this approach requires adapted tools to enable a relevant deciphering of the analysed TCR repertoire diversity. My thesis aimed to: i) deepen the concept of diversity of the lymphocyte repertoire, ii) develop an appropriate methodology to exploit optimally RepSeq data while taking into account the limits of this technology, and iii) develop a tool providing immunologists a thorough characterisation of their TCR repertoires of interest.

L’hépatite B chronique touche environ 257 millions de personnes dans le monde. La perte de l’antigène HBs (AgHBs), marqueur de guérison fonctionnelle, n’est que très rarement observée, même sous traitement antiviral (3-16 %). Les protéines d’enveloppe du virus de l’hépatite B (VHB), formant l’AgHBs, sont très variables et cruciales pour le pouvoir infectieux du virus de l’hépatite B (VHB) et la physiopathologie. Nous avons émis l’hypothèse que cette variabilité pourrait expliquer, au moins partiellement, l’évolution de l’infection par le VHB, évaluée par la clairance de l’AgHBs, chez des patients traités ou non par analogues nucléos(t)idiques anti-VHB. Chez 29 patients infectés par différents génotypes du VHB (A, C et D), présentant différents profils cliniques (infection aigüe ou chronique, co-infection VHB/VIH) et thérapeutiques, une très grande variabilité des protéines d’enveloppe du VHB a été mise en évidence. Chez ces patients, la persistance de l’AgHBs était corrélée avec la présence de mutations et délétions localisées dans des régions des protéines d’enveloppe virale jouant un rôle important dans la reconnaissance du virus par le système immunitaire. Ces résultats renforcent l’hypothèse que l’étude des protéines d’enveloppe du VHB pourrait mettre en évidence des signatures moléculaires influençant le fitness du VHB et par conséquent l’évolution clinique de la maladie liée à l’infection par le VHB / Chronic hepatitis B affects about 257 million people worldwide. The loss of HBS antigen (HBsAg), a marker of the functional cure, is very rarely observed, even on anti-HBV treatment (3-16%). The hepatitis B virus (HBV) envelope proteins (HBsAg) are highly variable and crucial for the viral infectivity and pathogeny. We hypothesized that the HBV variability in the envelope proteins could explain, at least partially, the evolution of HBV infection, evaluated by HBsAg clearance, in patients treated or not by anti-HBV nucleos(t)idic analogues. For 29 patients infected with different HBV genotypes (A, C and D), presenting different clinical profiles (acute or chronic infection, HBV/HIV co-infection) and therapies, a very high variability of HBV envelope proteins was observed. In these patients, the persistence of HBsAg was correlated with the presence of mutations and deletions located in areas that play a key role in the viral recognition by the immune system. These results reinforce the hypothesis that the study of HBV envelope proteins could highlight molecular signatures influencing HBV fitness which would subsequently modify the clinical evolution of HBV-related disease

Les mutations de résistance pour une molécule sont produites avant que la molécule en question ne soit utilisée, et c’est sous « pression de sélection » que la souche résistante va être sélectionnée. Des données récentes montrent que des variants résistants minoritaires (VRMs) peuvent être une source d’échec virologique. Les nouvelles techniques de séquençage sont bien plus sensibles que les techniques classiques de séquençage et permettent la détection des VRMs. Afin d’évaluer l’intérêt de l’utilisation de ces techniques, nous avons étudié les prélèvements de patients en situation d’échec virologique après traitement par deux combinaisons antirétrovirales très utilisées (tenofovir/emtricitabine/efarirenz et tenofovir/emtricitabine/rilpivirine). De nombreux variants de résistance supplémentaires ont été détectés, touchant principalement la classe des Inhibiteurs Nucléosidiques de la Transcriptase Inverse (INTIs), avec un impact potentiel sur le choix du traitement de relais. Nous avons également étudié la prévalence des mutations de résistance transmise sur le gène de la protéase et de la transcriptase inverse chez des patients naïfs chroniquement infectés, chez deux groupes de transmission : des patients hommes ayant des rapports avec d’autres hommes (HSH), et des patients hétérosexuels. Nous avons retrouvé une prévalence plus élevée de mutations touchant les INTIs dans le groupe des patients hétérosexuels. Parmi les patients HSH, ceux infectés par un virus de sous-type B étaient plus fréquemment infectés par un virus résistant. Cette thèse met en avant la puissance des ces techniques, dont les conditions d'utilisation ne sont pas encore complètement définies. / Resistance mutations for a given molecule are produced before the molecule is used, and it is under "selection pressure" that the resistant strain will be selected. Recent data show that minority resistant variants (MRV) can be a source of virologic failure. New sequencing techniques are much more sensitive than conventional sequencing techniques and allow MRV detection. To assess the value of these new techniques, we studied samples from patients experiencing virologic failure after treatment with two antiretroviral combinations widely used (tenofovir/emtricitabine/efarirenz et tenofovir/emtricitabine/rilpivirine). Many additional resistance variants affecting the class of nucleoside reverse transcriptase inhibitors (NRTIs) were detected, with a potential impact on the choice of the subsequent regimen. We also studied the prevalence of transmitted resistance mutations in the protease and reverse transcriptase genes among naive patients chronically infected, among two groups of transmission: patients of men who have sex with men (MSM) and heterosexual patients. We found a higher prevalence of NRTI mutations among the heterosexual group. Among MSM patients, those infected with subtype B viruses were more frequently infected with a resistant virus. This thesis highlights the power of these techniques, the conditions of use are not yet fully defined.

Herpes simplex virus type 1 (HSV-1) is a double stranded DNA virus that causes epithelial skin infections and persists through the life of the host by infecting neurons, where it can switch to a latent state to evade an immune response. In rare cases during primary infection or after reactivation, instead of undergoing lytic infection at the epithelial surface, it instead travels to the brain and causes herpes simplex virus encephalitis (HSVE) which can have a ≥70% mortality rate if untreated. As the virus takes over its host cell, it gains control of the host cell machinery and manipulates host gene expression in order to evade the immune system and to pool its resources into the replication of the virus. One aspect of the dysregulated gene expression involves microRNAs (miRNAs). MiRNAs are short, non-coding RNAs that bind to the 3' untranslated region (3'UTR) of messenger RNAs (mRNAs), leading to translational repression of the target. Dysregulated miRNAs are often down-regulated during infection as the virus takes over, but many miRNAs have also been found to be up-regulated as well1–5. The aim of this study is to observe the full cellular miRNA changes in the context of an acute viral encephalitic infection using HSV-1, and to further characterize selected up-regulated miRNAs to determine their function in the context of the disease state. Of particular note were miR-141 and miR-200c which showed anti-apoptotic effects on neuronal cell culture and did not impact cell viability during an over-expression of the miRNAs. MiR-141, miR-183 and miR-200a expression was enriched within specific areas of the brain during infection. In addition, the potential for miR-150 to bind to a bioinformatically predicted target site within the shared 3'UTR of the HSV-1 UL18, UL19 and UL20 genes was explored. Examining the changes in expression of this class of regulatory RNAs and investigating their potential functions may yield new insight into the relationship between host and virus during infection.

Les éléments Alu sont les retrotransposons les plus prolifiques chez l’humain avec plus d’1 million de copies occupant plus de 10% du génome. Afin de contrecarrer l’expansion des rétro-éléments, les organismes ont développés différents mécanismes pour préserver l’intégrité de leurs génomes. Le plus proéminent, également utilisé pour lutter contre la réinsertion d’ADN viral dans le génome hôte, est l’édition de l’ARN. Chez les mammifères, la plus courante est la déamination de l’adénine en inosine catalysée par la famille de protéine ADAR dont Les principales cibles sont les éléments Alu chez l’humain. L’édition des éléments Alu conduit à leur séquestration dans le noyau des cellules, mute leurs promoteurs internes, cible de l’ARN polymérase III (POLIII), et leurs queues poly-A, prévenant ainsi leur future rétrotransposition. Dans la première partie de cette étude, l’analyse de données de séquençage haut-débit révèle que ~40% des éléments Alu sont reconnus par POLIII, qu’ils sont présents en tant que petits ARN dans le cytoplasme et le noyau des cellules, que certain d’entre eux sont associés à la chromatine, et que la transcription des éléments Alu est un phénomène courant dans les tissus somatiques qui concorde avec l’expression d’éléments LINE1 fonctionnels. Ceci suggère que la rétrotransposition peut être un mécanisme normal dans la plupart des tissus humains. Enfin, l’analyse de l’expression des éléments Alu et LINE1 chez la souris montre que la transcription de rétrotransposons n’est pas spécifique de l’humain. Dans la seconde partie de cette étude, une nouvelle méthode a été développée pour explorer l’impact de l’édition de l’ARN sur le transcriptome en identifiant les ARN édités par séquençage haut-débit. Dans un premier temps, un anticorps ciblant ADAR a été utilisé pour extraire les ARN associés aux protéines de l’édition. Cette méthode n’étant pas suffisamment efficace, une autre stratégie, qui extrait directement les ARN contenant de l’inosine, a été développée : dans un premier temps, l’ARN est fixé à des billes magnétiques par leurs extrémités 3’, ensuite, les billes sont traitées au glyoxal/acide borique et à la RNAse T1 pour libérer la région 5’ des ARN contenant une ou plusieurs inosines, et enfin, les ARN libérés sont séquencés par séquençage haut débit. En utilisant cette méthode, 1822 sites d’éditions ont été identifiés dans l’ARN de cerveau de souris, incluant 28 nouveaux sites présents dans des séquences codantes qui conduisent à des mutations non-synonymes des futures protéines. Des sites d’éditions ont aussi été observés pour la première fois dans les ARN ribosomaux, les snoRNA et les snRNA. / The Alu repeats comprise more than 10% of the human genome. They spread in the genome by retrotransposition. As a response to this invasion, organisms developed mechanisms to preserve the integrity of their genome, such as RNA editing. The most abundant type of editing in mammals is A-to-I editing where the ADAR proteins transform adenosine into inosine and targets mainly Alu elements in human. Editing of the Alu elements leads to their sequestration in the nucleus and mutates their internal POLIII promoter and their poly-A tail, thus preventing their subsequent transposition. In the first part of this study, we challenged the view that Alu elements are dormant occupant of the genome by characterizing their activity. Deep-sequencing data analyses revealed that ~40% of Alu elements can bind POLIII, they present a definite localization in the cell and associate with chromatin and polysomes, and that Alu elements transcription is a widespread phenomenon in normal tissues which correlates with functional LINE1 elements expression. This suggested that Alu element retrotransposition may be a natural mechanism in most normal human tissues. Further analyses showed that SINE and LINE expression in somatic tissues was not exclusive to human but also occurs in mouse. Finally, attempts were made to identify tissue specific insertions in the human genome resulting from retrotransposition events. In the second part of this study, a new method was developed to understand the full impact of RNA editing on transcriptomes by characterizing the edited RNA in a high-throughput fashion. First, immunoprecipitation was attempted to pull-down RNA associated with the editing enzymes ADARs. Since this method was inefficient, another approach purifying directly the edited RNA was developed. First, the RNA was sequestered on magnetic beads. Then an inosine specific cleavage based on RNAseT1 treatment of RNA protected with glyoxal and borate allowed the separation of the edited RNA from the total RNA. Finally, deep sequencing was used to identify edited RNA. 1,822 editing sites were found in mouse brain RNA by this method, including 28 new editing sites modifying the coding sequences of genes and editing in rRNA, snoRNA and snRNA which were never observed before.

L'ovocyte se développe au sein d'un follicule, en contact étroit avec des cellules d'origine somatique, les cellules de cumulus (CC). Ces deux types cellulaires communiquent entre eux via des jonctions intercellulaires, permettant ainsi la régulation et la coordination du métabolisme pendant le développement et la maturation de l'ovocyte. Notre hypothèse est que l'expression et la régulation des gènes dans les CC joue un rôle crucial dans des fonctions essentielles pour la croissance de l'ovocyte et l'acquisition de sa compétence. Mes travaux de thèse comportent deux parties. Dans la première partie nous avons utilisé le séquençage haut débit pour examiner le répertoire des microARN (communément appelés miRNA) dans les cellules de cumulus et dans l'ovocyte. Les miRNA, séquences d'ARN non codantes dont la longueur varie entre 19 et 25 nucléotides, ont émergé récemment comme régulateurs majeurs de nombreux processus biologiques, dont le vieillissement. Nous avons identifié 32 miRNA spécifiquement dans les cellules de cumulus humains et seulement 3 dans l'ovocyte MII. Dans la seconde partie de nos travaux, nous avons analysé l'impact de l'âge maternel sur l'expression des gènes dans les cellules de cumulus. Alors qu'une baisse de la compétence de l'ovocyte avec l'avancement de l'âge maternel est bien établie, les bases moléculaires de ce phénomène demeurent peu connues. Dans une première étape pour aborder cette question, nous avons utilisé des puces à ADN pour analyser les profils d'expression des gènes des CC en fonction de l'âge maternel. De façon remarquable l'âge maternel impacte significativement l'expression de gènes qui sont critiques pour la maturation de l'ovocyte tels que les gènes impliqués dans l'angiogenèse, les voies de signalisation de TGF-ß et de l'insuline. Par l'utilisation d'outils bioinformatiques, nous avons aussi identifié des miRNA potentiels régulateurs de gènes impliqués dans des processus ou des voies impactés par l'âge ; ils pourraient constituer de nouveaux biomarqueurs pour prédire un vieillissement ovarien prématuré ainsi que la qualité et la compétence de l'ovocyte. / The oocyte develops into a follicle where it is in close contact with cumulus cells (CCs), of somatic origin. The two cell types undergo a bidirectional communication via gap junctions, which results in the regulation and coordination of the metabolism during oocyte development and maturation. We assume that gene expression and regulation in the CCs play a crucial role in functions that are essential for oocyte growth and competence acquisition. The present study may be subdivided in two parts. In the first part we used deep sequencing to investigate the repertoire of miRNAs in the cumulus cells and the oocyte. MicroRNAs that are noncoding RNA sequences whose length is approximately 19-25 nucleotides have emerged as important regulators in many biological processes including aging. Our data showed that 32 miRNAs were specifically expressed in human cumulus cells while only 3 miRNAs were identified in MII human oocyte. The impact of maternal age on gene expression in cumulus cells was addressed in a second part of my thesis work. While the correlation of oocyte competence decline with advancing maternal age is well established, little is known on its molecular basis. In a first attempt to address this issue, we used microarrays to study gene expression profiles of human cumulus cells according to maternal age. Remarkably, maternal age greatly impacted expression of genes that are critical for oocyte maturation such as genes involved in angiogenesis, TGF-β signaling, and insulin signaling pathways. Also, using bioinformatic tools, we identified miRNAs that potentially target some of the genes involved in the aging-impacted processes and pathways; this could candidate them as new biomarkers to predict premature ovarian aging and oocyte quality and competence.