Vitamin interdependencies predicted by metagenomics-informed network analyses validated in microbial community microcosms

Microbial communities are fundamental components of most ecosystems but little is known about the interactions that structure them. Metagenomic data have revolutionized our understanding of complex consortia, yet predicted interactions are rarely explored experimentally. We developed an organism abundance correlation network to investigate factors that control community organization in soil-derived laboratory microbial consortia grown under dozens of conditions. The network was overlaid with metagenomic information about functional capacities to generate testable hypotheses. We developed a metric to predict the importance of each node within its local network environments relative to correlated vitamin auxotrophs and predicted that a Variovorax species is a hub because it is a highly important source of thiamine. Quantification of thiamine during the growth of Variovorax in minimal media conditions showed extraordinarily high levels of production of thiamine, up to 100 mg/L. This over-production of thiamine explains why the auxotrophs correlated with Variovorax are reliant on this organism for thiamine, despite the presence of other predicted thiamine producers in the community. A few of these thiamine auxotrophs are predicted to produce pantothenate that we show that Variovorax requires to grow, implying that a subset of vitamin-dependent interactions are mutualistic. We found that Cryptococcus produces the B-vitamin pantothenate, and co-cultures led to a 90-130-fold fitness increase for both Cryptococcus and Variovorax. This study demonstrates the predictive power of metagenome-informed, microbial consortia-based network analyses for identifying microbial interactions that underpin the structure and functioning of microbial communities.