Abstract Liver abscesses (LA) are a polymicrobial infection and the causative agents include two subspecies of Fusobacterium necrophorum, necrophorum and funduliforme, and Trueperella pyogenes. Nucleic acid-based analysis of the purulent material of LA has indicated that Proteobacteria and Bacteroidetes were the dominant phyla, next to phylum Fusobacteria. The sporadic isolations of bacterial species of the Proteobacteria and Bacteroidetes in LA have been reported. An aspect of the pathogenesis that has not been investigated is whether bacterial pathogens in LA originate, besides the rumen, from the colon? Therefore, our objectives were to target detection and isolations of Salmonella enterica, Escherichia coli, Klebsiella pneumoniae, Pseudomonas aeruginosa and Bacteroides fragilis, in addition to routine isolations of F. necrophorum and T. pyogenes, from LA and also from ruminal and colonic epithelial tissues of the same animals. Liver abscesses, matched with ruminal and colonic epithelial tissue samples from 96 steers and heifers, originating from 15 feedlots, were collected at slaughter, transported to the laboratory and processed within 24 hours. Samples were subjected to anaerobic and aerobic culture methods, including enrichment and selective media for detection and isolation of S. enterica, E. coli, K. pneumonia, P. aeruginosa, and B. fragilis. Prevalence of bacterial species were shown in table 1. For each bacterium, two-sided Fisher's exact test was used to evaluate the association of its prevalence in LA with that in colon and rumen at 0.05 level. Overall, prevalence of subsp. necrophorum, subsp. funduliforme, T. pyogenes and S. enterica in liver abscesses were 86.4% (83/96), 21.8% (21/96), 35.4% (34/96) and 6.2% (6/96), respectively (Table 1). In ruminal tissue samples, prevalence of subsp. necrophorum, subsp. funduliforme, T. pyogenes and S. enterica were 6.6% (6/96), 40.6% (39/96), 11.4% (11/96) and 4.1% (4/96), respectively. Prevalence of subsp. necrophorum, subsp. funduliforme, T. pyogenes and S. enterica in colonic tissues were 18.7% (18/96), 36.4% (35/96), 0% (0/96) and 7.2% (7/96), respectively. Prevalence of E. coli, Klebsiella spp., and P. aeruginosa in LA samples were 70.8% (68/96), 92.7% (89/96), and 91.6% (88/96), respectively. Prevalence of E. coli, Klebsiella spp., and P. aeruginosa in rumen epithelial tissue samples were 17.7% (17/96), 50.0%, (48/96), and 15.6% (15/96), respectively. Prevalence of E. coli, Klebsiella spp., and P. aeruginosa in colonic tissue samples were 16.6% (16/96), 24.0% (26/96), and 12.5% (12/96), respectively. Prevalence of subsp. funduliforme in ruminal and colonic tissues was greater than subsp. necrophorum. None of the liver abscess samples yielded B. fragilis. In conclusion, E. coli was the second most dominant species detected and isolated from LA, next only to subsp. necrophorum. The greater frequencies of isolations of subsp. necrophorum and S. enterica from colonic epithelial than ruminal epithelial tissues suggest that hindgut could also be a source of pathogens involved in LA.