Low Dose Colonization of Broiler Chickens With ESBL-/AmpC- Producing Escherichia coli in a Seeder-Bird Model Independent of Antimicrobial Selection Pressure

Abstract

Extended-spectrum beta-lactamase- (ESBL-) and AmpC beta-lactamase- (AmpC-) producing Enterobacteriaceae pose a risk for both human and animal health. For livestock, highest prevalences have been reported in broiler chickens, which are therefore considered as a reservoir of multidrug-resistant bacteria. The possibility of transfer to humans either by a close contact to colonized broiler flocks or through contaminated retail meat results in the necessity to develop intervention measures for the entire broiler production chain. In this regard, a basic understanding of the colonization process is mandatory including the determination of the minimal bacterial load leading to a persistent colonization of broiler chickens. Therefore, we conducted a bivalent broiler colonization study close to real farming conditions without applying any antimicrobial selection pressure. ESBL- and AmpC- negative broiler chickens (Ross 308) were co- colonized on their third day of life with two strains: one CTX-M-15-producing Escherichia coli-ST410 and one CMY-2/mcr-1-positive E. coli-ST10. Colonization was assessed by cloacal swabs over the period of the trial, starting 24 h post inoculation. During the final necropsy, the contents of crop, jejunum, cecum, and colon were quantified for the occurrence of both bacterial strains. To define the minimal oral colonization dosage 10⁴ to 10¹ colony forming units (cfu) were orally inoculated to four separately housed broiler groups (each n = 19, all animals inoculated) and a dosage of already 10¹ cfu E. coli led to a persistent colonization of all animals of the group after 3 days. To assure stable colonization, however, a dosage of 10² cfu E. coli was chosen for the subsequent seeder-bird trial. In the seeder-bird trial one fifth of the animals (seeder, n = 4) were orally inoculated and kept together with the non-inoculated animals (sentinel, n = 16) to mimic the route of natural infection. After 35 days of trial, all animals were colonized with both E. coli strains. Given the low colonization dosage and the low seeder/sentinel ratio, the rapid spread of ESBL- and AmpC- producing Enterobacteriaceae in conventional broiler farms currently seems inevitably resulting in an urgent need for the development of intervention strategies to reduce colonization of broilers during production.

Keywords: AmpC; ESBL; Escherichia coli; antimicrobial selection pressure; broiler chicken; colonization; seeder bird model.

Figure 1

Equivalence testing of ESBL- E. coli 10716 and AmpC- E. coli 10717 in samples of crop (A), jejunum (B), cecum (C), and colon (D) in between the four colonization dosage groups (10¹-10⁴ cfu/E. coli) at necropsy. 95%CI = 95% confidence interval; all data shown are log10 transformed (log10 cfu/g); margin of <1 log of 95%CI of dosage groups 10¹, 10³, and 10⁴ cfu/E. coli to the mean value of the minimal colonization dosage group (10² cfu/E. coli) shows equivalence.

Figure 2

Comparison of the bacterial counts of ESBL- E. coli 10716 and AmpC- E. coli 10717 in samples of crop (A), jejunum (B), cecum (C), and colon (D) in between the four colonization dosage groups (10¹-10⁴ cfu/E. coli) at necropsy.

Figure 3

Comparison of the bacterial counts of ESBL- E. coli 10716 (A) and AmpC- E. coli 10717 (B) in samples of crop, jejunum, cecum, and colon at necropsy between the minimal colonization dosage group (10² cfu/E. coli; necropsy on day 14; all animals inoculated on third day of life) and the seeder-bird model (necropsy on day 35; inoculation ratio of 1:5 on third day of life). ***p < 0.001 (t-test).