Analysis of bovine rumen microbiota under different dietary regimens for identification of feedstock-targeted cellulolytic genes

Lead Project PIs: Eddy Rubin, Tao Zhang, Susannah Green Tringe, Roderick Mackie and Edward DePeters
Lead Campus: UC Berkeley

The lack of robust cellulolytic enzymes for efficient conversion of cellulosic biomass into fermentable sugars is a major bottleneck of ethanol production from biomass. Currently bioethanol production relies on chemical and physical pretreatments that are not only expensive but also generate waste products which represent major environmental pollutants. Yet in the gut environments of ruminants this enzymatic process takes place rapidly and at moderate pH and temperature. We aim to identify enzymes produced by the abundant microbes responsible for degradation of plant cell wall polymers in forage-feeding animals that can potentially be co-opted for cellulosic biomass conversion. The fistulated cow offers easy, nondestructive access to the rumen, making it possible to sample large volumes of material and even to isolate different feeds within the same animal using synthetic bags. In this study, switchgrass and control substrates in synthetic porous bags will be incubated in the bovine rumen and feedstuff-associated microbes will be sampled for DNA and RNA extraction.  High throughput sequencing of expressed genes will be used to identify gene sets specifically upregulated in switchgrass-bound organisms and therefore likely involved in lignocellulose breakdown.  A second phase of targeted sequencing and computational and biochemical investigation will further characterize gene products implicated in digestion of lignocellulosic substrates and evaluate their potential for industrial applications.

Background

Cellulosic crops such as switchgrass are promising feedstocks for biomass-to-ethanol conversion processes aimed at replacing petroleum as a transportation fuel.  However, several obstacles still stand in the way of widespread, economical use of biofuels, one of which is the relative inefficiency with which the recalcitrant carbohydrates in cellulosic feedstocks can be depolymerized into fermentable sugars.  Yet ruminant animals depend on such material as their primary energy source, suggesting that they possess enzymes capable of efficient depolymerization of cellulosic feedstocks.  As this ability is known to rely heavily on the intestinal microbiota, sequencing the transcribed genes of the gut microbial community in feedstock-fed ruminants will specifically identify enzymes that target those feedstocks.

Ruminants such as cows and sheep possess a large fermentation tank, the rumen, that harbors a dense and diverse microbial community, enabling the digestion of high-fiber plant matter such as grasses. Rumen contents can be readily sampled via surgically created fistulas fitted with removable cannulas, and synthetic nylon bags containing feedstuffs can be incubated in the rumen to monitor digestion in situ.  A key advantage of the rumen microcosm is the availability of large quantities of material, and the opportunity to sequester specific feedstuff-associated microbes within the compartment.  While the diverse microbial community inhabiting the rumen is a proven source of novel hydrolytic enzymes, most rumen microbes are not readily culturable so in vitro enrichment techniques such as growth on purified substrates are ineffective for capturing the full cellulolytic potential of the community. Thus we propose a nucleic-acid based metagenomic approach to identify microbial genes specifically expressed in organisms digesting a high-fiber substrate.  Because the community under study will be actively degrading plant biomass, we plan to focus on expressed genes in addition to the entire genomes as this should provide us with the complete picture of the microbial community present in the rumen and highlight specifically the enzymes involved in fiber degradation.

Our initial transcript sequencing will use a high-throughput next-generation sequencing technology, 454 pyrosequencing, to deeply sample the range of expressed genes in the rumen community.  These data will identify potential cellulolytic genes specifically expressed in microbes digesting switchgrass.  Since the relatively short reads provided by this technology (200-250 bp) will not encompass complete genes, a second experimental phase will screen large-insert libraries constructed from rumen community DNA for genes of interest identified in the first phase, followed by complete sequencing and functional characterization. A 16S/18S ribosomal RNA sequence analysis done in parallel with metatranscriptomic analysis will shed light on the community changes associated with different feedstocks and will enhance our knowledge of the organisms that produce the biocatalysts required for a complete breakdown of lignocellulosic biomass.

Questions to be addressed

1) What organisms and families of cellulolytic enzymes are involved in the digestion of feedstock substrates in the cow rumen?  Studying the microbes involved in feedstock degradation in a natural environment will lead us toward an understanding of the enzymatic requirements for lignocellulose breakdown.

2) What is the diversity of cellulolytic enzymes in the rumen, and what features do they share in common?  An expanded set of feedstock-targeted enzymes will aid in identifying key functional domains and features necessary for the digestion of specific substrates.