Global dissemination of npmA mediated pan-aminoglycoside resistance via a mobile genetic element in Gram-positive bacteria

Abstract

The npmA gene, encoding a 16S rRNA methyltransferase, confers resistance to all clinically available aminoglycosides, posing a significant threat to effective antibiotic therapy. We analyze 1,932,812 bacterial genomes to investigate the distribution and mobilization of npmA variants. npmA is not found in Gram-negative bacteria, where it was originally described, but is identified among Gram-positive bacteria, predominantly as the npmA2 variant in the globally distributed Clostridioides difficile ST11 lineage. We also detect npmA2 in two vancomycin-resistant Enterococcus faecium isolates from a Dutch hospital. Upon sequencing and phenotypic analysis, we determine that E. faecium isolates are pan-resistant to aminoglycosides. Genomic characterization links npmA2 to a composite transposon, Tn7734, which is integrated within a previously uncharacterized Integrative and Conjugative Element (ICE) Tn7740, present in both npmA2-carrying C. difficile and E. faecium clinical isolates. Tn7740-like, but not npmA2, appears across diverse taxa, including human microbiome members. Here, we show that Tn7740 likely facilitates cross-species npmA2 mobilization between these Gram-positive bacteria and emphasize the risk of mobile genetic elements transferring pan-aminoglycoside resistance between clinically important bacterial pathogens.

Fig. 1. Distribution, population structure and diversity of npmA-carrying C. difficile isolates.

a Global map of npmA-carrying C. difficile isolates included. b Minimum spanning tree (MST) representing the genetic relationships among npmA-carrying C. difficile isolates based on core-genome multilocus sequence typing (cgMLST). Tips are collapsed if there are not allelic differences and colored according to the sequence type (ST), and sizes correspond to the number of isolates. Numbers on the connecting lines indicate allelic differences between them. Colors correspond to sequence types as shown in the panel legend. c Bar graphs showing the distribution of isolates according to their source (human, livestock, or environmental), geographic origin, and the year of isolation, categorized by ST. The first two bar graphs are oriented horizontally, with the number of isolates on the x-axis and the STs/countries on the y-axis. The third bar graph is oriented vertically, with the collection year on the x-axis and the number of isolates on the y-axis. Source data are provided as a Source Data file.

Fig. 2. Phylogenetic analysis of Clostridioides difficile ST11 and ST54 lineages.

a Maximum likelihood phylogeny of 568 ST11 isolates (48 npmA-positive and 520 contextual genomes) based on non-recombinant core single-nucleotide polymorphisms (SNPs) across 1954 core-genome alleles. b Maximum likelihood phylogeny of 105 ST54 isolates (8 npmA-positive and 97 contextual genomes) based on non-recombinant core SNPs across 2096 core-genome alleles. Both trees are midpoint rooted. Metadata columns show isolation niche, country of origin, npmA variant and presence of mobile genetic elements Tn7734 and ICE Tn7740. Source data are provided in the Source Data file.

Fig. 3. Genomic characterization of mobile genetic elements (MGEs) in npmA2-carrying C. difficile isolates.

a Genetic context using a 5 kb upstream and downstream window to npmA2. Blocks of the same color indicate homology, and the gray blocks represent areas of non-homology within the wider dataset. b Genetic context in the 30 kb flanking regions of npmA2. The top panel displays the inverse cumulative distribution function (cdf), which represents the decay of structural similarity with distance from npmA2, helping to identify common positions where non-homologous structural variation is introduced. The bottom panel illustrates the homologous blocks and their structural arrangement. Each block is colored according to homology, and the gray blocks indicate regions of non-homology. The bar plot on the left-hand side shows the number of contigs sharing each structure, colored by the source of the isolate. c Genetic structure and comparison of the ICE Tn7740 variants. d Schematic representation of the npmA2 integration sites using as a reference C. difficile 630 (GenBank accession AM180355.1) by mapping the npmA2-carrying contigs. Key integration sites include a helix-turn-helix transcriptional regulator gene (CD630_15730, coordinates: 1,821,726-1,822,940) and a distinct site in German ST11 isolates (CD630_19010, 2,203,327-2,203,836). In addition, an integration into a metallophosphoesterase gene (CD630_06890, 833,702-835,627) in an ST11 isolate and within a helix-turn-helix transcriptional regulator gene (CD630_02920, 353,054-354,148) for ST161 and ST54 clusters, along with a unique site in the ST36 cluster within a fic family protein gene (CD630_06470, 769,216-769,530).

Fig. 4. Integrative genomic dynamics of MGEs in the C. difficile ST11 cluster.

a Genetic comparison of the ICE Tn7740 integration site using npmA2-carrying and npmA-negative ST11 contextual isolates. Labels on the left of the alignment indicate C. difficile BioSample IDs. b Schematic diagram of the potential of ISCld1 to retain a copy at the integration site through the generation of a putative circular intermediate.

Fig. 5. Conserved genetic context of npmA2 between C. difficile and E. faecium.

Comparison of the genetic environment surrounding npmA2 in C. difficile isolate SAMEA1710827 and E. faecium isolate SAMEA4885232, showing conserved synteny and shared mobile genetic elements.