Longitudinal study on antibiotic susceptibility in commensal E. coli from geese raised in free-range production systems

Abstract

The transmission of antimicrobial resistance bacteria from animals to humans has become an important concern. The extended-spectrum beta-lactamase (ESBL) -AmpC- producing Escherichia coli (ESBL-AmpC EC) and quinolones resistant E. coli are of particular interest. The present study aimed to evaluate the load and prevalence of antibiotic-resistant commensal E. coli along the goose production cycle on 2 free-range farms in central Italy. On A farm, oxytetracycline was administered, while the B farm did not use antibiotics during the geese productive cycle. One hundred geese of 1-day-old from the same batch were divided into the two farms. At hatching, the animals showed an average of E. coli loads was 6.83 ± 0.48 log CFU/g, and 0.28 ± 0.28, 0, 5.12 ± 0.54 log CFU/g for E. coli resistant to nalidixic acid (E. coli^nal), to cefotaxime (E. coli^cef) and to tetracyclines (E. coli^tet), respectively. The loads of E. coli, E. coli^nal, E. coli^cef and E. coli^tet on 224 environmental faecal pools were determined at 8 time points. Antimicrobial susceptibility and molecular characterization of E. coli^cef isolates were performed. The ANOVA was used to assess the difference in bacterial loads between the two farms. We described more than 50% of resistances for tetracyclines in both farms, and sulphonamides and cephazolin in the A farm. The loads of E. coli and E. coli^nal in faeces were estimated at approximately 6-7 log (CFU/g) and 5-6 log (CFU/g) in the two farms, respectively. The average load of extended-spectrum beta-lactamase Escherichia coli (ESBL EC) in goose faeces varied broadly along the production cycle: in the first weeks, a sharp increase was observed in both farms, while later on A farm, the burden of ESBL EC remained steady until the end of the production cycle and on B farm the load dramatically decreased from 6 wk of age onward. An increase in the proportion of E. coli^nal was observed on A farm shortly after the antibiotic administration. Our study shows that the dynamics of antibiotic-resistant E. coli in farmed geese are similar to the ones observed in broilers. However, the risk of the emergence of antibiotic-resistant commensal E. coli, might be mitigated by the adoption of good management practices, including prudent use of antibiotics.

Keywords: antibiotic susceptibility; commensal E. coli; free-range farm; goose.

Figure 1

Graphical representation of the A farm and B farm.

Figure 2

Results of quantitative analysis. (A) Averages of E. coli loads (log UFC/g); (B) averages of E. coli^cef loads (log UFC/g), (C) average of E. coli^tet (log UFC/g), (D) average of E. coli^nal (log UFC/g), in the A and B farms. Within each curve, different letters indicate significant differences between means (P ≤ 0.05). Whereas * indicate significant differences between farms. Bars mean ± SE (n = 14).

Figure 3

Averages proportion of E. coli^cef (A), E. coli^tet (B) and E. coli^nal (C) loads respect to the total E. coli population (%), in the A and B farms. Within each curve, different letters indicate significant differences between means (P≤ 0.05). Whereas * indicate significant differences between farms. Bars mean ± SE (n = 14).

Figure 4

Heatmap showing the molecular characterization and the antibiotic susceptibility profile of each E. coli^cef isolated on the two farms. Samples are coloured according to the antibiotic molecules results: resistant (red), intermediate (light red) and susceptible (pink). (Colored version of figure is available online.)

Figure 5

Distribution of isolates, divided according to the number of resistances, in each farm from T1 to T8. The number of isolates in the A and B farms, respectively, is shown.