RNA-Seq analysis of broiler intestinal transcriptome reveals new insight on the interaction between antibiotic growth promoters (AGPs) and the host

Purpose: Identify at the transcriptome level genes that could be involved in the mode of action of avilamycin, an AGP, when administered at non-therapeutic concentrations, and use this information in regard to development of alternatives to AGPs. Methods: Out of 120 broiler chickens fed either CONTROL feed or AGP feed (supplemented with avilamycin; Surmax® 100, 10 g/1000 kg of feed, Elanco), the RNA from the ileum tissue of five birds per group was isolated from day 35, profiled by RNA-sequencing, and results were verified via RT-qPCR. Results: Overall, 66 differentially expressed genes (DEGs; FDR ≤ 0.05 and fold change ≥ 2 or ≤ -2) were found in the ileum of chicken fed avilamycin in comparison to the control group. Only partial agreement was found on the DEGs (64%) when the reads were initially mapped to the former genome assembly Galgal5 instead of GRCg6a. Functional enrichment analysis showed that the down-regulated DEGs were associated with signaling by interleukins, and that IL-22 and certain antimicrobial peptides could play a central role in the intestinal host response to avilamycin at day 35. This was further corroborated on sixteen chickens per group via RT-qPCR. In addition, higher activity was detected in module of genes related to lipid metabolism and transport in the avilamycin group, as revealed by the co-expression network analysis. Conclusions: Improved feed efficiency was correlated with reduced immunological stress in the intestinal mucosa. The RNA-Seq approach allowed generating novel knowledge that could contribute to the development of AGP alternatives, with IL-22 identified as a potential marker. Ileum mRNA profiles from 35 day old chickens. For each tissue, 5 biological replicates (= 5 chickens) from Control and 5 from Avilamycin group were used for RNA-extraction. Two technical replicates of each of the biological replicates were used for library preparation. The final multiplexed (n=20) libraries were sequenced in two independant runs on a Illumina HiSeq 4000 machine (run1: Sample 1-9 in one lane; run2: Sample 10-20 in two lanes).