The prevalence and mechanism of triclosan resistance in Escherichia coli isolated from urine samples in Wenzhou, China

BACKGROUND: The widespread application of triclosan contributes to its residual deposition in urine, which provides an environment of long-term exposure to triclosan for the intestinal Escherichia coli. We determined the triclosan and antibiotic resistance characteristics of E. coli strains isolated from urine samples and further investigated the resistance mechanism and molecular epidemic characteristics of triclosan-resistant E. coli isolates. METHODS: A total of 200 non-repetitive E. coli strains were isolated from urine samples and then identified. The minimum inhibitory concentrations (MICs) of triclosan and antibiotics, fabI mutation, efflux pump activity, the expression of 14 efflux pump encoding genes, and epidemiological characteristics were determined by the agar dilution method, polymerase chain reaction (PCR), carbonyl cyanide 3-chlorophenylhydrazone (CCCP) inhibition test, quantitative real-time polymerase chain reaction (RT-qPCR), multilocus sequence typing (MLST), and pulse-field gel electrophoresis (PFGE) for all triclosan-resistant isolates. Furthermore, we also investigated the effect of triclosan exposure in vitro on antibiotic susceptibility and the efflux pump encoding gene expressions of triclosan-susceptible strains via serial passage experiments. RESULTS: Of the 200 E. coli isolates, 2.5% (n = 5) were found to be resistant to triclosan, and multidrug resistance (MDR) and cross-resistance phenotypes were noted for these triclosan-resistant strains. The triclosan-sensitive strains also exhibited MDR phenotypes, probably because of the high resistance rate to AMP, CIP, LVX, and GEN. Gly79Ala and Ala69Thr amino acid changes were observed in the triclosan-resistant strains, but these changes may not mediate resistance of E. coli to triclosan, because mutations of these two amino acids has also been detected in triclosan-susceptible strains. Moreover, except for DC8603, all other strains enhanced the efflux pumps activity. As compared with ATCC 25922, except for fabI, increased expressions were noted for all efflux pump encoding genes such as ydcV, ydcU, ydcS, ydcT, cysP, yihV, acrB, acrD, and mdfA among the studied strains with varying PFGE patterns and STs types. Unexpectedly, 5 susceptible E. coli isolates showed rapidly increasing triclosan resistance after exposure to triclosan in vitro for only 12 days, while MDR or cross-resistance phenotypes and the overexpression of efflux pump genes were recorded among these triclosan-induced resistant isolates. CONCLUSIONS: This is the first study to report that short-term triclosan exposure in vitro increases triclosan resistance in susceptible E. coli isolates. After acquiring resistance, these strains may present MDR or cross-resistance phenotypes. Moreover, triclosan resistance mainly involves the overexpression of fabI and efflux pumps in E. coli isolates.