Extensive screening reveals previously undiscovered aminoglycoside resistance genes in human pathogens

Abstract

Antibiotic resistance is a growing threat to human health, caused in part by pathogens accumulating antibiotic resistance genes (ARGs) through horizontal gene transfer. New ARGs are typically not recognized until they have become widely disseminated, which limits our ability to reduce their spread. In this study, we use large-scale computational screening of bacterial genomes to identify previously undiscovered mobile ARGs in pathogens. From ~1 million genomes, we predict 1,071,815 genes encoding 34,053 unique aminoglycoside-modifying enzymes (AMEs). These cluster into 7,612 families (<70% amino acid identity) of which 88 are previously described. Fifty new AME families are associated with mobile genetic elements and pathogenic hosts. From these, 24 of 28 experimentally tested AMEs confer resistance to aminoglycoside(s) in Escherichia coli, with 17 providing resistance above clinical breakpoints. This study greatly expands the range of clinically relevant aminoglycoside resistance determinants and demonstrates that computational methods enable early discovery of potentially emerging ARGs.

Fig. 1. The number and distribution of the predicted AME families, divided into known and new AMEs.

a AME families carried by pathogens. b AME families carried by all species. c Mobile AME families carried by pathogens. d Mobile AME families carried by all species.

Fig. 2. The number of AME families carried by pathogenic species that were associated with different combinations of genes relating to mobile genetic elements (MGEs), including conjugation systems, insertion sequences (IS), integrons, and/or other known mobile antibiotic resistance genes (ARGs).

The bars at the bottom indicate the distribution of genes predicted by each of the nine models within each category. a Families representing new AMEs and b Families representing known AMEs.

Fig. 3. Metadata about the isolates (n = 92,056) carrying new AMEs that were associated with mobile genetic elements in pathogens (representing 50 AME families).

The panels show the number of AME families divided based on their hosts’ isolation a environment (data available for 14,747 isolates (16%) representing 39 AME families (78%)), b continent (data available for 10,631 isolates (12%) representing 41 AME families (82%)), and c collection date (data available for 10,260 isolates (11%) representing 41 AME families (82%)). Note that a family can contain gene variants carried by multiple hosts from different sources.

Fig. 4. Results from disk diffusion tests using E. coli and 28 selected new AMEs.

Panels a–g show the mean inhibition zone diameter difference [mm] between clones carrying new AMEs (n = 3 for each gene and antibiotic) and susceptible controls (n = 6 for each antibiotic) for seven different aminoglycosides: amikacin (AK 30 µg), gentamicin (CN 30 µg), kanamycin (K 30 µg), neomycin (N 30 µg), netilmicin (NET 10 µg), spectinomycin (SH 25 µg), and tobramycin (TOB 30 µg). Individual data points are presented in Supplementary Fig. 6. Significantly increased growth (p-value < 0.01, one-sided two-sample t-test) is denoted by an asterisk above the bar, with red asterisks indicating a resistance level beyond the clinical breakpoint (amikacin, gentamicin, tobramycin) or ECOFF (neomycin). Standard deviations are displayed as error bars. Panel h shows an overview of the tested antibiotic resistance genes and the aminoglycoside(s) that each gene conferred significantly increased resistance to, with asterisks denoting clinical levels of resistance.