Thèses sur le sujet « Human metagenomic »

Co-evolving with the human host, gut microbiota establishes configurations, which vary under the pressure of inflammation, disease, ageing, diet and lifestyle. In order to describe the multi-stability of the microbiome-host relationship, we studied specific tracts of the bacterial trajectory during the human lifespan and we characterized peculiar deviations from the hypothetical development, caused by disease, using molecular techniques, such as phylogenetic microarray and next-generation sequencing. Firstly, we characterized the enterocyte-associated microbiota in breast-fed infants and adults, describing remarkable differences between the two groups of subjects. Successively, we investigated the impact of atopy on the development of the microbiome in Italian childrens, highlithing conspicuous deviations from the child-type microbiota of the Italian controls. To explore variation in the gut microbiota depending on geographical origins, which reflect different lifestyles, we compared the phylogenetic diversity of the intestinal microbiota of the Hadza hunter-gatherers of Tanzania and Italian adults. Additionally, we characterized the aged-type microbiome, describing the changes occurred in the metabolic potential of the gut microbiota of centenarians with respect to younger individuals, as a part of the pathophysiolology of the ageing process. Finally, we evaluated the impact of a probiotics intervention on the intestinal microbiota of elderly people, showing the repair of some age-related dysbioses. These studies contribute to elucidate several aspects of the intestinal microbiome development during the human lifespan, depicting the microbiota as an extremely plastic entity, capable of being reconfigured in response to different environmental factors and/or stressors of endogenous origin.

Co-evolving with the human host, gut microbiota establishes configurations, which vary under the pressure of inflammation, disease, ageing, diet and lifestyle. In order to describe the multi-stability of the microbiome-host relationship, we studied specific tracts of the bacterial trajectory during the human lifespan and we characterized peculiar deviations from the hypothetical development, caused by disease, using molecular techniques, such as phylogenetic microarray and next-generation sequencing. Firstly, we characterized the enterocyte-associated microbiota in breast-fed infants and adults, describing remarkable differences between the two groups of subjects. Successively, we investigated the impact of atopy on the development of the microbiome in Italian childrens, highlithing conspicuous deviations from the child-type microbiota of the Italian controls. To explore variation in the gut microbiota depending on geographical origins, which reflect different lifestyles, we compared the phylogenetic diversity of the intestinal microbiota of the Hadza hunter-gatherers of Tanzania and Italian adults. Additionally, we characterized the aged-type microbiome, describing the changes occurred in the metabolic potential of the gut microbiota of centenarians with respect to younger individuals, as a part of the pathophysiolology of the ageing process. Finally, we evaluated the impact of a probiotics intervention on the intestinal microbiota of elderly people, showing the repair of some age-related dysbioses. These studies contribute to elucidate several aspects of the intestinal microbiome development during the human lifespan, depicting the microbiota as an extremely plastic entity, capable of being reconfigured in response to different environmental factors and/or stressors of endogenous origin.

Studying the evolutionary history of human microbiome members is indispensable to understanding how our microbiota complexity has been shaped over time. While recent decades have witnessed remarkable progress in investigation of the human microbiome within different evolutionary contexts, reconstructing a time-resolved divergence history for human microbiome members is still extremely challenging. Longitudinal metagenomics studies have shown the potential of addressing such challenges, but it is limited in tracking the evolutionary change in a short term. By contrast, leveraging the microbial genomic information preserved in the ancient metagenomic samples from the long past is emerging as a powerful strategy to study the long-established evolutionary history of the human microbiome. In this thesis, I aimed to devise a novel methodology that allows for efficiently reconstructing a time-resolved evolutionary history of human microbiome species using ancient metagenomic data. To this end, I firstly started from comparing four newly excavated paleofeces samples to a large body of contemporary metagenomic datasets. I observed that our human gut microbiota has diverged from its ancestral state in both microbial composition and metabolic pathways. This could be related to the change of lifestyle during human history. To better understand the divergence time of microbiome members, I secondly developed a novel computational pipeline which can precisely reconstruct and date strain-level phylogenies for microbiome species using carbon dated ancient metagenomic samples as calibration under Bayesian molecular clocking framework. The application of this tool has uncovered the unprecedented evolutionary diversity, in the context of geography and time period, of three Methanobrevibacter species from the human oral microbiome, and reconstructed a delicate time-resolved evolutionary history for common gut microbial species, such as Prevotella copri, Eubacterium rectale, Methanobrevibacter smithii and others mainly populating in the human gut microbiota. This approach promises that more underlying evolutionary history about the human microbiome will be unveiled in the foreseen future.

La métagénomique virale est une approche prometteuse pour la détection et l’identification sans a priori de potentiels nouveaux pathogènes.Cependant, son utilisation reste encore marginale en raison de l’importante contamination des viromes par les séquences nucléiques de l’hôte.L'objectif de cette thèse était d’améliorer l’approche de métagénomique pour le diagnostic clinique de maladies infectieuses virales en augmentant le ratio de séquences pathogène/hôte par déplétion des acides nucléiques humains.Le premier chapitre consiste en une synthèse bibliographique des approches de métagénomique virale en recherche clinique et des challenges à relever dans ce domaine. Elle inclut également une revue sur les approches de capture/séquençage ciblées de certains pathogènes dans le domaine des maladies infectieuses humaines.Le deuxième chapitre propose une mise au point méthodologique permettant d’enrichir les métagénomes en séquences non-humaines basée sur l’hybridation et la capture de l’ensemble des acides nucléiques de l’hôte après hybridation avec des sondes ARN humaines biotinylées.Le troisième chapitre est divisé en deux sous-chapitres qui proposent l’application de ce protocole à la détection d’agents potentiellement impliqués (1) dans un cas fatal d’encéphalite et (2) dans un cas énigmatique d’endocardite infectieuse à hémoculture négative.Dans un quatrième chapitre, l’approche méthodologique que nous avons développée est discutée et les résultats sont replacés dans un contexte élargi d’émergence des maladies infectieuses et de lien de causalité entre l’agent détecté et la pathologie observée
Viral metagenomics, which is based on the random shotgun sequencing of all viral genomes present in a sample, is a promising approach for blind detection and identification of potential new pathogens. Its use is however still marginal because of the large proportion of human nucleic sequences. In this context, this thesis work aims at improving the metagenomic approach for the clinical diagnosis of viral infectious diseases by increasing the ratio of pathogen-to-host sequences trough depletion of human nucleic acids from the samples. The first chapter of this thesis consists in a bibliographic synthesis of viral metagenomic approaches in clinical research and the challenges we faced in this field. This bibliographic overview also includes a review article on targeted-enrichment sequencing approaches for pathogen detection in the field of human infectious diseases. The second chapter proposes a methodological development allowing the enrichment of non-human sequences from metagenomes through hybridization and capture of human nucleic acids with biotinylated human RNA probes. The third chapter is divided into two sub-chapters that propose the application of this protocol to the detection of putative pathogens in (1) a fatal case of encephalitis and (2) an enigmatic case of blood-culture negative infectious endocarditis. The methodological approach developed during this work is finally discussed in a fourth chapter, which also replaces the results obtained in the broader context of emerging infectious diseases and validation of the causal link between the agent detected and the observed pathology

It is well established that mixed microbial communities contain organisms which have not been studied by conventional culture-based methods. In the human oral cavity this number is estimated at around 50%. Commensal bacteria develop and maintain an intimate relationship with human cells without triggering proinflammatory mechanisms and this study aims to explore this by searching for bacterial proteins which facilitate binding to the human tongue dorsum and wider oral cavity. Metagenomic DNA from the human tongue dorsum of 9 volunteers was extracted and a phage display library created, to our knowledge the first to incorporate metagenomic DNA. Phage display is an elegant molecular technique involving fusion of fragmented DNA to a phagemid coat protein, such that inserted DNA is encoded by the phage and displayed on the phage surface. The affinity selection technique panning, then exploited the natural affinity and specificity of the fusion proteins to identify bacterial binding proteins using, in this case, three ligands: IgA, Fibronectin and BSA. IgA is of special interest to this group as it interacts with bacterial proteins and is poised to respond to bacterial numbers in human secretions such as saliva. Proteins from panning were analysed in silico, however, the majority were discarded due to the presence of stop codons in the protein sequences. Remaining phagemid displaying fusion proteins of interest were assessed for function and binding assays carried out to confirm binding specificity. Due to the biased nature of phage display library production, a 16S rRNA gene analysis was also carried out in order to assess metagenomic DNA diversity prior to library construction. Because phage display was used successfully by colleagues with the genomes of single organisms, it was believed that including metagenomic DNA in a phage display library would cast a wide net over the tongue dorsum allowing capture of many more binding proteins occurring in this environment from a wide range of bacteria.

The gut microbiota is a complex and diverse microbial community that is adapted to a carbohydrate-rich ecosystem. Plant cell wall components (cellulose, hemicelluloses and pectins), resistant starch and various oligosaccharides reach the colon by escaping digestion in the upper gastrointestinal tract. Fermentation of these dietary carbohydrates by the gut microbiota has well-recognised beneficial effects on host health. The microbial community in the human gut requires specific enzymes to efficiently degrade these carbohydrates. In this project, a culture-independent approach based on functional screening of genomic and metagenomic libraries using Escherichia coli and Lactococcus lactis as heterologous expression hosts, was used to isolate novel genes encoding glycoside hydrolase (GH) enzymes. The study identified several active GH enzymes involved in the breakdown of dietary polysaccharides such as starch, cellulose, xylan and β-glucan, recovered from the E. coli metagenomic library. The bioinformatic analysis of the insert from positive clones showed the presence of ORFs with the similarity to enzymes from GH families 13, 43 and 51 encoded by dominant bacterial genera from the human colon (Bacteroides sp., Roseburia sp., Ruminococcus sp.). A group of clones encoding potentially novel GH enzymes was also identified, emphasising the importance of functional-based study. One highly active clone was detected during screening of the L. lactis metagenomic library and showed fibrolytic activity on cellulose-, lichenanand xylan-containing plates. The insert contained a partial gene with the GH9 catalytic domain and identity to the protein from Coprococcus eutactus ART55/1. Further functional analysis established the fibrolytic activity of selected Coprococcus species. Moreover, several active clones were isolated from the Ruminococcus sp. 80/3 genomic library which encoded protein with the similarity to enzymes from GH families 2, 3 and 5. In this work, the traditional approach of expression in E. coli was complemented by using an alternative host – L. lactis. While this did not improve the screening efficiency in terms of number of recovered clones, differences in gene expression and protein export between E. coli and L. lactis were noted during this study which highlights the benefits of using different heterologous hosts in functional metagenomic approaches.

Viruses are biological agents that infect cellular organisms. Most viruses are bacteriophages, these are the most abundant biological entities on earth. Not much is known about virus diversity in the human mouth, including dental plaque, compared to other environments. A culture-independent based approach was tried using metagenomic analysis to characterize uncultured virus gene fragments in human dental plaque. The isolated viral genomes were amplified using a multiple displacement amplification method. Eighty, eleven and ten clones were sequenced from three volunteers, respectively. TBLASTX analysis showed that 44% of the sequences had significant identities to the GenBank databases. Of these 66% were viral; 12% human; 10% bacterial; 6% mobile and 6% eukarya. These sequences were sorted into six contigs and forty five single sequences. Four contigs and one single sequence were found to have a significant identity to a small region of a putative prophage in the Corynebcterium diphtheria genome. The gaps between these were filled by primer walking and PCR to give a continuous contig of 11554 bp. Two viruses A1 and A2 and their bacterial host were isolated from the human mouth. The 16S rRNA gene sequence of the host had a 99% identity to several Neisseria sp. The A1 virus was found to appear spontaneously on soft top agar plates, and might be a lysogenic virus. The A2 virus was a lytic virus. The two viruses have different morphological shapes. A1 has a varied isometric head size that ranges from 32 to 58 nm and no tail; it may belong to the Tectiviridae family. It has a linear dsDNA genome with a size between 12 kb and 23kb. A limited amount of the genome of the A1 virus was sequenced. The A2 virus has an icosohedral head with size of 60±3 nm and a sheathed rigid tail about 175 nm long with no detectable base plate or tail fibres. It can be classified into the order Caudovirales family Siphoviridae. The size of the A2 virus genome is estimated to be 35 to 40 kb. 31703 bp of unique sequence has been determined and sorted into three contigs and 14 single sequences. Further attempts at gap filling using primer walking and PCR were unsuccessful. It has a linear dsDNA genome, with a GC content of 49 mol%. A latent period of 25 min and a burst size of 25±2 particles were determined by a single step growth curve. Bioinformatic approaches were used to identify ORFs in the genome. A2 virion associated proteins were analysed by SDS–PAGE gel electrophoresis, and some proteins sequences were directly related to the translated genomic sequence.

Antibiotic resistance is currently one of the major global healthcare problems. Bacteria can become resistant by acquiring resistance genes from other bacteria. This process is usually facilitated by mobile genetic elements (MGEs), a type of DNA that can move from one site to another site within bacterial genome, and often between bacterial cells. The human oral cavity has been shown to harbour various antimicrobial resistance genes (ARGs). The aim of this research is to study the fundamental biology and the association between MGEs and ARGs present in human oral bacteria by both sequence and functional-based metagenomic assays. Using a PCR-based method, various genes predicted to confer antimicrobial resistance and other adaptive traits were identified on different MGEs (composite transposons, integrons and novel MGEs called translocatable units). This is the first report that showed ARGs in the human oral cavity were associated with these MGEs, especially in integron gene cassettes (GCs). Some of the integron gene cassettes were predicted to not contain any genes at all. They were predicted to have a regulatory function as a promoter, which could be important for the expression of other genes carried by integrons. Using an enzymatic reporter assay, it was proven that one of the functions of these GCs is as a promoter, which could allow bacteria to survive multiple stresses within the complex environment of the human oral cavity. Functional screening of a metagenomic library identified a clone that can confer resistance to two commonly used antiseptics agents. This was shown to be a result of UDP-glucose 4-epimerase enzyme derived from a common oral bacteria Veillonella parvula, which altered the cell’s surface charge to be more positive, presumably reducing the binding of positively charges antiseptics to the bacteria. To tackle the antibiotic resistance problems effectively, the understanding of the nature of MGEs is crucial. We have shown the presence of multiple novel MGEs, ARGs and a novel resistance mechanism. Those detected ARGs can be used for the surveillance and increase the understanding of MGEs in other environments.

Metagenomics is the study of the collective genomes of the members of a microbial community. In contrast to conventional culture methods, sequence-based metagenomics exclusively relies on sequence analysis without culturing microorganisms. This approach has revolutionised our understanding of the complexity of microbiomes. As such, microbial profiling, particularly of microbiomes in humans that appear to play key roles in numerous disease phenotype, may provide information to help define associated underlying aetiological mechanisms. However, a number of available metagenomic approaches have different biases in the identification and quantification of the microbial composition, resulting in misinterpretation of the accrued data which subsequently affect conclusions. Therefore, the aim of this study was to interrogate sources of error in various methodologies in sequence-based metagenomic analysis. In this study, genomic DNA of common bacterial pathogens representative of both Gram-positive and Gram-negative organisms mixed at defined quantitative proportions were used as a standardised metagenomics control material (MCM) in order to assess the comparative accuracy of different approaches, i.e. 16S ribosomal RNA (rRNA) profiling and metagenomic shot~un-sequencing. Sources of bias in 16S rRNA including primer-template mismatches, primer design, and bioinformatics analytical tools were identified. Whole genome sequencing generated a high precision of microbial profiling. Bias was also observed due to DNA extraction protocol when the whole cell material (WCM) containing a bacteriologically quantitated range of bacteria was used. This study also suggested that the MCM provided the opportunity to develop species specific assays to detect multiple bacterial pathogens collected from the clinical samples by using high-throughput quantitative PCR (ht-qPCR). In conclusion, the methodology applied to microbial profiling analysis must consider sources of error and methods of standardisation such as those described here. Moreover, ht-qPCR demonstrated the value of a high-throughput bacterial detection technique for clinical diagnostic applications. This thesis has thus applied the principles of metrology to generate, characterise and evaluate whole organism and DNA based quantitative control materials as an essential pre-requisite for the precise and accurate biological interpretation of both 16S profiling and metagenomic analysis of human diseases.

Microorganisms are an indispensable part of our ecosystem, yet the natural metabolic and ecological diversity of these organisms is poorly understood due to a historical reliance of microbiology on laboratory grown cultures. The awareness that this diversity cannot be studied by laboratory isolation, together with recent advances in low cost scalable sequencing technology, have enabled the foundation of culture-independent microbiology, or metagenomics. The study of environmental microbial samples with metagenomics has led to many advances, but a number of technological and methodological challenges still remain. A potentially diverse set of taxa may be represented in anyone environmental sample. Existing tools for representing the genetic composition of such samples sequenced with short-read data, and tools for identifying variation amongst them, are still in their infancy. This thesis makes the case that a new framework based on a joint-genome graph can constitute a powerful tool for representing and manipulating the joint genomes of population samples. I present the development of a collection of methods, called SCRAPS, to construct these efficient graphs in small communities without the availability or bias of a reference genome. A key novelty is that genetic variation is identified from the data structure using a probabilistic algorithm that can provide a measure of the confidence in each call. SCRAPS is first tested on simulated short read data for accuracy and efficiency. At least 95% of non-repetitive small-scale genetic variation with a minor allele read depth greater than 10x is correctly identified; the number false positives per conserved nucleotide is consistently better than 1 part in 333 x 103. SCRAPS is then applied to artificially pooled experimental datasets. As part of this study, SCRAPS is used to identify genetic variation in an epidemiological 11 sample Neisseria meningitidis dataset collected from the African meningitis belt". In total 14,000 sites of genetic variation are identified from 48 million Illumina/Solexa reads. The results clearly show the genetic differences between two waves of infection that has plagued northern Ghana and Burkina Faso.

Les virus sont les entités biologiques les plus abondantes et diversifiées sur Terre et leur diversité est encore très peu connue. Récemment, la métagénomique virale a facilité l'exploration de cette diversité. Néanmoins, la plupart des viromes environnementaux générés à ce jour proviennent de régions tempérées et la plupart des viromes humains proviennent d’échantillons de selles, sang ou de prélèvements oro-naso-pharyngés. L'objectif de mon travail de thèse était d’apporter de nouvelles connaissances sur les communautés virales d’environnements et d’échantillons humains les moins étudiés en utilisant une approche de métagénomique virale.La première partie de cette thèse est une revue des principaux outils d'analyse des métagénomes viraux. La deuxième partie présente la première étude de métagénomique virale dans le désert du Sahara. Dans la troisième partie de ma thèse, je présente de viromes associés a l'Homme: i) le premier métagénome viral issu d'un coprolithe humain du Moyen Âge; ii) la première étude de métagénomique virale sur de liquides péricardiques provenant de patients atteints d’une péricardite infectieuse d'origine inconnue; iii) une analyse fonctionnelle de métagénomes viraux précédemment publiés associés aux expectorations de patients atteints de mucoviscidose qui décrit les gènes de résistance aux antibiotiques portés par les bactériophages dans ces patients.Ce travail présente ainsi des données inédites sur certaines communautés virales peu étudiées et confirme le potentiel de la métagénomique virale pour étudier la diversité virale, révéler la présence de virus inattendus ou inconnus et comprendre leur rôle dans leur écosystème d’origine
Viruses are the most abundant and diverse organisms but little is known about their diversity. Recently, viral metagenomics has allowed performing broad unselective exploration of uncultivated viral communities, bypassing the limits of classical viral detection tools. However, most viral metagenomes are generated from temperate regions (for environmental studies) or from modern stool samples, sera/blood and naso-/oro- pharyngeal samples (for human-associated studies). Therefore, the purpose of my thesis is to study viral communities in the least investigated environments or human samples, using viral metagenomics.The first part of my thesis is a review of the main computational tools for the analysis of viral metagenomes. The second part of my thesis presents the first viromes generated from the Sahara desert. In the third part, I investigate human-associated viral communities: i) the first virome from a human coprolite; ii) the first viromes generated from human pericardial fluids, in idiopathic pericarditis cases; ii) a functional-level investigation of previously described viral metagenomes from cystic fibrosis patient sputa that focuses on antimicrobial resistance genes carried by bacteriophages to better understand the emergence of multidrug-resistance bacteria in the airways of cystic fibrosis patients.This thesis work provides original data on unexplored viral communities and shows the potential of viral metagenomics to give insights on viral diversity, reveal the presence of expected and unexpected viruses and decipher their role in the ecosystem

L’avènement du séquençage métagénomique aléatoire a révolutionné la microbiologie en permettant la caractérisation sans culture préalable de communautés microbiennes complexes telles que le microbiote intestinal humain. Des outils bioinformatiques récemment développés atteignent une résolution au niveau de la souche en recensant des gènes accessoires ou en capturant des variants nucléotidiques (SNPs). Toutefois, ces outils sont limités par l’étendue des génomes de référence disponibles qui sont loin de couvrir toute la variabilité microbienne. En effet, de nombreuses espèces n’ont pas encore été séquencées ou sont représentées par seulement quelques génomes.La création de catalogues de gènes non redondants par assemblage de novo suivie du regroupement des gènes co-abondants révèlent une partie de la matière noire microbienne en reconstituant le répertoire de gènes d’espèces potentiellement inconnues. Bien que les méthodes existantes identifient avec précision les gènes core présents dans toutes les souches d’une espèce, elles omettent de nombreux gènes accessoires ou les divisent en petits groupes de gènes qui ne sont pas associés aux core génomes. Or, capturer ces gènes accessoires est indispensable en recherche clinique et épidémiologique car ces derniers assurent des fonctions spécifiques à certaines souches telles que la pathogénicité ou la résistance aux antibiotiques.Lors de cette thèse, nous avons développé MSPminer, un logiciel performant qui reconstitue et structure des pan-génomes d’espèces métagénomiques (ou MSPs pour Metagenomic Species Pan-genomes) en regroupant les gènes co-abondants dans un ensemble d’échantillons métagénomiques. MSPminer s’appuie sur une nouvelle mesure robuste de la proportionnalité couplée à un classificateur empirique pour regrouper et distinguer les gènes core mais aussi les gènes accessoires des espèces microbiennes.Grâce à MSPminer, nous avons structuré un catalogue de 9,9 millions de gènes du microbiote intestinal humain en 1 661 MSPs. L’homogénéité de l’annotation taxonomique, de la composition nucléotidique ainsi que la présence de gènes essentiels indiquent que les MSPs ne correspondent pas à des chimères mais à des objets biologiquement cohérents regroupant des gènes provenant de la même espèce. Parmi ces MSPs, 1 301 (78%) n’ont pas pu être annotées au niveau espèce montrant que de nombreux microorganismes colonisant l’intestin humain demeurent inconnus malgré les progrès substantiels des techniques de culture microbienne. Remarquablement, les MSPs capturent bien plus de gènes que les clusters générés par les outils existants tout en garantissant une spécificité élevée.Cet ensemble de MSPs peut d’ores et déjà être utilisé pour le profilage taxonomique et la découverte de biomarqueurs dans des échantillons de selles humaines. Ainsi, nous tirons parti des MSPs pour comparer l’impact sur le microbiote intestinal des deux principaux types de chirurgie bariatrique, la gastrectomie par laparoscopie (LSG) et la dérivation gastrique de Roux-en-Y (LRYGB). Enfin, les MSPs ouvrent la voie à des analyses au niveau souche. Dans une autre cohorte, nous avons mis en évidence l’existence de sous-espèces associées à l’origine géographique de l’hôte en étudiant les profils de présence/absence des gènes accessoires groupés dans les MSPs
The advent of shotgun metagenomic sequencing has revolutionized microbiology by allowing culture-independent characterization of complex microbial communities such as the human gut microbiota. Recently developed bioinformatics tools achieved strain-level resolution by making a census of accessory genes or by capturing nucleotide variants (SNPs). Yet, these tools are hampered by the extent of available reference genomes which are far from covering all the microbial variability. Indeed, many species are still not sequenced or are represented by only few genomes.Building of non-redundant gene catalogs followed by the binning of co-abundant genes reveals a part of the microbial dark matter by reconstituting the gene repertoire of species potentially unknown. While existing methods accurately identify core genes present in all the strains of a species, they miss many accessory genes or split them into small gene groups that remain unassociated to core genomes. However, capturing these accessory genes is essential in clinical research and epidemiology because they provide functions specific to certain strains such as pathogenicity or antibiotic resistance.In this thesis, we developed MSPminer, a computationally efficient software tool that reconstitutes Metagenomic Species Pan-genomes (MSPs) by binning co-abundant genes across metagenomic samples. MSPminer relies on a new robust measure of proportionality coupled with an empirical classifier to group and distinguish not only species core genes but accessory genes also.With MSPminer, we structured a catalog made up of 9.9 million genes of the human gut microbiota in 1 661 MSPs. The homogeneity of the taxonomic annotation, of the nucleotide composition as well as the presence of essential genes indicate that the MSPs do not correspond to chimeras but to biologically consistent objects grouping genes from the same species. Among these MSPs, 1 301 (78%) could not be annotated at species level showing that many microorganisms colonizing the human intestinal tract are still unknown despite the substantial improvements of microbial culture techniques. Remarkably, MSPs capture more genes than clusters generated by existing tools while ensuring high specificity.This set of MSPs can be readily used for taxonomic profiling and biomarkers discovery in human gut metagenomic samples. In this way, we take advantage of the MSPs to compare the impact of two main types of surgeries, the laparoscopic sleeve gastrectomy (LSG) and the Roux-En-Y gastric bypass (LRYGB). Finally, the MSPs open the way to strain-level analyses. In another cohort, we identified subspecies associated the host geographical origin by studying presence/absence patterns of the accessory genes grouped in the MSPs

Made available in DSpace on 2015-05-14T16:53:22Z (GMT). No. of bitstreams: 0 Previous issue date: 2014-08-22Bitstream added on 2015-05-14T16:58:56Z : No. of bitstreams: 1 000828058_20160820.pdf: 387489 bytes, checksum: 00df71de2a21eb5c6e83a5f95951a8c4 (MD5) Bitstreams deleted on 2016-08-22T16:41:54Z: 000828058_20160820.pdf,. Added 1 bitstream(s) on 2016-08-22T16:42:31Z : No. of bitstreams: 1 000828058.pdf: 892357 bytes, checksum: b80ec59e6b4e026208848788bf2c37c2 (MD5)
Zoonoses emergentes com origem em animais selvagens representam a mais significante e crescente ameaça à saúde global. Os primatas não humanos contribuem com 20% das principais doenças humanas, são considerados o principal modelo experimental e também servem como sentinelas para diferentes doenças que acometem o homem. Os objetivos da presente tese foram desenvolver protocolos de metagenômica viral para detectar sequências de vírus a partir de plasma de primatas não humanos do Novo Mundo (Saimiri sciuratus, Aotus ainfulatus, Alouatta guariba e Sapajus apella) mantidos em cativeiro no Centro Nacional de Primatas (Belém, PA), no Refúgio Biológico Bela Vista (Foz do Iguaçu, PR) e no Parque Ecológico do Tietê (São Paulo, SP). A tecnologia Illumina (MiSeq) foi usada para o sequenciamento massivo e um fluxograma bioinformático foi desenvolvido para analisar as sequências obtidas. De forma geral, genomas correspondentes a diferentes famílias virais foram identificadas e ênfase foi dada ao torque teno vírus, ao vírus da hepatite G (tipos A e B), parvovírus e papilomavírus. Uma nova sequência de genoma completo do gênero Hepacivirus, família Flaviviridae, foi identificado em primatas do sul do Brasil. Os resultados demonstram a possibilidade de variedade de genomas virais presentes em primatas do Novo Mundo, considerados clinicamente saudáveis e destaca a importância de estudos que objetivam identificar possíveis vírus emergentes. Palavras-chave: doenças infecciosas emergentes, primatas não humanos do Novo Mundo, metagenômica viral, sequenciamento massivo e um fluxograma bioinformático foi desenvolvido para analisar as sequências obtidas. De forma geral,genomas correspondentes a diferentes famílias virais foram identificadas e ênfase foi dada ao torque teno vírus, ao vírus da hepatite G (tipos A e B), parvovírus e papilomavírus. Uma nova sequência de genoma completo do gênero Hepacivirus, família ...
Wild animals-borne emerging zoonoses represent the most significant and growing threat to global health. Non-human primates contribute with 20% of major human diseases, as they are considered the main experimental model and also can be considered sentinels for different diseases that affect humans. The aims of the present thesis were to develop a viral metagenomics protocol to detect sequences from viruses in plasma from captive New World non-human primate species (Saimiri sciuratus, Aotus ainfulatus, Alouatta guariba e Sapajus apella) from National Primate Center (Belém, PA), Bela Vista Biological Sanctuary (BVBS, Foz do Iguaçu, PR), and Tietê Ecological Park (São Paulo, SP). The Illumina technology (MiSeq) was used for deep sequencing, and a bioinformatics pipeline was developed in order to analyze the sequences obtained. Overall, genomes corresponding to different viral families were identified from plasma, and emphasis was given to torque tenus virus, hepatitis G virus types A and B, parvovirus, and pappilomavirus. A new whole-genome sequence from Hepacivirus genus, Flaviviridae family, was identified in monkeys from Southern Brazil. The results demonstrate the wide variety of viral genomes present on clinically healthy New World monkeys and highlight the importance of studies that aim to identify possible emerging viruses

Ce travail de thèse a pour but de modéliser la capacité de dégradation des polysaccharides non digestibles par le microbiote intestinal humain. Nous exploitons pour cela des données métagénomiques. Il s'agit de données d'abondances de séquences de nucléotides dans 1408 échantillons dont les fonctions métaboliques sont assignées par annotation contre une base de données. Les séquences sont annotées par des marqueurs fonctionnels. Après une étape de sélection manuelle de 86 marqueurs fonctionnels pertinents à l'activité de métabolisation des polysaccharides, nous étudions leurs variations d'abondances parmi les échantillons métagénomiques.Nous proposons une approche de modélisation écologique du microbiote intestinal humain et considérons principalement la sélection fonctionnelle intense de cet écosystème pour faire l'hypothèse que des regroupements identiques de fonctions métaboliques sont présents en proportions différentes dans tous les microbiotes intestinaux humains. Nous proposons le terme d'assemblage fonctionnel qui rend compte de la co-occurrence spatiale et temporelle d'un groupement de fonctions. Ces assemblages sont en pratiques déterminés par leur composition en marqueurs fonctionnels, et peuvent s'interpréter comme une combinaison de traits fonctionnels agrégés au niveau des microorganismes composant l'assemblage.Les assemblages fonctionnels sont inférés par le biais d'une factorisation en matrice positive aussi nommée NMF de l'anglais Non-Negative Matrix Factorisation. Cette méthode permet de déterminer les assemblages fonctionnels, à la fois concernant leur composition et à la fois concernant leur abondance dans chacun des 1408 échantillons. Nous exploitons par ailleurs une information métabolique provenant de 190 génomes microbiens et de la bibliographie qui permet de préciser la composition de ces assemblages fonctionnels. Cette information se traduit sous forme d'une contrainte.Nous trouvons 4 assemblages en considérant un consensus entre différents critères. L'utilisation de l'information métabolique nous permet d'interpréter biologiquement ces assemblages. Les métadonnées associées aux 1408 échantillons nous permettent d'observer un comportement différent pour les échantillons provenant d'individus atteints de la maladie de Crohn. Nous validons cette observation sur des données extérieures.Nous avons proposé une approche réductionniste permettant de représenter un processus métabolique important à l'échelle du microbiote. Nous trouvons un nombre réduit de 4 assemblages fonctionnels qui sont biologiquement vraisemblables et permettent de bien approcher les 1408 échantillons métagénomiques
The purpose of this work of thesis is to model the capacity of degradation of non-digestible polysaccharides by the human intestinal microbiote. To this end we exploit metagenomic data. We use abundances of nucleotide sequences in 1408 samples whose metabolic function are assigned by annotation against a database. The sequences are annotated with functional markers. Upon manual selection of 86 functional markers relevant to the activity of metabolisation of polysaccharides, we their abundances variation among the metagenomic samples are studied.We propose an ecological approach in modeling the human intestinal microbiote. We consider the intense functional selection of this ecosystem and assume that identical cluster of metabolic functions can be found in different proportions in every human gut microbiota. We propose the term of functional assembly as to account for spacial and temporal co-occurence of functional cluster. In practice, theses assemblies are determined by their composition and can be interpreted as combinations of functional traits aggregated at the levels of the cluster of microorganisms composing each assembly. Functional assemblies are inferred by the means of Non-Negative Matrix Factorization (NMF). This method allows to determine the composition of functional assemblies and their abundance in each of the 1408 metagenomic sample.Furthermore, we exploit metabolic information from bibliographic resources and 190 microbial genomes in order to specify the composition of these functional assemblies. This information is translated in the form of a constraint.We find 4 assemblies by considering a consensus between various criteria. The use of metabolic information allow to interpret theses assemblies biologically. By exploiting the metadata of the 1408 samples, we observe a different behaviour for the samples coming from individuals suffering from Crohn disease. We validate this observation on external data.We proposed a reductionistic approach allowing to represent an important metabolic process at the level of the microbiota. We find a small number of 4 functional assemblies which are biologically likely and approach well the 1408 metagenomic samples

Viruses, and particularly bacteriophages, are key players in many microbial ecosystems and can profoundly influence the human microbiome and its impact on human health. While the bacterial and archaeal fraction of the human microbiome can now be profiled at an unprecedented resolution via cultivation-free metagenomics, viral metagenomics is still extremely challenging. The lack of universal viral genetic markers limits the de-novo discovery of viral entities, and the low number of available viral reference genomes from cultivation studies does not cover well the phage diversity in human microbiome samples. Viral-like particle (VLP) purification has been proposed as a set of experimental tools to concentrate viruses in samples prior to sequencing, but it remains unclear how efficient and reproducible such tools are in practice. In this thesis we aim to address some of these challenges and better exploit the potential of viral metagenomics in the context of the human microbiome. First, we performed and studied the performance of VLP procedures on freshwater and sediment samples. We found that bacteria can still be abundant at the end of the filtration process, thus lowering the efficiency of the enrichment. Analyzing samples with a low enrichment may lead to inconsistent conclusions, as the residual bacterial contamination might misdirect the computational analysis. To better quantify the extent of non-viral contamination in VLP sequencing, we designed ViromeQC, a novel open-source tool able to assess and rank viromes by their viral purity directly from the raw reads. In ViromeQC, rRNA genes and bacterial single-copy proteins are used as a proxy to estimate non-viral contamination. With the ViromeQC, we conducted the largest meta-analysis on the degree of enrichment of thousands of viral metagenomes, and concluded that the vast majority of them are three-fold less enriched than a standard metagenome. ViromeQC was then used to select the human gut viromes that had the highest enrichment as a starting point for a novel reference-free pipeline for the discovery of previously uncharacterized viral entities. The approach included metagenomic assembly of the enriched viromes as well as extensive mining of many thousands of assembled metagenomes, and led to a catalog of 162,876 sequences of highly-trusted viral origin. Most of these predicted viral sequences had no match against any known virus in RefSeq even though some of them showed a prevalence in gut metagenomes of up to 70%. Our analyses and publicly available tools and resources are helping to uncover the still hidden virome diversity and improve the support for current and future investigations of the human virome.

Background During the first stages of development human infants are either fed human milk or human milk substitutes (infant formulas). The composition of infant formulas and human milk differ drastically, including a difference in protein constituents and bacterial load. Due to the high global frequency of infant formula use, the humanization of infant formulas to better reflect the complex nature of human milk is warranted. To better understand the role of human milk components, the fate and function of a key bacterial sensor in human milk, soluble CD14, was determined. Additionally, the microbiome of human milk was analyzed from a metagenomic standpoint in an attempt to determine which types of bacteria are present in human milk and what their potential biological function might be. Results In rodent models, ingested sCD14 persisted in the gastrointestinal tract and was transferred intact into the blood stream. Once transferred to the blood, ingested sCD14 retained its ability to recognize lipopolysaccharide and initiate an immune response in pups. This transfer of sCD14 across the epithelial barrier was also observed in human cells in vitro, where it appears to be dependent on Toll-like receptor 4. Using Illumina sequencing and the MG-RAST pipeline, the human milk metagenome of ten mothers was sequenced. DNA from human milk aligned to over 360 prokaryotic genera, and contained 30,128 open reading frames assigned to various functional categories. The DNA from human milk was also found to harbor immune-modulatory DNA motifs that may play a significant role in immune development of the infant. Conclusions Given the complex nature of human milk in comparison to its bovine or plant based substitutes, the results presented in this thesis warrant future modification of infant formulas to include non-nutritive bioactive components. Current human milk components not yet present in infant formulas include the diverse microbiome of human milk, the immune-modulatory DNAs which those microbes harbor, and bioactive human proteins such as sCD14.

Antimicrobial resistance genes are harboured by bacteria in the human oral cavity and ruminant faeces and they are shed in particularly high abundances in calf faeces. Furthermore, bacteriocin (antimicrobial peptide) producing bacteria have been isolated from these environments. In recent times bacteriocins have received much attention as potential alternatives to antibiotics. Human saliva and calf faeces harbour ‘yet-to-be cultured bacteria’ that can only be studied by analysing their DNA. To this end, two metagenomic libraries were created from human saliva and calf faeces metagenomic DNA with the aim of identifying novel antimicrobial resistance and bacteriocin genes. Screening these libraries for tetracycline resistance identified two tetracycline resistant clones. Clone PS9 was also tigecycline resistant and contained a 7,765 bp insert that encoded two half-ABC transporter genes; subcloning of these genes showed that they were responsible for the observed resistance phenotype. As the ABC transporter conferred resistance only to tetracyclines and its putative amino acid sequence showed < 80 % identity to known tetracycline resistance proteins, it was named TetAB(60). Clone TT31 contained a 14,226 bp insert. 7, 216 bp of the insert had 97 % nucleotide identity to Tn916 and contained part of tet(M) and a full length tet(L) gene. This gene organisation has not been described in Tn916-like elements and it may represent a novel Tn916-like element. The human saliva library was also screened for antiseptic resistance revealing a CTAB resistant clone. Random transposon mutagenesis of the 19.1 Kb insert and subcloning of a UDP-glucose 4-epimerase revealed it to be solely required for the observed resistance. This study identified novel tetracycline, tigecycline and CTAB resistance genes from the human saliva metagenome, demonstrating the importance of this environment as a source of resistance genes that may compromise the effectiveness of these antibiotics and antimicrobials. Additionally, this work highlights the relevance of house-keeping genes to the development of antimicrobial resistance.

The human microbiome - defined as the microbial communities that live in and on our bodies - is emerging as a key factor in human diseases. The expanding research field that investigates the role of the microbiome on human cancer development, termed oncobiome, has led to important discoveries such as the role of Fusobacterium nucleatum in colorectal cancer carcinogenesis and tumor progression. Motivated by these discoveries, this thesis studied the oncobiome from different perspectives, investigating whether alterations to microbial profiles were associated with disease status or an adverse response to treatment. We used both biopsy tissue samples and 16S rRNA amplicon sequencing (N = 36), as well as privately and publicly available fecal whole metagenomes (N = 764) to investigate microbiome-colorectal cancer (CRC) associations. We observed significant increases in species richness in CRC, regardless of sample type or methodology, which was partially due to expansions of species typically from the oral cavity, as well as an overabundance of specific taxa such as Bacteroides fragilis, Fusobacterium, Desulfovibrio and Bilophila in CRC. Functional potential analysis of CRC metagenomes revealed that the choline trimethylamine-lyase (cutC) gene was over-abundant in CRC, with the strength of association dependent on four identified sequence variants, pointing at a novel potential mechanism of CRC carcinogenesis. Predictive microbiome signatures trained on the combination of multiple datasets showed very high and consistent performances on distinct cohorts (average AUC 0.83, minimum 0.81). To investigate the microbiomes role in response to treatment, we profiled microbial communities of gastric wash samples in gastric cancer patients (N = 36) before and after neoadjuvant chemotherapy through 16S rRNA amplicon sequencing. Gastric wash microbial communities presented remarkably high inter-individual variation, with significant decreases in richness and phylogenetic diversity after treatment and associations with pH, pathological response and sample collection. The most abundant genera found in patients before or after chemotherapy treatment included Streptococcus, Prevotella, Rothia and Veillonella. Despite limitations inherent to differing experimental choices, this thesis provides microbiome signatures that can be the basis for clinical prognostic tests and hypothesis-driven mechanistic studies, as well as supporting the role of the human oral microbiome in whole-body diseases.
O microbioma humano - definido como as comunidades microbianas que vivem sobre e dentro do corpo humano - está se tornando um fator cada vez mais importante em doenças humanas. O campo de estudo que investiga o papel do microbioma no desenvolvimento do câncer humano, denominado oncobioma, está crescendo e já levou a importantes descobertas como o papel da espécie Fusobacterium nucleatum na carcinogênese e progressão tumoral de tumores colorretais. Motivado por estas descobertas, esta tese de doutorado analisou o oncobioma por diferentes perspectivas, investigando se alterações nos perfis microbianos estavam associados à presença da doença ou a uma resposta adversa ao tratamento. Usamos tanto amostras de tecidos de biópsias e o sequenciamento do gene 16S rRNA (N = 36), quanto metagenomas fecais públicos e privados (N = 764), para investigar associações entre o microbioma e o câncer colorretal (CCR). Observamos um aumento significativo da riqueza microbiana no CCR, independentemente do tipo da amostra ou metodologia, que era em parte, devido ao aumento de espécies tipicamente presentes na cavidade oral. Observamos também um aumento da abundância de táxons específicos no CCR, que incluíam Bacteroides fragilis, Fusobacterium, Desulfovibrio e Bilophila. Analisando o potencial funcional dos metagenomas, encontramos um aumento significativo da enzima liase colina trimetilamina (cutC) no CCR, cuja associação era dependente de 4 variantes de sequência, demonstrando ser um possível novo mecanismo de carcinogênese no CCR. Assinaturas preditivas do microbioma treinadas na combinação dos estudos demonstraram ser altamente preditivas e consistentes nos diferentes estudos (média de AUC 0.83, mínimo de 0.81). Para investigar o possível papel do microbioma na resposta ao tratamento, analisamos os perfis microbianos do suco gástrico de pacientes com câncer gástrico (N = 36) antes e depois do tratamento quimioterápico neoadjuvante. As comunidades microbianas apresentaram uma variabilidade inter-individual notavelmente grande, com diminuições significativas na riqueza e diversidade filogenética pós tratamento, além de estarem associadas principalmente ao pH, mas também à resposta patológica e ao tempo da coleta. Os gêneros mais abundantes encontrados nos pacientes antes ou depois da quimioterapia incluíam Streptococcus, Prevotella, Rothia e Veillonella. Apesar das limitações inerentes às escolhas experimentais, esta tese proporciona assinaturas do microbioma que podem servir de base para testes clínicos prognósticos e estudos mecanísticos, além de dar mais suporte ao papel do microbioma oral em doenças humanas.

Antibiotic resistance is a serious threat to public health. Horizontal gene transfer (HGT) plays a critical role in the dissemination of antibiotic resistance, however, knowledge about the source of the resistance determinants and their mobile vectors is still limited. Integrons are natural gene capture and expression systems and are known for their role in dissemination of antibiotic resistance genes (ARGs) in clinically important pathogens. The human oral cavity is a reservoir of ARGs many of which were found on plasmids and transposons. However, the association of these ARGs with integrons have not been investigated before. In this study, a PCR-based metagenomic approach was used to investigate the presence of integrons carrying ARGs in the oral cavity of healthy human individuals from the UK (n=11) and Bangladesh (n=10). PCR primers targeting the mobile integrons (class 1, 2 and 3) as well as chromosomal integrons of Treponema were used to amplify integrons and associated array of gene cassettes (GCs) in the oral metagenome. By analysing the libraries of PCR amplicons, a large pool of GCs including the cassettes located at the first position of an integron were identified and most of them were found to be novel. The cassettes were predicted to carry open-reading frames encoding proteins of a diverse range of functions including antibiotic resistance, competence, plasmid stability and adaptation to stress. Two novel variants of D-alanine-D-alanine ligase (ddl) were identified and located in the first position within an integron for the first time. It was found that expression of ddls increase the resistance of the surrogate host (Escherichia coli) to D-cycloserine (D-cycloserine), an antibiotic used to treat multi-drug-resistant (MDR) and extensively drug-resistant (XDR) tuberculosis. A SNP at c.777 of the ddl variants was found to be responsible for alteration of minimum inhibitory concentration (MIC) of D-cycloserine. Cloning and sequencing of upstream sequence showed that the putative host of the Ddl encoding integron is likely to be a strain of Treponema denticola, one of the causative agents of periodontitis. The Ddl proteins encoded by the cassette genes were expressed and purified and their specific ligase activity was confirmed. The predicted 3D structures were also determined using I-TASSER tools and the generated 3D models were used to test the hypothesis that plant-based flavonoids could play an important role in the evolution ddls as integron GCs. It was found that the flavonoids namely quercetin and apigenin can bind to both ATP and D-alanine binding sites of the Ddls. This study shows that a PCR-based metagenomic approach can recover novel GCs including functional genes that confer resistance to clinically important antibiotics. The evidence for HGT of integron GCs and a hypothesis for evolution of ddls within integrons is also presented.

Submitted by Sara Ribeiro (sara.ribeiro@ucb.br) on 2018-11-07T17:34:33Z No. of bitstreams: 1 HenriqueCesarTeixeiraVerasDissertacao2013.pdf: 26147875 bytes, checksum: 3014f6ec0d791f46e4e1ee8c49894c2f (MD5)
Approved for entry into archive by Sara Ribeiro (sara.ribeiro@ucb.br) on 2018-11-07T17:35:09Z (GMT) No. of bitstreams: 1 HenriqueCesarTeixeiraVerasDissertacao2013.pdf: 26147875 bytes, checksum: 3014f6ec0d791f46e4e1ee8c49894c2f (MD5)
Made available in DSpace on 2018-11-07T17:35:09Z (GMT). No. of bitstreams: 1 HenriqueCesarTeixeiraVerasDissertacao2013.pdf: 26147875 bytes, checksum: 3014f6ec0d791f46e4e1ee8c49894c2f (MD5) Previous issue date: 2013-09-06
Humans live in constant association with microorganims. The amount of microorganims present in the human body exceeds our own cell number. That community of microorganisms has deep influence in health and disease. The use of high-throughput DNA sequencing technologies and culture independent approaches have been enlarging the understanding concerning the communities of microorganisms and the association of these with the host. The human gastrointestinal tract contains one of the most complex bacterial communities. It was proposed recently that the microbiome can be classified in three enterotypes. In our study, we used data metagenomics quantitative search to identify phylogenetic patterns in the intestinal microbiome to develop prediction models for the enterotypes. To reach this aim, statistical tests were applied to the data regarding abundance of bacteria in level taxonomic corresponding to genus. We identified genus significantly with the abundance different and important correlations. Besides the ratio among genus to be used as parameter bioindicator of the respectives enterotypes. Through the logistic regression test we identified that the prediction model for ET1 was influenced significantly by the ratio of Bacteroides / (Prevotella + Ruminococcus). In the model for prediction of ET2, it was the ratio of Prevotella / Bacteroides with such as characteristic significance. And for the model of ET3, we identified the ratio of (Akkermansia + Alistipes) / (Bacteroides + Prevotella) as significant parameter. These models were assessed against two groups of independent data and associated with the value of cut-off 5%; 20% and 95% respectively. Besides the value of cutoff for each models, the crossed validation allowed the association of the model with the measures of PPV for ET1, specificity for ET2 and PNV for ET3. We propose the experimental validation of these models for the qPCR technique. And with that methodology established, it would be possible to do the diagnosis of the enterotype individually.
Humanos vivem em constante associação com microrganismos. A quantidade de microrganismos presentes no corpo humano ultrapassa o nosso próprio número de células. Essa comunidade de microrganismos tem profunda influência na saúde e doenças. As tecnologias de sequenciamento de DNA de alta capacidade e abordagens moleculares independente de cultura têm ampliado a compreensão acerca das comunidades de microrganismos e a associação destes com o hospedeiro. O trato gastrointestinal humano abriga uma das mais complexas comunidades bacterianas. Foi proposto recentemente que o microbioma pode ser categorizado em três enterotipos. No nosso estudo, utilizamos dados metagenômicos quantitativos buncando identificar padrões filogenéticos no microbioma intestinal para desenvolver modelos de predição para os enterotipos. Para alcançar este objetivo, testes estatísticos foram aplicados aos dados referente a abundância de bactérias em nível taxonômico correspondente a gênero. Identificamos gêneros com a abundância significativamente diferente e correlações importantes. Além da razão entre gêneros para ser utilizada como parâmetro bioindicativo dos respectivos enterotipos. Através do teste de regressão logística identificamos que o modelo de predição para o ET1 foi influenciado significativamente pela razão de Bacteroides / (Prevotella + Ruminococcus). No modelo para predição do ET2, foi a razão de Prevotella / Bacteroides que apresentou significância. E para o modelo do ET3, identificamos a razão de (Akkermansia + Alistipes) / (Bacteroides + Prevotella) como parâmetro significativo. Estes modelos foram avaliados contra dois conjuntos de dados independentes e associados com o valor de cut-off 5%; 20% e; 95% respectivamente. Além do valor de cut-off para cada modelos, a validação cruzada permitiu a associação do modelo com as medidas de PPV para o ET1, especificidade para o ET2 e PNV para o ET3. Propomos a validação experimental destes modelos pela técnica de qPCR. E com essa metodologia estabelecida, seria possível fazer o diagnóstico do enterotipo individualmente.

Orientador: João Pessoa de Araújo Junior
Coorientador: Alexander Welker Biondo
Banca: Maria Inês de Moura Campopos Pardini
Banca: Deilson Elgui de Oliveira
Banca: Jean Carlos Ramos da Silva
Banca: Walfrido Kuhl Svoboda
Resumo: Zoonoses emergentes com origem em animais selvagens representam a mais significante e crescente ameaça à saúde global. Os primatas não humanos contribuem com 20% das principais doenças humanas, são considerados o principal modelo experimental e também servem como sentinelas para diferentes doenças que acometem o homem. Os objetivos da presente tese foram desenvolver protocolos de metagenômica viral para detectar sequências de vírus a partir de plasma de primatas não humanos do Novo Mundo (Saimiri sciuratus, Aotus ainfulatus, Alouatta guariba e Sapajus apella) mantidos em cativeiro no Centro Nacional de Primatas (Belém, PA), no Refúgio Biológico Bela Vista (Foz do Iguaçu, PR) e no Parque Ecológico do Tietê (São Paulo, SP). A tecnologia Illumina (MiSeq) foi usada para o sequenciamento massivo e um fluxograma bioinformático foi desenvolvido para analisar as sequências obtidas. De forma geral, genomas correspondentes a diferentes famílias virais foram identificadas e ênfase foi dada ao torque teno vírus, ao vírus da hepatite G (tipos A e B), parvovírus e papilomavírus. Uma nova sequência de genoma completo do gênero Hepacivirus, família Flaviviridae, foi identificado em primatas do sul do Brasil. Os resultados demonstram a possibilidade de variedade de genomas virais presentes em primatas do Novo Mundo, considerados clinicamente saudáveis e destaca a importância de estudos que objetivam identificar possíveis vírus emergentes. Palavras-chave: doenças infecciosas emergentes, primatas não humanos do Novo Mundo, metagenômica viral, sequenciamento massivo e um fluxograma bioinformático foi desenvolvido para analisar as sequências obtidas. De forma geral,genomas correspondentes a diferentes famílias virais foram identificadas e ênfase foi dada ao torque teno vírus, ao vírus da hepatite G (tipos A e B), parvovírus e papilomavírus. Uma nova sequência de genoma completo do gênero Hepacivirus, família ...
Abstract: Wild animals-borne emerging zoonoses represent the most significant and growing threat to global health. Non-human primates contribute with 20% of major human diseases, as they are considered the main experimental model and also can be considered sentinels for different diseases that affect humans. The aims of the present thesis were to develop a viral metagenomics protocol to detect sequences from viruses in plasma from captive New World non-human primate species (Saimiri sciuratus, Aotus ainfulatus, Alouatta guariba e Sapajus apella) from National Primate Center (Belém, PA), Bela Vista Biological Sanctuary (BVBS, Foz do Iguaçu, PR), and Tietê Ecological Park (São Paulo, SP). The Illumina technology (MiSeq) was used for deep sequencing, and a bioinformatics pipeline was developed in order to analyze the sequences obtained. Overall, genomes corresponding to different viral families were identified from plasma, and emphasis was given to torque tenus virus, hepatitis G virus types A and B, parvovirus, and pappilomavirus. A new whole-genome sequence from Hepacivirus genus, Flaviviridae family, was identified in monkeys from Southern Brazil. The results demonstrate the wide variety of viral genomes present on clinically healthy New World monkeys and highlight the importance of studies that aim to identify possible emerging viruses
Doutor

La fermentation des fibres alimentaires est l’une des fonctions majeures du microbiote intestinal humain. Les bactéries fibrolytiques synthétisent un grand nombre d’enzymes, appelées Glycoside Hydrolases (GH), indispensables à la déconstruction de la grande variété structurelle des polysaccharides pariétaux que nous ingérons. Au cours de ce travail, nous avons exploré, grâce à une approche de métagénomique fonctionnelle, l’organisation et les propriétés des systèmes enzymatiques bactériens impliqués dans la dégradation des glycanes de parois végétales dans l’intestin humain.En premier lieu, nous avons cherché à déterminer si les bactéries de la muqueuse iléale étaient capables de dégrader les fibres pariétales dans un contexte sain. Cette fonction étant généralement décrite pour le microbiote colique par extrapolation de travaux menés à partir de selles humaines, nos connaissances de la dégradation des fibres dans la partie haute du tractus digestif sont donc très limitées. Un total de 20 000 clones issus du métagénome bactérien d’une partie saine de la muqueuse iléale d’un individu a été criblé pour des activités de dégradation de la carboxymethylcellulose et du xylane, deux substrats modèles des polysaccharides pariétaux. Douze clones métagénomiques positifs nous ont permis de mettre en évidence un arsenal de gènes bactériens codant pour des GH et d’autres protéines impliquées dans le métabolisme des fibres alimentaires dont certains organisés en PUL (Polysaccharide Utilization Loci), des clusters de gènes spécialisés dans la dégradation des polysaccharides complexes. Ces gènes proviennent de chromosomes bactériens assignés au genre Bacteroides ou à des espèces de Clostridiales, et codent pour des enzymes capables de dégrader également des β-glucanes à liaisons mixtes. L’étude de la prévalence de ces gènes dans les métagénomes de référence indique que plusieurs d’entre eux proviendraient de souches bactériennes plutôt spécifiques de la muqueuse iléale. De plus, certaines enzymes présentent des propriétés inédites potentiellement intéressantes dans le domaine biotechnologique. Nos recherches ont donc permis de revisiter la fonction fibrolytique du microbiote intestinal chez l’Homme et de proposer une localisation de cette fonction dès l’intestin grêle.Dans un second temps, en utilisant une approche méthodologique similaire, nous avons étudié la capacité du microbiote fécal d’un individu obèse à dégrader des polysaccharides pariétaux complexes, en général moins consommés par les individus obèses. Au total, nous avons identifié 50 clones appartenant à 14 espèces bactériennes des phyla suivants : Bacteroidetes, Firmicutes et Actinobacteria. Les inserts métagénomiques portent des gènes codant pour différentes familles de GH, impliquées dans la dégradation des polysaccharides d’intérêt. Les premières analyses de la prévalence de ces gènes chez plus d’une centaine d’individus (obèses ou non), par interrogations des catalogues de gènes microbiens de référence, suggèrent des associations avec le statut phénotypique « obèse »
Among the crucial functions of the intestinal microbiota, extracting energy from food such as dietary fibres is of major importance. Facing the huge diversity of incoming complex carbohydrates, the fibrolytic bacteria synthesize a set of diversified Carbohydrate-Active Enzymes (CAZymes) including Glycoside Hydrolases (GH) that specifically disrupt complex polysaccharides. Here, using functional metagenomic approaches, we explored the organization and properties of bacterial enzymatic systems involved in the breakdown of plant cell wall (PCW) glycans in the intestinal tract.Firstly, we investigated the capacity of the microbiota associated to the human ileum mucosa to degrade complex non-starch polysaccharides in a healthy context. This function has never been investigated in this part of the intestine, but it has been rather associated to microorganisms inhabiting the colon, due to more accessible fecal samples. Using a fosmid library derived from a healthy part of the human ileal mucosa, we screened 20,000 metagenomic clones for their activities against carboxymethylcellulose and xylan chosen as models of the major PCW polysaccharides from dietary fibres. Twelve positive clones revealed a broad range of CAZyme encoding genes from Bacteroides to Clostridiales species, as well as Polysaccharide Utilization Loci (PUL). Functional GH genes were identified and break-down products examined from different polysaccharides including mixed-linkage β-glucans. Revealed CAZymes and PUL were also examined for their prevalence in human gut microbiomes. Part of them belongs to unidentified strains rather specifically established in the ileum. Others were enzymes unclassified in identified GH families or with original properties addressing novel candidates for biotechnological applications. Thus, we evidenced for the first time that the ileal mucosa associated-microbiota encompasses the enzymatic potential for PCW complex polysaccharide degradation that might start in the small intestine.In a second time, by using the same methodology, we harvested the enzymatic capacities of the fecal microbiota from an obese person to disrupt complex polysaccharides from dietary fibres usually consumed in lower quantity in obese people. This study aimed at examining the links between genes encoding enzymes specifically dedicated to PCW complex carbohydrates and the obese phenotypic status using reference microbial gene catalogs. We screened a fecal metagenomic library from an obese individual on representative PCW substrates and identified 50 clones belonging to 14 different species from the Bacteroidetes, Firmicutes and Actinobacteria phyla. The metagenomic inserts harbor genes encoding enzymes from GH families specific from complex carbohydrate degradation. First querying of the prevalence of these genes in hundreds individuals (obese and control), using catalogs of reference microbial genes, suggest associations with the "obese" phenotypic status

Il est maintenant bien établi que le microbiote du tractus digestif humain joue un rôle important dans la santé humaine. Pourtant, nous commençons à peine à comprendre les mécanismes moléculaires par lesquels les bactéries agissent sur les cellules hôtes, des connaissances qui pourraient fournir des nouvelles orientations dans le traitement et la prévention de maladies. Cette dernière décennie a vu un développement rapide des études du microbiote intestinal, et à présent des quantités importantes de données métagénomiques ainsi que des centaines de séquences génomiques de bactéries commensales sont disponibles. Ensemble, ces données fournissent une plateforme pour des approches in silico pour l'identification de molécules bactériennes impliquées dans la communication moléculaire avec l'hôte. Le défi consiste à développer des stratégies efficaces d'exploration de données et de validation, permettant de passer de corrélations et prédictions à des interactions bactérie - hôte fonctionnelles, validées expérimentalement. Le travail présenté dans cette thèse vise à démontrer l'importance d'analyses in silico afin d'élargir nos connaissances sur les interactions bactéries - hôtes. Il montre également comment cette information peut être appliquée dans des études fonctionnelles visant à identifier des molécules effectrices bactériennes fonctionnelles. Les principaux résultats peuvent être divisés en trois parties. La première partie traite de l'élaboration et de la validation d'un système hôte - vecteur pour des études de (méta)génomique fonctionnelle. La deuxième partie décrit une étude fonctionnelle où un certain nombre d'effecteurs candidats ont été identifiés parmi les protéines sécrétées et de surface de bactéries à Gram positif par une approche d'exploration in silico. Il décrit également l'application du nouveau système hôte - vecteur pour l'évaluation du rôle de ces candidats dans l'immuno-modulation. Enfin, dans la troisième partie, nous présentons une étude in silico qui a permis l'identification de fonctions bactériennes sur- ou sous-représentées dans une sélection de bactéries à Gram positif du tractus digestif humain
It is now well established that the human gastrointestinal tract microbiota plays an intricate role in human health. However, we are only beginning to understand the molecular mechanisms by which bacteria act on the host cells, knowledge that could provide new directions in treating and preventing disease. The last decade has seen a rapid development of the gut microbiota field, and presently abundant metagenome data and hundreds of genome sequences of individual commensal bacteria are available. Together, these data provide a platform for in silico mining approaches to identify bacterial molecules involved in communication with the host. The challenge is to develop efficient mining and validation strategies, in order to move from correlations and predictions to experimentally validated functional bacteria – host relationships. The work presented in this thesis aims to demonstrate the importance of in silico analyses to broaden our knowledge on bacteria - host interactions. It also shows how this information can be applied in functional studies aiming to identify functional bacterial effector molecules. The main results can be divided in three parts. The first part deals with the development and validation of a host - vector system for functional (meta)genomics studies. The second part describes a functional study where a number of candidate effectors were identified among secreted and surface-exposed proteins from Gram-positive bacteria using an in silico mining approach. It also describes the application of the newly developed host - vector system to evaluate the role of these candidates in immune modulation. Finally, in the third part we present an in silico study that identified new bacterial functions over- or under-represented in a selection of Gram-positive human gut bacteria

Le système immunitaire muqueux, et plus particulièrement les réponses intestinales IgA sont essentielles non seulement à la défense contre les agents pathogènes, mais aussi au façonnement de la flore intestinale commensale. Dans les modèles murins de déficit en IgA, on observe une dysbiose intestinale majeure associée à une inflammation muqueuse, réversibles après restauration des IgA. Le but de ce travail est de décrire l'impact de l'absence d'IgA chez l'homme sur la composition du microbiote intestinal ainsi que ses conséquences locales et systémiques. L'étude comparative par analyse métagénomique des selles de 17 sujets déficitaires en IgA et de 34 donneurs sains retrouve l'absence de différence majeure en termes de répartition des phyla dominants, de diversité et de richesse génique bactériennes entre les deux groupes. En revanche, en analysant à l'échelon des espèces, on observe dans le déficit en IgA une surreprésentation d'espèces pro-inflammatoires et une sous-représentation d'espèces anti-inflammatoires. En outre, en l'absence d'IgA, nous observons la présence de réponses IgM qui opsonisent partiellement les genres ciblés par l'IgA, mais semblent maintenir la diversité au sein des Actinobactéries. Les patients présentent un biais phénotypique lymphocytaire T circulant (TH17) associé à des stigmates de translocation bactérienne. Enfin, l'absence d'IgA s'associe à une perturbation du réseau bactérien minimal "obligatoire". Ces résultats suggèrent que le déficit en IgA humain s'accompagne d'une dysbiose modérée associée à une altération de l'architecture du réseau bactérien induisant une hyperactivation du système immunitaire, malgré la présence de réponses IgM
IgA responses play a key role in gut mucosa, defending host against pathogens but also shaping the commensal flora. In order to get insights into the specific contributions of IgA to host/microbial symbiosis in humans, we explored patients that lack only IgA, using gut microbial metagenomics and systems immunology. Microbiota composition was compared between 34 healthy controls and 17 selective IgA deficiency (sIgAd) patients. Contrary to what was observed in murine models of IgA deficiency, we show that human sIgAd is not associated with massive perturbations of gut microbial ecology, regarding phyla distribution, bacterial diversity and gene richness. A clear gut microbial signature is however associated to sIgAd: we found 19 over-represented MGS mainly described to be pro-inflammatory, but also 14 under-represented MGS, mainly known to be beneficial. We also explored local consequences of IgA deficiency, particularly whether IgM could replace IgA at host/bacterial interface. Using a combination of bacterial flow sorting and DNA sequencing, we therefore analysed the composition of IgM-coated microbiomes observed in sIgAd. We show that IgM only partially supply IgA deficiency, as not all typical IgA targets can also be opsonized by IgM, but nevertheless contribute to maintain Actinobacteria diversity. IgA deficiency is associated with a skewed circulating CD4+ T cell profile towards TH17, as well as markers of bacterial translocation. Finally, sIgAd is associated with a perturbation of the minimal bacterial network. Altogether our results suggest that human IgA deficiency is associated with a mild dysbiosis associated to systemic inflammation despite the presence of IgM

La flore microbienne est un ensemble de micro-organismes comme les bactéries, archées, eucaryotes inférieurs et virus, jouant un rôle important dans l’équilibre d’un écosystème. Cette flore reste encore mal définie car en 2012, seules ~30% des micro-organismes de la flore intestinale humaine étaient caractérisés, et moins de 50% des micro-organismes de la flore fromagère traditionnelle étaient caractérisés au niveau fonctionnel. Depuis 2006, les séquenceurs à ADN de seconde génération permettent d’analyser directement le contenu génique d’un prélèvement de flore sans contrainte d’isolement ou de culture. Toutefois, les séquences d’ADN générées ne sont pas structurées en génome et sont hautement fragmentées, freinant considérablement l’analyse et l’exploitation de ces données. Dans ce travail, nous avons développé de nouvelles méthodes dites de métagénomiques quantitatives, car elles permettent de regrouper les courtes séquences d’ADN ayant la même abondance au sein de plusieurs échantillons métagénomiques, pouvant provenir d’une même espèce microbienne.Au niveau du microbiote intestinal humain, nous avons structuré un catalogue de 3,9 millions de gènes de la flore intestinale humaine en 741 unités ou « clusters » correspondant à des génomes de bactéries, d’archées et d’eucaryotes, que nous appelons espèces métagénomiques (MGS) et 6640 unités correspondant principalement à des génomes de virus, plasmides et divers ilots génomiques comme des CRISPR, que nous appelons unités métagénomiques (MGU). Cette méthode de structuration a ensuite été utilisée pour faciliter des analyses d’associations de la composition de la flore intestinale avec des maladies humaines comme la maladie de Crohn, l’obésité ou le diabète de type 2. Enfin, au niveau des flores alimentaires, nos méthodes ont été utilisées pour constituer un catalogue de 134 génomes d’espèces bactériennes fromagères et caractériser la flore de surface de fromages traditionnels. Ceci nous a permis de détecter la présence de nouvelles bactéries alimentaires, pouvant enrichir la liste des bactéries à possible intérêt technologique dans les produits laitiers fermentés
The microbial flora is a micro-organism complex containing for example bacteria, archaea, lower eukaryotes and viruses, which play an important role in ecosystem equilibrium. This flora remains poorly defined as in 2012, only ~30% of the intestinal flora micro-organisms have been characterized, and less than 50% of traditional cheese floras were characterized at the functional level. Since 2006, the second generation of DNA sequencers have allowed the direct analysis of the genetic content from a microbial flora sample without isolation or culture limitation. However, the DNA reads generated are not structured with respect to genomes and also are highly fragmented, slowing down dramatically the exploitation and analysis of these data.In this work, a new methodology based on quantitative metagenomic are described., This allows the clustering of short DNA sequences with the same abundance in multiple metagenomic samples, which should originate from the same microbial species. A 3.9 million gene catalog has been built from the human intestinal tract microbiota and divided into 741 units or clusters corresponding to bacteria, archaea and eukaryote genomes. These have been defined as metagenomic species (MGS) and 6640 units of them corresponds mainly to viral genomes, plasmids and genetic islands like CRISPR, with the sub-name of metagenomic units (MGU). This methodology was then used to facilitate the association analysis of the intestinal flora composition with human diseases such as Crohn’s disease, obesity or type 2 diabetes. Within, the alimentary flora, our methods have also been used to constitute a 134 genomic catalog of cheese bacteria and characterize them from the surface of traditional cheeses. This allowed the detection of new alimentary bacteria, that will enriched the list of bacteria with potential interest for the commercial exploitation of fermented products

Le microbiote intestinal humain abrite un grand nombre de micro-organismes qui jouent un rôle crucial dans la santé humaine. Cependant, le nombre de bactéries détectées par des méthodes quantitatives indépendantes de la culture s'est révélé beaucoup plus élevé que celui des bactéries cultivées sur des géloses. Cet écart est connu sous le nom de la « Great plate count anomaly ». L'objectif principal de ce travail était d'expliquer ''l’incultivabilité" des bactéries intestinales et de maintenir leur viabilité. Dans ce travail, nous avons constaté que l'exposition à l'oxygène pendant plus d'une heure réduisait la cultivabilité des bactéries à 50%. De plus, lorsque les échantillons ont été exposés à l'oxygène pendant moins de 2 minutes, la cultivabilité a augmenté à 87%. Ce dernier résultat suggère que la non-cultivabilité pourrait être due au fait que les bactéries sensibles à l'oxygène étaient à l'état viable mais non cultivable, ou qu'elles étaient blessées ou mortes. Ce résultat a été confirmé lorsque nous avons séquencé les bactéries vivantes, blessées et mortes triées par le FACS. 28 % des UTOs correspondaient à des bactéries mortes, dont environ deux tiers étaient inconnues, et la majorité de ces bactéries étaient sensibles à l’oxygène. D'autre part, notre nouveau milieu protecteur a démontré son efficacité sur les échantillons fécaux et les bactéries sensibles à l'oxygène. En conclusion, notre étude nous a permis d’explorer la « dark matter » du microbiote intestinal humain et a révélé que la métagénomique et l'approche culturomics sont nécessaires pour bien comprendre la diversité et la richesse des bactéries cultivables et non cultivables
The human gut microbiota harbors a wide range of microorganisms that play a crucial role in human health. However, the number of bacterial cells detected by quantitative culture-independent methods was found to be much higher compared to that of cultured bacteria on agar plates. That discrepancy is known as the “ Great plate count anomaly ”. The main goal of this work was to investigate the “ unculturability ” of gut bacteria and maintain their viability. we found that exposure to oxygen for more than 1 hour decreased the culturability of bacteria to 50%. More importantly, when samples were exposed to oxygen for less than 2 min, the culturability increased to 87%. This result suggested that the non-culturability might be due to the fact that oxygen-sensitive cells were in the viable but non-culturable state, or either injured or dead. This funding was confirmed when we sequenced the FACS sorted live, injured and dead bacteria, where, 28% of of bacterial OTUs in total fecal samples were exclusively found dead and/or injured. Among these non-live bacteria, about two-thirds were unknown, thus a large amount were anaerobic. In the other hand, our new protectant medium showed its effectiveness on fecal samples and oxygen sensitive bacteria. In conclusion, our work has confirmed the importance of sample conditioning and processing to obtain the best culture conditions and isolation rates. In addition, our study allowed us to shed light on the dark matter of the human gut microbiota and revealed that both metagenomics and culturomics approach are needed for full insight into the diversity and richness of culturable and unculturable bacteria in the human gut microbiota

Les inhibiteurs des protéases à sérine (Serpins) constituent une classe d'enzymes très peu étudiée chez les bactéries. Dans ce travail de thèse nous nous sommes intéressés à l'étude des serpins provenant du microbiote intestinal et l'investigation de leur potentiel anti-inflammatoire pour le traitement des maladies inflammatoires chroniques de l'intestin (MICI) chez l'homme. Pour cela nous avons identifié les serpins provenant du microbiote intestinal humain et analysé leur diversité ainsi que leur distribution entre les individus malades et sains. Ces données nous ont permis d'isoler les serpins significativement associées aux MICI. La purification de quarte d'entre elles nous a amené à démontrer qu'elles inhibent les protéases humaines impliquées dans les MICI. L'analyse biochimique et cinétique approfondie de ces protéines a montré qu'elles possèdent des propriétés originales notamment leur efficacité d'inhibition élevée. L'étude de l'effet protecteur de trois serpins chez un modèle animal de colite a démontré pour la première fois l'efficacité des serpins in vivo démontrant ainsi leur potentiel thérapeutique
Serine protease inhibitors (Serpins) are a class of proteins that reamin poorly studied in bacteria. In this thesis we are interested in the study of serpins originating from the intestinal microbiota and the investigation of their anti-inflammatory potential for the treatment of inflammatory bowel diseases (IBD) in humans. For this we have identified serpins from the human gut microbiota and analyzed their diversity as well as their distribution between healthy and IBD patients. These data allowed isolating serpins significantly associated with IBD. The purification of four of them led us to demonstrate that they inhibit human proteases involved in IBD. Biochemical and kinetic analysis of these proteins showed that they exhibit original properties, in particular their high inhibition efficiency. The study of the protective effect of three serpins in an animal model of colitis demonstrated for the first time the efficacy of serpins in vivo demonstrating thus their therapeutic potential

The human gut is home for thousands of microbes that are important for human life. As most of these cannot be cultivated, metagenomics is an important means to understand this important community. To perform comparative metagenomic analysis of the human gut microbiome, I have developed SMASH (Simple metagenomic analysis shell), a computational pipeline. SMASH can also be used to assemble and analyze single genomes, and has been successfully applied to the bacterium Mycoplasma pneumoniae and the fungus Chaetomium thermophilum. In the context of the MetaHIT (Metagenomics of the human intestinal tract) consortium our group is participating in, I used SMASH to validate the assembly and to estimate the assembly error rate of 576.7 Gb metagenome sequence obtained using Illumina Solexa technology from fecal DNA of 124 European individuals. I also estimated the completeness of the gene catalogue containing 3.3 million open reading frames obtained from these metagenomes. Finally, I used SMASH to analyze human gut metagenomes of 39 individuals from 6 countries encompassing a wide range of host properties such as age, body mass index and disease states. We find that the variation in the gut microbiome is not continuous but stratified into enterotypes. Enterotypes are complex host-microbial symbiotic states that are not explained by host properties, nutritional habits or possible technical biases. The concept of enterotypes might have far reaching implications, for example, to explain different responses to diet or drug intake. We also find several functional markers in the human gut microbiome that correlate with a number of host properties such as body mass index, highlighting the need for functional analysis and raising hopes for the application of microbial markers as diagnostic or even prognostic tools for microbiota-associated human disorders
Der menschliche Darm beheimatet tausende Mikroben, die für das menschliche Leben wichtig sind. Da die meisten dieser Mikroben nicht kultivierbar sind, ist „Metagenomics“ ein wichtiges Werkzeug zum Verständnis dieser wichtigen mikrobiellen Gemeinschaft. Um vergleichende Metagenomanalysen durchführen zu können, habe ich das Computerprogramm SMASH (Simple metagenomic analysis shell) entwickelt. SMASH kann auch zur Assemblierung und Analyse von Einzelgenomen benutzt werden und wurde erfolgreich auch das Bakterium Mycoplasma pneumoniae und den Pilz Chaetomium thermophilum angewandt. Im Zusammenhang mit der Beteiligung unserer Arbeitsgruppe am MetaHIT (Metagenomics of the human intestinal tract) Konsortium, habe ich SMASH benutzt um die Assemblierung zu validieren und die Fehlerrate der Assemblierung von 576.7 Gb Metagenomsequenzen, die mit der Illumina Solexa Technologie aus der fäkalen DNS von 124 europäischen Personen gewonnen wurde, zu bestimmen. Des Weiteren habe ich die Vollständigkeit des Genkatalogs dieser Metagenome, der 3.3 Millionen offene Leserahmen enthält, geschätzt. Zuletzt habe ich SMASH benutzt um die Darmmetagenome von 39 Personen aus 6 Ländern zu analysieren. Hauptergebnis dieser Analyse war, dass die Variation der Darmmikrobiota nicht kontinuierlich ist. Anstatt dessen fanden wir so genannte Enterotypen. Enterotypen sind komplexe Zustände der Symbiose zwischen Wirt und Mikroben, die sich nicht durch Wirteigenschaften, wie Alter, Body-Mass-Index, Erkrankungen und Ernährungseigenschaften oder ein mögliches technisches Bias erklären lassen. Das Konzept der Enterotypen könnte weitgehende Folgen haben. Diese könnten zum Beispiel die unterschiedlichen Reaktionen auf Diäten oder Medikamenteneinahmen erklären. Weiterhin konnten wir eine Anzahl an Markern im menschlichen Darmmikrobiome finden, die mit unterschiedlichen Wirtseigenschaften wie dem Body-Mass-Index korrelieren. Dies hebt die Wichtigkeit dieser Analysemethode hervor und erweckt Hoffnungen auf Anwendung mikrobieller Marker als diagnostisches oder sogar prognostisches Werkzeug für menschliche Erkrankungen in denen das Mikrobiom eine Rolle spielt

Microbial species (bacteria and archaea) in the gut are important for human health in various ways. Not only does the species composition vary considerably within the human population, but each individual also appears to have its own strains of a given species. While it is known from studies of bacterial pan-genomes, that genetic variation between strains can differ considerably, such as in Escherichia coli, the extent of genetic variation of strains for abundant gut species has not been surveyed in a natural habitat. This is mainly due to the fact that most of these species cannot be cultured in the laboratory. Genetic variation can range from microscale genomic rearrangements such as small nucleotide polymorphism (SNP) to macroscale large genomic rearrangements like structural variations. Metagenomics offers an alternative solution to study genetic variation in prokaryotes, as it involves DNA sequencing of the whole community directly from the environment. However, most metagenomic studies to date only focus on variation in gene abundance and hence are not able to characterize genetic variation (in terms of presence or absence of SNPs and genes) of gut microbial strains of individuals. The aim of my doctorate studies was therefore to study the extent of genetic variation in the genomic sequence of gut prokaryotic species and its phenotypic effects based on: (1) the impact of SNP variation in gut bacterial species, by focusing on genes under selective pressure and (2) the gene content variation (as a proxy for structural variation) and their effect on microbial species and the phenotypic traits of their human host. In the first part of my doctorate studies, I was involved in a project in which we created a catalogue of 10.3 million SNPs in gut prokaryotic species, based on metagenomes. I used this to perform the first SNP-based comparative study of prokaryotic species evolution in a natural habitat. Here, I found that strains of gut microbial species in different individuals evolve at more similar rates than the strains within an individual. In addition, I found that gene evolution can be uncoupled from the evolution of its originating species, and that this could be related to selective pressure such as diet, exemplified by galactokinase gene (galK). Despite the individuality (i.e. uniqueness of each individual within the studied metagenomic dataset) in the SNP profile of the gut microbiota that we found, for most cases it is not possible to link SNPs with phenotypic differences. For this reason I also used gene content as a proxy to study structural variation in metagenomes. In the second part of my doctorate studies, I developed a methodology to characterize the variability of gene content in gut bacterial species, using metagenomes. My approach is based on gene deletions, and was applied to abundant species (demonstrated using a set of 11 species). The method is sufficiently robust as it captures a similar range of gene content variability as has been detected in completely sequenced genomes. Using this procedure I found individuals differ by an average of 13% in their gene content of gut bacterial strains within the same species. Interestingly no two individuals shared the same gene content across bacterial species. However, this variation corresponds to a lower limit, as it is only accounts for gene deletion and not insertions. This large variation in the gene content of gut strain was found to affect important functions, such as polysaccharide utilization loci (PULs) and capsular polysaccharide synthesis (CPS), which are related with digestion of dietary fibers. In summary, I have shown that metagenomics based approaches can be robust in characterizing genetic variation in gut bacterial species. I also illustrated, using examples both for SNPs and gene content (galK, PULs and CPS), that this genetic variation can be used to predict the phenotypic characteristics of the microbial species, as well as predicting the phenotype of their human host (for example, their capacity to digest different food components). Overall, the results of my thesis highlight the importance of characterizing the strains in the gut microbiome analogous to the emerging variability and importance of human genomics
Mikrobielle Arten (Bakterien und Archaeen) im menschlichen Darm sind wichtige Begleiter für unsere Gesundheit. Jedoch gibt es nicht nur starke Unterschiede zwischen individuellen Wirten in der Artenzusammensetzung des Darmmikrobioms, sondern es scheint sogar Individuen-spezifische Bakterienstämme zu geben. Analysen von Bakterien wie z.B. Escherichia coli haben schon früh gezeigt, dass die Genome von Bakterienstämmen derselben Art große Unterschiede aufzeigen können; jedoch wurden diese Unterschiede bisher noch nicht in einer natürlichen Umgebung gezeigt. Genetische Variation kann viele Ausprägungen haben und reicht von kleinen Veränderungen wie „small nucleotide polymorphism“ (SNP) zu makroskopischen Veränderung, wie z.B. chromosomalen Restrukturierungen. All diese genetischen Variationen wurden bis jetzt nicht in der natürlichen Umgebung der Bakterien studiert, vorallem bedingt durch fehlende Methoden um die meisten dieser Bakterien um Labor zu kultivieren. Metagenomische Studien können hier helfen, da sie unabhängig von Kultivierungen jegliche DNS aus einer natürlichen Bakteriengemeinschaft untersuchen. Jedoch wurde dies in den meisten bisher veröffentlichten metagenomischen Studien nicht ausgenutzt da diese hauptsächlich auf die Anzahl der gefunden Gene ausgerichtet waren. Das Ziel meiner Doktorarbeit war es, die genetische Variation in Darmbakterien zu beschreiben und phenotypische Veränderungen zu untersuchen. Dies habe ich umgesetzt durch die Erforschung (1) der SNP-Varianz in Darmbakterien, mit besonderem Augenmerk auf Gene, die unter einem selektivem Druck stehen und (2) der Variationen in der Genzusammensetzung eines Genomes (als eine Annäherung an strukturelle Variationen) und welchen Effekt dies auf Mikrobenarten und Wirtsphenotypen hat. Im ersten Kapitel meiner Doktorarbeit beschreibe ich meine Arbeit in einem Projekt unserer Gruppe, in dem wir basierend auf metagenomischen Daten 10 Millionen SNPs in menschlichen Darmbakterien beschrieben haben. Diesen Datensatz habe ich verwendet um die erste SNP-basierte, vergleichende Studie der Bakterienevolution in einem natürlichen Habitat zu realisieren. Ich entdeckte, dass Bakterienstämme unabhängig vom Wirt ähnliche evolutionäre Raten haben. Genauer gesagt, die evolutionäre Rate für eine Art ist stabiler zwischen Wirten, als die von verschiedenen Spezies innerhalb eines Wirtes. Ausserdem fand ich heraus, dass die Evolution von einzelnen Genen unabhängig vom restlichen Genom einer Spezies ist. Dies könnte durch einen Selektionsdruck wie z.B. die Ernährung des Wirtes ausgelöst werden, was ich am Beispiel des Galactokinasegenes (galK) gezeigt habe. Obwohl wir zeigen konnten, dass das SNP-Profil der Darmbakterien spezifisch für den jeweiligen Wirt ist, konnten wir keine Assoziation zwischen SNPs und Wirtsphänotypen finden. Auch aus diesem Grund habe ich mich in meiner weiteren Arbeit verstärkt auf makroskopische Genomvariationen konzentriert. Im zweiten Teil meiner Doktoarbeit entwickelte ich eine neue Methode, um Variationen in der genomische Zusammensetzung von einzelnen Bakterienarten zu beschreiben, wieder basierend auf metagenomischen Daten. Hierbei fokussiere ich mich insbesondere auf Gene, die in unseren metagenomischen Daten im Verglich zum Referengenom fehlen und wende dies auf die 11 dominantesten Bakterienspezies an. Diese neue Methode ist robust, da die gefundene Genomvarianz in unseren metagenomischen Daten übereinstimmt mit Daten aus komplett sequenzierten Genomen. So konnte ich herausfinden, dass im Durchschnitt 13% der Gene einer Bakterienart zwischen einzelen Wirten varieren. Besonders interessant ist hier, dass wir keine zwei Wirte gefunden haben, die für eine Bakterienart genau diesselben Gene haben. Jedoch ist die erwarte Varianz aller Wahrscheinlichkeit nach noch größer, da ich mit dieser Methode nur fehlende Gene beschreiben kann, aber nicht neu hinzugekommende. Diese Varianz kann auch wichtige bakterielle Funktionen betreffen, z.B. Gene für „polysaccharide utilization loci“ (PULs) und „capsular polysaccharide synthesis“ (CPS), welche wichtig sind um Ballaststoffe in der Nahrung zu verwerten. Zusammenfassend konnte ich in dieser Arbeit zeigen, dass metagenomische Methoden robust genug sind um die genetische Varianz von Darmbakterien zu beschreiben. Ausserdem konnte ich zeigen, dass die beschriebene Varianz benutzt werden kann, um phenotypische Veränderungen von Bakterien vorherzusagen (demonstriert für die galK, PULs and CPS-Gene). Dies wiederrum könnte benutzt werden um Vorhersagen für den Wirt über z.B. seine Ernährung zu machen. Meine Doktorarbeit zeigt wie wichtig es ist, einzelne Bakterienstämme zu charakterisieren, ganz analog zu der Bedeutsamkeit der genetischen Varianz des menschlichen Genomes

Tese de mestrado em Bioinformática e Biologia Computacional, apresentada à Universidade de Lisboa, através da Faculdade de Ciências, 2016
O microbiota humano (a soma de todos os microrganismos que colonizam o corpo humano) ´e composto aproximadamente por 1e+14 células bacterianas, que abrangem vários taxa, e colonizam principalmente a pele, mucosas, tecido conjuntivo, e o tracto gastrointestinal, nomeadamente o cólon. O somatório de todos os genomas microbianos que lhe dizem respeito é frequentemente denominado microbioma. O conjunto de genes que codificam virulência (eventualmente conferindo patogenicidade à bactéria) são frequentemente codificados em elementos genéticos móveis. Deste modo, muitos factores de virulência (VF) bacteriana conseguem ser facilmente disseminados em populações bacterianas por transferência horizontal de genes (movimento de material genético entre células), convertendo batérias mutualistas ou comensais em potenciais patogenes. Analogamente, a coleção de genes cujos determinantes (produtos génicos) conferem resistência a antibióticos (AR), existentes tanto em bactérias patogénicas como não-patogénicas, também se apresenta repetidamente codificada em elementos genéticos móveis, os quais sob pressão selectiva, se conseguem disseminar por entre comunidades bacterianas através do processo de transferência horizontal de genes, atravessando por vezes as barreiras taxonómicas de espécie e género. Esta característica permite que a comunidade sobreviva, e persista como um todo, comportando-se como um reservatório de genes conferentes de resistência. Microbiomas ambientais, como os que se encontram presentes no solo, são descritos como reservatórios abundantes de genes de resistência a antibióticos. Estes codificam para determinantes de resistência a todas as classes de antibióticos descritas até hoje. Apesar da antibioticoterapia ser direcionada a bactérias patogénicas, esta também afeta muitas espécies bacterianas não-patogénicas que fazem parte do microbiota dos indivíduos sujeitos a este tipo de terapia medicamentosa. Efeito que também se verifica em bactérias ambientais que se encontrem expostas a este tipo de pressão selectiva consequente de más práticas agrárias e da pecuária, ou simplesmente poluição antropológica. Por conseguinte, o microbioma humano detém genes de resistência passiveis de transmissão a estirpes patogénicas, tornando-o num repertório de determinantes de resistência a antibióticos altamente diversificado. O estilo de vida virulento, característico de bactérias patogénicas, tem sido consecutivamente associado a fenótipos de resistência a antibióticos. No entanto, esta associação nem sempre tem sido direta e previsível. Por um lado, o crescente uso de antibióticos tem vindo a seleccionar bactérias detentoras de fenótipos resistentes, sejam elas patogénicas ou não, criando reservatórios genéticos de resistência em diversos microbiomas. Porventura não é claro se a diversidade de genes conferentes de virulência se correlaciona com a diversidade dos que conferem resistência a antibióticos. Em contrapartida, existem inúmeros relatos bibliográficos de estirpes altamente virulentas e multi-resistentes (resistentes a mais do que uma classe de antibiótico) que têm vindo a disseminar-se por todo o globo. Tendo em conta que atualmente existe uma grande disponibilidade de antibióticos, e em alguns casos, administração não supervisionada, podemos concluir que os microbiotas humanos, bem como os seus respectivos microbiomas, estão sujeitos a diferentes graus de pressões seletivas impostas pelos referidos compostos. Neste contexto, podemos inferir que para alguns patogenes, em ordem a sobreviver e colonizar o hospedeiro, codificar apenas para virulência pode não ser suficiente, se se encontrarem na presença de antibióticos. Por outras palavras, sob o efeito de pressões seletivas impostas pela administração de antibióticos, a seleção de elementos genéticos móveis que codifiquem para resistência aos referidos compostos juntamente com características genotípicas que confiram virulência irá ocorrer, tendo como consequência a sua disseminação dentro de comunidades bacterianas pertencentes ao microbiota humano. Tanto quanto sabemos, não existem registos bibliográficos sobre a dinâmica evolutiva que dita a epidemiologia de genes conferentes de resistência a antibióticos, e os que conferem virulência, colectivamente. Assim sendo, a presente dissertação pergunta se sob o efeito de pressões selectivas exercidas por antibióticos, os determinantes de resistência e de virulˆencia se encontram co-representados tanto em diversos microbiomas ambientais, como em microbiomas provenientes do trato gastrointestinal humano. Deste modo, foram escolhidos metagenomas a fim de abordar esta temática por várias razões. A mais preeminente prende-se com o facto das bactérias serem organismos sociais, vivendo em comunidades. Um metagenoma corresponde à panóplia de material genético isolado de uma comunidade, e posteriormente sequenciado, pelo que caracteriza o repertório completo de genes envolvidos em processos metabólicos, fisiológicos e ecológicos, como por exemplo, na adaptação ao ambiente pelas comunidades microbianas presentes na respectiva amostra sequenciada. Subsequentemente, a prospeção de genes em metagenomas surge como uma metodologia fidedigna no que toca ao estudo das pressões seletivas a que uma dada população bacteriana foi sujeita, assim como ao estudo da co-seleção de características genéticas do microbiota amostrado como um todo. No presente trabalho utilizamos 64 metagenomas ambientais, referentes a 12 biomas diversos, bem como 110 metagenomas do trato gastrointestinal humano, originários de indivíduos pertencentes a várias comunidades dos Estados Unidos da América, Venezuela, e Malawi, caracterizando várias faixas etárias, diferentes culturas, hábitos alimentares, bem como diferentes graus de acesso a saneamento básico, a cuidados médicos e antibióticos. Todos os metagenomas encontram-se publicamente disponíveis para download no servidor MG-RAST, tendo sido descarregados em ficheiros individuais em formato FASTA, nos dias 3 de Abril de 2015 (metagenomas do trato gastrointestinal humano) e 17 de Novembro de 2015 (metagenomas ambientais). Cada ficheiro compreende sequências proteicas previamente agrupadas a 90% de homologia pela pipeline de formatação de ficheiros do servidor MG-RAST, contendo assim sequências traduzidas não redundantes, e representando deste modo a diversidade proteica de cada metagenoma. O programa BLASTP foi utilizado a fim de inferir homologia de sequências proteicas envolvidas no fenómeno de resistência a antibióticos, bem como de virulência, de entre os metagenomas escolhidos, fazendo uso de duas bases-de-dados públicas: Resfams AR Proteins database (base-de-dados de proteínas bacterianas conferentes de resistência a antibióticos), e VFDB (base-de-dados de proteínas bacterianas envolvidas em virulência). Este programa permite inferir homologia entre sequências comparadas por via de um algoritmo de alinhamento de sequências derivado do algoritmo original de Smith-Waterman. De entre os vários critérios de seriação aplicáveis foi escolhido um limiar de E-value = 1e-15, com um filtro posterior que apenas considera o melhor alinhamento para cada sequência proteica, e que satisfaça os requisitos mínimos de 60% de homologia sob 75% de alinhamento entre sequências comparadas. Ulteriormente ainda se removeram os alinhamentos resultantes de sequências proteicas que tanto eram homologas de determinantes de resistência a antibióticos como de factores de virulência, de modo a eliminar um viés na análise estatística consecutivamente implementada. Seguidamente, de modo a aferir o tipo de relação entre os caracteres genéticos em causa, as contagens das diferentes sequências proteicas homologas de determinantes de resistência a antibióticos (ARd) foram correlacionadas numa primeira fase com o número de sequências proteicas presentes em cada metagenoma, previamente agrupadas a 90% (tamanho do metagenoma), procedendo de igual forma para as contagens de diversidade de sequências proteicas homologas de factores de virulência (VFd), para ambos os grupos de metagenomas considerados. Posteriormente as contagens ARd e VFd foram correlacionadas entre si, após a estandardização das mesmas. Em ordem a medir o grau de associação entre as correlações previamente descritas recorreu-se a medidas estatísticas como o coeficiente de correlação e o rho de Spearman, bem como o seu P-value. Foram também geradas todas as possíveis associações entre as contagens de ARd e VFd para subfamílias proteicas funcionais caracterizadas nas bases-de-dados mencionadas, efetuando uma correção aos P-values resultantes do rho de Spearman pelo procedimento de Benjamini-Hochberg. Em ordem a testar a dissemelhança entre médias provenientes dos rácios estandardizados de ARd/VFd em função da idade dos indivíduos pertencentes ao grupo de metagenomas do trato gastrointestinal humano, foram aplicados Welch Two Sample t-tests por pares, de acordo com os respectivos países de origem. Os nossos resultados mostram que os determinantes de resistência a antibióticos, bem como os factores de virulência, se encontram amplamente disseminados tanto em microbiomas ambientais, como em microbiomas do trato gastrointestinal humano pertencentes aos 110 indivíduos saudáveis originários de países diferentes. Em segundo lugar, também sugerem que, apesar das comunidades bacterianas ambientais possuírem maior variação de ARd e VFd tendo em conta o tamanho dos metagenomas, as comunidades que habitam o trato gastrointestinal humano detêm uma dependência linear muito forte no que toca à distribuição de ARd de acordo com o tamanho dos metagenomas, e uma relação linear forte entre VFd e o tamanho dos mesmos. Adicionalmente constatamos que as contagens estandardizadas de ARd e VFd apresentam uma correlação muito forte entre si nos metagenomas de origem ambiental, sendo que estas contagens também se mostraram fortemente correlacionadas no grupo de metagenomas provenientes do trato gastrointestinal humano. Entre os metagenomas do grupo anterior, os referentes aos indivíduos originários dos Estados Unidos da América, apresentam uma ampla diversidade de associações, ao passo que as amostras provenientes de indivíduos Venezuelanos não possuem uma associação estatisticamente relevante. No entanto, os metagenomas pertencentes a indivíduos Malauianos retratam a correlação e associação linear mais forte de entre os três países amostrados, possuindo também duas vezes mais contagens de ARd por VFd que os outros dois países. Referindo-nos ainda ao mesmo conjunto de metagenomas, conseguimos verificar que as contagens estandardizadas de ARd e VFd aparentam decrescer com o aumento da idade dos indivíduos, enquanto que os rácios ARd/VFd afiguram-se relativamente estáveis, evidenciando um incremento comum durante o primeiro ano de vida. Relativamente a todas as associações geradas entre subfamílias de proteínas que codificam para AR e as que codificam VFs, aquelas que se relacionam com o envelope celular bacteriano apresentaram as melhores correlações e estatísticas correspondentes. É de salientar que os resultados descritos neste trabalho apenas fornecem evidência para a co-representação de determinantes de AR e VF entre os metagenomas ambientais e do trato gastrointestinal humano que foram amostrados. Visto que a inclusão ou proximidade de determinantes AR e VF nos mesmos elementos genéticos m´oveis não se prende com os objectivos da presente dissertação, os nossos resultados não podem confirmar que a mobilização dos demais determinantes esteja a ocorrer conjuntamente. De qualquer modo, a natureza co-representativa dos nossos resultados reforça a noção, bem como a hipótese de co-seleção dos referidos determinantes.
Genes contributing to the pathogenicity of a particular bacterial species are often grouped in pathogenicity islands, and encoded on mobile genetic elements such as plasmids or phages, as happens with some genes coding for resistance to antibiotics. Pathogenic bacteria have gradually become resistant to antibiotics as a result of intense selective pressure they are subjected to. Here we provide evidence that further reinforces the hypothesis on which, under antibiotics selective pressure, resistance and virulence traits are co-selected amongst bacterial communities naturally occurring in the human gut microbiome. Through means of metagenome mining, we have studied 64 environmental metagenomes from 12 diverse biomes, as well as 110 human gut metagenomes issuing from individuals belonging to different human populations across the world, having contrastive cultural, dietary and sanitary lifestyles, along with different medical access to antibiotics. Our results demonstrate that there is a great diversity of antibiotic resistance (AR) and virulence factor (VF) genetic traits amongst metagenomes in general. In the human gut there are less AR and VF genetic traits than in more versatile environments, yet the correlations between the latter determinants are still strong, advocating that in the human gut microbiome, there appears to be co-selection of these traits, remaining well established and long lasting in the foregoing host’s microbiome. In the USA human gut metagenomes there are a few examples of AR determinants per VF accumulation, suggesting a possible consequence of antibiotic consumption abuse. In Malawi, a very poor African country, where there is a high prevalence of unattended antibiotic consumption, the correlation between AR determinants and VFs is very strong, as opposed to the scenario portrayed by the metagenomes pertaining to Amerindians (native populations of the Venezuelan Amazon) where there is neither reports of pharmaceutical-grade antibiotics consumption nor correlation at all, thus allowing us to link this effect to antibiotic exposure. Furthermore, the best correlations gathered between AR and VF protein sub-families amidst both metagenomic cohorts relate to the bacterial cell envelope, namely multidrug efflux pump components (AR determinants), along with secretion systems, adhesion proteins and iron-acquisition systems (VFs), most of which are known to be encoded within mobile genetic elements.

The human microbiome, the complex and dynamic ecosystem that populates our body, performs essential functions such as aiding digestion and protecting us from harmful pathogens. An increasing number of diseases are found to be associated with a shift of balance - or dysbiosis - of the microbiome. However, we still know little about this delicate balance and how it depends on different microbial functions. In this thesis project, I used metagenomic sequencing data to study the variability of microbes and their functions in different areas of the human body. First, in an attempt to characterize the dysbiosis associated with periodontitis, I examined the microbial community of the oral cavity in presence and absence of this chronic inflammatory disease. Specifically, I catalogued the phylogenetic signatures composed of tetramer nucleotide frequencies and observed that the disease state occupies a much narrower region than the healthy one. This result suggests that upon onset of the disease, through host cell invasion, pathogenic bacteria may find a more consistent environment for their parasitic lifestyle. Motivated by these findings, I sought further evidence of an environment-specific use of metabolic functions in the oral and gut communities. Rather than focusing on the abundance of individual metabolic functions, I evaluated their diversity, i.e., the extent to which these functions are performed by different classes of organisms. My hypothesis was that such diversity may confer increased robustness to taxonomic variability. Using metagenomic sequencing data and NCBI's Protein Clusters database, I characterized the multiplicity of gene families associated with a given metabolic function. I found that different human body sites display different degrees of metabolic functional diversity, as assessed by Shannon entropy. For some well-studied gene functions, such as those involved in glycolytic pathways, I found entropy signatures consistent with the known degree of oxygen availability of different environmental niches. Conversely, in an unsupervised analysis, I identified functions with nontrivial entropy signatures. These results pave the way for a new way to inspect human microbiome activity, and could help understand its functional resilience and suggest ways to shift its balance towards healthy configurations.

博士
國立交通大學
生物資訊及系統生物研究所
102
The human body plays host to a vast array of bacteria which are harmful or beneficial. Next generation sequencing technology have increased its accuracy in identifying bacteria. This work develops a novel platform for Metagenomics analysis, bacterial disease risk evaluation model construction and microbiome characterization in human using high-throughput 16S rRNA sequencing. A database that recorded the bacteria 16s rRNA sequences from literature and public database was constructed, along with quality filtering, chimeric sequence detection, sequence clustering and assigning the taxonomy of the sequenced 16S rRNA sequences. A bacteria disease risk model for seven diseases was constructed based on 174 samples and 78 bacterial disease risk biomarkers. Applicability of the proposed platform is demonstrated by collecting the microbiome in human gut of 81 samples and 53 samples with BMI. For analyzing bacteria related disease, this work expects to collect more bacterial disease biomarkers from human body. Additionally, we design case-control study to finding novel association between bacteria and disease for extending bacterial disease risk evaluation model. The proposed platform provides a relatively easy means of identifying a certain amount of bacteria and evaluating bacterial disease risk based on Taiwanese control group for clinical microbiology applications. Detecting how bacteria inhabit humans and affect their health significantly contributes to develop a diagnosis and treatment method.

The microbial communities that live inside the human gastrointestinal tract -the human gut microbiome- are important for host health and wellbeing. Characterizing this new “organ”, made up of as many cells as the human body itself, has recently become possible through technological advances. Metagenomics, the high-throughput sequencing of DNA directly from microbial communities, enables us to take genomic snapshots of thousands of microbes living together in this complex ecosystem, without the need for isolating and growing them. Quantifying the composition of the human gut microbiome allows us to investigate its properties and connect it to host physiology and disease. The wealth of such connections was unexpected and is probably still underestimated. Due to the fact that most of our dietary as well as medicinal intake affects the microbiome and that the microbiome itself interacts with our immune system through a multitude of pathways, many mechanisms have been proposed to explain the observed correlations, though most have yet to be understood in depth. An obvious prerequisite to characterizing the microbiome and its interactions with the host is the accurate quantification of its composition, i.e. determining which microbes are present and in what numbers they occur. Historically, standard practices have existed for sample handling, DNA extraction and data analysis for many years. However, these were generally developed for single microbe cultures and it is not always feasible to implement them in large scale metagenomic studies. Partly because of this and partly because of the excitement that new technology brings about, the first metagenomic studies each took the liberty to define their own approach and protocols. From early meta-analysis of these studies it became clear that the differences in sample handling, as well as differences in computational approaches, made comparisons across studies very difficult. This restricts our ability to cross-validate findings of individual studies and to pool samples from larger cohorts. To address the pressing need for standardization, we undertook an extensive comparison of 21 different DNA extraction methods as well as a series of other sample manipulations that affect quantification. We developed a number of criteria for determining the measurement quality in the absence of a mock community and used these to propose best practices for sampling, DNA extraction and library preparation. If these were to be accepted as standards in the field, it would greatly improve comparability across studies, which would dramatically increase the power of our inferences and our ability to draw general conclusions about the microbiome. Most metagenomics studies involve comparisons between microbial communities, for example between fecal samples from cases and controls. A multitude of approaches have been proposed to calculate community dissimilarities (beta diversity) and they are often combined with various preprocessing techniques. Direct metagenomics quantification usually counts sequencing reads mapped to specific taxonomic units, which can be species, genera, etc. Due to technology-inherent differences in sampling depth, normalizing counts is necessary, for instance by dividing each count by the sum of all counts in a sample (i.e. total sum scaling), or by subsampling. To derive a single value for community (dis-)similarity, multiple distance measures have been proposed. Although it is theoretically difficult to benchmark these approaches, we developed a biologically motivated framework in which distance measures can be evaluated. This highlights the importance of data transformations and their impact on the measured distances. Building on our experience with accurate abundance estimation and data preprocessing techniques, we can now try and understand some of the basic properties of microbial communities. In 2011, it was proposed that the space of genus level variation of the human gut microbial community is structured into three basic types, termed enterotypes. These were described in a multi-country cohort, so as to be independent of geography, age and other host properties. Operationally defined through a clustering approach, they are “densely populated areas in a multidimensional space of community composition”(source) and were proposed as a general stratifier for the human population. Later studies that applied this concept to other datasets raised concerns about the optimum number of clusters and robustness of the clustering approach. This heralded a long standing debate about the existence of structure and the best ways to determine and capture it. Here, we reconsider the concept of enterotypes, in the context of the vastly increased amounts of available data. We propose a refined framework in which the different types should be thought of as weak attractors in compositional space and we try to implement an approach to determining which attractor a sample is closest to. To this end, we train a classifier on a reference dataset to assign membership to new samples. This way, enterotypes assignment is no longer dataset dependent and effects due to biased sampling are minimized. Using a model in which we assume the existence of three enterotypes characterized by the same driver genera, as originally postulated, we show the relevance of this stratification and propose it to be used in a clinical setting as a potential marker for disease development. Moreover, we believe that these attractors underline different rules of community assembly and we recommend they be accounted for when analyzing gut microbiome samples. While enterotypes describe structure in the community at genus level, metagenomic sequencing can in principle achieve single-nucleotide resolution, allowing us to identify single nucleotide polymorphisms (SNPs) and other genomic variants in the gut microbiome. Analysis methodology for this level of resolution has only recently been developed and little exploration has been done to date. Assessing SNPs in a large, multinational cohort, we discovered that the landscape of genomic variation seems highly structured even beyond species resolution, indicating that clearly distinguishable subspecies are prevalent among gut microbes. In several cases, these subspecies exhibit geo-stratification, with some subspecies only found in the Chinese population. Generally however, they present only minor dispersion limitations and are seen across most of our study populations. Within one individual, one subspecies is commonly found to dominate and only rarely are several subspecies observed to co-occur in the same ecosystem. Analysis of longitudinal data indicates that the dominant subspecies remains stable over periods of more than three years. When interrogating their functional properties we find many differences, with specific ones appearing relevant to the host. For example, we identify a subspecies of E. rectale that is lacking the flagellum operon and find its presence to be significantly associated with lower body mass index and lower insulin resistance of their hosts; it also correlates with higher microbial community diversity. These associations could not be seen at the species level (where multiple subspecies are convoluted), which illustrates the importance of this increased resolution for a more comprehensive understanding of microbial interactions within the microbiome and with the host. Taken together, our results provide a rigorous basis for performing comparative metagenomics of the human gut, encompassing recommendations for both experimental sample processing and computational analysis. We furthermore refine the concept of community stratification into enterotypes, develop a reference-based approach for enterotype assignment and provide compelling evidence for their relevance. Lastly, by harnessing the full resolution of metagenomics, we discover a highly structured genomic variation landscape below the microbial species level and identify common subspecies of the human gut microbiome. By developing these high-precision metagenomics analysis tools, we thus hope to contribute to a greatly improved understanding of the properties and dynamics of the human gut microbiome
Die mikrobiellen Gemeinschaften innerhalb des menschlichen Darmtrakts – das menschliche Darm-Mikrobiom - sind wichtig für das Wohlbefinden und die Gesundheit des Wirts. Die Charakterisierung dieses neuen “Organs”, welches aus ähnlich vielen Zellen besteht wie der menschliche Körper, ist in jüngster Zeit durch technologische Fortschritte möglich geworden. Die Metagenomik, die direkte Hochdurchsatz-Sequenzierung mikrobieller DNA, ermöglicht die Aufnahme “genomischer Schnappschüsse” tausender verschiedener, in einem komplexen Ökosystem zusammenlebender Bakterien, ohne dafür auf deren Isolierung und Wachstum angewiesen zu sein. Die Quantifizierung des menschlichen Mikrobioms erlaubt es uns, seine Eigenschaften zu untersuchen und Verbindungen zu Wirtsphysiologie und -krankheiten zu knüpfen. Der Reichtum dieser Informationen ist unerwartet hoch und wahrscheinlich noch immer unterbewertet. Aufgrund der Tatsache, dass der Großteil unserer Ernährung und unseres Medikamentenkonsums unser Mikrobiom, welches wiederum selbst über verschiedene Arten mit unserem Immunsystem interagiert, beeinflusst, wurden viele Mechanismen vorgeschlagen, um die beobachteten Korrelationen zu erklären. Die meisten davon sind jedoch noch nicht vollständig verstanden. Eine offensichtliche Komponente zur Charakterisierung des Mikrobioms und dessen Interaktionen mit dem Wirt ist eine akkurate Quantifizierung seiner genauen Zusammensetzung, womit sowohl die Anwesenheit von bestimmten Bakterien als auch deren Anzahl gemeint ist. Obwohl etablierte Standardprozeduren zur Probenbehandlung, DNA- Extrahierung und Datenanalyse existieren, sind sie nicht immer für metagenomische Studien anwendbar, da sie für isolierte Bakterienkulturen entwickelt worden. Deswegen und auch wegen der Begeisterung, die neuartige Technologien mit sich bringen, nahmen sich die ersten metagenomischen Studien jeweils die Freiheit, ihre eigenen Protokolle und Herangehensweisen zu definieren. Die Metaanalyse dieser Studien zeigte, dass Unterschiede sowohl in der Probenbehandlung als auch in der statistischen Auswertung den Vergleich zwischen Studien sehr schwierig machen. Das wiederum beschneidet unsere Fähigkeit, Entdeckungen zu bestätigen und Daten über Studien hinweg zu kombinieren. Um die zwingend notwendige Standardisierung voranzutreiben haben wir einen umfassenden Vergleich von 21 verschiedenen DNA-Extraktionsmethoden sowie verschiedener weiterer Probenbehandlungen, welche Quantifizierungen beeinflussen, vorgenommen. Wir haben eine Reihe von Kriterien entwickelt, um die Messqualität in Abwesenheit von Mock-Kontrollen zu bestimmen und schlagen anhand dieser Methoden für Probenbeschaffung, DNA-Extraktion und Library- Generierung optimale Verfahren vor. Wenn diese als Standard akzeptiert werden, würde das eine stark verbesserte Vergleichbarkeit zwischen Studien ermöglichen und damit sowohl einen extremen Zuwachs an statistischer Power als auch unserer Fähigkeit, generelle Schlüsse über das Mikrobiom zu ziehen, zur Folge haben. Die meisten metagenomischen Studien teilen ihre Datensätze auf um Vergleiche anzustellen, z.B. zwischen Stuhlproben gesunder und erkrankter Menschen. Eine Vielzahl verschiedener Ansätze, welche wiederum oft mit verschiedenen Datenvorbehandlungen kombiniert werden, wurden vorgeschlagen, um Dissimilarität zwischen Gemeinschaften (Beta-Diversität) zu berechnen. Um metagenomische Daten auf Spezies-, Genus- und höheren Ebenen zu quantifizieren werden üblicherweise reads auf Referenzgenome bestimmter taxonomischer Einheiten aligniert und gezählt. Aufgrund technologieabhängiger Unterschiede in Sequenziertiefe müssen reads normalisiert werden, z.B. indem man alle counts durch die Gesamtanzahl der counts einer Sequenzierung teilt (total sum scaling), oder durch subsampling. Für die Messung der Gemeinschafts(dis)similarität wurden viele Distanzmaße vorgeschlagen. Da es schwierig ist diese Ansätze theoretisch zu vergleichen, haben wir ein biologisch motiviertes Konzept entwickelt, mit dem man Distanzmaße evaluieren kann. Dies unterstreicht die Wichtigkeit der Datentransformation und dessen Einwirkung auf Distanzmaße. Aufbauend auf unserer Erfahrung mit Häufigkeitsabschätzungen und Techniken zur Datenvorbehandlung können wir nun versuchen, grundlegende Eigenschaften mikrobieller Gemeinschaften zu verstehen. 2011 wurde vorgeschlagen, dass sich die Variation auf Genusebene im menschlichen Darm auf drei grundlegende Typen beschränkt, welche Enterotypen getauft wurden. Diese wurden in Datensätzen verschiedener Länder als unabhängig von Herkunft, Alter und anderer Wirtseigenschaften beschrieben. Die Enterotypen sind durch einen Cluster-Ansatz als „dicht besiedelte Bereiche in einem multidimensionalen Raum der Gemeinschaftszusammensetzung“ definiert und wurden als grundlegende Stratifikatoren für die menschlichen Population vorgeschlagen. Spätere Studien, welche dieses Konzept auf andere Datensätze anwandten, erhoben Zweifel bezüglich der optimalen Anzahl an Clustern und an der generellen Robustheit des Ansatzes. Dies leitete erneut eine langanhaltende Debate über die Existenz von Strukturen und die besten Wege, diese zu bestimmen und einzufangen, ein. Hier überdenken wir, in Anbetracht der stark gestiegenen Anzahl an verfügbaren Daten, das Enterotypen-Konzept. Wir schlagen ein überarbeitetes Konzept vor, in welchem die verschiedenen Enterotypen als schwache Attraktoren im multidimensionalen Raum verstanden werden und implementieren einen Ansatz zur Berechnung des Attraktors, der dem Datensatz am ähnlichsten ist. Dafür trainieren wir einen Klassifizierer auf einen Referenz- Datensatz, um neue Datensätze zuzuordnen. Damit ist Enterotypisierung nicht mehr datensatzabhängig und der Effekt von sampling bias ist minimiert. Indem wir ein Modell nutzen für das wir die Existenz dreier Enterotypen (definiert durch die selben Genera wie ursprünglich postuliert) annehmen, zeigen wir die Relevanz dieser Stratifikation und schlagen es in einem klinischen Zusammenhang als potentiellen Marker für Krankheitsfortschritt vor. Außerdem glauben wir, dass diese Attraktoren verschiedene Regeln mikrobieller Zusammensetzung widerspiegeln und schlagen vor, sie bei der Analyse von mikrobiellen Daten zu berücksichtigen. Während Enterotypen Struktur in der Gemeinschaft auf Genusebene beschreiben, kann metagenomische Sequenzierung prinzipiell Auflösung auf Nukleotidebene erreichen, womit single nucleotide polymorphisms (SNPs) und andere genomische Variationen im Darm- Mikrobiom identifiziert werden können. Analysemethoden für dieses Auflösungsniveau wurden erst kürzlich entwickelt und bis heute wurden diese erst wenig erforscht. Wir zeigen, dass die Landschaft an genomischer Variation von SNPs in einer großen, multinationalen Kohorte sogar über die Speziesebene hinaus geht und hochgradig strukturiert ist, was das Vorkommen klar abgrenzbarer Subspezies unter Darmmikroben suggeriert. In mehreren Fällen zeigen diese Subspezies geographische Stratifikation, wobei einige Subspezies nur in chinesischen Populationen vorkommen. Im Allgemein zeigen Sie jedoch nur eine geringfügige Beschränkung der Dispersion und sind in der Mehrzahl der Populationen vorhanden. Innerhalb eines Individuums dominiert häufig eine bestimmte Subspezies, nur selten dominieren verschieden gemeinsam im gleichen Ökosystem. Eine Analyse von Zeitreihenexperimenten deutet darauf hin, dass die dominante Subspezies über Zeiträume von mehr als drei Jahren stabil bleibt. Wenn man ihre funktionalen Eigenschaften untersucht findet man viele Unterschiede, von denen bestimmte relevant für den Wirt erscheinen. Zum Beispiel identifizieren wir eine Subspezies von E. rectale, welcher das Flagellum-Operon fehlt, die signifikant assoziiert ist mit geringerem BMI und geringerer Insulinresistenz ihres Wirts; sie korreliert zudem mit höherer mikrobieller Diversität. Diese Assoziationen konnten auf Speziesebene nicht gesehen werden (auf der mehrere Subspezies überlagert sind), was die Wichtigkeit dieser erhöhten Auflösung für ein umfassenderes Verständnis mikrobieller Interaktionen innerhalb des Mikrobioms und mit dem Wirt illustriert. Zusammenfassend bieten unsere Ergebnisse eine präzise Grundlage für vergleichende Metagenomik des menschlichen Darms, einschließlich Empfehlungen über experimentelles Sampling und statistische Analysen. Weiterhin verfeinern wir das Konzept der Enterotypen- Stratifikation in Gemeinschaften, entwickeln referenzbasierte Ansätze für Enterotypen- Zuordnung und bieten überzeugende Beweise für ihre Relevanz. Indem wir die volle Auflösung metagenomischer Sequenzierungen nutzen entdecken wir eine Landschaft hochgradig strukturierter genomischer Variation unterhalb der Speziesebene und identifizieren gemeinsame Subspezies des menschlichen Darm-Mikrobioms. Durch die Entwicklung dieser hochpräzisen metagenomischen Untersuchungsansätze tragen wir zu einem verbesserten

Metagenomics, or genomic sequence of the community of microbiota (bacteria, fungi, virus), enables an investigation of the full complement of genetic material, including virulence, antibiotic resistance, and strain differentiating markers. The granularity to distinguish between closely related strains is important as within one species, these strains possess distinct functions and relationships to a host. To analyze metagenomic samples, I developed a reference-based approach that utilizes both single nucleotide variants and genetic content to assign species and strain-level designations. After refining this approach with complex simulated communities, I utilized it to analyze the microbial communities present in skin samples from healthy and diseased individuals. First, to investigate strain-level heterogeneity in healthy adults, I focused on the common skin commensals Propionibacterium acnes and Staphylococcus epidermidis with well-documented sequence variation. Results indicated that an individual’s strains of P. acnes are shared across multiple sites of his or her body, and that those strains are more similar within than between individuals. For S. epidermidis, in addition to individual site similarities, there were also site-specific strains. Overall these results emphasize that both individuality and site specificity shape our bodies’ microbial communities. Based on longitudinal data, an individual’s strain signatures remain stable for up to a year despite external, environmental perturbations. I then used metagenomic data to explore microbial temporal dynamics in atopic dermatitis (AD; eczema), an inflammatory skin disease commonly associated with Staphylococcal species. Species-level investigation of AD flares demonstrated a microbial dichotomy in which S. aureus predominated on more severely affected patients while S. epidermidis predominated on less severely affected patients. Strain-level analysis determined that S. aureus-predominant patients were monocolonized with distinct S. aureus strains, while all patients had heterogeneous S. epidermidis strain communities. To assess the host immunologic effects of these species, I topically applied patient-derived strains to mice. AD strains of S. aureus were sufficient to elicit a skin immune response, characteristic of AD patients. This suggests a model whereby staphylococcal strains contribute to AD progression through activation of the host immune system. Overall, this strain-level analysis of healthy and disease communities provides previously unexplored resolution of human skin microbiome.
2018-03-24T00:00:00Z

Histologically, colorectal cancer (CRC) resides in the abnormal proliferation of epithelial cells of the colon mucosa, progressing from adenoma to adenocarcinoma. This cancer continues to be the third with the highest incidence and mortality worldwide. It is caused by an accumulation of genetic mutations and epigenetic silencing, in addition to other intrinsic and extrinsic risk factors. Due to the high rates of incidence and mortality, diagnostic and prevention tools have been created, implemented, and optimized. However, the need to develop tools that provide an increasingly early, rigorous and sensitive diagnosis remains a pressing need. In this sense, the objectives of this dissertation were: (1) to develop the state of the art about CRC diagnostic tools, (2) to summarize "omic" applications in order to identify microbial biomarkers related to CRC and (3) to compare the microbiome of Portuguese patients with CRC and healthy individuals. Currently, the diagnosis of CRC has been driven by more (e.g., colonoscopy) or less (e.g., imaging techniques, molecular biomarkers) invasive procedures. Recently, the search for microbial biomarkers through "omic" tools has been an alternative, mainly due to the relevance of the microbiome in the metabolic and physiological homeostasis, as well as in the functioning of the host immune system. Thus, the intestinal microbiome has been assigned an active role in the evolution of CRC, being able to influence or be influenced by the disease. In particular, the metagenomic and metabolomic analysis of the CRC-associated microbiome in stool samples has stimulated the scientific community in the search for sensitive, reliable, differential, stable, and early biomarkers in the non-invasive detection of the disease. However, these advances lack representativeness for several geographic areas, given the cultural, genetic, and environmental impact on the incidence of this disease. In this sense, the microbiome (with a focus on Bacteria) was analyzed in feces of two clinical groups constituted by Portuguese individuals (patients with CRC and healthy individuals), through the sequencing of the 16S rRNA gene using Ilumina MiSeq. This study is a contribution to fill the gap of existing knowledge about the microbiome associated with CRC in the Portuguese population. Although the structure of the fecal microbiome assumes homogeneous patterns among individuals of the same clinical group, there was some variability in the abundance of taxa between these groups and at different stages of CRC. For example, Prevotella, Alloprevotella, Sutterella, Desulfovibrio and Olsenella observed in CRC samples can serve as microbial biomarkers. In the future, the study will be extended to larger population samples, as well as to other types of human samples and clinical groups, in order to identify sensitive and specific microbial signatures that can translate the development of CRC, thus reducing incidence rates and mortality.
Do ponto de vista histológico, o cancro coloretal (CCR) reside na proliferação anormal de células epiteliais da mucosa do cólon, progredindo de adenoma a adenocarcinoma. Este cancro continua a ser o terceiro com maior incidência e mortalidade mundialmente. É causado por um acúmulo de mutações genéticas e silenciamento epigenético, para além de outros fatores de risco intrínsecos e extrínsecos. Devido às altas taxas de incidência e mortalidade, têm vindo a ser criadas, implementadas e otimizadas ferramentas de diagnóstico e prevenção. No entanto, continua premente a necessidade de desenvolver ferramentas que forneçam um diagnóstico cada vez mais precoce, rigoroso e sensível. Neste sentido, os objetivos desta dissertação consistiram em (1) desenvolver o estado da arte acerca das ferramentas de diagnóstico de CCR, (2) resumir as aplicações “ômicas” para indentificar biomarcadores microbianos relacionados com CCR e (3) comparar o microbioma de pacientes portugueses com CCR e indivíduos saudáveis. Atualmente, o diagnóstico de CCR tem vindo a ser conduzido por procedimentos mais (e.g., colonoscopia) ou menos (e.g., técnicas de imagem, biomarcadores moleculares) invasivos. Muito recentemente, a procura de biomarcadores microbianos através de ferramentas “ómicas” tem sido uma alternativa, principalmente devido à relevância do microbioma nas homeostase metabólica e fisiológica, assim como no funcionamento do sistema imunitário do hospedeiro. Assim, ao microbioma intestinal tem sido atribuído um papel ativo na evolução do CCR, podendo influenciar ou ser influenciado pela doença. Em particular, a análise metagenómica e metabolómica do microbioma associado a CCR em amostras de fezes tem estimulado a comunidade científica na procura de biomarcadores sensíveis, fidedignos, diferenciais, estáveis e precoces na deteção não invasiva da doença. Contudo, estes avanços carecem de uma representatividade para diversas áreas geográficas, dada o impacto cultural, genético e ambiental na incidência desta doença. Neste sentido, realizou-se a análise do microbioma (com enfoque em Bacteria) em fezes de dois grupos clínicos constituídos por indivíduos Portugueses (pacientes com CCR e indivíduos saudáveis), através da sequenciação do gene 16S rRNA usando Ilumina MiSeq. Este estudo é um contributo para colmatar a lacuna de conhecimento existente sobre o microbioma associado a CCR na população Portuguesa. Apesar da estrutura do microbioma de fezes assumir padrões homogéneos entre indivíduos do mesmo grupo clínico, houve alguma variabilidade na abundância de taxa entre esses grupos e em diferentes estádios do CCR. Por exemplo, maiores abundâncias de Prevotella, Alloprevotella, Sutterella, Desulfovibrio e Olsenella observadas em amostras de CCR podem servir como biomarcadores microbianos. No futuro, o estudo será alargado a amostras populacionais maiores, assim como a outro tipo de amostras humanas e grupos clínicos, no sentido de identificar assinaturas microbianas sensíveis e específicas, que possam traduzir o desenvolvimento de CCR, reduzindo, assim, as taxas de incidência e mortalidade.
Mestrado em Biologia Molecular e Celular