A tuberculose é uma doença infectocontagiosa causada por bactérias do complexo Mycobacterium tuberculosis (MTBC) que afeta humanos e/ou animais. Membros desse complexo evoluíram clonalmente e possuem grande similaridade genômica, diferenciando-se por polimorfismos de nucleotídeo único (SNPs) e regiões de diferença (RDs). Dentre os patógenos da tuberculose em animais, Mycobacterium bovis, causador da tuberculose bovina, é o membro do MTBC de maior importância global. Desta maneira, o presente estudo tem por objetivo o sequenciamento, a anotação e a análise da estirpe brasileira SP38 de M. bovis, seguido da genômica comparativa desse com outros genomas de M. bovis depositados no GenBank. Mycobacterium bovis SP38 apresenta um genoma tradicional de micobactéria tuberculosa, sendo esse único, circular com 4.347.646 pb, alto conteúdo de GC (65,6%) e 4.216 genes, incluindo 154 pseudogenes, 3 genes de rRNA (RNA ribossomal), 45 de tRNA (RNA transportador), 2 de ncRNA (RNA não codificante), 1 tmRNA (RNA transferência-mensageiro) e 4.011 sequências de DNA codificante (CDSs) (NZ_CP015773.1). Dentre as CDSs, a maioria (2.805 - 69,93%) foi anotado com função e 1.206 (30,07%) como hipotéticos. Para a genômica comparativa, os 31 genomas completos ou em drafts de M. bovis depositados no GenBank, 32 genomas de Mycobacterium bovis BCG e 23 genomas de Mycobacterium tuberculosis foram selecionados. Análises in silico dos padrões de RD resultaram na exclusão de três genomas anotados equivocadamente como M. bovis virulentos. A análise de genes ortólogos sugere que M. bovis está sob processo de decay genômico. A quantificação de sítios polimórficos indica uma maior variabilidade genética em números totais (8.335 em M. tuberculosis, 3.448 em M. bovis virulentos, e 1.088 em M. bovis BCGs) e comparações par-a-par (p ≤0,05) de M. tuberculosis em relação a M. bovis virulentos e BCGs, indicando uma maior pressão evolutiva sob M. tuberculosis, contrastando com o fato de que M. bovis é capaz de infectar um maior número de espécies hospedeiras que M. tuberculosis. A maioria desses sítios polimórficos estão localizados em CDSs hipotéticos (31,7% - 51,3%), sendo associados a família gênica PE/PPE, e apresentam uma proporção de mutações não sinônimas crescentes pela ordem M. bovis BCG, M. bovis virulentos e M. tuberculosis (48,90%, 51,92% e 59,52%, respectivamente). Essa menor proporção de mutações não sinônimas e a categorização funcional dissimilar entre CDSs contendo sítios polimórficos, indica que M. bovis BCG está sujeito a diferentes pressões seletivas quando comparado a M. bovis virulentos e M. tuberculosis. Por fim, a análise filogenética baseada em sítios polimórficos indica agrupamentos filogenéticos de M. bovis suportados pela classificação dos Complexos Clonais (CCs) e não por hospedeiros de origem dos isolados, confirmando que sítios polimórficos podem ser utilizados para classificação filogenética de linhagens genéticas desta espécie bacteriana. Além do mais, 2/28 (7,14%) genomas de M. bovis não puderam ser classificados nos CCs atualmente descritos, sugerindo a existência de complexos ainda não determinados. Este estudo representa o primeiro genoma de uma estirpe nacional de M. bovis a ser completamente sequenciado e a primeira análise de genômica comparativa de genomas desta espécie bacteriana. / Tuberculosis is an infectious disease caused by bacteria of the Mycobacterium tuberculosisComplex (MTBC) that affects human beings and/or animals. Members of this complex clonally evolved and have high genomic similarity, differentiated by single nucleotide polymorphisms (SNPs) and regions of difference (RDs). Among the animal tuberculosis pathogens, Mycobacterium bovis, the causative agent of bovine tuberculosis, is the MTBC member of greatest global importance. Therefore, the aim of the present study is to sequence, assemble and annotate the genome of the Brazilian strain SP38 of M. bovis, followed by the comparative genomics with other M. bovis genomes available in GenBank. Mycobacterium bovis SP38 has a traditional mycobacteria genome. It has a single and circular chromosome with 4,347,646 bp, high GC content (65.6%), and 4,216 genes, including 154 pseudogenes, 3 rRNA genes (ribosomal RNA), 45 tRNA (transfer RNA), 2 ncRNA (non-coding RNA), 1 tmRNA (transfer-messenger RNA), and 4,011 coding DNA sequences (CDSs) (NZ_CP015773.1). The majority of CDSs (2,805 - 69,93%) was annotated with function and 1,206 (30,07%) are hypothetical. For the comparative genomics analyses, the 31 genomes (complete and drafts) of M. bovis available in GenBank, 32 Mycobacterium bovis BCG and, 23 of Mycobacterium tuberculosis were chosen. In silico analysis of the RDs patterns resulted in the exclusion of three genomes, mistakenly annotated as virulent M. bovis. Orthologous gene analysis suggests that strains of M. bovis are under genomic decay. The quantification of polymorphic sites indicates the greater variability in absolute numbers (8,335 in M. tuberculosis, 3,448 in virulent M. bovis, and 1,088 in M. bovis BCG) and in pairwise comparisons (p≤0,05) of M. tuberculosis compared to virulent M. bovis and M. bovis BCG, suggesting that M. tuberculosis is under high evolutionary pressure. This is in contrast to the fact that M. bovis is capable of infecting a higher number of host species than M. tuberculosis. Most of these polymorphic sites are located in hypothetical CDSs (31.7% - 52.3%), being associated with PE/PPE family, and demonstrating a nonsynonymous mutations proportion of the following increasing order: M. bovis BCG, virulent M. bovis and M. tuberculosis (48.90%, 51.92% and 59.52%, respectively). This lower proportion of nonsynonymous mutations and the dissimilar functional categorization of CDSs with polymorphic sites indicates that M. bovis BCG is subjected to different selective pressure when compared to virulent M. bovis and M. tuberculosis. Finally, the phylogenetic analysis based on polymorphic sites indicates that the phylogenetic grouping of M. bovis is supported by Clonal Complexes (CCs), and not by the host of M. bovis isolates, confirming that polymorphic sites can be used for phylogenetic classification of genetic lineages of this bacterial species. Furthermore, 2/28 (7.14%) genomes of M. bovis could not be classified in the currently described CCs, suggesting the existence of complexes yet to be determined. This study represents the first genome of a Brazilian strain of M. bovis to be completely sequenced and the first comparative genomic analysis of the genomes of this bacterial species.

O cancro cítrico é uma doença que afeta diversas espécies de Citrus, cujo agente causal é Xanthomonas axonopodis pv citri (XAC). O genoma desta fitobactéria consiste de um cromossomo de ~5 Mpb e dois plasmídeos, que juntos codificam 4313 CDS (seqüências codificadoras), das quais 2710 apresentam similaridade com proteínas conhecidas. Neste trabalho realizamos uma análise comparativa detalhada do genoma de XAC com genomas de três fitopatógenos, Xanthomonas campestris campestris, Xylella fastidiosa 9a5c e Xylella fastidiosa temecula. Com esta análise identificamos genes espécie e gênero-específicos, potencialmente relevantes para adaptação aos seus respectivos nichos ou hospedeiros, além de ilhas de inserção e deleção genômica putativas. Também identificamos vias metabólicas relacionadas com osmoproteção/osmorregulação e com degradação de compostos aromáticos em XAC, que possivelmente são determinantes na eficácia de sua interação com o hospedeiro. Analisamos o nível de expressão de 9 CDS após crescimento de XAC em diferentes concentrações de glicose e verificamos que este açúcar modula positivamente a expressão de CDS relacionadas à síntese de goma e ao sistema de osmoproteção. Além disso, descrevemos a construção de microarranjos de DNA representando 2760 CDS de XAC, constituindo-se uma nova ferramenta para estudos de genômica comparativa e expressão gênica deste fitopatógeno. / Xanthomonas axonopodis pv citri (XAC) is the bacterial pathogen that causes citrus canker disease in several species of Citrus plants. XAC genome consists of a main cromosome of ~5 Mpb and two plasmids that together encode 4313 CDS (coding sequences). Approximately 63% of the CDS have assigned biological functions. In this work, we present a detailed genomic comparison between the genomes of XAC and of three other phytopathogens, X. campestris campestris, Xylella fastidiosa 9a5c and X. fastidiosa Temecula. Based on this analysis, we identified species and genus-specific genes that might be relevant for adaptation to their niches and hosts. We mapped putative insertion/deletion regions in the XAC genome possibly related to gene gains and losses during the divergence of the four bacterial lineages. We have identified the metabolic pathways related to osmoprotection/osmoregulation and aromatic compound degradation important for XAC efficient host colonization and interaction. Expression levels of 9 CDS were analyzed after XAC growth under different glucose concentrations revealing that this sugar upregulates the expression of CDS related to gum synthesis and to osmoregulation. In addition, we describe here the construction of a DNA microarray representing 2760 CDS of XAC as a new tool for comparative genomic and gene expression studies in this phytopathogen.

A tuberculose é uma doença infectocontagiosa causada por bactérias do complexo Mycobacterium tuberculosis (MTBC) que afeta humanos e/ou animais. Membros desse complexo evoluíram clonalmente e possuem grande similaridade genômica, diferenciando-se por polimorfismos de nucleotídeo único (SNPs) e regiões de diferença (RDs). Dentre os patógenos da tuberculose em animais, Mycobacterium bovis, causador da tuberculose bovina, é o membro do MTBC de maior importância global. Desta maneira, o presente estudo tem por objetivo o sequenciamento, a anotação e a análise da estirpe brasileira SP38 de M. bovis, seguido da genômica comparativa desse com outros genomas de M. bovis depositados no GenBank. Mycobacterium bovis SP38 apresenta um genoma tradicional de micobactéria tuberculosa, sendo esse único, circular com 4.347.646 pb, alto conteúdo de GC (65,6%) e 4.216 genes, incluindo 154 pseudogenes, 3 genes de rRNA (RNA ribossomal), 45 de tRNA (RNA transportador), 2 de ncRNA (RNA não codificante), 1 tmRNA (RNA transferência-mensageiro) e 4.011 sequências de DNA codificante (CDSs) (NZ_CP015773.1). Dentre as CDSs, a maioria (2.805 - 69,93%) foi anotado com função e 1.206 (30,07%) como hipotéticos. Para a genômica comparativa, os 31 genomas completos ou em drafts de M. bovis depositados no GenBank, 32 genomas de Mycobacterium bovis BCG e 23 genomas de Mycobacterium tuberculosis foram selecionados. Análises in silico dos padrões de RD resultaram na exclusão de três genomas anotados equivocadamente como M. bovis virulentos. A análise de genes ortólogos sugere que M. bovis está sob processo de decay genômico. A quantificação de sítios polimórficos indica uma maior variabilidade genética em números totais (8.335 em M. tuberculosis, 3.448 em M. bovis virulentos, e 1.088 em M. bovis BCGs) e comparações par-a-par (p ≤0,05) de M. tuberculosis em relação a M. bovis virulentos e BCGs, indicando uma maior pressão evolutiva sob M. tuberculosis, contrastando com o fato de que M. bovis é capaz de infectar um maior número de espécies hospedeiras que M. tuberculosis. A maioria desses sítios polimórficos estão localizados em CDSs hipotéticos (31,7% - 51,3%), sendo associados a família gênica PE/PPE, e apresentam uma proporção de mutações não sinônimas crescentes pela ordem M. bovis BCG, M. bovis virulentos e M. tuberculosis (48,90%, 51,92% e 59,52%, respectivamente). Essa menor proporção de mutações não sinônimas e a categorização funcional dissimilar entre CDSs contendo sítios polimórficos, indica que M. bovis BCG está sujeito a diferentes pressões seletivas quando comparado a M. bovis virulentos e M. tuberculosis. Por fim, a análise filogenética baseada em sítios polimórficos indica agrupamentos filogenéticos de M. bovis suportados pela classificação dos Complexos Clonais (CCs) e não por hospedeiros de origem dos isolados, confirmando que sítios polimórficos podem ser utilizados para classificação filogenética de linhagens genéticas desta espécie bacteriana. Além do mais, 2/28 (7,14%) genomas de M. bovis não puderam ser classificados nos CCs atualmente descritos, sugerindo a existência de complexos ainda não determinados. Este estudo representa o primeiro genoma de uma estirpe nacional de M. bovis a ser completamente sequenciado e a primeira análise de genômica comparativa de genomas desta espécie bacteriana. / Tuberculosis is an infectious disease caused by bacteria of the Mycobacterium tuberculosisComplex (MTBC) that affects human beings and/or animals. Members of this complex clonally evolved and have high genomic similarity, differentiated by single nucleotide polymorphisms (SNPs) and regions of difference (RDs). Among the animal tuberculosis pathogens, Mycobacterium bovis, the causative agent of bovine tuberculosis, is the MTBC member of greatest global importance. Therefore, the aim of the present study is to sequence, assemble and annotate the genome of the Brazilian strain SP38 of M. bovis, followed by the comparative genomics with other M. bovis genomes available in GenBank. Mycobacterium bovis SP38 has a traditional mycobacteria genome. It has a single and circular chromosome with 4,347,646 bp, high GC content (65.6%), and 4,216 genes, including 154 pseudogenes, 3 rRNA genes (ribosomal RNA), 45 tRNA (transfer RNA), 2 ncRNA (non-coding RNA), 1 tmRNA (transfer-messenger RNA), and 4,011 coding DNA sequences (CDSs) (NZ_CP015773.1). The majority of CDSs (2,805 - 69,93%) was annotated with function and 1,206 (30,07%) are hypothetical. For the comparative genomics analyses, the 31 genomes (complete and drafts) of M. bovis available in GenBank, 32 Mycobacterium bovis BCG and, 23 of Mycobacterium tuberculosis were chosen. In silico analysis of the RDs patterns resulted in the exclusion of three genomes, mistakenly annotated as virulent M. bovis. Orthologous gene analysis suggests that strains of M. bovis are under genomic decay. The quantification of polymorphic sites indicates the greater variability in absolute numbers (8,335 in M. tuberculosis, 3,448 in virulent M. bovis, and 1,088 in M. bovis BCG) and in pairwise comparisons (p≤0,05) of M. tuberculosis compared to virulent M. bovis and M. bovis BCG, suggesting that M. tuberculosis is under high evolutionary pressure. This is in contrast to the fact that M. bovis is capable of infecting a higher number of host species than M. tuberculosis. Most of these polymorphic sites are located in hypothetical CDSs (31.7% - 52.3%), being associated with PE/PPE family, and demonstrating a nonsynonymous mutations proportion of the following increasing order: M. bovis BCG, virulent M. bovis and M. tuberculosis (48.90%, 51.92% and 59.52%, respectively). This lower proportion of nonsynonymous mutations and the dissimilar functional categorization of CDSs with polymorphic sites indicates that M. bovis BCG is subjected to different selective pressure when compared to virulent M. bovis and M. tuberculosis. Finally, the phylogenetic analysis based on polymorphic sites indicates that the phylogenetic grouping of M. bovis is supported by Clonal Complexes (CCs), and not by the host of M. bovis isolates, confirming that polymorphic sites can be used for phylogenetic classification of genetic lineages of this bacterial species. Furthermore, 2/28 (7.14%) genomes of M. bovis could not be classified in the currently described CCs, suggesting the existence of complexes yet to be determined. This study represents the first genome of a Brazilian strain of M. bovis to be completely sequenced and the first comparative genomic analysis of the genomes of this bacterial species.

<p>Lymphomas are a heterogeneous group of neoplasias originating from B- or T-lymphocytes. In this thesis, we determined the genetic and immunophenotypic characterization of DLBCL and their prognostic impact. Moreover, genomic alterations associated with the transformation to DLBCL from Hodgkin lymphoma (HL) and follicular lymphoma (FL) were elucidated. </p><p>In order to outline the impact of cytogenetic as well as immunophenotypic prognostic markers in DLBCL, we firstly studied a series of 54 DLBCL tumors using comparative genomic hybridization (CGH) and we identified several frequently occurring chromosomal imbalances. Loss of 22q was more often found in the diagnostic tumors with a more advanced clinical stage, while gain of 18q21 was more commonly identified in relapses. Secondly, we correlated the expression patterns of CD10, bcl-6, IRF-4 and bcl-2 with clinical parameters in a series of 173 de novo DLBCL patients. Patients with a germinal center (GC) phenotype displayed a better survival than the non-GC group. Expression of bcl-6 and CD10 was correlated with a better survival while bcl-2 expression was associated with a poor prognosis.</p><p>In approaching the HL transformation, two novel B-cell lines (U-2932 and U-2940), derived from patients with DLBCL following HL, were characterized. Interestingly, a translocation with materials from 2q and 7q as well as loss of material on 6q was found in both cell lines. For FL transformation, we assessed chromosomal alterations in a panel of 28 DLBCL patients with a previous history of FL. The DLBCL tumors displayed more chromosomal imbalances compared to FL tumors. Loss of 6q16-21 and gain of 7pter-q22 were more commonly found in the DLBCL counterparts, suggesting the chromosomal location of putative genes that may be involved in the transformation process.</p>

Lymphomas are a heterogeneous group of neoplasias originating from B- or T-lymphocytes. In this thesis, we determined the genetic and immunophenotypic characterization of DLBCL and their prognostic impact. Moreover, genomic alterations associated with the transformation to DLBCL from Hodgkin lymphoma (HL) and follicular lymphoma (FL) were elucidated. In order to outline the impact of cytogenetic as well as immunophenotypic prognostic markers in DLBCL, we firstly studied a series of 54 DLBCL tumors using comparative genomic hybridization (CGH) and we identified several frequently occurring chromosomal imbalances. Loss of 22q was more often found in the diagnostic tumors with a more advanced clinical stage, while gain of 18q21 was more commonly identified in relapses. Secondly, we correlated the expression patterns of CD10, bcl-6, IRF-4 and bcl-2 with clinical parameters in a series of 173 de novo DLBCL patients. Patients with a germinal center (GC) phenotype displayed a better survival than the non-GC group. Expression of bcl-6 and CD10 was correlated with a better survival while bcl-2 expression was associated with a poor prognosis. In approaching the HL transformation, two novel B-cell lines (U-2932 and U-2940), derived from patients with DLBCL following HL, were characterized. Interestingly, a translocation with materials from 2q and 7q as well as loss of material on 6q was found in both cell lines. For FL transformation, we assessed chromosomal alterations in a panel of 28 DLBCL patients with a previous history of FL. The DLBCL tumors displayed more chromosomal imbalances compared to FL tumors. Loss of 6q16-21 and gain of 7pter-q22 were more commonly found in the DLBCL counterparts, suggesting the chromosomal location of putative genes that may be involved in the transformation process.

O cancro cítrico é uma doença que afeta diversas espécies de Citrus, cujo agente causal é Xanthomonas axonopodis pv citri (XAC). O genoma desta fitobactéria consiste de um cromossomo de ~5 Mpb e dois plasmídeos, que juntos codificam 4313 CDS (seqüências codificadoras), das quais 2710 apresentam similaridade com proteínas conhecidas. Neste trabalho realizamos uma análise comparativa detalhada do genoma de XAC com genomas de três fitopatógenos, Xanthomonas campestris campestris, Xylella fastidiosa 9a5c e Xylella fastidiosa temecula. Com esta análise identificamos genes espécie e gênero-específicos, potencialmente relevantes para adaptação aos seus respectivos nichos ou hospedeiros, além de ilhas de inserção e deleção genômica putativas. Também identificamos vias metabólicas relacionadas com osmoproteção/osmorregulação e com degradação de compostos aromáticos em XAC, que possivelmente são determinantes na eficácia de sua interação com o hospedeiro. Analisamos o nível de expressão de 9 CDS após crescimento de XAC em diferentes concentrações de glicose e verificamos que este açúcar modula positivamente a expressão de CDS relacionadas à síntese de goma e ao sistema de osmoproteção. Além disso, descrevemos a construção de microarranjos de DNA representando 2760 CDS de XAC, constituindo-se uma nova ferramenta para estudos de genômica comparativa e expressão gênica deste fitopatógeno. / Xanthomonas axonopodis pv citri (XAC) is the bacterial pathogen that causes citrus canker disease in several species of Citrus plants. XAC genome consists of a main cromosome of ~5 Mpb and two plasmids that together encode 4313 CDS (coding sequences). Approximately 63% of the CDS have assigned biological functions. In this work, we present a detailed genomic comparison between the genomes of XAC and of three other phytopathogens, X. campestris campestris, Xylella fastidiosa 9a5c and X. fastidiosa Temecula. Based on this analysis, we identified species and genus-specific genes that might be relevant for adaptation to their niches and hosts. We mapped putative insertion/deletion regions in the XAC genome possibly related to gene gains and losses during the divergence of the four bacterial lineages. We have identified the metabolic pathways related to osmoprotection/osmoregulation and aromatic compound degradation important for XAC efficient host colonization and interaction. Expression levels of 9 CDS were analyzed after XAC growth under different glucose concentrations revealing that this sugar upregulates the expression of CDS related to gum synthesis and to osmoregulation. In addition, we describe here the construction of a DNA microarray representing 2760 CDS of XAC as a new tool for comparative genomic and gene expression studies in this phytopathogen.

Parmi les arthropodes, les tiques sont les plus importants vecteurs de pathogènes en termes de diversité et sont la première cause de transmission de maladies vectorielles en Europe et Amérique du Nord. Si ces pathogènes font l'objet de nombreux travaux, les tiques hébergent aussi d'autres symbiotes qui contribuent de manière importante à leur phénotype. Au cours des dernières années, de nombreuses bactéries symbiotiques ont ainsi été recensées chez les tiques. Parmi celles-ci des bactéries présentant une forte homologie avec Coxiella burnetii ont été découvertes. Contrairement à C. burnetii, l'agent responsable de la fièvre Q, les Coxiella-like endosymbiotiques (Coxiella-LE) ne semblent pas capables d'infecter d'autres hôtes que les tiques. Elles font partie des symbiotes à transmission maternelle les plus répandus chez les espèces de tiques et pourraient jouer un rôle important dans la biologie de ces arthropodes. Des premiers éléments suggèrent en effet, que les Coxiella-LE pourraient avoir un rôle nutritionnel en synthétisant vitamines B et cofacteurs absents de l'alimentation de leur hôte tique. Pour comprendre les relations entretenues entre les Coxiella-LE et leurs hôtes je me suis intéressé,au cours de mes travaux de thèse, a l'évolution du genre Coxiella. Pour cela des approches d'analyses phylogénétiques et de génomique comparative ont été utilisées. J'ai ainsi participe à l'établissement de la phylogénie du genre Coxiella par Multi-Locus Sequence Typing (MLST), qui a permis de mettre en évidence la diversité de ce genre bactérien. De manière intéressante C. burnetii émerge au sein d'un de ces clades de bactéries endosymbiotiques de tiques, ce qui semble témoigner d'une récente transition vers la pathogénie. Nous avons séquencé deux nouveaux génomes de Coxiella-LE afin de réaliser une étude de génomique comparative. Tous les génomes de Coxiella étudiés, y compris ceux de C. burnetii, possèdent les gènes nécessaires à la biosynthèse des vitamines B et des cofacteurs, retrouvés habituellement chez les symbiotes nutritionnels d'arthropodes hématophages. Cette découverte renforce l'idée d'un rôle important des Coxiella-LE pour leur hôte tique et, d'après la phylogénie, l'ancêtre commun de ces bactéries serait donc un endosymbiote mutualiste de tique. Pourtant des traces de gènes impliqués dans la virulence de Coxiella burnetii ont été retrouvées dans des génomes appartenant à des clades distincts de Coxiella-LE, ce qui semble plutôt indiquer des pertes récurrentes de la virulence. De plus, différents niveaux d'érosion génomique sont retrouvés dans les génomes de Coxiella-LE étudiés, ce qui indiquerait de fréquents transferts d'hôtes. De tels transferts expliqueraient l'absence de co-cladogenese entre la phylogénie des Coxiella-LE et celle de leur hôte, une caractéristique originale pour un symbiote qui semble obligatoire. Par ailleurs, plusieurs symbiotes à transmission maternelle peuvent être retrouvés chez les tiques, le deuxième axe de ma thèse s'intéresse à l'impact de ces co-infections. Pour cela une population de tiques appartenant à l'espèce Dermacentor marginatus a été étudiée. Cette espèce est, en effet, fréquemment infectée par des bactéries Coxiella-LE, Rickettsia et Spiroplasma et différents statuts d'infection peuvent être observés chez les individus. Aucune compétition n'a été démontrée entre ces bactéries, puisqu’aucune n'interfère avec la transmission et la densité des autres. Néanmoins, en cas de triple infection, la valeur adaptative des hôtes est fortement diminuée avec une réduction importante de leur taille (10%). La transmission verticale de ces symbiotes n'étant pas complète, il devient alors difficile de comprendre comment ces bactéries atteignent de si fortes prévalences au sein de la population avec des coûts associés aussi importants [etc…] / Among arthropods, ticks are the most important vectors of pathogens in terms of diversity and are the leading cause of transmission of vector-borne diseases in Europe and North America. While these pathogens are the most studied, ticks also harbor other symbionts that contribute significantly to their phenotype. Recently many symbiotic bacteria have been described in ticks. Among them, bacteria exhibiting strong homology with Coxiella burnetii have been discovered. Unlike C. burnetii, the causative agent of Q fever, Coxiella-Like Endosymbiont (Coxiella-LE) seems unable to infect other hosts than ticks. They are among the most widespread maternally-inherited symbionts in tick species and could play an important role in their biology. Coxiella-LE may indeed have a nutritional role by synthesizing B vitamins and cofactors absent from their host's diet. To understand the interaction between Coxiella-LE and their hosts, my thesis work focused on the evolution of the Coxiella genus. For this purpose, phylogenetic analyzes and comparative genomic approaches have been carried out. I have participated in the establishment of the phylogeny of the Coxiella genus by Multi-Locus Sequence Typing (MLST), which highlights the diversity of this bacterial genus. Interestingly C. burnetii emerges within one of these clads of tick endosymbiotic bacteria, which may suggest a recent transition towards pathogenicity. Two new genomes of Coxiella-LE were sequenced to perform comparative genomic analyses. All Coxiella genomes studied, including those of C. burnetii, possess the genes encoding for the biosynthesis of B vitamins and cofactors, as usually found in nutritional symbionts of blood-sucking arthropods. This result strengthens the idea of an important role of Coxiella-LE for their host ticks and, according to the phylogeny, the common ancestor of these bacteria was therefore a mutualistic tick endosymbiont. However, traces of genes involved in the virulence of C. burnetii have been found in genomes belonging to distinct clads of Coxiella-LE, which rather indicate recurrent losses of virulence. Moreover, different levels of genomic erosion are found in the genomes of Coxiella-LE studied, which could indicate different transitions towards the mutualistic way of life. Such recurrent transfers would explain the absence of cocladogenesis between Coxiella-LE and their host phylogeny, an uncommon feature for an obligatory symbiont. As several maternally-inherited symbionts can be found in ticks, the second axis of my thesis has focused on the impact of co-infections. For this purpose, a population of ticks belonging to the species Dermacentor marginatus was studied. This species is frequently infected with Coxiella-LE, Rickettsia and Spiroplasma bacteria and different infection status can be observed in individuals from a single population. No competition has been demonstrated between these bacteria, since none interferes with the transmission and density of the others. However, in case of triple infection, the fitness of the host appears greatly reduced with a significant reduction in size (10%). Since vertical transmission of these symbionts is incomplete, understanding how these symbionts and co-infections are maintained despite this significant cost remains an open question. If the symbiotic strategies of these symbionts are still unknown, it is likely that their transmission is not only maternal, but also horizontal [etc…]

Uropathogenic E. coli (UPEC) is the most frequent cause of urinary tract infection (UTI), being responsible for up to 85% of community acquired and 40% of nosocomial cases. UPEC strains harbour various virulence factors that contribute to their ability to cause disease. The high prevalence across the globe of multidrug resistant UPEC is a significant threat to therapy. Virulent and resistant UPEC strains have been recognised as belonging to major lineages and we have only recently begun to understand the factors contributing to their successful global dissemination. Work in this thesis was carried out to identify the population structure of E. coli isolates recovered from urosepsis and biliary sepsis, to reveal any differences in genetic background. A total of 100 isolates from the blood and urine of 50 patients presenting with urosepsis and 27 isolates from cases of biliary sepsis were subjected to genotypic and phenotypic analysis, including MLST, virulence gene detection and antibiogram and metabolic profiling. Urosepsis paired isolates showed identical genotypes and antimicrobial resistance profiles. However, several pairs of isolates showed discrepant metabolic activity profiles suggesting niche specific regulation of metabolism. Members of the ST131 clone were significantly associated with antibiotic resistance and ST38 isolates were associated with the highest level of metabolic activity. An in vivo infection model was used to investigate the virulence potential of isolates from the major UPEC lineages. Galleria mellonella larvae inoculated with ST69 and ST127 isolates showed significantly higher mortality rates than those infected with other strains. However, one isolate of ST127 (strain EC18) was avirulent and comparative genomic analyses with a single virulent ST127 strain revealed an IS1 mediated deletion in the O-antigen cluster in strain EC18, which is likely to explain the lack of virulence in the larvae and demonstrates the importance of this cell surface molecule in the model system. Finally, a total of 202 UPEC isolates were recovered from community and hospital urine samples from a tertiary care hospital in Riyadh, Saudi Arabia. Molecular epidemiological investigation of the strains was carried out to examine the overall UPEC population structure, for the first time in any part of Saudi Arabia. The most common lineages were ST131 (17.3%), ST73 (11.4%), ST38 (7.4%), ST69 (7.4%) and ST10 (6.4%). The findings highlight the successful spread of multidrug resistant, CTX-M positive ST38, ST131 and ST405 UPEC in Saudi Arabia. The high proportion (35%) of ESBL producing E. coli isolates is a particular concern and is driving frequent prescription of carbapenem antibiotics. A total of four isolates of ST38 were positive for aggR, which is a virulence marker of enteroaggregative E. coli (EAEC); ST38 strains that cause UTI but have an EAEC genetic background are becoming recognised as novel UPEC and this clonal group warrants further study.

No description available.

Smith-Magenis syndrome (SMS) is a multiple congenital abnormalities intellectual disability syndrome that results from a deletion of chromosome 17p11.2 or mutation of the retinoic acid inducted one gene (RAI1). SMS is characterized by a multitude of phenotypic features including craniofacial defects, short stature, obesity, intellectual disability, self-abusive behavior, sleep disturbance and behavioral abnormalities. Interestingly, although SMS is a clearly defined syndrome with a known molecular change at its foundation, ~40% of all candidate cases sent to the Elsea lab for evaluation do not have a mutation or deletion of RAI1. We hypothesize that at least one other locus must be responsible for this Smith-Magenis-like (SMS-like) phenotype. To address this hypothesis, we first compiled a cohort of 52 subjects who had been referred to the Elsea lab for a clinical diagnosis of SMS. Once these individuals were confirmed to not have an RAI1 mutation or deletion, their phenotypes were compiled and statically analyzed to distinguish whether SMS and SMS-like cohorts are different in the prevalence of the core phenotypes of SMS such as, but not limited to, sleep disturbance, self-abusive behavior and developmental delay. SMS-like and SMS cohorts are not different in prevalence for these core features. Next, all SMS-like subjects were sent for whole genome array comparative genomic hybridization (aCGH) to identify duplications or deletions of each individual’s genome which contribute to the phenotype observed. We identified 6 pathogenic copy number variants (CNVs) in six individuals which contribute directly to the clinical phenotype, including two del(2)(q37). This study enabled us to draw relationships between SMS and other syndromes that had never been appreciated before and helped to identify pathways in which RAI1 may function. Using the data from our SMS-like study we were able to further characterize two known syndromes; Deletion 2q37 syndrome (brachydactyly mental retardation syndrome) and deletion 2q23 syndrome. With regard to deletion 2q37, syndrome we used genomic data from known and new deletion 2q37 subjects to refine the critical region to one gene: the histone deacetylase 4 gene (HDAC4). Using both clinical and molecular clues, we were able to identify one subject from our SMS-like cohort who has an insertion in HDAC4 which results in a premature stop codon. We conclude from this study that mutation of HDAC4 results in brachydactyly mental retardation syndrome. With regard to deletion 2q23 syndrome there were only five known cases in the published literature to which we were able to add two more. Using as similar approach to our del2q37 study we refined the critical region for this syndrome to one gene, the methyl binding domain 5 gene (MBD5). Using a molecular and clinical approach we were able to conclude that haploinsufficiency of MBD5 results in the core phenotypes seen in del2q23 syndrome including microcephaly, intellectual disabilities, severe speech impairment, and seizures. Using all the data generated from the three previous studies we set out to characterize the molecular function of RAI1. We hypothesize that RAI1 is a transcription factor that regulates gene expression of core genes involved in development, neurological function, and circadian rhythm. Using a ChIP-chip based approach we identified 257 transcripts we believe RAI1 regulates. Following up on these transcripts, using in vitro and in vivo methods, we have been able to conclude that RAI1 is a positive regulator of CLOCK, the master regulator of the central circadian cycle. Taken together, these studies have given us insight into the specific molecular changes that contribute to SMS and SMS-like syndromes. We have unveiled pathways and genes which are important to normal human development and behavior and identified novel functions of RAI1. These studies will provide the foundation for the future discovery of the pathways affected.