Identification and Metabolic Characterization of a Novel Strain of Prevotella Ruminicola Using Starch As Substrate in Anaerobic Batch Cultures

The bovine rumen is a complex and diverse microbial ecosystem composed of bacteria, archaea, protozoa, and fungi. Bacteria are the predominant populations, play a vital role in the fermentation of feedstuffs into SCFAs that can be utilized by their host as a main energy source. In intensive management systems cereal grains, which consist of starch as primary component, are used as diet ingredients to provide an easily fermentable source of carbohydrates for ruminants. In animals fed concentrate diets, rumen amylolytic bacteria break down starch to glucose, which is then used for growth and synthesis of microbial proteins. Based on reported metagenomics studies, uncharacterized amylolytic bacteria far outnumber starch utilizers that have been analyzed to date. This indicates that there are significant gaps in our knowledge about a functional group of rumen microorganisms that are critical to current production strategies. To gain further insight, the primary objective of this research was to identify uncharacterized bacteria involved in metabolizing starch using a 16S rRNA-based approach, then predict their metabolic capacity using metagenomics. For this purpose, 3 treatment replicate cultures from rumen fluid supplemented with starch (ADM Corn processing Clinton, Iowa) were compared to 2 replicate cultures that received no supplementation. PCR amplification of the 16S rRNA gene (V1-V3 region), followed by Illumina Miseq 2×300 sequencing, were used to determine the microbial composition of treatment and control cultures. One species-level Operational Taxonomic Unit (OTU), corresponding to an uncultured strain of Prevotella ruminicola was found to be enriched on days 7 (15.6±3.22) and 14 (17.7±5.02) in the treatment group cultures. When compared to control cultures (0.17±0.08), differences were found to be statistically significant (ANOVA, p<0.05). Shotgun metagenomics on one of the enriched day 7 starch sample generated 8.4 million sequence reads, which were assembled into contigs using the de novo assembly program ABYSS. From a total of 1,041 contigs of length greater than 3,000kb that were obtained, 440 contigs that were assigned to the phylum Bacteroidetes, and thus predicted to belong to the OTU of interest, were selected for further analysis. Annotation using RAST revealed genes that were expected to be involved in starch utilization encoding enzymes such as alpha-amylase, lactate dehydrogenase, as well as ethanol dehydrogenase. A better understanding of the metabolic capabilities of novel strains will provide new avenues for developing supplements or feed additives that can improve animal health and increase the production efficiency of the ruminant livestock.