Rapid Evolution of Reduced Susceptibility against a Balanced Dual-Targeting Antibiotic through Stepping-Stone Mutations

Abstract

Multitargeting antibiotics, i.e., single compounds capable of inhibiting two or more bacterial targets, are generally considered to be a promising therapeutic strategy against resistance evolution. The rationale for this theory is that multitargeting antibiotics demand the simultaneous acquisition of multiple mutations at their respective target genes to achieve significant resistance. The theory presumes that individual mutations provide little or no benefit to the bacterial host. Here, we propose that such individual stepping-stone mutations can be prevalent in clinical bacterial isolates, as they provide significant resistance to other antimicrobial agents. To test this possibility, we focused on gepotidacin, an antibiotic candidate that selectively inhibits both bacterial DNA gyrase and topoisomerase IV. In a susceptible organism, Klebsiella pneumoniae, a combination of two specific mutations in these target proteins provide an >2,000-fold reduction in susceptibility, while individually, none of these mutations affect resistance significantly. Alarmingly, strains with decreased susceptibility against gepotidacin are found to be as virulent as the wild-type Klebsiella pneumoniae strain in a murine model. Moreover, numerous pathogenic isolates carry mutations which could promote the evolution of clinically significant reduction of susceptibility against gepotidacin in the future. As might be expected, prolonged exposure to ciprofloxacin, a clinically widely employed gyrase inhibitor, coselected for reduced susceptibility against gepotidacin. We conclude that extensive antibiotic usage could select for mutations that serve as stepping-stones toward resistance against antimicrobial compounds still under development. Our research indicates that even balanced multitargeting antibiotics are prone to resistance evolution.

Keywords: antibiotic resistance; genome engineering; gepotidacin.

FIG 1

Binding site analysis of gepotidacin based on directed evolution and molecular modeling. (A) Gepotidacin, a novel triazaacenaphthylene antibiotic candidate, inhibits DNA gyrase and topoisomerase IV in Gram-negative bacteria. (B and C) Ribbon representation of E. coli’s DNA gyrase and topoisomerase IV in complex with gepotidacin, based on molecular dynamics simulations. Inset shows the closeup view of gepotidacin (yellow) and its interacting residues (red) in a stick model. DNA is shown in magenta and blue. (D) The workflow of in vivo directed evolution of reduced susceptibility to gepotidacin in K. pneumoniae ATCC 10031.

FIG 2

Fitness cost and virulence of bacteria with reduced susceptibility to gepotidacin. (A) In vitro competition between mutant and wild-type Klebsiella pneumoniae. Isogenic mutants of Klebsiella pneumoniae ATCC 10031 carrying either a mutation combination conferring reduced susceptibility to gepotidacin or its single-step constituents (red), or clinically occurring fluoroquinolone resistance-associated mutations (blue) competed against the wild-type strain. In competition assays, a competitive index of <0 indicates that the wild-type population outcompetes the mutant population, and conversely, a competition index of >0 indicates that the mutant population outcompetes the wild-type population. Error bars indicate the standard deviation (SD) based on five biological replicates. (B) Virulence of Klebsiella pneumoniae mutants and the wild-type strain in a murine thigh infection model. Shown are the bacterial burdens in infected thigh tissues after 26 h of infection caused by wild-type Klebsiella pneumoniae ATCC 10031 (black) or isogenic mutants carrying either mutations causing reduced susceptibility to gepotidacin or its single-step constituents (red), or clinically occurring mutations associated with fluoroquinolone resistance (blue). The bacterial burden was assayed in CFU per gram of tissue (n = 5 animals per data point, see Materials and Methods for details). No mutants were found to display a significant difference in virulence compared to the wild-type strain (t test, P > 0.05 for all pairwise comparisons).

FIG 3

Adaptive laboratory evolution of Klebsiella pneumoniae and Escherichia coli under ciprofloxacin stress. (A) Antibiotic concentrations at which K. pneumoniae ATCC 10031 as well as E. coli K-12 MG1655 wild-type and ΔmutS hypermutator strains were able to grow under increasing ciprofloxacin stress as a function of time (number of cell generations). Dashed line represents the clinical breakpoint of ciprofloxacin resistance according to EUCAST (40). (B) Relative resistance level of the evolved lines against ciprofloxacin (gray bars) and gepotidacin (yellow bars) compared to the wild type at the end of the adaptive laboratory evolution. Asterisk (*) indicates that the evolved line was subjected to whole-genome sequence analysis to uncover mutational processes behind the reduced susceptibility.