A genomic framework for tracking antibiotic resistance genes: global dissemination of cfr-family genes

The global spread of antimicrobial resistance (AMR) calls for advanced surveillance frameworks capable of tracking high-risk resistance genes across genomic, temporal, and ecological environments. To address this, we designed an integrated genomic epidemiology workflow, encompassing antibiotic resistance gene sequence alignment and phylogenetic classification, integration of whole-genome functional annotation, and individual analysis of key bacterial strains. Applying this framework, we obtained 1,026 chloramphenicol-florfenicol resistance (cfr)-like sequences through homology searches, and phylogenetic analysis resolved them into 11 monophyletic clusters, including the newly identified cfr(F) and cfr(G) variants. From the screening of 702,252 bacterial genomes, we identified 5,901 cfr-positive isolates comprising 11 distinct cfr variants. The earliest variant, cfr(G), was detected in a soil isolate dating to 1893. Genomic localization analysis revealed that clbA, clbC, cfr(E), cfr(F), and cfr(C) were putatively predominantly chromosomal, whereas cfr(G), cfr(B), cipA, and clbB were putatively mainly plasmid-borne. Importantly, cfr-like genes frequently co-localize with other last-resort antibiotic resistance genes (ARGs), including vanA, optrA, and carbapenemase genes, on the same isolate, promoting multidrug resistance. Beyond gene detection, this study clarifies the context-dependent transmission risks of resistance determinants and provides a strategic framework for anticipating resistance convergence and informing preemptive AMR surveillance.