Summary
A bioRxiv preprint analysing 1,501 publicly available Enterobacterales genomes found 241 combinations of genes from five carbapenemase families. It also reports that cross-family combinations increased over time in the dataset.
A bioRxiv preprint analysing 1,501 publicly available genomes has mapped how often Enterobacterales bacteria carry genes for more than one carbapenemase, an enzyme that can inactivate carbapenem antibiotics. The researchers found 241 distinct combinations involving five carbapenemase families: IMP, KPC, NDM, OXA-48-like and VIM.
Carbapenems are an important class of antibiotics used to treat serious bacterial infections. When bacteria carry carbapenemase genes, these drugs can become less effective against them. The study examines bacterial genomes, rather than patients or treatment outcomes, and is a preprint.
Which gene combinations were most common?
The most frequent groups were NDM with OXA-48-like genes (25.6%), multiple KPC genes (25.3%), KPC with NDM (18.8%) and multiple NDM genes (10.9%). Overall, 56.2% of the genomes encoded carbapenemases from different families, while 36.2% carried copies of the same allele, a particular version of a gene.
The genomic location of the genes also varied. In 59.2% of genomes, the carbapenemase genes were carried on two plasmids; in 23.5%, they were on one. Plasmids are DNA molecules separate from a bacterium’s chromosome and can carry genes, including genes associated with antibiotic resistance. The authors also describe several routes by which multiple carbapenemase genes may arise, including plasmid co-integration and gene duplication. Nearly half of the genomes (48.0%) belonged to lineages the study classified as high-risk.
Resistance patterns and surveillance
The researchers found that plasmids carrying multiple carbapenemases had more antimicrobial-resistance genes, and genes associated with resistance to more antimicrobial classes, than plasmids carrying a single carbapenemase. They also report that cross-family combinations—particularly KPC/NDM and NDM/OXA-48-like—became more common over time in the genomes analysed. NDM-5/OXA-181 and NDM-4/OXA-181 combinations were reported as enriched 3.3-fold and 8.9-fold, respectively.
These findings add detail to the genetic patterns that surveillance programmes can track: not only which resistance genes are present, but which combinations occur and how they are arranged on plasmids. The proportions describe the publicly available genomes included in the analysis; they are not estimates of the prevalence of these bacteria among people or of clinical outcomes. As a preprint, the work has not yet completed peer review.