Fundamental metabolic strategies of heterotrophic bacteria

Abstract

Through their metabolism, heterotrophic microbes drive carbon cycling in many environments. These microbes consume (and produce) hundreds to thousands of different metabolic substrates, begging the question of what level of description is required to understand the metabolic processes structuring their communities: do we need to account for the detailed metabolic capabilities of each organism, or can these capabilities be understood in terms of a few well-conserved carbon utilization strategies that could be more easily interpreted and more robustly predicted? Based on the high-throughput phenotyping of a diverse collection of marine bacteria, we showed that the fundamental metabolic strategy of heterotrophic microbes can be understood in terms of a single axis of variation, representing their preference for either glycolytic (sugars) or gluconeogenic (amino and organic acids) carbon sources. Moreover, an organism's position on this axis is imprinted in its genome, allowing us to successfully predict metabolic strategy across the bacterial tree of life. Our analysis also unveils a novel and general association between metabolic strategy and genomic GC content, which we hypothesize results from the difference in C:N supply associated with typical sugar and acid substrates. Thus, our work reveals a fundamental constraint on microbial evolution that structures bacterial genomes and communities and can be leveraged to understand diversity in functional terms, beyond catalogs of genes and taxa.

Metabolic similarity and the predictability of microbial community assembly

“…Because these traits are quantitatively similar in phylogenetically related taxa, related species can fulfill equivalent functional roles leading to communities with similar coarse-grain phylogenetic compositions despite differences at the species level (Martiny et al 2015; Aguirre de Cárcer 2019; Estrela, Vila, et al 2022). Additionally, recent studies involving high-throughput measurements of microbial growth phenotypes and pairwise inter-specific interactions in different environments have shown that different environments can sometimes be ’coarse- grained’ and grouped together based on the chemical class of substrate present (Tong et al 2020; Kehe et al 2021; Gralka, Pollak, and Cordero 2022). These two observations suggest that similar initial metabolic environments may lead to quantitative similarities in metabolic traits, which we reasoned would lead to selection for quantitatively similar communities.…”

“…S4B). Recent studies have classified carbon sources based on whether they enter metabolism via glycolysis (glycolytic) or the TCA cycle (gluconeogenic) (Gralka et al 2022, Borer et al 2022). When we classify our 19 carbon sources along this binary, we find that all eight carbon sources on which E+ grew significantly faster than P+ were glycolytic, whereas all nine carbon sources on which P+ grew significantly faster than E+ were gluconeogenic (Fig.…”

The architecture of metabolic networks constrains the evolution of microbial resource hierarchies

“…The questions about the ecology and evolution of metabolic hierarchies have received renewed a ention in recent years (Bajic and Sanchez 2020;Okano et al 2021), as part of the ongoing effort to understand the drivers of microbial community assembly and coexistence (Chang et al 2022;Estrela et al 2022;Gralka et al 2022). Recent theoretical work (Posfai et al 2017;Goyal et al 2018;Pacciani-Mori et al 2020;Wang et al 2021;Bloxham et al 2023) and experiments with model communities (Pacciani-Mori et al 2020;Bloxham et al 2022) have shown that differences in metabolic hierarchies can impact ecology, e.g.…”

Statistical mechanics of phenotypic eco-evolution: from adaptive dynamics to complex diversification

“…Consequently, evolutionary effects cannot be decoupled from ecological ones, as both can occur at contemporary timescales [68, 69] and microbial communities often exhibit a very large fine-scale diversity in the form of multiple co-occurring phenotypes [70]. Importantly, such a diversity is nowadays accessible to experiments owing to technological advances in determining single-cell traits [31, 71, 72] and metabolic functions [73–76]. All this calls for the development of novel eco-evolutionary frameworks —extending existing ones— to analyze complex microbial communities, with individuals distributed in phenotypic space and evolving on ecological time scales.…”

“…Importantly, such a diversity is nowadays accessible to experiments owing to technological advances in determining single-cell traits [31,71,72] and metabolic functions [73][74][75][76]. All this calls for the development of novel ecoevolutionary frameworks -extending existing ones-to analyze complex microbial communities, with individuals distributed in phenotypic space and evolving on ecological time scales.…”