Salmonella Excludes Salmonella in Poultry: Confirming an Old Paradigm Using Conventional and Barcode-Tagging Approaches

Abstract

Salmonella is one of the major foodborne bacterial pathogens, and the consumption of contaminated chicken meats isa primary route of Salmonella transmission into human food chains. However, the mechanism of Salmonella transmission within the chicken flock is not fully understood, including competition among Salmonella strains during chicken infection. The purpose of the present study was to evaluate the competitive exclusion (CE) between different or same Salmonella species consecutively challenged through the oral route. Two different approaches were used to evaluate the CE effect, including tracking Salmonella colonization by wild-type strains with difference in natural antibiotic resistance or DNA barcode-tagged isogenic strains. When day-of-hatch chicks were administered by wild-type S. Typhimurium (ST) on day 1, followed by infection on day 2 by S. Enteritidis (SE) or vice versa, most of the birds were colonized only by the first strains administered (82% by ST or 83% by SE). When similar experiments were performed using two different isogenic barcode-tagged SE strains, Illumina sequencing analysis of the barcode region showed that the first barcode-tagged strains administered were dominant strains, ranging from 92 to 99% of the Salmonella recovered from ceca. These results provide quantitative evidence supporting the CE theory that oral administration of Salmonella will produce predominant inhibition over the subsequent colonization of ceca by the following administration one day later by different or same Salmonella species. We also showed that the use of barcode-tagged isogenic strains in combination with deep profiling of barcodes by Illumina sequencing can serve as a quantitative method for studying complex dynamics of Salmonella infection, transmission and colonization in poultry.

Keywords: Salmonella; barcode-tagged isogenic strain; competitive exclusion; intestinal colonization; poultry.

Figure 1

Relative abundance of Salmonella barcode-tagged strains (BC1 and BC2) in cecal samples in Experiment 2. The relative abundance of each barcode-tagged strain (BC1 in blue bars, and BC2 in red bars) in % (left Y axis) was determined from MiSeq data. Black dots (right Y axis) indicate CFU/g of cecal contents as determined by plating of serial dilutions. The black dots corresponding to 0 CFU/g indicate the samples in which Salmonella level was lower than detection limit. Day-of-hatch chickens (n = 12/group) were orally gavaged in Group 1 (G1) with BC1 (Day 1) and saline (Day 2); in G2 with BC2 (Day 1) and saline (Day 2); in G3 with saline (Day 1) and BC1 (Day 2); in G4, with saline (Day 1) and BC2 (Day 2), in G5 with BC1(Day 1) and BC2 (Day 2), and in G6 with BC2 (Day 1) and BC1 (Day 2). For all oral gavage with either BC1 or BC2, each chick received 2.5 × 10⁴ CFU. Asterisk (*) on the top of the bars indicates significant difference between BC1 and BC2 within the same group at P < 0.01.

Figure 2

Relative abundance of Salmonella barcode-tagged strains (BC1 and BC2) in liver/spleen samples in Experiment 2. The relative abundance of each barcode-tagged strain (BC1 in blue bars, and BC2 in red bars) in % (left Y axis) was determined from MiSeq data. Black dots (right Y axis) indicate CFU/g of liver/spleen samples as determined by plating of serial dilutions. The black dots corresponding to 0 CFU/g indicate the samples in which Salmonella level was lower than detection limit. The treatment groups G1-G6 are the same to those in Figure 1. Asterisk (*) on the top of the bars indicates significant difference between BC1 and BC2 within the same group at P < 0.01.