An official website of the United States government.

Official websites use .gov
A .gov website belongs to an official government organization in the United States.

Secure .gov websites use HTTPS
A lock ( ) or https:// means you've safely connected to the .gov website. Share sensitive information only on official, secure websites.

In Vivo Models to Investigate the Development of Antibiotic Resistance

Veterinary Laboratories Agency, UK
Start date
End date
  • Establish oral models of colonisation with Campylobacter jejuni, E. coli, Enterococcus faecium, Salmonella entertidis or Salmonella typhimurium in chickens with or without growth promoters.
  • Determine the effects of test antibiotics on antibiotic resistance of the colonising target organisms in the vivo poultry model.
  • Establish molecular methods to detect changes in gut flora and determine the effects of growth promoters on such flora.
  • Determine the genetic basis of any observed antibiotic resistance in target organisms.
More information
Final report summary: The emergence and evolution of antibiotic resistance in food-borne zoonotic pathogens is an acknowledged public health risk, with intensive farming regimes implicated in the development of such resistance. Nevertheless, such treatments are often necessary for animal welfare purposes. The present study describes the development of a generic chick model to investigate the potential for growth promoting and therapeutic antimicrobials to induce resistance in vivo. Existing models of colonisation of the avian gut were modified to do this. One day old Specific Pathogen Free chicks were orally administered a mixed dose of test strains of Enterococcus faecium, Escherichia coli, Salmonella enterica serovar Enteritidis, Salmonella enterica serovar Typhimurium and Campylobacter jejuni. Strains selected for dosing were chosen on the basis of colonisation potential, distinct PFGE profile, known pathogenesis for humans and sensitivity to test antimicrobials. Chicks were fed commercial food with or without the growth promoter MAXUS at the manufacturers recommended dose. Subsequent experiments involved administering therapeutic antimicrobials, also at the manufacturers recommended dose. Gut contents were sampled over 5 weeks, and cultured using selective media. All test strains colonised at levels expected in birds dosed with a single strain, suggesting that this approach is useful as a model of colonisation. In each case, the PFGE profiles of strains recovered from treated and untreated birds were identical to the original dosing strain. Administration of the growth promoter avilamycin (Maxus G200) did not result in increased resistance to avilamycin in any recovered isolates.

Incorporation of the therapeutic antimicrobial enrofloxacin (Baytril) into drinking water at the recommended treatment level of 50 ppm for 5 days, led to a 5 to 7-fold increase in MIC to ciprofloxacin, enrofloxacin and nalidixic acid in recovered C. jejuni isolates within 72 h of drug administration. Despite an initial 2-logarithm reduction in colonisation levels, administration of enrofloxacin did not result in a long-term reduction of C. jejuni. The mechanism of resistance was determined to be the Thr86Ile substitution previously reported to be the predominant mutation conferring resistance to fluoroquinolones in C. jejuni. Colonisation of S. Typhimurium and S. Enteritidis was reduced to undetectable levels within the duration of the treatment period, but were observed to re-colonise to maximum levels at a later stage, suggesting that the S. enterica was eradicated from the caecal contents by enrofloxacin treatment, but later reinfected the birds, possibly due to the recycling of organisms from litter to gut.

A total of 300 chickens were separated into 10 groups in two separate experiments and treated with different therapeutic levels of enrofloxacin (ranging from 12 ppm to 500 ppm) in the drinking water. Caecal and liver levels of enrofloxacin and its primary metabolite ciprofloxacin were measured using HPLC to investigate the correlation between resistance and levels of fluoroquinolone to which campylobacters colonising the intestines of each groups of chickens were exposed following commencement of treatment. Only treatment with 500 ppm enrofloxacin, for 3 days reduced C. jejuni below the level of detection levels from 48 hours of treatment. Chicks drank the medicated water and behaved normally throughout the treatment period and for up to 35 days following treatment. HPLC data demonstrated that enrofloxacin and ciprofloxacin levels peaked at the end of the treatment period then fell below the level of detection (ƒÝg/g or ƒÝg/ml ) within 96 hours. This was supported by the data obtained using Reduction of therapeutic doses (12 and 25 ppm), to allow colonisation of campylobacter, but prevent resistance did not prevent fluoroquinolone resistance occurring at 3 days following commencement of treatment as before. When chicks were challenged with C. jejuni 7 days following completion of enrofloxacin treatment the chicks were rapidly colonised by susceptible organisms, supporting the observations of rapid clearance of fluoroquinolones from the caeca following completion of therapy.

Treatment with the manufacturer¡¦s recommended dosing strategy for the macrolide tylosin (Tylan), did not reduce colonisation levels of S. enterica or E. coli and the genotypes and MICs to the panel of antimicrobials were unchanged from strains used in the original dose. This was the expected observation, as these organisms are intrinsically resistant to macrolides. Colonisation of both the C. jejuni and C. coli dose strains was reduced to undetectable levels by 48 hours following commencement of therapy with tylosin. This time the campylobacter strains did not re-colonise, an interesting observation, given that resistance to macrolides is also conferred by a point mutation, this time in the 23 S rRNA gene.

Apart from a temporary reduction in C. jejuni colonisation levels following treatment with aureomycin, the antimicrobials aureomycin and amoxypen had no lasting effect on colonisation of the dose strains in the model. Treatment did not result in increased resistance to the active agents of these therapeutic antimicrobials, tetracycline and amoxicillin, respectively, or to the other antimicrobials under test.

This in vivo approach has been extremely successful in modelling the development of resistance in gut pathogens and may be useful as a tool in developing sustainable control strategies for the future.

Moreover, by addition of a 6th organism to inoculum, Bacillus subtilis, we were able to examine the interaction of the spores on the 5 previously-mentioned organisms of public health importance. We were not able to demonstrate a probiotic effect of the B. subtilis on the colonisation of the challenge strains, but did observe a slight reduction in shedding of these organisms. We can report that this in vivo model has the potential to assess other phenomena, apart from antimicrobials, on the 5 strain inoculum.

Funding Source
Dept. for Environment, Food and Rural Affairs
Project number
Meat, Poultry, Game