Exploring the presence, genomic traits, and pathogenic potential of extended spectrum ß-lactamase Escherichia coli in freshwater, wastewater, and hospital effluents.

AIMS: The purpose of this work was to study extended-spectrum ß-lactamase-producing E. coli (ESBL-EC) in freshwaters, hospital effluents and wastewaters during two sampling campaigns in 2021. METHODS AND RESULTS: Water sampling was performed in 24 stations of the Ourthe watershed in Belgium. A total of 644 ESBL (n = 642) and AmpC (n = 2) E. coli strains were isolated. Disk-diffusion assays were performed following the EUCAST's recommendations. All strains were tested for the presence of blaCTX-M-1, blaCTX-M-2 and blaCTX-M-9 gene's group by PCR. Genes belonging to blaCTX-M-1 and blaCTX-M-9 groups were detected respectively in 73.6% and 14.9% of the strains. No blaCTX-M-2 group's gene was found. A subset of strains (n = 40) was selected for whole genome sequencing. E. coli serotype O18: H7 ST 1463 was predominant (n = 14) in the sequenced strains and showed pathogenicity in the Galleria mellonella larvae model. ß-lactamase genes identified were blaCTX-M (n = 21), with blaCTX-M-15 mostly represented (n = 15), as well as blaTEM (n = 11), blaOXA (n = 7) and blaSHV (n = 9) and carbapenemase (CP) genes were observed in several strains-blaKPC-3 (n = 19), blaNDM-1 (n = 1), blaVIM-1 (n = 2) and blaOXA-244 (n = 2)-even from freshwaters. CONCLUSIONS: ESBL-EC are widely distributed in the aquatic environment in Belgium and contain a variety of ESBL and CP genes.

Antimicrobial resistance of Escherichia coli isolated from freshwaters and hospital effluents in Belgium.

The purpose of this work was to evaluate the level of antimicrobial resistant Escherichia coli isolates in freshwaters and hospital effluents in Belgium. The samples were collected from 24 locations along the Ourthe, Vesdre, Amblève and Meuse rivers and in the wastewater effluents of several hospitals. The sampling stations in rivers were classified according to the dominant land covers of the rivers (rural, urban and forest areas). Two sampling campaigns were organized in May and October 2019 to highlight a possible seasonal effect. A total of 938 E. coli strains were isolated on Chromogenic Selective Tryptone Bile X-glucuronide (TBX) and TBX supplemented with amoxicillin (TBX+AMX) media. Disk diffusion assays were performed following the EUCAST's recommendations to assess the antimicrobial resistance against 12 antibiotics. A total of 32·7% of strains were at least resistant to one antibiotic and 24·6% were multiple antimicrobial resistant strains on TBX. The highest resistance rates were found for ampicillin (AMP), amoxicillin coupled with clavulanic acid (AMC) and sulfamethoxazole/trimethoprim (SXT). The lowest resistance rates were observed for meropenem (MEM) and ertapenem (ETP), which are last resort antibiotics. No significant difference was observed between both campaigns for the resistance rate to antibiotics.

Antimicrobial resistant Escherichia coli prevalence in freshwaters in Belgium and human exposure risk assessment.

The objective of this work was to evaluate the antimicrobial resistant (AR) E. coli prevalence in recreational waters in Belgium and to assess the exposure risk for bathers. Nine stations were sampled during the 2021 bathing season. A total of 912 E. coli strains were isolated and tested by the disk diffusion method in accordance with EUCAST recommendations, including Extended-Spectrum Beta-Lactamase (ESBL) production. AR E. coli were counted at each bathing sites, 24% of strains were resistant to at least one antibiotic and 6% were Multi-Drug Resistant (MDR). A Multiple Antibiotic Resistance (MAR) index was calculated to compare the bathing sites. The Lesse river had the highest MAR index as well as the highest E. coli absolute abundance and the largest number of ESBL-producing E. coli. Conversely, the 3 lakes showed lower E. coli contamination levels and AR rates. A human health risk assessment of exposure to AR E. coli, based on the calculation of measured prevalence, was performed considering four different dose-response model scenarios. The human health risk (Pd) ranged from 10-9 to 0.183 (children). The exposure probabilities were low, except for scenario 3 (E. coli O157:H7), which is the most severe.