093 16S characterization of liver abscesses in Western United States feedlot cattle.

Liver abscesses are a major economic burden to beef producers. Although different causative organisms have been isolated from purulent material, there is a knowledge gap concerning the complete characterization of the polymicrobial community of a liver abscess. The objective of this study was to characterize liver purulent material across common beef cattle production systems in the western United States using the 16S rRNA gene. Sixteen pens of cattle, from 5 different feedlots, were identified to represent 2 exposure groups: tylosin fed and no tylosin (8 pens per exposure group). From each pen, up to 5 liver abscesses were collected at harvest. Due to sampling limitations (some pens had fewer than 5 abscesses) and the variability in purulent material volume (resulting from the stage and size of the abscess), DNA from 36 liver samples was successfully extracted. Libraries from DNA samples were prepared and sequenced on an Illumina HiSeq 2500 using the 515F/806R primer set for amplification of the V4 region of the 16S rRNA. Sequenced forward and reverse reads were concatenated into 1 contig and raw sequencing reads were clustered into operational taxonomic units. Through 16S characterization, 5 phyla, 11 classes, and 13 orders were identified in the DNA extracted from an abscess. The phyla identified were Bacteroidetes (30.4%), Fusobacteria (30.1%), Proteobacteria (21.9%), Firmicutes (10.6%) and Actinobacteria (5%), with the remaining 2% unclassified. Fusobacterium and Trueperella were both present in the abscess communities, and both are known causative agents of liver abscesses. Of note, the top 3 phyla, which make up 82% of the abscess community, were Gram-negative. This result is surprising, as tylosin, a macrolide, is primarily active against Gram-positive bacteria. Although the isolation of Gram-negative bacteria from liver abscesses is commonly observed, the concentration of Gram-negative bacteria in the microbial community suggests that conventional mechanisms of prevention may not always result in the absence of an abscess. Despite significant effects of variables such as treatment (P = 0.026) and geographic region (P = 0.002) on the microbiome phylogenetic composition, the overall composition at the phylum level was very similar across samples and the 5 most abundant taxa were present in all samples. Because our goal was to describe overall microbiome of purulent material, treatment and region were not considered for final summary analyses. This is the first time the complex community of liver purulent material has been described. These data may suggest a more targeted approach to liver abscess prevention and treatment.