S159: Metagenomics and proteomic analysis of feedstock-adapted thermophilic enrichments

Biomass degrading communities provide a rich source of novel lignocellulolytic enzymes. In previous work, we demonstrated the feasibility of deriving a range of full-length, active cellulases and hemicellulases from metagenomic sequences of a switchgrass degrading compost community, but limited sequence assembly of complex metagenome data proved a significant bottleneck. Here, we explore a more targeted and efficient approach, focusing on lower-complexity enrichment communities that are easier to assemble and more likely to produce enzymes adapted to relevant feedstocks and conditions, guided by activity assays of the community as a whole, and of the mixture of enzymes they secrete. Metagenomic sequencing of the most active enrichment community resulted in a Gigabase of 454 and Illumina shotgun sequence data, which assembled into nearly 9000 contigs between 1kb and 140kb, and 37,000 proteins with function predictions. Phylogenetic binning identified several thermophilic species, as well as a fairly complete genome for a new member of the Gemmatimonadetes, a large phylum with only a single sequenced representative. We identified hundreds of proteins in the secretome based on LC-MS/MS proteomics. The secretome contained far more hemicellulases than cellulases – a trend which is also present in the metagenome itself – as well as sugar isomerases that may play a role in relieving product inhibition. The presence of superoxide dismutase and catalase may indicate oxidative stress dues to active bacterial oxidative degradation of lignin. A variety of proteins from this enrichment are currently in the pipeline to be expressed and tested for enzymatic activity.