The Serres lab focuses on annotating genome sequences of selected microbes with the goal of understanding their metabolic capabilities. We compare the metabolic potentials between phylogenetically related organisms as well as between organisms that co-inhabit specific environments. Through these comparative studies we aim to understand what metabolic capabilities are selected for and how these capabilities are diversified as a result of environmental factors and evolutionary forces. For our curation work we make use of bioinformatics tools to identify sequence similarities and characteristics as well as the primary literature for experimentally derived functional data. The genome annotations are maintained in BioCyc pathway genome databases that include tools for displaying metabolic pathways and overview diagrams as well as visualizing and analyzing omics datasets (i.e. proteome, transcriptome, and metabolome).

We have created and curated pathway genome databases for genomes of the Shewanella genus as well as of autotrophs (Cyanobacteria) and heterotrophs (Chloroflexi) known to co-exist in the environment or as co-cultures in the lab. The microbial interaction studies are part of the Pacific Northwest National Laboratory lead Scientific Focus Area; Biological Systems Interactions. The project is funded by Department of Energy and addresses interactions that are occurring between microbes and their environment and how these interactions infer stability, robustness and efficiency to the community. Alkaline phototrophic and Fe(II)-oxidizing mat communities of Yellowstone National Park and a phototrophic mat community of central Washington have been chosen for study. In addition representative strains are co-cultured in order to understand and model their interactions. Bioinformatics-based predictions and experimental data generated by several co-investigators contribute to these interaction studies. Our pathway genome databases are maintained via Bay Paul Center servers where they are accessed by our collaborators.

A bioinformatics based comparison of the metabolic capabilities of two microbes, the autotroph Synechococcus sp. PCC 7002 and the heterotroph Shewanella sp. W3-18-1 has been done by our group. These organisms were chosen as a model to develop and test predictions of molecular exchanges during co-culture growth. They have been co-cultured without the supplement of nutrients supporting that their metabolisms can be coupled. The cellular overview diagrams below shows the gene products and metabolic pathways that have been annotated in the pathway genome databases. Highlighted in red are the reactions that are common to both of the organisms. Reactions shown in blue are those that are present in the organism but absent from the other organism. The majority of the core metabolic pathways are shared by in both organisms. Shewanella specific genes are involved in biosynthesis of the siderophore putrebactin, the signaling molecule AI-2, synthesis and degradation of arabinose and several amino acids. Synechococcus specific genes are involved in biosynthesis of the siderophore synechobactin, secondary metabolites, and the compatible solutes glucosyl glycerol and glucosylglycerate. Strain-specific functions may indicate traits that contribute to the ability of organisms to co-exist and not consume the same resources. For example, Shewanella is capable of arabinose degradation and this sugar has been detected in the EPS of several cyanobacteria. As part of our analysis we also compared the transport functions encoded by these two organisms and identified compounds that could be transported by both or by only one of the strains. These compounds provide additional clues to what compounds may be involved in the metabolic coupling during co-culturing. They include vitamins, N-sources, fermentation products, and sugars. Gene expression data is being generated of the co-cultures by our collaborators, and this dataset will be analyzed against our predictions. We will be comparing additional co-existing microorganisms in the future as well as incorporating protein and metabolite measurements in our analyses.