A Changing Landscape of Carbapenem-Resistant Escherichia coli in Hong Kong: Emergence of bla NDM-5 -Carrying ST69 across Clinical and Food Sources
Discuss this preprint
Start a discussion What are Sciety discussions?Listed in
This article is not in any list yet, why not save it to one of your lists.Abstract
Carbapenem-resistant Escherichia coli (CR- E. coli ) is an emerging One Health threat, but recent shifts in predominant lineages and genomic links between clinical and food reservoirs in Hong Kong remain poorly defined. We analyzed 271 CR- E. coli isolates from four hospitals (2022-2026) and 585 isolates recovered from 4,917 retail food samples (2022-2025). Isolates underwent antimicrobial susceptibility testing, whole-genome sequencing, multilocus sequence typing, resistance-gene and plasmid profiling, core-genome SNP phylogenetics, and comparative genomics. Food isolates were mainly from raw pork (268/585, 45.8%) and raw chicken (231/585, 39.5%). bla NDM-5 was detected in 527/585 (90.1%) food and 241/271 (88.9%) clinical isolates. ST69 was the most frequent defined sequence type in both collections, representing 44/585 (7.5%) food and 36/271 (13.3%) clinical isolates, in contrast to the heterogeneous lineages and carbapenemases previously reported in Hong Kong. Applying a predefined ≤50-pairwise-SNP threshold for close genomic relatedness, core-genome phylogeny of 80 ST69 isolates identified two major mixed-source clusters collectively comprising 28 clinical and 27 food isolates. Clustered isolates showed similar antimicrobial resistance profiles, carried bla NDM-5 and bla TEM-1 , and were associated with IncI1 MLST | ST136 plasmids. Comparative analyses showed >99.85% average nucleotide identity and broad conservation of the bla NDM-5 -associated plasmid backbone across sources. These findings indicate the emergence of bla NDM-5 -carrying ST69 as a prominent CR- E. coli lineage in Hong Kong and demonstrate close genomic relatedness between selected clinical and retail food isolates. Although transmission direction have not been inferred yet, the findings support integrated One Health surveillance and source-tracing across clinical, food, animal, and environmental sectors.