A Computational Odyssey into Microbial Communities Across Ecosystems

This thesis uses mainly computational data-driven approaches to study variation, taxonomy and functional characterization, interactions between members and underlying principles that govern microbial communities across ecosystems. The importance to study microbes in the natural context of complex systems of communities was highlighted in Chapter 1. In Chapter 2 we studied the variation of the microbial community composition during spontaneous in vitro wine fermentation of riesling must. We made the following observations: (i) There is a general influence of the vineyard on microbial composition with a striking differential abundance of Metschnikowia. (ii) There is a decrease in biodiversity during alcoholic fermentation. Unexpectedly, the fraction of Micrococcus increased in one vineyard during alcoholic fermentation. (iii) There is a relation between stuck fermentations and the abundance of Starmerella. In Chapter 3, we continued to explore microbial communities during wine fermentation. Specifically, the role of the adjunct Lactobacillus plantarum in the malolactic conversion of industrial wine fermentation was investigated and this species was found to thrive better on white than on red wine fermentations. We obtained experimental evidence to support the hypothesis that a successful introduction of this species in a community was in the case of wine determined by the composition of the must, and possibly by the presence of grape skins during fermentation. In Chapter 5 a simple experimental coloring method is presented which distinguishes colonies of the yeasts Lachancea thermotolerans and Saccharomyces cerevisiae on agar media. It does so by the addition of bromocresol purple which induces Lachancea colonies to develop a brown color, whereas Saccharomyces colonies remain white. In Chapter 3 we also studied a kefir community using genome sequences of isolated and sequenced Bacteria representative strains. We found that Lactobacillus kefiranofaciens, a dominant organism in kefir, stands out among the lactobacilli because it potentially has a high number of amino acid auxotrophies. Also, the only organism in kefir that had genes for flagellar assembly and chemotaxis was Acetobacter. The presence of flagella in Acetobacter was experimentally confirmed. In Chapter 4 we study pairwise interactions between microbes using measures based on genes involved in metabolic processes. In the case of microbes from the urinary tract, a number of putative metabolic interactions were identified that could explain the experimentally obtained pairwise growth effects. We found that members of Enterococcus may be complemented in their metabolism by the other members of the community. In Chapter 6 we investigated the metabolism of a novel Mycobacterium species, which was found to be dominant in a microbial community residing in an acidic biofilm attached to the wall of a sulfur cave in Romania. This Mycobacterium species expresses a full suite of enzymes involved in methanotrophic growth. Growth experiments using methane as the sole carbon- and free energy source verify its methanotrophic niche. To our knowledge, this is the first report about a methanotrophic Mycobacterium of Actinobacteria. In Chapter 7 a key question in microbial ecology is asked - how large biodiversity can be maintained on a few resources. To address this question a highly diverse microbial community was investigated, which grew for 15 years in an anoxic bioreactor on benzene as the main carbon and free energy source and nitrate as an electron acceptor. We found evidence that many different niches are present and while only a few community members seem to degrade benzene, the majority of species seems to feed on metabolic left-overs, microbial necromass or even autotrophically using anaerobic ammonium oxidation for free energy transduction and carbon fixation. An additional succession experiment verified that the same few community members are the actual drivers of benzene degradation