Genetic Resistance Determinants, In Vitro Time-Kill Curve Analysis and Pharmacodynamic Functions for the Novel Topoisomerase II Inhibitor ETX0914 (AZD0914) in Neisseria gonorrhoeae

Abstract

Resistance in Neisseria gonorrhoeae to all available therapeutic antimicrobials has emerged and new efficacious drugs for treatment of gonorrhea are essential. The topoisomerase II inhibitor ETX0914 (also known as AZD0914) is a new spiropyrimidinetrione antimicrobial that has different mechanisms of action from all previous and current gonorrhea treatment options. In this study, the N. gonorrhoeae resistance determinants for ETX0914 were further described and the effects of ETX0914 on the growth of N. gonorrhoeae (ETX0914 wild type, single step selected resistant mutants, and efflux pump mutants) were examined in a novel in vitro time-kill curve analysis to estimate pharmacodynamic parameters of the new antimicrobial. For comparison, ciprofloxacin, azithromycin, ceftriaxone, and tetracycline were also examined (separately and in combination with ETX0914). ETX0914 was rapidly bactericidal for all wild type strains and had similar pharmacodynamic properties to ciprofloxacin. All selected resistant mutants contained mutations in amino acid codons D429 or K450 of GyrB and inactivation of the MtrCDE efflux pump fully restored the susceptibility to ETX0914. ETX0914 alone and in combination with azithromycin and ceftriaxone was highly effective against N. gonorrhoeae and synergistic interaction with ciprofloxacin, particularly for ETX0914-resistant mutants, was found. ETX0914, monotherapy or in combination with azithromycin (to cover additional sexually transmitted infections), should be considered for phase III clinical trials and future gonorrhea treatment.

Keywords: DNA topoisomerase II inhibitor; ETX0914; antimicrobial resistance; gonorrhea; pharmacodynamics; time-kill curve analysis; treatment.

FIGURE 1

Time-kill curves of six different Neisseria gonorrhoeae strains with ETX0914: H041 (A), WHO F (B), WHO O (C), OM-5 (D; in vitro selected resistant mutant from WHO O), WHO P (E), and PM-4 (F; in vitro selected resistant mutant from WHO P). Untreated controls are shown in red, each line corresponds to a doubling concentration of the antimicrobial, and circles indicate each measured time point. The limit of detection in the assay was 100 colony forming units (CFUs)/mL.

FIGURE 2

Pharmacodynamic functions based on the time-kill curve data for ETX0914 and six different Neisseria gonorrhoeae strains: H041 (A), WHO F (B), WHO O (C), OM-5 (D; in vitro selected resistant mutant from WHO O), WHO P (E), and PM-4 (F; in vitro selected resistant mutant from WHO P). The bacterial growth rates per hour derived from linear regression on time-kill curves for each ETX0914 concentration are plotted. Growth in absence of ETX0914 was plotted as a red circle, at a tenth of the lowest concentration to allow plotting on a log scale, and decreasing ETX0914 concentrations are shown from dark to light blue color.

FIGURE 3

Comparisons of pharmacodynamic functions based on the time-kill curve data for ETX0914 (A), ciprofloxacin (B), azithromycin (C), ceftriaxone (D), and tetracycline (E) and six different Neisseria gonorrhoeae strains: H041, WHO F, WHO O, OM-5 (in vitro selected resistant mutant from WHO O), WHO P, and PM-4 (in vitro selected resistant mutant from WHO P). The bacterial growth rates per hour derived from linear regression on time-kill curves for each antimicrobial concentration are plotted.

FIGURE 4

Pharmacodynamic properties of ETX0914 in direct comparison to ciprofloxacin, azithromycin, ceftriaxone and tetracycline. (A) Shows a hierarchical clustering for the minimal bacterial net growth rate at high concentrations of antimicrobial (parameter ψ_min), (B) for the steepness of the slope of the pharmacodynamic function (Hill coefficient, κ), (C) shows the maximal growth rate in the absence of antimicrobial (parameter ψ_max) for each strain calculated as mean and standard error from all independent experiments (Supplementary Table S1) and (D) the correlation between MICs measured with conventional methods (Etest or agar dilution technique; values below <0.002 mg/L and >32 mg/L (Etest detection limit) were excluded from the analysis) and pharmacodynamic MIC (zMIC). The linear regression line is drawn in black and the confidence interval shaded in gray. The heatmap was produced by plotting the numeric values of ψ_min and the Hill coefficient (κ) in a color gradient from red (high values) to blue (low values). The hierarchical clustering was performed using the complete linkage algorithm as implemented in the package heatmap3 (Zhao et al., 2014). The distances in the resulting dendrogram represented similarities between strains and antimicrobials.