Multi-faceted approaches to discovering and predicting microbial nutritional interactions

Abstract

Nearly all microbes rely on other species in their environment to provide nutrients they are unable to produce. Nutritional interactions include not only the exchange of carbon and nitrogen compounds, but also amino acids and cofactors. Interactions involving cross-feeding of cobamides, the vitamin B₁₂ family of cofactors, have been developed as a model for nutritional interactions across species and environments. In addition to experimental studies, new developments in culture-independent methodologies such as genomics and modeling now enable the prediction of nutritional interactions in a broad range of organisms including those that cannot be cultured in the laboratory. New insights into the mechanisms and evolution of microbial nutritional interactions are beginning to emerge by combining experimental, genomic, and modeling approaches.

Figure 1.. Structural and functional diversity of cobamides.

A. Structure of cobalamin (vitamin B₁₂) and various upper and lower ligands of cobamides. Cobamides consist of a central corrin ring, an upper ligand (R), and a lower ligand (boxed), covalently attached via the nucleotide loop. Upper ligands confer chemical activity by generating a radical or donating a methyl group. Lower ligands are diverse and often do not directly participate in chemical reactions, but confer enzyme specificity. B. Examples of specificity of microbes for three cobamides with the lower ligands shown. Check mark indicates cobamides that can be used; X indicates cobamides that poorly support growth or enzyme activity; blanks indicate cobamides that were not tested [–29,31,33,47].

Figure 2.. Uneven distribution of cobamide production and use in bacteria.

Distribution of the de novo cobamide biosynthesis pathways and cobamide-dependent pathways among all bacteria (left) and for the four most abundant phyla in the dataset. Percentages larger than 1 are rounded to the closest integer. Adapted from [25]. Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0).

Figure 3.. Multi-layered framework for modeling the evolution of cross-feeding in a co-culture undergoing successive gene loss.

A. Three-layered framework to model evolution of cross-feeding interactions in microbial metabolic networks. Flux balance analysis (FBA) is employed to iteratively predict instantaneous growth, metabolite uptake and metabolite secretion rates of each species (bottom). The behavior of the individual species is fed into a co-culture model to simulate community growth in a shared environment allowing for metabolite exchange (middle). Reductive evolution is performed by iteratively deleting metabolic genes at random (excluding genes with major fitness defects) until no more genes that meet this condition can be removed from either species (top). B. Schematic illustration of the development of cross-feeding (left) and metabolic missed opportunities (right). Panel A adapted from [44]. Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0).