Triazolopyrimidines Target Aerobic Respiration in Mycobacterium tuberculosis

ABSTRACT We previously identified a series of triazolopyrimidines with antitubercular activity. We determined that Mycobacterium tuberculosis strains with mutations in QcrB, a subunit of the cytochrome bcc-aa3 supercomplex, were resistant. A cytochrome bd oxidase deletion strain was more sensitive to this series. We isolated resistant mutants with mutations in Rv1339. Compounds led to the depletion of intracellular ATP levels and were active against intracellular bacteria, but they did not inhibit human mitochondrial respiration. These data are consistent with triazolopyrimidines acting via inhibition of QcrB.

activity resulted in higher sensitivity to the three compounds (Table 2). Taken together, these data strongly support the hypothesis that the target of the series is QcrB.
We have previously demonstrated that QcrB inhibitors lead to the depletion of intracellular ATP that is independent of the inhibition of growth and is consistent with disruption of the electron transport chain. We determined the effect of compounds on ATP levels (Fig. 2). M. tuberculosis was exposed to compounds for 24 h; ATP levels were measured using the BacTiter-Glo assay kit (Promega). Growth was measured by the optical density at 590 nm (OD 590 ). Q203 caused depletion of ATP levels at concentrations lower than the MIC (Fig. 2F). Similarly, ATP levels were reduced in a dose-dependent fashion on exposure to TZP molecules at concentrations which did not inhibit growth ( Fig. 2A to D). Depletion of ATP was not seen with the protein synthesis inhibitor kanamycin (Fig. 2E). These data further support the disruption of the electron transport chain as the mechanism of action of the TZP series.
We wanted to determine if there were additional targets or mechanism(s) of resistance, so we isolated and characterized mutants resistant to the series. We selected compounds from our original set with the lowest liquid MIC and determined the MIC against M. tuberculosis H37Rv ATCC 25618 on solid medium (Table 3). We selected two  compounds and plated ;10 8 bacteria onto 5Â and 10Â solid MIC as described (4). We isolated colonies and confirmed resistant mutants by measuring the MIC on solid medium; we obtained nine resistant isolates for TPN-0006239 and five resistant isolates for TPN-0006267 (Table 3). We sequenced the entire QcrB gene in all 14 isolates, but none of them had mutations (Table 3). We had previously seen mutations in Rv1339 leading to resistance to other QcrB inhibitors (5,9), so we sequenced Rv1339. We found the same mutation in 11 strains (P121L); 1 strain had the mutation S120P (Table 3). Two strains had no mutations in Rv1339. We have previously linked Rv1339 mutations to resistance to other QcrB inhibitor series, namely, the imidazopyridines and the phenoxyalkylimidazoles (5,9). Recent work in the related organism Mycobacterium smegmatis suggests that Rv1339 is an atypical class II cAMP phosphodiesterase that has been linked to antibiotic tolerance (13). In addition, a P94L mutation in Rv1399 led to increased persistence in animal models and increased resistance to external stress in Mycobacterium canetti, which was proposed to be due to changes in cell wall permeability (14). It is possible that the mutations we obtained lead to decreased compound permeation leading to resistance. However, it is unusual that resistance is seen largely with QcrB inhibitors, not as a general phenomenon; an alternative explanation for resistance could be changes in the intracellular ATP pool due to decreased turnover of cAMP.
We had demonstrated previously that this series had bacteriostatic activity against replicating M. tuberculosis but bactericidal activity against nonreplicating bacteria (1). We have noted this biological activity profile for other QcrB inhibitors, and thus, it is  Triazolopyrimidines Target Respiration in Mycobacterium tuberculosis Antimicrobial Agents and Chemotherapy consistent with it being an inhibitor of aerobic respiration (5,9,12). Since other QcrB inhibitors are active against intracellular bacteria, we tested the TZP series for activity against M. tuberculosis in human THP-1 macrophages. Macrophages were infected with M. tuberculosis expressing luciferase (15) at a multiplicity of infection of ;1 for 24 h, washed to remove extracellular bacteria, and then exposed to the compound for 72 h. Bacterial growth was measured by fluorescence. We tested five representative molecules, and all had potent activity with an 50% inhibitory concentration (IC 50 ) of ,1 mM (Table 4).
Since we identified the target of the TZP series as aerobic respiration, we determined whether the series might also inhibit mitochondrial respiration. We determined cytotoxicity against HepG2 cells cultured in Dulbecco's modified Eagle's medium (DMEM) with galactose as the carbon source to force the cells into using mitochondrial respiration (16). HepG2 cells were exposed to the compound for 72 h, and viability was measured using CellTiterGlo (Promega) (1). Of eight compounds, six showed some cytotoxicity (Table 5), although they still had a good selectivity index (activity was more potent against M. tuberculosis). We compared the IC 50 s under this condition to those generated when HepG2 cells were cultured in glucose when mitochondrial respiration is not active (1). There was less than a 2-fold difference in the cytotoxicity, confirming that molecules are not inhibiting eukaryotic respiration.
In conclusion, we have determined that the most likely target of the triazolopyrimidine series in M. tuberculosis is QcrB, a component of the electron transport chain. We demonstrated that mutations in either the target QcrB or the putative phosphodiesterase Rv1339 lead to resistance. This information adds another series of interest to the list of known or proposed QcrB inhibitors, which include the imidazopyridine amides   (5), morpholino thiophenes (6), quinolinyl acetamides (19), pyrazolopyridines (20), and arylvinylpiperazine amides (21). Since QcrB is a clinically validated target (22), this series is an attractive one to develop further. a HepG2 cells were cultured in medium containing either galactose or glucose as the carbon source. b IC 50 , the concentration required to reduce cell number by 50%, was determined after 3 days of exposure to compounds.