Reconstruction of Metabolic Models and Prediction of Flux Profile for Complex Electrosynthetic Microbial Community Based on Metatranscriptome Data

Microbial electrosynthesis systems (MESs) are a new emerging route for the production of biofuels and potentially high-value organic compounds using CO₂and electricity as feedstocks. This mechanism for biosynthesis can serve as a productive outlet for excess electrical energy produced during non-peak hours or due to development of natural energy sources that are distant from regions of high electricity demand. Thermodynamic and biological challenges exist in the development of single-species microbial electrosynthesis processes, but mixed microbial communities confer high potential production rates and versatile production profiles compared to other MESs.  

We have collected deep metagenome and metatranscriptome data for a stable MES comprised of 11 microbes [1], none of which have been successfully cultured in isolation. We assembled 11 complete genomes and constructed genome-scale metabolic models based on our metagenome reads [2]. We then mapped our metatranscriptome data to these genomes and applied novel approaches to gapfill and fit our draft models to this gene expression data, maximizing agreement between active reactions in our models and actively expressed genes in our genomes. We demonstrate how transcriptome data may be used in combination with metabolic models to improve gapfilling of models during reconstruction and to permit the qualitative prediction of flux profiles. This approach is particularly valuable in analysis of multi-species communities, where annotations, genomes, and models are generally lower quality.

Additionally, we applied our approach to fit our entire microbiome model to our metatranscriptome data, predicting the web of trophic interactions between the species comprising our community during both heterotrophic and electrosynthetic growth.  Our predictions provide an accurate estimate of carbon fixers in the community, as well as interdependencies between community members. These methods demonstrate how a metagenome and metatranscriptome can generally be applied to produce a compartmentalized model of a microbial community, capable of predicting community flux profiles, trophic interactions between community members, and overall community consumption and production of nutrients. All tools and approaches described have been integrated into the DOE Knowledgebase and are available for general use.

1.         Marshall, C.W., et al., Electrosynthesis of commodity chemicals by an autotrophic microbial community. Appl Environ Microbiol, 2012. 78(23): p. 8412-20.

2.         Henry, C.S., et al., High-throughput generation, optimization, and analysis of genome-scale metabolic models. Nature Biotechnology, 2010. Nbt.1672: p. 1-6.