Virulence factors and antimicrobial resistance in Escherichia coli strains isolated from hen egg shells.

Eggs may contain extraintestinal pathogenic (ExPEC) and diarrheogenic (DEC) Escherichia coli which in addition may carry antibiotic resistance. The wide use of biocides and disinfectants in the food industry may induce biocide tolerance in bacteria. The aim of the present study was to evaluate biocide tolerance and antibiotic resistance in E. coli from hen egg shells. A total of 27 isolates obtained from a screening of 180 eggs were studied. Seven isolates carried both eae and bfpA genes of typical enteropathogenic E. coli (EPEC) strains, while 14 isolates only carried eae associated with atypical EPEC strains. Shiga toxin genes stx and stx2 were detected in four isolates. Heat-stable and heat-labile enterotoxin genes as well as aggR were also detected. Several isolates had minimum inhibitory concentrations (MICs) that were higher than the wild-type for the biocide hexadecylpyridinium chloride (HDP, 18.52%) or the commercial disinfectant P3 oxonia (OX, 14.81%). Antibiotic resistance was detected for ampicillin (37.03%), streptomycin (37.03%), tetracycline (37.03%), chloramphenicol (11.11%), nalidixic acid (18.51%) and trimethoprim-sulfamethoxazole (14.81%). Eight isolates (29.63%) were biocide tolerant and antibiotic resistant. Efflux pump genes detected included acrB (96.29%), mdfA (85.18%) and oxqA (37.03%), in addition to quaternary ammonium compound (QAC) resistance genes qacA/B (11.11%) and qacE (7.40%). Antibiotic resistance genes detected included blaCTX-M-2 (22.22%), blaTEM (3.70%), blaPSE (3.70%), tet(A) (29.63%), tet(B) (29.63%), tet(C) (7.40%), tet(E) (11.11%), aac(6')-Ib (3.70%), sul1 (14.81%), dfrA12 (3.70%) and dfrA15 (3.70%). Most isolates (96.30%) carried more than one genetic determinant of resistance. The most frequent combinations were efflux pump components acrB and mdfA with tetracycline resistance genes (33.33% of isolates). Isolates carrying QAC resistance genes also carried between 4 and 8 of the additional antimicrobial resistance genes investigated. Regardless of biocide tolerance and antibiotic resistance, all isolates were sensitive to carvacrol (0.25%), thymol (0.125%) and trisodium phosphate (1 to 1.5%), but they exhibited a heterogeneous response to sodium lactate and lysozyme-EDTA combinations that apparently were not related with antibiotic resistance. Results from the study reveal not only a low incidence of biocide tolerance but also the presence of multiple resistance strains carrying multiple genetic determinants of resistance.