Metabolic profiling and genome-scale modelling uncover mechanistic drivers of microbiome stability in synthetic maize root community

Stability is a desirable property for agricultural microbiomes, but there is a poor understanding of the mechanisms that mediate microbial community stability. A representative bacterial synthetic community from maize roots has been proposed by Niu et al. (2017, PNAS, 114:E2450) as a model system to study microbiome stability. This SynCom assembles stably when all seven members are present, but community diversity collapses without the keystone E. ludwigii strain. In this study, we used complementary in vitro experiments and in silico metabolic modelling to assess the role of metabolites for the stability of this SynCom, by defining the metabolic niches occupied by each strain, as well as their cross-feeding phenotypes and B-vitamin dependencies. We show that the individual member strains occupy complementary metabolic niches, measured by the depletion of distinct metabolites in exometabolomic experiments, as well as contrasting growth phenotypes on diverse carbon substrates, patterns which are largely recapitulated by computational simulations. Minimal medium experiments show that the established seven-member community comprises a mixture of prototrophic and auxotrophic strains. Correspondingly, experimental and in silico cross-feeding phenotypes showed that spent media harvested from the prototrophic strains can sustain growth of two auxotrophs and let to the identification of B-vitamin dependencies. Altogether, this study highlights the complementary power of in vitro and in silico approaches and suggests that the metabolic mechanisms of this SynCom can serve as design principles to inform the rational assembly of stable plant-associated microbial communities. This Zenodo submission provides all LC-MS datafiles associated with the study (.MZXML format).