Age-dependent differences in gut microbial communities and resistomes of veal calves

Authors: SALAHEEN, SEARAJUST - University Of Maryland; SEON-WOO, KIM - University Of Maryland; HOVINGH, ERNEST - Pennsylvania State University; Haley, Bradd; Van Kessel, Jo Ann

Submitted to: American Society for Microbiology
Publication Type: Abstract Only
Publication Acceptance Date: 5/30/2018
Publication Date: 6/4/2018
Citation: Salaheen, S., Seon-Woo, K., Hovingh, E., Haley, B.J., Van Kessel, J.S. 2018. Age-dependent differences in gut microbial communities and resistomes of veal calves. American Society for Microbiology. p. 1.

Interpretive Summary:

Technical Abstract: Background: Age-associated differences in the carriage of antibiotic resistant (AR) bacteria have been demonstrated using traditional culturing methods for a variety of food animals. In this study, we used shotgun metagenomic sequencing followed by network analysis and machine-learning approaches to investigate age-associated differences in the microbial communities and resistomes of feces collected from veal calves. Methods: Fecal samples were collected from milk-fed calves on four operations on two sampling dates which were approximately three months apart (n = 24 samples). Shotgun sequencing was conducted on extracted DNA using the Illumina NextSeq 500 platform. Random Forests, a supervised machine-learning technique, was used to differentiate fecal community composition within and between younger and older calves. Results: High levels of alpha (Chao1, ACE, Shannon, Simpson, InvSimpson, and Fisher) and beta diversity in fecal composite samples demonstrated that microbial profiles of veal calves were quite diverse. NMDS plots and PERMANOVA analysis on Bray-Curtis, Jaccard, and Euclidean distances among the microbial taxa and the resistomes revealed clear separation of microbial communities as well as resistomes when samples were grouped by age (PERMANOVA F=1.71-4.13 R2=0.07-0.16, P < 0.05). Discriminatory taxa and AR genes that separated the microbial communities and resistomes in younger- and older veal-claves were predicted using Random Forests (out-of-bag estimate of error rate 8.33%). A correlation matrix was generated by calculating all possible pairwise Spearman’s rank correlations between the predicted taxa and AR genes and visualized in a network interface that identified two major module classes (Module A and B) with >80% predicted taxa and AR genes. Kruskal-Wallis rank sum test indicated significantly higher relative abundance of module A in younger vs older calves that mainly consisted of Proteobacteria, bacteriophages, and AR genes conferring resistance to aminoglycosides, beta-lactams, polymyxin, and sulfonamides ('2 99.30-266.78, P<0.0001). Whereas, significantly higher relative abundance of module B was observed in older calves that mainly consisted of Bacteroidetes, Firmicutes, limited Proteobacteria, and AR genes conferring resistance to glycopeptides, lincosamides, macrolides, trimethoprim, and tetracyclines ('2 318.36-424.11, P<0.0001). Conclusions: Findings from this study suggested an age-dependent shift in the fecal microbial community and resistome structures in veal calves.