Dynamics of extended-spectrum beta-lactamase-producing Enterobacterales colonization in long-term carriers following travel abroad

Abstract

Travel to tropical regions is associated with high risk of acquiring extended-spectrum beta-lactamase-producing Enterobacterales (ESBL-E) that are typically cleared in less than 3 months following return. The conditions leading to persistent carriage that exceeds 3 months in some travellers require investigation. Whole-genome sequencing (Illumina MiSeq) was performed on the 82 ESBL-E isolates detected upon return and 1, 2, 3, 6 and 12 months later from the stools of 11 long-term (>3 months) ESBL-E carriers following travel abroad. One to five different ESBL Escherichia coli strains were detected per traveller upon return, and this diminished to one after 3 months. Long-term carriage was due to the presence of the same ESBL E. coli strain, for more than 3 months, in 9 out of 11 travellers, belonging to epidemic sequence type complexes (STc 10, 14, 38, 69, 131 and 648). The mean carriage duration of strains belonging to phylogroups B2/D/F, associated with extra-intestinal virulence, was higher than that for commensal-associated A/B1/E phylogroups (3.5 vs 0.5 months, P=0.021). Genes encoding iron capture systems (fyuA, irp), toxins (senB, sat), adhesins (flu, daaF, afa/nfaE, pap, ecpA) and colicin (cjrA) were more often present in persistent strains than in transient ones. Single-nucleotide polymorphism (SNP) analysis in persistent strains showed a maximum divergence of eight SNPs over 12 months without signs of adaptation. Genomic plasticity was observed during the follow-up with the loss or gain of mobile genetic elements such as plasmids, integrons and/or transposons that may contain resistance genes at different points in the follow-up. Long-term colonization of ESBL-E following travel is primarily due to the acquisition of E. coli strains belonging to epidemic clones and harbouring 'virulence genes', allowing good adaptation to the intestinal microbiota.

Trial registration: ClinicalTrials.gov NCT01526187.

Keywords: E. coli; ESBL; carriage; long-term carriage; persistence; travel; whole-genome sequencing.

Fig. 1.

Follow-up of ESBL E. coli carriage in travellers after return.

Fig. 2.

Estimated duration of carriage depending on the phylogenetic group of the ESBL E. coli strains. The phylogenetic tree was generated using the maximum-likelihood method. The tree was rooted on Escherichia fergusonii. The bootstrap values (1000 replicates) are shown near the nodes. Scale bar represents the number of nucleotide substitutions per site.

Fig. 3.

Mutations or frameshift events observed in the genome of persistent ESBL E. coli strains during follow-up. We created, for each traveller, a maximum-likelihood unrooted tree relating all isolates sampled during the follow-up period. For each traveller, the length of the branches is proportional to the number of mutations separating two samples. The diameters of circles are proportional to the numbers of isolates that are identical. Genes in red had non-synonymous mutation, genes in black had synonymous mutation and genes in blue had nonsense mutation or frameshift. This representation does not follow any time scale.

Fig. 4.

Genetic similarity between the different isolates of the nine persistent ESBL E. coli strains. The tree was constructed from the similarity matrix generated by the k-mer comparison approach. The tree was unrooted. Scale bar represents the distance between isolates. Phylogroup and sequence type are mentioned in parentheses for each persistent strain.