Molecular epidemiology of antibiotic resistant Escherichia coli from intensively-produced poultry in a farm-to-fork continuum in KwaZulu-Natal, South Africa.

Abstract

The increased use of antibiotics in intensively produced food animals has resulted in the selection of drug-resistant bacteria across the farm-to-fork continuum. There is a risk of transfer of this resistance to humans and as such a public health risk. The aim of this study was to investigate the molecular epidemiology of antibiotic resistant Escherichia coli from intensively produced poultry in the uMgungundlovu district of Kwa-Zulu Natal, South Africa. This was a longitudinal descriptive study with the aim to determine the epidemiology of antibiotic resistance of E.coli from hatching through to the final retail product from an intensive poultry farm house. The farm reported the use of zinc bacitracin and Salinomycin included in the feed, but no therapeutic antibiotics used in this batch of chickens. However, the following antibiotics were used on the farm in the previous 12 months: Doxycycline, Sulfadiazine and Trimethoprim, Enrofloxacin, Ceva olaquindox 10%, Avilamycin, Tylosin 10% and Kitasamycin tartate. During the first five weeks, ten samples from litter and faeces were collected. During transfer from the house to abattoir ten swabs from transport trucks and transport crates were taken. At the abattoir ten samples from carcass wash were collected. After slaughter and dressing ten caecums, whole chickens, thighs and necks were collected. Again, during house washing, ten samples were collected. E.coli was putatively identified using Eosin Methylene Blue agar followed by Sorbitol MacConkey agar and confirmed by identification of the uidA gene by polymerase chain reaction. Susceptibility to a panel of antibiotics recommended by the World Health Organization Advisory Group on the Integrated Surveillance of Antimicrobial Resistance (WHO-AGISAR) was ascertained by the Kirby-Bauer disk diffusion method for 20 antibiotics according to CLSI guidelines. Realtime PCR was used to test for resistance genes tetA, tetB, qnrB, qnrS, aac(6)-lb-cr, sul1, sul2, sul3, blaSHV, blaCTX-M, blaTEM conferring resistance to tetracyclines, quinolones, sulphonamides and cephalosporin antibiotics. Clonal similarities were investigated using ERIC-PCR. A total of 266 E.coli isolates constituted the sample size with a non-susceptibility profile of ampicillin 48.1%, tetracycline 27.4%, nalidixic acid 20.3%, trimethoprim-sulphamethoxazole 13.9%, chloramphenicol 11.7%, cefalexin 4.5%, ciprofloxacin 4.1%, amoxycillin-clavulanic acid 3.4%, gentamicin 1.9%, cefoxitin 1.1%, cefepime 1.1%, cefotaxime 1.1%, amikacin 1.1%, ceftriaxone 0.8% and azithromycin 0.8%. Isolates were fully susceptible to ceftazidime, imipenem, meropenem and tigecycline. Of the 266 isolates 6.4% were multidrug resistant (resistant to one or more antibiotics in three or more distinct antibiotic classes). The most frequently observed resistance genes were blaCTX-M (100%), sul1(80%), tetA(77%), tetB(71%). Using ERIC-PCR the isolates were grouped into 27 clusters with a 75% similarity. Eight clusters comprised of isolates from only one sample. xiv There was an increase in MDR and resistance genes over the farm to fork continuum with lowest and highest levels seen in transport and waste-water samples respectively. ERIC-PCR did not indicate the transmission of clones across the farm-to-fork continuum. There instead appeared to be de novo or evolution of resistance genes or the introduction of plasmids over the time period. As the only antimicrobials used in this flock were salinomycin and zinc bacitracin it is postulated that the resistance observed could be attributed to the co-selection of resistance genes and/or horizontal gene transfer from the environment, insects, chicken food and workers. Overall resistance levels were low over the six weeks of the study, MDR and the prevalence of resistance genes increased over time. The diverse clonality shown by the ERIC PCR results did not support the transmission of clones across the farm-tofork continuum but indicated a de novo evolution of resistance genes and/or the loss or gain of plasmids over the time period.