Plasmid Backbones Drive Global E. Coli Resistome Structure

Investigators report a plasmidome-resolved resistome analysis of 9700 Escherichia coli genomes, examining where resistance genes localize on plasmids versus chromosomes and how recurring plasmid “backbone” families shape those patterns. Across the collection, the authors describe a structured architecture in which a restricted set of backbone families repeatedly co-localize with clinically important antimicrobial resistance genes (ARGs), including last-resort determinants such as blaNDM-5 and mcr-1.1, rather than these genes appearing diffusely across many unrelated plasmid types. The report centers on backbone families as recurring vehicles observed across diverse phylogroups and sequence types.

To build these associations, the authors identified plasmid replicon (backbone) hits in genome assemblies and mapped ARGs with lineage context, treating plasmid-associated and chromosomal occurrences as separate genomic “compartments.” They compared how often specific ARGs appeared on contigs carrying plasmid replicons versus contigs without replicon signals, generating gene-by-gene localization summaries across phylogroups and sequence types. The paper also introduces a Mobility Potential Index (MPI), defined as the proportion of a given ARG’s detections that are plasmid-borne, as a compact way to describe mobility between compartments without relying on any single lineage. This localization-and-mobility framing sets up the paper’s backbone–gene pairings.

Within that framework, the authors highlight strong backbone associations for carbapenemase and colistin resistance determinants, particularly blaNDM-5 (147/159 plasmid-associated occurrences; 92.6% on IncX3) and mcr-1.1 (271/323; 83.9% linked to IncX4). They report blaNDM-5 was detected 246 times, with 159 detections (64.6%) located on plasmids versus 87 (35.4%) on the chromosome in this dataset, and describe IncX3 as the dominant backbone linked to plasmid-borne blaNDM-5, with additional associations involving IncF and IncI types. For mcr-1 variants (mostly mcr-1.1), the authors similarly report a plasmid bias, with 323 of 423 detections (76.4%) located on plasmids, and emphasize IncX4 as a primary backbone family carrying plasmid-borne mcr-1.1, with other plasmid families contributing smaller shares. In compartment-level summaries, plasmid versus chromosomal localization is described as varying across phylogroups and sequence types, including pandemic lineages (e.g., ExPEC clones such as ST131 and ST410) that exhibit reduced plasmid dependency and frequent chromosomal integration of extended-spectrum β-lactamase genes.

The authors also define six High-Risk Plasmid (HRP) groups based on combined abundance, ARG burden, and lineage breadth: IncFIB(AP001918), IncFIA, IncFIC(FII), IncI1_1_Alpha, IncX3, and IncX4. In the plasmid-positive subset, HRP groups were detected in 5027 of 7758 genomes, and IncFIB(AP001918) was reported in 4311 genomes. The co-occurrence analyses presented in the article describe IncF/IncI backbones as forming dense multidrug-resistance “hub” modules with multiple ARG categories, whereas IncX backbones are presented as more specialized vehicles concentrated around blaNDM-5 (IncX3) and mcr-1.1 (IncX4). The authors frame backbone families and HRP groupings as coherent, trackable units in genomic surveillance discussions.

Key Takeaways:

• The authors report that a limited set of plasmid backbone families disproportionately encode certain clinically important determinants, with recurring backbone–gene modules observed across diverse lineages, while most ARGs were chromosomal.
• Six High-Risk Plasmid groups are defined in the report: IncFIB(AP001918), IncFIA, IncFIC(FII), IncI1_1_Alpha, IncX3, and IncX4.
• The analysis separates plasmid-associated from chromosomal ARG occurrences and uses the Mobility Potential Index (MPI) to summarize how frequently specific ARGs are detected on plasmids versus chromosomes.