Dynamics of extended-spectrum cephalosporin resistance genes in Escherichia coli from Europe and North America

Abstract

Extended-spectrum cephalosporins (ESCs) are critically important antimicrobial agents for human and veterinary medicine. ESC resistance (ESC-R) genes have spread worldwide through plasmids and clonal expansion, yet the distribution and dynamics of ESC-R genes in different ecological compartments are poorly understood. Here we use whole genome sequence data of Enterobacterales isolates of human and animal origin from Europe and North America and identify contrasting temporal dynamics. AmpC β-lactamases were initially more dominant in North America in humans and farm animals, only later emerging in Europe. In contrast, specific extended-spectrum β-lactamases (ESBLs) were initially common in animals from Europe and later emerged in North America. This study identifies differences in the relative importance of plasmids and clonal expansion across different compartments for the spread of different ESC-R genes. Understanding the mechanisms of transmission will be critical in the design of interventions to reduce the spread of antimicrobial resistance.

Fig. 1. Frequency and diversity of ESC-R genes in E. coli by country and source.

a Percentage of ESC-R gene by country and b source (total n = 1,660 ESC-R + ESC-susceptible genomes*). The bars are coloured by ESC-R gene, as shown in the inset legend. c Diversity indices (SR: species richness; SE Shannon entropy, SD Simpson diversity, and reciprocal BP Berger-Parker) of ESC-R profile by compartment (country + source), where only compartments with minimum sample size of n = 30 were included in the analysis (total n = 1361 ESC-R genomes*). The shape of the dots is linked with the source and the colours with the country. d Common and unique ESC-R genes among countries and e compartments (total n = 1548 ESC-R genomes*). Heatmaps are coloured as per inset scale bars. *Genomes from Pietsch et al. (n = 158; 148 ESC-R genomes + 10 ESC-susceptible genomes) were excluded. Source data are available in the Source Data file. ESC-R: extended-spectrum cephalosporin resistance.

Fig. 2. Distribution and trend over time of ESC-R genes in E. coli by compartment.

a Trends in percentage and b count for ESC-R types over time in Canada and France across humans and farm animals (cattle, chickens, pigs and turkeys). The lines and bars are coloured by ESC-R types. c Trends over time of the percentage of the major ESC-R gene in main compartments (colour key is in inset legend). Source data are available in the Source Data file. ESC-R extended-spectrum cephalosporin resistance, ESBL extended-spectrum β-lactamases.

Fig. 3. Presence and dynamics of AMR determinants in E. coli.

a Presence of resistance determinants by antimicrobial class and compartment (country + source) for n = 1189 E. coli isolates. Two-sided Fisher’s Exact Test was used to compare the proportion of AMR class between compartments, and the significance levels are indicated with asterisks: *p < 0.01, **p < 0.001 and ***p < 0.0001. The bars are coloured by country and the pattern by source. b Odds ratios and 95% confidence intervals (coloured as per inset legend) demonstrating change in AMR determinant presence over the study period by compartment; significance cut-off is 0.00125 after Bonferroni adjustment for multiple hypothesis testing. Source data are available in the Source Data file. Q-FQ quinolone-fluoroquinolone.

Fig. 4. Core-gene phylogenetic tree for 1818 E. coli isolates.

Maximum likelihood tree with eight main clusters linked with phylogroup. Bootstrap support values > 95 are represented by blue dots for main clusters. Metadata are represented next to the tree by four rings: phylogroup, ESC-R gene, source and country (coloured as per inset legend). The number of AMR genes per genome is plotted on the outside as a bar plot (light green colour).*ST117 cluster contains mainly ST117 but also other closely related STs. Source data are available in the Source Data file. ESC-R extended-spectrum cephalosporin resistance.