Streptococcus agalactiae is a pathogen of multiple hosts. The bacterium, an aetiological agent of septicaemia and meningo-encephalitis in freshwater and saltwater fish species, is considered a major threat to the aquaculture industry, particularly for tilapia. Cattle and humans are however the main known reservoirs for S. agalactiae. In humans, the bacterium (commonly referred to as Group B Streptococcus or GBS) is a member of the commensal microflora of the intestinal and genito-urinary tracts, but it is also a major cause of neonatal invasive disease and an emerging pathogen in adults. In cattle, S. agalactiae is a well-recognized causative agent of mastitis. Numerous studies focusing on S. agalactiae from human and bovine origins have provided insight into the population structure of the bacterium, as well as the genome content and pathogenic mechanisms through identification of virulence determinants. Concerning S. agalactiae from aquatic origins, scientific information mainly focused on case reporting and/or experimental challenges, with a limited or absence of information in terms of pathogenesis, virulence determinants and genotypes of the strains involved. The objective of this study was to enhance our understanding of the molecular epidemiology, host-adaptation and pathogenicity of S. agalactiae in aquatic species, with particular emphasis on tilapia. Firstly, a collection of 33 piscine, amphibian and sea mammal isolates originating from several countries and continents was assembled, with the aim of exploring the population structure and potential host specificity of aquatic S. agalactiae. Isolates were characterised using pulsed-field gel electrophoresis (PFGE), multi-locus sequence typing (MLST), and a standardised 3-set genotyping system comprising molecular serotypes, surface protein gene profiles and mobile genetic element profiles. Two major subpopulations were identified in fish. The first subpopulation consisted of non-haemolytic isolates that belonged to sequence type (ST) 260 or 261, which are STs that have been reported only from teleosts. These isolates exhibited a low level of genetic diversity by PFGE and clustered with other STs that have been reported only in fish. Another common feature was the absence of all surface protein genes or mobile genetic elements targeted as part of the 3-set genotyping and that are usually found in human or bovine isolates. The second subpopulation consisted of β-haemolytic isolates recovered from fish, frogs and sea mammals, and that exhibited medium to high genetic diversity by PFGE. STs identified among these isolates have previously been identified from strains associated with asymptomatic carriage and invasive disease in humans. The human pathogenic strain ST7 serotype Ia was detected in fish from Asia. Moreover, ST283 serotype III-4 and its novel single locus variant ST491 detected in fish from Southeast Asia shared a 3-set genotype identical to that of an emerging ST283 clone associated with invasive disease of adult humans in Asia. These observations suggested that some strains of aquatic S. agalactiae may present a zoonotic or anthroponotic hazard. STs found among the seal isolates (ST23) have also been reported from humans and numerous other host species, but never from teleosts. This work provided an excellent basis for exploration of the virulence of selected strains in experimental challenges. The virulence of two strains of S. agalactiae was experimentally investigated by intra-peritoneal infection of Nile tilapia (Oreochromis niloticus), using an isolate originally recovered from fish and belonging to ST260, and an isolate originating from a grey seal and belonging to ST23. The clinical signs, the in vivo distribution of viable bacteria and bacterial antigens, and the gross and histopathological lesions that developed during the time course of the infection were investigated. The ST260 strain was highly virulent, whereas no major clinical sign or mortalities occurred in the fish challenged with the ST23 strain. After injection, both strains however gained access to the bloodstream and viable bacteria were recovered from all organs under investigation. During the early stages of infection, bacteria were mostly found within the reticulo-endothelial system of the spleen and kidney. Thereafter, the ST260 demonstrated a particular tropism for the brain and the heart, but granulomatous inflammation and associated necrotic lesions were observed in all organs. ST23 was responsible for a mixed inflammatory response associated with the presence of bacteria in the choroid rete and in the pancreatic tissue only. After 7 days post-challenge and for both strain, the formation or containment of bacteria within granulomata or other encapsulated structures appeared to be a major component of the fish response. However, the load of viable bacteria remained high within organs of fish infected with ST260, suggesting that, unlike ST23, this strain is able to survive within macrophages and/or to evade the immune system of the fish. This work demonstrates that the lack of report of ST23 strains in fish is possibly not due to a lack of exposure but to a lack of virulence in this host. The two strains, which differ in prevalence and virulence in fish, provide an excellent basis to investigate genomic differences underlying the host-association of distinct S. agalactiae subpopulations. The genome of the ST260 strain used in challenge studies was sequenced. We therefore provided the first description for the genome sequence of a non-haemolytic S. agalactiae isolated from tilapia (strain STIR-CD-17) and that belongs by multi-locus sequence typing (MLST) to clonal complex (CC) 552, which corresponds to a presumptive fish-adapted subgroup of S. agalactiae. The genome was compared to 13 S. agalactiae genomes of human (n=7), bovine (n=2), fish (n=3) and unknown (n=1) origins. Phylogenetic analysis based on the core genome identified isolates of CC552 as the most diverged of all S. agalactiae studied. Conversely, genomes from β-haemolytic isolates of CC7 recovered from fish were found to cluster with human isolates of CC7, further supporting the possibility that some strains may represent a zoonotic or anthroponotic hazard. Comparative analysis of the accessory genome enabled the identification of a cluster of genes uniquely shared between CC7 and CC552, which encode proteins that may provide enhanced fitness in specific niches. Other genes identified were specific to STIR-CD-17 or to CC552 based on genomic comparisons; however the extension of this analysis through the PCR screening of a larger population of S. agalactiae suggested that some of these genes may occasionally be present in isolates belonging to CC7. Some of these genes, occurring in clusters, exhibited typical signatures of mobile genetic elements, suggesting their acquisition through horizontal gene transfer. It is not possible to date to determine whether these genes were acquired through intraspecies transfer or through interspecies transfer from the aquatic environment. Finally, general features of STIR-CD-17 highlighted a distinctive genome characterised by an absence of well conserved insertion sequences, an abundance of pseudogenes, a smaller genomic size than normally observed among human or bovine S. agalactiae, and an apparent loss of metabolic functions considered conserved within the bacterial species, indicating that the fish-adapted subgroup of isolates (CC552) has undergone niche restriction. Finally, genes encoding recognised virulence factors in human S. agalactiae were selected and their presence and structural conservation was evaluated within the genome of STIR-CD-17.

La phagocytose est un processus cellulaire par lequel de larges particules sont internalisées dans une vésicule, le phagosome. Lorsque formé, le phagosome acquiert ses propriétés fonctionnelles à travers un processus complexe de maturation nommé la biogénèse du phagolysosome. Cette voie implique une série d’interactions rapides avec les organelles de l’appareil endocytaire permettant la transformation graduelle du phagosome nouvellement formé en phagolysosome à partir duquel la dégradation protéolytique s’effectue. Chez l’amibe Dictyostelium discoideum, la phagocytose est employée pour ingérer les bactéries de son environnement afin de se nourrir alors que les organismes multicellulaires utilisent la phagocytose dans un but immunitaire, où des cellules spécialisées nommées phagocytes internalisent, tuent et dégradent les pathogènes envahissant de l’organisme et constitue la base de l’immunité innée. Chez les vertébrés à mâchoire cependant, la transformation des mécanismes moléculaires du phagosome en une organelle perfectionnée pour l’apprêtement et la présentation de peptides antigéniques place cette organelle au centre de l’immunité innée et de l’immunité acquise. Malgré le rôle crucial auquel participe cette organelle dans la réponse immunitaire, il existe peu de détails sur la composition protéique et l’organisation fonctionnelles du phagosome. Afin d’approfondir notre compréhension des divers aspects qui relient l’immunité innée et l’immunité acquise, il devient essentiel d’élargir nos connaissances sur les fonctions moléculaire qui sont recrutées au phagosome. Le profilage par protéomique à haut débit de phagosomes isolés fut extrêmement utile dans la détermination de la composition moléculaire de cette organelle. Des études provenant de notre laboratoire ont révélé les premières listes protéiques identifiées à partir de phagosomes murins sans toutefois déterminer le ou les rôle(s) de ces protéines lors du processus de la phagocytose (Brunet et al, 2003; Garin et al, 2001). Au cours de la première étude de cette thèse (Stuart et al, 2007), nous avons entrepris la caractérisation fonctionnelle du protéome entier du phagosome de la drosophile en combinant diverses techniques d’analyses à haut débit (protéomique, réseaux d’intéractions protéique et ARN interférent). En utilisant cette stratégie, nous avons identifié 617 protéines phagosomales par spectrométrie de masse à partir desquelles nous avons accru cette liste en construisant des réseaux d’interactions protéine-protéine. La contribution de chaque protéine à l’internalisation de bactéries fut ensuite testée et validée par ARN interférent à haut débit et nous a amené à identifier un nouveau régulateur de la phagocytose, le complexe de l’exocyst. En appliquant ce modèle combinatoire de biologie systémique, nous démontrons la puissance et l’efficacité de cette approche dans l’étude de processus cellulaire complexe tout en créant un cadre à partir duquel il est possible d’approfondir nos connaissances sur les différents mécanismes de la phagocytose. Lors du 2e article de cette thèse (Boulais et al, 2010), nous avons entrepris la caractérisation moléculaire des étapes évolutives ayant contribué au remodelage des propriétés fonctionnelles de la phagocytose au cours de l’évolution. Pour ce faire, nous avons isolé des phagosomes à partir de trois organismes distants (l’amibe Dictyostelium discoideum, la mouche à fruit Drosophila melanogaster et la souris Mus musculus) qui utilisent la phagocytose à des fins différentes. En appliquant une approche protéomique à grande échelle pour identifier et comparer le protéome et phosphoprotéome des phagosomes de ces trois espèces, nous avons identifié un cœur protéique commun à partir duquel les fonctions immunitaires du phagosome se seraient développées. Au cours de ce développement fonctionnel, nos données indiquent que le protéome du phagosome fut largement remodelé lors de deux périodes de duplication de gènes coïncidant avec l’émergence de l’immunité innée et acquise. De plus, notre étude a aussi caractérisée en détail l’acquisition de nouvelles protéines ainsi que le remodelage significatif du phosphoprotéome du phagosome au niveau des constituants du cœur protéique ancien de cette organelle. Nous présentons donc la première étude approfondie des changements qui ont engendré la transformation d’un compartiment phagotrophe à une organelle entièrement apte pour la présentation antigénique. / Phagocytosis is a cellular process by which large particulate material are internalized in a newly formed vesicule, the phagosome. Once formed, the phagosome acquires its functional properties through a complex maturation process called phagolysosome biogenesis. This pathway involves a series of rapid interactions with organelles of the endocytic apparatus, enabling the gradual transformation of newly formed phagosomes into phagolysosomes in which proteolytic degradation occurs. The amoeba Dictyostelium discoideum uses phagocytosis as a predation mechanism for feeding, whereas multicellular organisms utilize this process as an immune mechanism where specialized cells named phagocytes internalize, kill and degrade phatogens found through the host, forming the basis of innate immunity. In jawed verterbrates however, the phagosome links innate and adaptive immunity by processing and presenting antigenic peptides. Despite its crucial role in immunity, little is known about the composition and the functional organization of the phagosome. It is therefore essential to characterize in details the functional properties that are recruited to the phagosome. High-throughput proteomics analysis of isolated phagosomes has been tremendously helpful for the molecular comprehension of this organelle. Studies of our lab notably have revealed the first proteomics identification of mouse phagosomes without determining the roles of these proteins through the complex process of phagocytosis (Brunet et al, 2003; Garin et al, 2001). In the first study of this thesis (Stuart et al, 2007), we characterized the functions of the entire drosophila phagosome proteome by combining high-throughput proteomics, interactive networks and RNAi. By applying this strategy, we’ve identified 617 phagosomal proteins by mass spectrometry from which we’ve expanded this list by building the phagosome interactome. The contribution of each protein to bacterial internalization was tested and validated by RNAi and led to the identification of a new regulator of phagocytosis, the exocyst complex. In generating this 'systems-based model', we show the power of applying this approach to the study of complex cellular processes and organelles and expect that this detailed model of the phagosome will provide a new framework for studying host-pathogen interactions and innate immunity. In the second study of this thesis (Boulais et al, 2010), we characterized some of the key steps that contributed to the remodeling of phagosomes functional properties during evolution. To do so, we isolated this organelle from three distant organisms: the amoeba Dictyostelium discoideum, the fruit fly Drosophila melanogaster, and mouse (Mus musculus) that use phagocytosis for different purposes. By performing and comparing proteomics and phosphoproteomics analyses of isolated phagosomes from the three species, we identified an ancient core of phagosomal proteins around which the immune function of this organelle have likely organized. Our data indicate that a larger proportion of the phagosome proteome, has been acquired through gene duplication at periods coinciding with the emergence of innate and adaptive immunity. Our study also characterizes in detail the acquisition of novel proteins and the significant remodeling of the phagosome phosphoproteome that contributed to modify the core constituents of this organelle in evolution. Our work thus provides the first thorough analysis of the changes that enabled the transformation of the phagosome from a phagotrophic compartment into an organelle fully competent for antigen presentation.

Avec plus de 1000 génomes complets disponibles (la grande majorité venant de bactéries), les analyses comparatives de génomes deviennent indispensables pour leurs annotations fonctionnelles, ainsi que pour la compréhension de leur structure et leur évolution, et s'appliquent par exemple en phylogénomique ou au design des vaccins. L'une des approches de plus utilisées pour comparer des génomes est l'alignement de leurs séquences d'ADN, i.e. alignement de génomes complets, c'est à dire identifier les régions de similarité en s'affranchissant de toute annotation. Malgré des améliorations significatives durant les dernières années, des outils performants pour cette approche ainsi que des méthodes pour l'estimation de la qualité des résultats qu'elle produit, en particulier sur les génomes bactériens, restent encore à développer. Outre leurs grandes tailles qui rendent les solutions classiques basées sur la programmation dynamique inutilisables, l'alignement de génomes complets posent des difficultés supplémentaires dues à leur évolution particulière, comprenant: la divergence, qui estompe les similarités entre les séquences, le réordonnancent des portions génomiques (réarrangements), ou l'acquisition de matériel génétique extérieur, qui produit des régions non alignables entres les séquences, e.g. transfert horizontal des gènes, phages. En conséquence, les solutions pour l'alignement de génomes sont des heuristiques, dont la plus commune est appelée stratégie basée sur des ancres. Cette stratégie commence par identifier un ensemble initial de régions de similarité (phase 1). Ensuite une phase de chaînage sélectionne un sous-ensemble (non-chevauchantes et généralement colinéaires) de ces similarités de poids maximal, nommées ancres (phase 2). Les phases 1 et 2 sont appliquées de manière récursive sur les régions encore non-alignées (phase 3). La dernière phase consiste en l'application systématique des outils d'alignement classiques sur toutes les régions courtes qui n'ont pas encore été alignées. Cette thèse adresse plusieurs problèmes liés à l'alignement de génomes complets dont: l'évaluation de la qualité des résultats produits par les outils d'alignement et l'amélioration de la stratégie basée sur des ancres. Premièrement, nous avons créé un protocole pour évaluer la qualité des résultats d'alignement, contenant des mesures de calcul quantitatives et qualitatives, dont certaines basées sur des connaissances biologiques. Une analyse de la qualité des alignements produits par deux des principaux outils existants sur des paires de génomes bactériens intra-espèces révèle leurs limitations: des similarités non détectées et des portions d'alignement incorrectes. À partir de ces résultats, qui suggèrent un manque de sensibilité et spécificité, nous proposons un nouvel outil pour l'alignement deux à deux de génomes complets, YOC, qui implémente une version simplifiée de la stratégie basée sur des ancres, contenant seulement deux phases. Dans la phase 1, YOC améliore la sensibilité en utilisant comme ancres, pour la première fois dans cette stratégie, des similarités locales basées sur des graines espacées, capables de détecter des similarités plus longues dans des régions plus divergentes. Cette phase est suivie par une méthode de chainage adaptée aux similarités locales, un nouveau type de chaînage colinéaire, permettant des chevauchements proportionnels. Nous avons donné une formulation de ce nouveau problème et réalisé un premier algorithme. L'algorithme, qui adopte une approche de programmation dynamique basée sur le paradigme de la ``sweep-line'', donne une solution optimale, i.e. est exacte, et s'exécute en temps quadratique. Nous avons montré que cet algorithme, comparé au chainage colinéaire classique, améliore les résultats sur des génomes bactériens, tout en restant aussi efficace en pratique. / With more than 1000 complete genomes available (among which, the vast majority come from bacteria), comparative genomic analysis become essential for the functional annotation of genomes, the understanding of their structure and evolution and have applications in phylogenomics or vaccine design. One of the main approaches for comparing genomes is by aligning their DNA sequences, i.e. whole genome alignment (WGA), which means identifying the similarity regions without any prior annotation knowledge. Despite the significant improvements during the last years, reliable tools for WGA and methodology for estimating its quality, in particular for bacterial genomes, still need to be designed. Besides their extremely large lengths that make classical dynamic programming alignment methods unsuitable, aligning whole genomes involves several additional difficulties, due to the mechanisms through which genomes evolve: the divergence, which let sequence sim ilarity vanish over time, the reordering of genomic segments (rearrangements), or the acquisition of external genetic material generating regions that are unalignable between sequences, e.g. horizontal gene transfer, phages. Therefore, whole genome alignment tools implement heuristics, among which the most common is the anchor based strategy. It starts by detecting an initial set of similarity regions (phase 1), and, through a chaining phase (phase 2), selects a non-overlapping maximum-weighted, usually collinear, subset of those similarities, called anchors. Phases 1 and 2 are recursively applied on yet unaligned regions (phase 3). The last phase (phase 4) consists in systematically applying classical alignment tools to all short regions still left unaligned.This thesis addresses several problems related to whole genome alignment: the evaluation of the quality of results given by WGA tools and the improvement of the classical anchor based strategy. We first designed a protocol for evaluating the quality of alignment results, based on both computational and biological measures. An evaluation of the results given by two state of the art WGA tools on pairs of intra-species bacterial genomes revealed their shortcomings: the failure of detecting some of the similarities between sequences and the misalignment of some regions. Based on these results, which imply a lack in both sensitivity and specificity, we propose a novel, pairwise whole genome alignment tool, YOC, implementing a simplified two-phase version of the anchor strategy. In phase 1, YOC improves sensitivity by using as anchors, for the first time, local similarities based on spaced seeds that are capable of detecting larger similarity regions in divergent sequences. This ph ase is followed by a chaining method adapted to local similarities, a novel type of collinear chaining, allowing for proportional overlaps. We give a formulation for this novel problem and provide the first algorithm for it. The algorithm, implementing a dynamic programming approach based on the sweep-line paradigm, is exact and runs in quadratic time. We show that, compared to classical collinear chaining, chaining with overlaps improves on real bacterial data, while remaining almost as efficient in practice. Our novel tool, YOC, is evaluated together with other four WGA tools on a dataset composed of 694 pairs of intra-species bacterial genomes. The results show that YOC improves on divergent cases by detecting more distant similarities and by avoiding misaligned regions. In conclusion, YOC should be easier to apply automatically and systematically to incoming genomes, for it does not require a post-filtering step to detect misalignment and is less complex to calibrate.

B. diazoefficiens CPAC 7 e B. japonicum CPAC 15 são estirpes brasileiras de Bradyrhizobium que apresentam grande relevância para o cultivo da soja, pois são capazes de fornecer nitrogênio para a produção desta leguminosa através do processo de fixação biológica de nitrogênio (FBN), uma técnica sustentável e de baixo custo. Por esse motivo, tais bactérias são de grande interesse, e seu estudo contribui na compreensão do processo complexo e orquestrado por um conjunto de genes específicos que culmina no estabelecimento da simbiose. A estirpe CPAC 7 possui maior eficiência em fixar N2 , e a CPAC 15 destaca-se pela sua competitividade. Recentemente, o genoma de cada uma foi sequenciado na tentativa de conhecer seu conteúdo gênico e identificar os fatores genéticos responsáveis pelas diferenças no desempenho simbiótico. Apesar de ter sido encontrado alguns rearranjos, os genoma mostraram-se sintênicos na sua maioria. Entretanto, o fato de haver muitas transposases ao redor dos genes, principalmente na ilha simbiótica, e devido a presença de muitos genes hipotéticos, representando uma limitação no conhecimento, nos motivou a realizar o presente estudo, onde exploramos estes dois genomas. Portanto, os objetivos deste estudo foram de definir a população de elementos de transposição (TEs) que compõe estes genomas, avaliar se os elementos completos podem estar impactando os genes de alguma forma; explorar as proteínas hipotéticas, tentando identificar novas funções que possam estar associadas com a interação soja-Bradyrhizobium e apontá-las para estudos experimentais futuros; e ainda explorar os genes exclusivos das regiões atípicas dos genomas, sendo que para isso, nós também desenvolvemos uma nova metodologia, baseada na máxima entropia (ME), que pode ser utilizada em novos estudos genômicos a partir da simples sequência nucleotídica. Todas as análises deste estudo foram realizadas in silico. Estudando os TEs, identificamos 33 novas sequências de inserção, sendo que algumas destacaram-se por terem potencial impacto nos genes associados com a simbiose destas bactérias, como nopAN, nopAG, rhcU, modC e hypB. Explorar as proteínas hipotéticas nos permitiu reduzir a porcentagem de hipotéticas dos genomas. Adicionamos novas informações à 1.204 proteínas, das quais muitas apresentaram similaridade com proteínas comprovadamente associadas com a interação planta-bactéria, em condições de simbiose e/ou patogenicidade, como proteínas envolvidas na motilidade e adesão celular, fatores de virulência, proteínas secretoras e efetoras, entre outras. Além disso, a metodologia ME, desenvolvida neste estudo com o intuito de direcionar análises genômicas para regiões atípicas, quando comparada com outras ferramentas existentes, mostrou-se superior em termos de eficiência e tempo de execução computacional. Nas regiões genômicas apontadas pela ME nos dois genomas de interesse, identificamos 269 genes exclusivos de CPAC 7 e 368 de CPAC 15, sendo que destacamos aqueles com potencial relação com as diferenças simbióticas das estirpes, como o gene fixW, noeE, rtxA e nex18. Assim, os resultados obtidos neste trabalho vêm expandir nosso conhecimento sobre os genomas destas estirpes. Destacando ainda, importantes diferenças que podem estar associadas com a habilidade simbiótica de cada bactéria. / B. diazoefficiens CPAC 7 and B. japonicum CPAC 15 are Brazilian Bradyrhizobium strains of great importance for soybean cultivation, since when in a symbiotic state they provide nitrogen for the crop through the biological nitrogen fixation process (BNF), a sustainable technique and low cost. For this reason, such bacteria represent great interest and have been widely studied, once the symbiotic establishment is a complex process and orchestrated by a specific set of genes. The CPAC 7 strain has a higher efficiency to fix N2 , while CPAC 15 stands out for its competitiveness. Recently, their genomes were sequenced in an attempt to gain knowledge about their gene content and to identify the genetic factors responsible for differences in their symbiotic performance. Despite having identified some rearrangements, the majority of genomes showed syntenic. However, the fact that there are many transposases around the genes, especially in symbiotic island, and due to the presence of many hypothetical genes, representing a limitation on knowledge, motivated us to conduct this study, which explored these two important genomes. Therefore, the objectives of this study were to define the population of transposable elements (TEs) present in these genomes and to verify whether such TEs could be impacting the genes somehow; to study the hypothetical proteins, trying to identify new features that may be associated with the soybean-Bradyrhizobium interaction and point them for future experimental studies; and to explore the exclusive genes from atypical regions of both genomes, and for that, we have also developed a new methodology, based on maximum entropy (ME), which can be used in new genomic studies. All analyzes in this study were performed in silico. Studying the TEs, we identified 33 new insertion sequences, and some stood out for having potential impact on genes associated with the symbiosis of these bacteria, such as nopAN, nopAG, rhcU, modC and hypB. As a consequence of improving the annotation of hypothetical proteins we were able to reduce the hypothetical percentage. Among these, we add new information to 1,204 proteins, many of which had similarity to proteins with involvement in the plant-bacteria interaction, in symbiosis and/or pathogenicity conditions, such as proteins involved in cell motility and adhesion, virulence factors, secretion proteins, effectors, among others. Moreover, the ME methodology developed in this study to direct genomic analysis to atypical regions, compared with other existing tools, it was superior in efficiency and execution time. In the genomic regions identified by the ME in both Bradyrhizobium genomes, we identified 269 exclusive genes of CPAC 7 and 368 of CPAC 15, we highlighted those with potential involvement with symbiotic differences of strains, as fixW, noeE, rtxA and nex18. Thus, the results obtained in this study come to expand our knowledge about the genomes of these important bacteria. Finally, differences were identified as potential targets to be associated with the symbiotic ability of each strain to be futher studied.

Les angiospermes (ou plantes à fleurs) regroupent environ 350 000 espèces ayant divergé il y a 150 à 200 millions d’années en deux familles botaniques principales, les monocotylédones (les orchidées, les palmiers, les bananiers, les joncs, les graminées...) et les eudicotylédones (les Brassicaceae, les Rosaceae, les légumineuses...) représentant respectivement 20% et 75% des plantes à fleurs. Les angiospermes font l’objet de nombreux travaux de recherche, en particulier en génomique depuis 2000 avec le séquençage du premier génome de plantes (Arabidopsis thaliana) qui a précédé le décryptage des génomes d’un nombre important d’autres espèces modèles et/ou d’intérêt agronomique (environ 100 aujourd’hui). L’accès croissant à la séquence des génomes de plantes a permis de mettre à jour une importante diversité structurale de leur génome, en termes de taille physique, de nombre de chromosomes, de nombre de gènes et de richesse en éléments transposables. Les forces évolutives ayant permis une telle diversité structurale des génomes au cours de l’évolution sont au cœur des travaux de cette thèse. La paléogénomique se propose d’étudier à travers la reconstruction de génomes ancestraux, comment ces espèces ont divergé à partir d’ancêtres communs et quels mécanismes ont contribué à une telle plasticité de structure génomique. Dans cet objectif, les travaux de cette thèse ont mis en œuvre des méthodes basées sur la génomique comparée permettant l’étude de l’évolution structurale des génomes via la reconstruction des génomes ancestraux fondateurs des espèces modernes. Ainsi, un génome ancestral des angiospermes a été reconstruit constitué de 5 chromosomes et porteur de 6707 gènes ordonnés sur ceux-ci, permettant d’intégrer dans un même modèle les monocotylédones et les eudicotylédones et élucider leur histoire évolutive, notamment pour les espèces d’intérêt agronomique majeur telles que les céréales, les rosids et les Brassicaceae. L’inférence de ces génomes ancestraux des plantes modernes a permis l’identification et l’étude de l’impact des évènements de polyploïdie (doublement génomique), ubiquitaires chez les plantes. Nous avons montré que les génomes tendent à revenir à une structure diploïde suite à un évènement de polyploïdie. Cette diploïdisation structurale se fait au niveau caryotypique (par le biais de réarrangements chromosomiques impliquant la perte des centromères et télomères ancestraux) mais aussi géniques (par le biais de pertes de gènes ancestraux en double copies). Il a été montré que cette perte se faisait préférentiellement sur un des sous-génomes post-polyploïdie, menant au phénomène de « dominance des sous-génomes ». Ces biais de plasticité structurale (on parle de compartimentation de la plasticité) se font différentiellement entre les espèces, les chromosomes, les compartiments chromosomiques mais aussi les types de gènes, aboutissant à la diversité structurale observée entre les génomes modernes de plantes. Ces travaux qui rentrent dans le cadre de la recherche fondamentale ont également un fort aspect appliqué à travers la recherche translationnelle en ayant permis de créer des passerelles entre les différentes espèces travaillées en agriculture. Le passage d’une espèce à une autre via les génomes ancestraux fondateurs reconstruits permet notamment le transfert de connaissances des gènes ou de régions d’intérêt des espèces modèles aux espèces cultivées. Les travaux de thèse, par la reconstruction d’ancêtres, permettent une comparaison de haute-résolution des génomes de plantes et in fine l’étude de leur plasticité acquise au cours de l’évolution, et revêtent donc à la fois un aspect fondamental (pour comprendre l’évolution des espèces) mais aussi appliqué (pour l’amélioration des espèces d’intérêt agronomique à partir des modèles). / Angiosperms (or flowering plants) consist in approximatively 350 000 species that have diverged 150 to 200 million years ago in two main families, monocots (orchids, palm trees, banana, bulrushes, grasses...) and dicots (Brassicaceae, Rosaceae, legumes...) representing respectively 20% and 75% of flowering plants. Angiosperms are the subject of intense researches, in particular in genomics since 2000 with the sequence release of the first plant genome (Arabidopsis thaliana) preceding a large number of genomes of plant models and/or species of agronomical interest (around 100 today). Increasing access to plant genome sequences has allowed the identification of their structural diversity, in terms of genome size, number of chromosomes and genes as well as transposable element content. The evolutionary forces that have shaped such structural genomic divergence are at the center of this thesis. Our paleogenomics approach will investigate, through ancestral genome reconstructions, how modern species have diverged from common ancestors and which mechanisms have contributed to such present-day genome plasticity. In this thesis, we have developed methods based on comparative genomics to study plant genome evolution and reconstruct ancestral genomes, extinct progenitors of the modern angiosperm species. An ancestral angiosperm genome has been reconstructed made of 5 chromosomes and 6707 ordered genes allowing the integration in the same model of monocots and eudicots and finally elucidating evolutionary trajectories for species of major agricultural interest such as cereals, rosids and Brassicaceae. The reconstructed paleohistory of modern flowering plants enabled the identification as well as the investigation of the impact of polyploidy events (WGD, whole genome duplications), ubiquitous in plants, as a major driver of the observed structural plasticity of angiosperms. We established that genomes tend to return to a diploid status following a polyploidy event. This structural diploidization is performed at the karyotypic level through chromosomal rearrangements (involving ancestral centromeres and telomeres losses) as well as the gene level (through ancestral duplicates loss). It has been shown that this diploidization is preferentially done on one of the post-polyploidy subgenome, leading to the "sub-genome dominance" phenomenon. This structural plasticity bias (also referenced as plasticity partitioning) is acting differentially between species, chromosomes, chromosomal compartments, gene types, resulting in the structural diversity observed between the present-day plant genomes. This thesis is clearly within the scope of fundamental researches but also has a strong applied objective through translational research in creating bridges between species of major relevance for agriculture. The comparison of one species to another through the reconstructed ancestral genomes allows transferring knowledge gained on genes or any region of interest from model species to crops. Paleogenomics, in reconstructing ancestral genome and unveiling the forces driving modern plant genome plasticity, is therefore of fundamental (toward understanding species evolution) but also applied (toward improving orphan species from knowledge gained in models) objectives.

El trabajo que aquí se presenta, estudia el reconocimiento de las señales que delimitan y definen los genes que codifican para proteínas, así como su aplicabilidad en los programas de predicción de genes. La tesis que aquí se presenta, también explora la utilitzación de la genómica comparativa para mejorar la identificación de genes en diferentes especies simultaniamente. También se explica el desarrollo de dos programas de predicción computacional de genes: geneid y sgp2. El programa geneid identifica los genes codificados en una secuencia anónima de DNA basandose en sus propiedades intrínsecas (principalmente las señales de splicing y el uso diferencial de codones). sgp2 permite utilitzar la comparación entre dos genomas, que han de estar a una cierta distancia evolutiva óptima, para mejorar la predicción de genes, bajo la hipotesis que las regiones codificantes están mas conservadas que las regiones que no codifican para proteínas. / The motivation of this thesis is to give a little insight in how genes are encoded and recognized by the cell machinery and to use this information to find genes in unannotated genomic sequences. One of the objectives is the development of tools to identify eukaryotic genes through the modeling and recognition of their intrinsic signals and properties. This thesis addresses another problem: how the sequence of related genomes can contribute to the identification of genes. The value of comparative genomics is illustrated by the sequencing of the mouse genome for the purpose of annotating the human genome. Comparative gene predictions programs exploit this data under the assumption that conserved regions between related species correspond to functional regions (coding genes among them). Thus, this thesis also describes a gene prediction program that combines ab initio gene prediction with comparative information between two genomes to improve the accuracy of the predictions.

La phagocytose est un processus cellulaire par lequel de larges particules sont internalisées dans une vésicule, le phagosome. Lorsque formé, le phagosome acquiert ses propriétés fonctionnelles à travers un processus complexe de maturation nommé la biogénèse du phagolysosome. Cette voie implique une série d’interactions rapides avec les organelles de l’appareil endocytaire permettant la transformation graduelle du phagosome nouvellement formé en phagolysosome à partir duquel la dégradation protéolytique s’effectue. Chez l’amibe Dictyostelium discoideum, la phagocytose est employée pour ingérer les bactéries de son environnement afin de se nourrir alors que les organismes multicellulaires utilisent la phagocytose dans un but immunitaire, où des cellules spécialisées nommées phagocytes internalisent, tuent et dégradent les pathogènes envahissant de l’organisme et constitue la base de l’immunité innée. Chez les vertébrés à mâchoire cependant, la transformation des mécanismes moléculaires du phagosome en une organelle perfectionnée pour l’apprêtement et la présentation de peptides antigéniques place cette organelle au centre de l’immunité innée et de l’immunité acquise. Malgré le rôle crucial auquel participe cette organelle dans la réponse immunitaire, il existe peu de détails sur la composition protéique et l’organisation fonctionnelles du phagosome. Afin d’approfondir notre compréhension des divers aspects qui relient l’immunité innée et l’immunité acquise, il devient essentiel d’élargir nos connaissances sur les fonctions moléculaire qui sont recrutées au phagosome. Le profilage par protéomique à haut débit de phagosomes isolés fut extrêmement utile dans la détermination de la composition moléculaire de cette organelle. Des études provenant de notre laboratoire ont révélé les premières listes protéiques identifiées à partir de phagosomes murins sans toutefois déterminer le ou les rôle(s) de ces protéines lors du processus de la phagocytose (Brunet et al, 2003; Garin et al, 2001). Au cours de la première étude de cette thèse (Stuart et al, 2007), nous avons entrepris la caractérisation fonctionnelle du protéome entier du phagosome de la drosophile en combinant diverses techniques d’analyses à haut débit (protéomique, réseaux d’intéractions protéique et ARN interférent). En utilisant cette stratégie, nous avons identifié 617 protéines phagosomales par spectrométrie de masse à partir desquelles nous avons accru cette liste en construisant des réseaux d’interactions protéine-protéine. La contribution de chaque protéine à l’internalisation de bactéries fut ensuite testée et validée par ARN interférent à haut débit et nous a amené à identifier un nouveau régulateur de la phagocytose, le complexe de l’exocyst. En appliquant ce modèle combinatoire de biologie systémique, nous démontrons la puissance et l’efficacité de cette approche dans l’étude de processus cellulaire complexe tout en créant un cadre à partir duquel il est possible d’approfondir nos connaissances sur les différents mécanismes de la phagocytose. Lors du 2e article de cette thèse (Boulais et al, 2010), nous avons entrepris la caractérisation moléculaire des étapes évolutives ayant contribué au remodelage des propriétés fonctionnelles de la phagocytose au cours de l’évolution. Pour ce faire, nous avons isolé des phagosomes à partir de trois organismes distants (l’amibe Dictyostelium discoideum, la mouche à fruit Drosophila melanogaster et la souris Mus musculus) qui utilisent la phagocytose à des fins différentes. En appliquant une approche protéomique à grande échelle pour identifier et comparer le protéome et phosphoprotéome des phagosomes de ces trois espèces, nous avons identifié un cœur protéique commun à partir duquel les fonctions immunitaires du phagosome se seraient développées. Au cours de ce développement fonctionnel, nos données indiquent que le protéome du phagosome fut largement remodelé lors de deux périodes de duplication de gènes coïncidant avec l’émergence de l’immunité innée et acquise. De plus, notre étude a aussi caractérisée en détail l’acquisition de nouvelles protéines ainsi que le remodelage significatif du phosphoprotéome du phagosome au niveau des constituants du cœur protéique ancien de cette organelle. Nous présentons donc la première étude approfondie des changements qui ont engendré la transformation d’un compartiment phagotrophe à une organelle entièrement apte pour la présentation antigénique. / Phagocytosis is a cellular process by which large particulate material are internalized in a newly formed vesicule, the phagosome. Once formed, the phagosome acquires its functional properties through a complex maturation process called phagolysosome biogenesis. This pathway involves a series of rapid interactions with organelles of the endocytic apparatus, enabling the gradual transformation of newly formed phagosomes into phagolysosomes in which proteolytic degradation occurs. The amoeba Dictyostelium discoideum uses phagocytosis as a predation mechanism for feeding, whereas multicellular organisms utilize this process as an immune mechanism where specialized cells named phagocytes internalize, kill and degrade phatogens found through the host, forming the basis of innate immunity. In jawed verterbrates however, the phagosome links innate and adaptive immunity by processing and presenting antigenic peptides. Despite its crucial role in immunity, little is known about the composition and the functional organization of the phagosome. It is therefore essential to characterize in details the functional properties that are recruited to the phagosome. High-throughput proteomics analysis of isolated phagosomes has been tremendously helpful for the molecular comprehension of this organelle. Studies of our lab notably have revealed the first proteomics identification of mouse phagosomes without determining the roles of these proteins through the complex process of phagocytosis (Brunet et al, 2003; Garin et al, 2001). In the first study of this thesis (Stuart et al, 2007), we characterized the functions of the entire drosophila phagosome proteome by combining high-throughput proteomics, interactive networks and RNAi. By applying this strategy, we’ve identified 617 phagosomal proteins by mass spectrometry from which we’ve expanded this list by building the phagosome interactome. The contribution of each protein to bacterial internalization was tested and validated by RNAi and led to the identification of a new regulator of phagocytosis, the exocyst complex. In generating this 'systems-based model', we show the power of applying this approach to the study of complex cellular processes and organelles and expect that this detailed model of the phagosome will provide a new framework for studying host-pathogen interactions and innate immunity. In the second study of this thesis (Boulais et al, 2010), we characterized some of the key steps that contributed to the remodeling of phagosomes functional properties during evolution. To do so, we isolated this organelle from three distant organisms: the amoeba Dictyostelium discoideum, the fruit fly Drosophila melanogaster, and mouse (Mus musculus) that use phagocytosis for different purposes. By performing and comparing proteomics and phosphoproteomics analyses of isolated phagosomes from the three species, we identified an ancient core of phagosomal proteins around which the immune function of this organelle have likely organized. Our data indicate that a larger proportion of the phagosome proteome, has been acquired through gene duplication at periods coinciding with the emergence of innate and adaptive immunity. Our study also characterizes in detail the acquisition of novel proteins and the significant remodeling of the phagosome phosphoproteome that contributed to modify the core constituents of this organelle in evolution. Our work thus provides the first thorough analysis of the changes that enabled the transformation of the phagosome from a phagotrophic compartment into an organelle fully competent for antigen presentation.

Advances in genomics allow researchers to measure the complete set of transcripts in cells. These transcripts include messenger RNAs (which encode for proteins) and microRNAs, short RNAs that play an important regulatory role in cellular networks. While this data is a great resource for reconstructing the activity of networks in cells, it also presents several computational challenges. These challenges include the data collection stage which often results in incomplete and noisy measurement, developing methods to integrate several experiments within and across species, and designing methods that can use this data to map the interactions and networks that are activated in specific conditions. Novel and efficient algorithms are required to successfully address these challenges. In this thesis, we present probabilistic models to address the set of challenges associated with expression data. First, we present a novel probabilistic error correction method for RNA-Seq reads. RNA-Seq generates large and comprehensive datasets that have revolutionized our ability to accurately recover the set of transcripts in cells. However, sequencing reads inevitably contain errors, which affect all downstream analyses. To address these problems, we develop an efficient hidden Markov modelbased error correction method for RNA-Seq data . Second, for the analysis of expression data across species, we develop clustering and distance function learning methods for querying large expression databases. The methods use a Dirichlet Process Mixture Model with latent matchings and infer soft assignments between genes in two species to allow comparison and clustering across species. Third, we introduce new probabilistic models to integrate expression and interaction data in order to predict targets and networks regulated by microRNAs. Combined, the methods developed in this thesis provide a solution to the pipeline of expression analysis used by experimentalists when performing expression experiments.

Methicillin-resistant Staphylococcus aureus (MRSA) is an important human pathogen that is endemic in hospitals all over the world. It has more recently emerged as a serious threat to the general public in the form of community-acquired MRSA. MRSA has been implicated in a wide variety of diseases, ranging from skin infections and food poisoning to more severe and potentially fatal conditions, including; endocarditis, septicaemia and necrotising pneumonia. Treatment of MRSA disease is complicated and can be unsuccessful due to the bacterium's remarkable ability to develop antibiotic resistance. The considerable economic and public health burden imposed by MRSA has fuelled attempts by researchers to understand the evolution of virulent and antibiotic resistant strains and thereby improve epidemiological management strategies. Central to MRSA transmission management strategies is the implementation of active surveillance programs, via which unique genetic fingerprints, or genotypes, of each strain can be identified. Despite numerous advances in MRSA genotyping methodology, there remains a need for a rapid, reproducible, cost-effective method that is capable of producing a high level of genotype discrimination, whilst being suitable for high throughput use. Consequently, the fundamental aim of this thesis was to develop a novel MRSA genotyping strategy incorporating these benefits. This thesis explored the possibility that the development of more efficient genotyping strategies could be achieved through careful identification, and then simple interrogation, of multiple, unlinked DNA loci that exhibit progressively increasing mutation rates. The baseline component of the MRSA genotyping strategy described in this thesis is the allele-specific real-time PCR interrogation of slowly evolving core single nucleotide polymorphisms (SNPs). The genotyping SNP set was identified previously from the Multi-locus sequence typing (MLST) sequence database using an in-house software package named Minimum SNPs. As discussed in Chapter Three, the genotyping utility of the SNP set was validated on 107 diverse Australian MRSA isolates, which were largely clustered into groups of related strains as defined by MLST. To increase the resolution of the SNP genotyping method, a selection of binary virulence genes and antimicrobial resistance plasmids were tested that were successful at sub typing the SNP groups. A comprehensive MRSA genotyping strategy requires characterisation of the clonal background as well as interrogation of the hypervariable Staphylococcal Cassette Chromosome mec (SCCmec) that carries the β-lactam resistance gene, mecA. SCCmec genotyping defines the MRSA lineages; however, current SCCmec genotyping methods have struggled to handle the increasing number of SCCmec elements resulting from a recent explosion of comparative genomic analyses. Chapter Four of this thesis collates the known SCCmec binary marker diversity and demonstrates the ability of Minimum SNPs to identify systematically a minimal set of binary markers capable of generating maximum genotyping resolution. A number of binary targets were identified that indeed permit high resolution genotyping of the SCCmec element. Furthermore, the SCCmec genotyping targets are amenable for combinatorial use with the MLST genotyping SNPs and therefore are suitable as the second component of the MRSA genotyping strategy. To increase genotyping resolution of the slowly evolving MLST SNPs and the SCCmec binary markers, the analysis of a hypervariable repeat region was required. Sequence analysis of the Staphylococcal protein A (spa) repeat region has been conducted frequently with great success. Chapter Five describes the characterisation of the tandem repeats in the spa gene using real-time PCR and high resolution melting (HRM) analysis. Since the melting rate and precise point of dissociation of double stranded DNA is dependent on the size and sequence of the PCR amplicon, the HRM method was used successfully to identify 20 of 22 spa sequence types, without the need for DNA sequencing. The accumulation of comparative genomic information has allowed the systematic identification of key MRSA genomic polymorphisms to genotype MRSA efficiently. If implemented in its entirety, the strategy described in this thesis would produce efficient and deep-rooted genotypes. For example, an unknown MRSA isolate would be positioned within the MLST defined population structure, categorised based on its SCCmec lineage, then subtyped based on the polymorphic spa repeat region. Overall, by combining the genotyping methods described here, an integrated and novel MRSA genotyping strategy results that is efficacious for both long and short term investigations. Furthermore, an additional benefit is that each component can be performed easily and cost-effectively on a standard real-time PCR platform.

Les bactéries pectinolytiques des genres Pectobacterium (ancien nom Erwinia carotovora) et Dickeya (ancien nom Erwinia chrysanthemi) sont les agents des maladies de la jambe noire et de la pourriture molle. Ils provoquent des dommages aux cultures et des pertes économiques élevées. Les pertes causées par les bactéries pectinolytiques sont évaluées à environ 2 à 10% du rendement de pommes de terre, en fonction de l'année. En 2009, les pertes en pommes de terre en Europe ont été estimées à 250 millions d'euros. Au cours des dernières années, des souches de Dickeya ont été de plus en plus souvent isolées de plantes malades en Pologne, en France et d'autres pays européens. Le genre Dickeya est un groupe très diversifié, qui, selon la nomenclature actuelle contient sept espèces: D. aquatica, D. chrysanthemi, D. dadantii, D. dianthicola, D. paradisiaca, D. solani et D. zeae. Les résultats récents, obtenus dans différents pays européens, indiquent qu'un nouveau groupe de souches de Dickeya peut infecter efficacement les plantes de pomme de terre et causer des symptômes de la maladie en climat tempéré. Les souches de D. solani sont considérés comme plus agressives que les autres bactéries causant la jambe noire. Une analyse préliminaire a suggéré qu’elles ont besoin de plus faibles températures optimales pour le développement de la maladie ainsi que de niveaux d'inoculum inférieurs pour la propagation de l'infection. Elles semblent avoir une plus forte capacité à coloniser les racines de plantes de pomme de terre et à se propager à travers le système vasculaire de la plante. Les souches de D. solani produisent une large gamme d’enzymes dégradant de la paroi cellulaire végétale, qui sont les principaux facteurs de virulence. Les objectifs de l'étude étaient les suivants: 1) la caractérisation phénotypique et génotypique des souches de D. solani isolées dans des pays ayant des conditions climatiques différentes: Pologne, Finlande et Israël, 2) l'étude de l’influence d'extraits de pomme de terre sur l'expression de quelques gènes sélectionnés de D. solani: pelD, pelL, tssk, lfaA, 3) la génomique comparative de dix souches de D. solani, basée sur 4 génomes séquencés pour cette étude et 6 séquences génomiques disponibles dans la base de données GenBank. En conclusion, toutes les études génomiques ont montré que les souches de D. solani forment un groupe très homogène. Cependant, leur analyse phénotypique révèle une certaine variabilité entre les souches provenant de différentes conditions climatiques. La raison des variations observées dans les traits phénotypiques peut être liée à la régulation de l'expression des gènes codant les facteurs de virulence qui peuvent être influencés par la température, le pH, la carence en fer ou en oxygène et la disponibilité en azote, ainsi que par la présence de composés spécifiques des tissus végétaux. / Dickeya solani is a species consisting of newly emerged plant pathogenic bacteria that cause blackleg and soft rot diseases. They are responsible for great damages to potato plantations in most of European countries. D. solani strains produce a wide range of plant cell-wall degrading enzymes which are the main virulence factors. The aims of the study were: 1) phenotypic and genotypic characterizations of the D. solani strains isolated in countries with different climatic conditions: Poland, Finland and Israel, 2) study of the potato tuber extract influence on the expression of a few selected D. solani genes : pelD, pelL, tssK, lfaA,3) comparative genomics of ten D. solani strains, performed on 4 genomes sequenced for this study and 6 genome sequences available in the GenBank databases. The results showed that the strains from different climatic conditions have identical profiles in rep-PCR (with three different primers) and in Restriction Fragments Lenght Polymorphism-Pulse Field Gel Electrophoresis. However, they do differ phenotypically, especially in the activity of plant cell-wall degrading enzymes. Polish strains have higher activities of pectinolytic, cellulolytic and proteolytic enzymes than Finnish and Israeli strains. D. solani mutants in the pelD, pelL, tssK, lfaA genes were constructed by site-specific mutagenesis. The highest induction by plant extracts was observed for the lfaA gene. The expression of pelL is also induced by plant derived signal(s), but not that of pelD and tssK. Comparative genomics helped to elucidate the D. solani pangenome. The 10 D. solani strains genomes are coding for a total of 41 947 proteins which were grouped into 5 045 Orthologous Groups, 3 809 belonging to the core genome, 413 to the accessory genome and 823 to the unique genome. Some pathogenicity-related genes as well as their regulators were selected on the basis of the knowledge available for D. dadantii 3937, the most studied Dickeya strain, which belongs to a closely related species. Analysis of their protein sequence showed no difference in the sequence of those genes within the 10 genomes. All the genetic studies proved that D. solani strains form a very homogenous group. On the other hand, the phenotypic analysis showed some variability among strains from different climatic conditions. The observed variations in the phenotypic traits can results from a different regulation of the expression of the genes encoding virulence factors which are influenced by temperature, pH, iron deprivation, oxygen and nitrogen availability, as well as by the presence of plant compounds.