Molecular epidemiology of Antibiotic-Resistant Escherichia coli from companion animals attending veterinary practices in Durban, KwaZulu-Natal, South Africa.

Abstract

Background: Companion animals are globally documented to harbour antibiotic-resistant E. coli. This study aimed to investigate the molecular epidemiology of antibiotic-resistant E. coli from companion animals presenting at veterinary practices in Durban, KwaZulu-Natal, South Africa. Methods: E. coli were isolated on selective media from rectal swabs sampled from dogs and cats attending veterinary practices in Durban, KwaZulu-Natal, South Africa. All isolates were confirmed using real-time polymerase chain reaction (PCR) of the uidA gene. Antibiotic susceptibility testing was done against 20 antibiotics using the Kirby-Bauer disk diffusion method. Selected antibiotic-resistance genes (ARGs) that confer resistance to third-generation cephalosporins (blaTEM, blaSHV, and blaCTX-M), tetracycline (tetA, and tetB), and tigecycline (tetX/X2, tetX3, and tetX4), were detected using conventional PCR. PCR amplicons were confirmed by DNA sequencing and bioinformatics analysis. Enterobacterial Repetitive Intergenic Consensus Polymerase Chain Reaction (ERIC-PCR) was carried out to determine the clonality of E. coli (101) isolates that showed resistance to at least one antibiotic. Results: A total of 330 E. coli isolates from dogs (234) and cats (96) formed the study sample. Overall resistance was high in tetracycline (24.2%), ampicillin (18.8%), trimethoprim-sulfamethoxazole (14%), cephalexin (11.2%) and nalidixic acid (9.7%). Whilst lower resistance was observed in amikacin (0.3%), ceftazidime (0.3%), and piperacillin-tazobactam (0.6%). Third-generation cephalosporin-resistant E. coli retrieved from cats (26%) was more prevalent compared to dogs (9.8%). E. coli from dogs (2.1%) and cats (2%) were resistant to forth-generational cephalosporins. E. coli (3%) retrieved from dogs was resistant to tigecycline, which is regarded as a medically important antimicrobial (MIA) in human medicine. No resistance was observed against carbapenems. Thirty-five (10.6%) E. coli were multidrug-resistant (MDR) and exhibited twenty-two different phenotypic patterns. Amongst the E. coli that were not susceptible to third-generation cephalosporin, and tetracycline, it was observed that the blaCTX-M-15 (8%), and tetA (24%) were the most prevalent resistance genes. Thirty-one (9.3%) isolates were non-susceptible to third-generation cephalosporins and had the corresponding extended-spectrum beta-lactamase (ESBL) genes. The blaCTX-M-15 type gene was prevalent in all 25 E. coli isolates that tested positive for the blaCTX-M. The blaTEM-1 (17) was the second most prevalent β-lactamase gene. A total of 80/330 (24%) isolates were phenotypically not susceptible to tetracycline and carried either one, or both of tetA and tetB resistance genes. Only one tetracycline-resistant E. coli isolate did not harbour either tetA, or tetB genes. The blaSHV, tetX/X2, tetX3, and tetX4 were not detected in all the isolates. Using a 75% similarity cut-off, forty-eight clusters with isolates from both dogs and cats were identified. The ERIC-PCR types depicted a variety of clusters within veterinary practices in Durban, indicating that a high diversity of E. coli is in circulation in Durban, South Africa. Conclusion: Companion animals are reservoirs of antibiotic-resistant E. coli and ARGs. However, there was no evidence of transmission of antibiotic-resistant E. coli in Durban, South Africa. Resistance of E. coli from companion animals to MIA for humans is of particular concern and requires measures to control the spread of antibiotic-resistant bacteria, and ARGs between companion animals, veterinary practice personnel, and owners.