A tremendous number of microorganisms are known to interact with their animal hosts. The outcome of the interactions between microbes and their animal hosts range from modulating the maintenance of homeostasis to the establishment of processes leading to pathogenesis. Of the numerous species known to inhabit humans, the great majority live on mucosal surfaces which are highly defended. Despite their importance in human health, little is known about the molecular and cellular basis of most host-microbe interactions across the tremendous diversity of mucosal-adapted microorganisms. The ever-increasing availability of genome sequence data allows systematic comparative genomics studies to identify proteins with potential important molecular functions at the host-microbe interface. In this study, a genome-wide analysis was performed on 3,021,490 protein sequences derived from 867 complete microbial genome sequences across the three domains of cellular life. The ability of microbes to thrive successfully in a mucosal environment was examined in relation to functional genomics data from a range of publicly available databases. Particular emphasis was placed on the extracytoplasmic proteins of microorganisms that thrive on human mucosal surfaces. These proteins form the interface between the complex host-microbe and microbe-microbe interactions. The large amounts of data involved, combined with the numerous analytical techniques that need to be performed makes the study intractable with conventional bioinformatics. The lack of habitat annotations for microorganisms further compounds the problem of identifying the microbial extracytoplasmic proteins playing important roles in the mucosal environments. In order to address these problems, a distributed high throughput computational workflow was developed, and a system for mining biomedical literature was trained to automatically identify microorganisms’ habitats. The workflow integrated existing bioinformatics tools to identify and characterise protein-targeting signals, cell surface-anchoring features, protein domains and protein families. This study successfully demonstrated a large-scale comparative genomics approach utilising a system called Microbase to harness Grid and Cloud computing technologies. A number of conserved protein domains and families that are significantly associated with a speiii iv cific set of mucosa-inhabiting microorganisms were identified. These conserved protein regions of which their functions were either characterised or unknown, were quite narrow in their coverage of taxa distribution, with only a few protein domains more widely distributed, suggesting that mucosal microorganisms evolved different solutions in their strategies and mechanisms for their survival in the host mucosal environments. Metabolic and biological processes common to many mucosal microorganisms included: carbohydrate and amino acid metabolisms, signal transduction, adhesion to host tissues or contents in mucosal environments (e.g. food remnants, mucins), and resistance to host defence mechanisms. Invasive or virulence factors were also identified in pathogenic strains. Several extracytoplasmic protein families were shared among prominent bacterial members of gut microbiota and microbial eukaryotes known to thrive in the same environment, suggesting that the ability of microbes to adapt to particular niches can be influenced by lateral gene transfer. A large number of conserved regions or protein families that potentially play important roles in the mucosa-microbe interactions were revealed by this study. Several of these candidates were proteins of unknown function. The identified candidates were subjected to more detailed computational analysis providing hypothesis for their function that will be tested experimentally in order to contribute to our understanding of the complex host-microbe interactions. Among the candidates of unknown function, a novel M60-like domain was identified. The domain was deposited in the Pfam database with accession number PF13402. The M60-like domain is shared amongst a broad range of mucosal microorganisms as well as their vertebrate hosts. Bioinformatics analyses of the M60-like domain suggested a potential catalytic function of the conserved motif as gluzincins metalloproteases. Targeting signals were detected across microbial M60-likecontaining proteins. Mucosa-related carbohydrate-binding modules (CBMs), CBM32 was also identified on several proteins containing M60-like domains encoded by known mucosal commensals and pathogens. The co-occurrence of the CBMs and M60-like domain, as well as annotated potential peptidase function unveiled a new functional context for the CBM, which is typically connected with carbohydrate processing enzymes but not proteases. The CBM domains linked with members of different protease families are likely to enable these proteases to bind to specific glycoproteins from host animals further highlighting the importance of proteases and CBMs (CBM32 and CBM5_12) in host-microbe interactions.

Zinc is an essential trace element in biological systems. For example, it activates many enzymes, acts as a cellular membrane stabiliser and is a constituent of the zinc finger proteins that bind specific DNA sequences, playing a critical role in gene expression and genome modification. The present study has focused on the influence of zinc on cell physiology of the yeast Saccharomyces cerevisiae. Zinc uptake by industrial strains of S. cerevisiae, including brewing strains, and the subsequent utilisation of this key metal during fermentation was studied. Yeast strains take up most of the available zinc very quickly following pitching, with some strains increasing their cellular zinc ten-fold at the onset of fermentation. Zinc content of yeast cell walls was found to remain constant during fermentation and zinc was localised in the vacuole. These and other findings indicated that most of the initial zinc uptake was metabolism-dependent, rather than via a cell surface biosorption phenomenon. After initial periods of cellular zinc accumulation, and during the course of the subsequent fermentation, zinc became virtually undetectable in wort. As yeast cells grew during this period, zinc was distributed to daughter cells at cell division and this effectively lowered their individual cellular zinc concentration. Depending on the extent of yeast growth during fermentation, this may result in the generation of zinc-depleted biomass at the time of yeast harvesting. A brewing yeast strain was also investigated following exposure to stresses typically encountered during the brewing process and a relationship existed between zinc loss from cells and decrease in cell viability during various environmental insults. In order to extend initial laboratory studies to industrial scale, yeast fermentative performance was investigated under various zinc concentrations, firstly in 1 L conical vessels and subsequently in 200 L brewing fermenters. A cell physiological study was also made of zinc-limitation in the haploid reference strain S. cerevisiae CEN.PK113-7D in chemostat continuous culture cultivations, in order to pave the way for more in-depth studies aimed at identifying zinc-responsive molecular biomarkers in yeasts. The ultimate aim of such a study was to provide a valuable and rapid way to determine the status of cellular zinc in yeast during fermentation. The consequences of this for efficient industrial processes are discussed including implications for brewing fermentation optimisation based on control of wort zinc bioavailability. Overall, this research has provided new insights into the influence of zinc on yeast cell physiology and the fundamental information gained has practical implications for yeast-based biotechnologies.

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.

Au sein des sols forestiers, la richesse taxonomique et le rôle des microorganismes eucaryotes (en grande partie des champignons) restent encore largement méconnus. L’espèce d’arbre est un des facteurs qui structurent les communautés de ces microorganismes. Nous avons étudié l’impact de l’essence forestière (hêtre et épicéa) sur la diversité taxonomique et fonctionnelle de ces communautés par une approche métatranscriptomique et une approche biochimique (focalisée sur les champignons ectomycorhiziens). Nous avons montré un effet de la séquence étudiée (ADNr 18S, ADNc) sur la distribution taxonomique des communautés et développé un nouveau marqueur moléculaire mitochondrial pour l’étude des communautés de champignons métaboliquement actifs. L’identification de gènes d’intérêt écologique et industriel par séquençage systématique des banques métatranscriptomiques ainsi que l’identification fonctionnelle d’une nouvelle famille de transporteursmembranaires montrent l’intérêt de l’approche métatranscriptomique. L’approche biochimique a consisté en un dosage à haut débit, sur des extrémités racinaires ectomycorhizés, d’activités enzymatiques liées à la dégradation de la matière organique et à la mobilisation de l’azote et du phosphore du sol. L’ensemble de ces approches a permis de montrer un impact de l’essence forestière sur la nature des espèces présentes plutôt que sur la richesse taxonomique et une préférence d’hôte de certains groupes fongiques ectomycorhiziens. L’approche biochimique a montré une redondance fonctionnelle importante pour certaines activités enzymatiques tandis qu’une autre activité enzymatique était spécifique d’un groupe taxonomique fongique. / In forest soils, taxonomic richness and functional diversity of eukaryotic microorganisms (mainly Fungi) remain largely unknowned. Tree species is one of the main factors that structure eukaryotic microbial communities. We have studied the impact of tree species (beech and spruce) on taxonomic and functional diversity of these communities by using a metatranscriptomic approach and a biochemical one focusing on ectomycorrhizal fungi. We showed an effet of different sequences (18S rDNA, cDNA) on taxonomic composition of eukaryotic microbial communities and we developped anew mitochondrial molecular marker for the study of metabolically active fungal communities. Identification of ecologically and industrially important genes by the shotgun sequencing of metatranscriptomic libraries and also identification of a new family of transmembrane transporter demonstrate the great potential of the metatranscriptomic approach. The biochemical approachconsisted in a multiple enzymatic test carried out on ectomycorrhizal roots, of enzyme activities linked to organic matter degradation and phosphorus and nitrogen mobilization. All these approaches revealed an impact of tree species on the microbial species composition but not on taxonomic richness and also host preference for some ectomycorrhizal taxonomic groups. The biochemical approach showed a high functional redundancy for some enzyme activities while one activity was very specific of an ectomycorrhizal taxonomic group.