Generation of genome-scale metabolic reconstructions for 773 members of the human gut microbiota

Abstract

1 r e s o u r c eChanges in the composition of the human gut microbiota have been associated with the development of chronic diseases including type 2 diabetes, obesity, and colorectal cancer 1 . Gut bacterial functions, such as synthesis of amino acids and vitamins 2 , breakdown of indigestible plant polysaccharides 3 , and production of metabolites involved in energy metabolism 4 , have been linked to human health. The use of 'omics approaches to study human microbiome communities has led to the generation of enormous data sets whose interpretations require systems biology tools to shed light on the functional capacity of gut microbiomes and their interactions with the human host 5 .In order to infer the metabolic repertoire of a gut metagenome data set, researchers usually map sequenced genes or organisms onto metabolic networks derived from the Kyoto Encyclopedia of Genes and Genomes (KEGG) 6 , and functional annotations from KEGG ontologies 7 . However, this approach cannot identify the contribution of each bacterial species to the metabolic repertoire of the whole gut microbiome, nor can it infer the effects of different gut microbial communities on host metabolism.A technique that can bridge this gap is constraint-based reconstruction and analysis (COBRA) 8 using genome-scale metabolic reconstructions (GENREs) of individual human gut microbes. GENREs are assembled using the genome sequence and experimental information 9 . These reconstructions form the basis for the development of condition-specific metabolic models whose functions are simulated and validated by comparison with experimental results. The models can be used to investigate genotype-phenotype relationships 10 , microbe-microbe interactions 11 , and host-microbe interactions 11 . Numerous tools can be used to automatically generate draft GENREs but such models contain errors 12 and are incomplete.Manual curation of draft reconstructions is time consuming because it involves an extensive literature review and experimental validation of metabolic functions 9 .To provide an extensive resource of GENREs for human gut microbes, we developed a comparative metabolic reconstruction method that enables any refinement to one metabolic reconstruction to be propagated to others. This accelerates reconstruction and improves model quality. We generated AGORA, which includes 773 gut microbes, comprising 205 genera and 605 species. All reconstructions were based on literature-derived experimental data and comparative genomics. The metabolic predictions of two AGORA reconstructions and their derived metabolic models were validated against experimental data. RESULTS Metabolic reconstruction pipelineWe devised a comparative metabolic reconstruction method (Fig. 1a,c), which is analogous to the comparative microbial genome annotation approach 13 that has enabled accelerated annotation by propagation of refinements to one genome to others. First, we downloaded draft GENREs using Model SEED 14 and KBase (US Department of Energy Systems Biology Knowledgebase, http:/...

Fast automated reconstruction of genome-scale metabolic models for microbial species and communities

“…In a recent study from our group, the growth of 96 phylogenetically diverse gut bacteria was characterized across 19 different media (15 defined and 4 complex media compositions) (43). From this list, a total of 47 bacteria are also included in the AGORA collection, a recently published collection of 773 semi-manually curated models of human gut bacteria (57). These models, however, could not reproduce the metabolic needs experimentally observed by Tramontano et al ., highlighting the need for the use of validated growth media for species that are distant from model organisms.…”

The Relationship Between the Human Genome and Microbiome Comes into View

“…Humans acquire riboflavin (vitamin B 2 ) both through their diet and from riboflavin-producing gut microbes. The machinery required for riboflavin synthesis has been found in the genomes of most Bacteroidetes, Fusobacteria, and Proteobacteria examined, while a complete riboflavin operon was only present in about half of the Firmicutes and almost no Actinobacteria (44, 61). Riboflavin can be used as a redox mediator by Faecalibacterium prausnitzii to facilitate extracellular electron transfer, which consequently promotes its growth (34).…”

Gut microbiota, metabolism and psychopathology: A critical review and novel perspectives

“…The analysis of associations between these metabolite patterns and psychiatric symptoms calls for application of state of the art systems biology approaches and subsequent intervention studies to validate candidate metabolites. Recent advances in analysis of the complete metabolic repertoire of 753 bacterial strains [49] already make it possible to predict which metabolites are produced in ensembles of bacteria allowing development of targeted metabolomics.…”