Rates and mechanisms of resistance development in Mycobacterium tuberculosis to a novel diarylquinoline ATP synthase inhibitor

Abstract

R207910 (also known as TMC207) is an investigational drug currently in clinical studies for the treatment of multidrug-resistant (MDR) tuberculosis. It has a high degree of antimycobacterial activity and is equally effective against drug-susceptible and MDR Mycobacterium tuberculosis isolates. In the present study, we characterized the development of resistance to R207910 in vitro. Ninety-seven independent R207910-resistant mutants were selected from seven different clinical isolates of M. tuberculosis (three drug-susceptible and four MDR isolates) at 10x, 30x, and 100x the MIC. At a concentration of 0.3 mg/liter (10x the MIC), the mutation rates ranged from 4.7 x 10(-7) to 8.9 x 10(-9) mutations per cell per division, and at 1.0 mg/liter (30x the MIC) the mutation rate ranged from 3.9 x 10(-8) to 2.4 x 10(-9). No resistant mutants were obtained at 3 mg/liter (100x the MIC). The level of resistance ranged from 0.12 to 3.84 mg/liter for the mutants identified; these concentrations represent 4- to 128-fold increases in the MICs. For 53 of the resistant mutants, the atpE gene, which encodes a transmembrane and oligomeric C subunit of the ATP synthase and which was previously shown to be involved in resistance, was sequenced. For 15/53 mutants, five different point mutations resulting in five different amino acid substitutions were identified in the atpE gene. For 38/53 mutants, no atpE mutations were found and sequencing of the complete F0 ATP synthase operon (atpB, atpE, and atpF genes) and the F1 ATP synthase operon (atpH, atpA, atpG, atpD, and atpC genes) from three mutants revealed no mutations, indicating other, alternative resistance mechanisms. Competition assays showed no measurable reduction in the fitness of the mutants compared to that of the isogenic wild types.

FIG. 1.

Mutation rates for seven different M. tuberculosis strains at three different drug concentrations: 10×, 30×, and 100× the MIC of the susceptible parent strains. Independent mutants were selected from six clinical isolates (isolates 1 to 6) and reference strain H37Rv (isolate 7) at three selection concentrations. Parent strains 4, 6, and 7 were drug susceptible (Susc) and strains 1 to 3 and 5 were MDR, as determined with a Bactec radiometric system (29).

FIG. 2.

MICs of R207910-resistant mutants. The MICs of 69 independent mutants selected with R207910 at 10× MIC and 27 selected with R207910 at 30× MIC were determined by the use of serial dilutions (0.015 to 3.84 mg/liter) on solid Middlebrook 7H9 agar. The MIC was defined as the first concentration with no visible growth. The MIC₅₀ for all mutants was 0.48 mg/liter, and the highest MIC was 3.84 mg/liter. The arrow and line mark the MIC level of the R207910-susceptible parent isolates (0.03 to 0.06 mg/liter). MDR, mutants selected from MDR parent strains; Susc, mutants selected from drug-susceptible parent strains.

FIG. 3.

Relative fitness of six different mutant classes. Competition assays between each mutant and its corresponding isogenic, R207910-susceptible parent strain were conducted in Middlebrook 7H9 broth; and the colonies were counted by plating on selective and nonselective Middlebrook 7H10 agar on days 0, 2, 4, 6, 8, and 10. The left y axis depicts the growth of the mutant and parent strains as the total numbers of CFU/ml. The right y axis depicts the ratio of ln numbers of CFU of resistant mutants (R)/ln numbers of CFU of susceptible parents (S). Due to difficulties with assessment of the Asp28Val mutation in the competition assays, single-culture growth curves were determined for the mutant and its isogenic parent strain.