Abstract
Over a 3-month period, we monitored the population of extended-spectrum β-lactam-resistant Escherichia coli (ESBL-EC) associated with the patients, staff, and environment of an intensive care unit (ICU) in Guangzhou, China. Thirty-four clinical isolates were obtained from the same hospital 12 months later. A total of 165 isolates were characterized and whole-genome sequenced, with 24 isolates subjected to long-read sequencing. The diverse population included representatives of 59 different sequence types (STs). ICU patient and environmental isolates were largely distinct from staff isolates and clinical isolates. We observed five instances of highly similar isolates (0 to 13 single nucleotide polymorphisms [SNPs]) being obtained from different patients or bed unit environments. ESBL resistance in this collection was largely conferred by bla (CTX-M) genes, which were found in 96.4% of all isolates. The contexts of bla (CTX-M) genes were diverse, situated in multiple chromosomal positions and in various plasmids. We identified bla (CTX-M)-bearing plasmid lineages that were present in multiple STs across the surveillance, staff, and clinical collections. Closer examination of ISEcp1-bla (CTX-M) transposition units shed light on the dynamics of their transmission, with evidence for the acquisition of chromosomal copies of bla (CTX-M) genes from specific plasmid lineages and for the movement of bla (CTX-M-55) from a ST1193 chromosome to a small mobilizable plasmid. A carbapenem-resistant ST167 strain isolated from a patient that had been treated with meropenem and piperacillin-tazobactam contained seven copies of bla (CMY-146), which appears to have been amplified by IS1. Our data revealed limited persistence and movement of ESBL-EC strains in the ICU environment, but we observed circulating plasmid lineages playing an essential and ongoing role in shaping the cephalosporin-resistance landscape in the population examined. IMPORTANCE ESBL resistance significantly impacts clinical management of E. coli infections in hospitals globally. It is important to understand the structures of ESBL-EC populations carried by hospital patients and staff, their capacity to persist in hospital environments, and the dynamics of mobile genes that drive the spread of ESBL resistance. In our 3-month study, ESBL-EC strains found in the ICU environment were strongly associated with patient carriage but distinct from strains found in staff. However, plasmid lineages carrying bla (CTX-M) genes were found across the ICU populations and in a collection of clinical isolates obtained 1 year later. By examining their content and contexts, we have traced the recent histories of chromosomal and plasmid-borne ISEcp1-bla (CTX-M) transposition units in the ICU population. This information allowed us to implicate specific plasmid lineages in the acquisition of chromosomal bla (CTX-M) genes, even when the plasmids were no longer present, and to detect recent transposition of bla (CTX-M-55) from a chromosome to a mobilizable plasmid. Similar high-resolution approaches to the study of mobile genetic elements will be essential if the transmission routes associated with the spread of ESBL resistance are to be understood and subjected to interventions.