Discovery of Novel Oral Protein Synthesis Inhibitors of Mycobacterium tuberculosis That Target Leucyl-tRNA Synthetase

Abstract

The recent development and spread of extensively drug-resistant and totally drug-resistant resistant (TDR) strains of Mycobacterium tuberculosis highlight the need for new antitubercular drugs. Protein synthesis inhibitors have played an important role in the treatment of tuberculosis (TB) starting with the inclusion of streptomycin in the first combination therapies. Although parenteral aminoglycosides are a key component of therapy for multidrug-resistant TB, the oxazolidinone linezolid is the only orally available protein synthesis inhibitor that is effective against TB. Here, we show that small-molecule inhibitors of aminoacyl-tRNA synthetases (AARSs), which are known to be excellent antibacterial protein synthesis targets, are orally bioavailable and effective against M. tuberculosis in TB mouse infection models. We applied the oxaborole tRNA-trapping (OBORT) mechanism, which was first developed to target fungal cytoplasmic leucyl-tRNA synthetase (LeuRS), to M. tuberculosis LeuRS. X-ray crystallography was used to guide the design of LeuRS inhibitors that have good biochemical potency and excellent whole-cell activity against M. tuberculosis Importantly, their good oral bioavailability translates into in vivo efficacy in both the acute and chronic mouse models of TB with potency comparable to that of the frontline drug isoniazid.

FIG 1

In vitro structure-activity relationships. M.tb, M. tuberculosis; NT, not tested.

FIG 2

Compound 1-resistant mutants bear mutations in the editing domain of LeuRS. (A) Domain map of M. tuberculosis LeuRS. (B) Amino acid alignment of part of the editing domain of LeuRS from M. tuberculosis and M. smegmatis. Identical residues are shown in blue, and nonidentical residues are shown in red, with arrows indicating where the mutations were found.

FIG 3

In vivo efficacy of compound 2 (Cmp 2) in a murine GKO (C57BL/6-Ifngtm1ts) model of acute TB. Oral treatment was started 15 days (start) after infection with a low-dose aerosol of M. tuberculosis Erdman lux and continued for 9 consecutive daily treatments until day 23, when mice were euthanized on day 24 (end). CFU in lungs and spleens were determined, and means from five mice for drug-treated groups and from 6 mice per group for the untreated controls are shown. *, P < 0.001 by pairwise multiple-comparison procedures (Tukey test) compared to controls.

FIG 4

X-ray cocrystal structure of LeuRS with compound 2. (A) Crystal structure of the M. tuberculosis LeuRS editing domain in complex with compound 2 (carbon atoms are shown in green)-AMP (carbon atoms are shown in magenta). Color coding is the same throughout all figures, with blue for nitrogen, red for oxygen, pink for boron, orange for phosphorus, and yellow for sulfur. (B) Zoomed view into the editing site of M. tuberculosis LeuRS showing the compound 2-AMP adduct and the key residues establishing important hydrogen bonds (red dashed lines), with only the H bond from the 3-aminomethyl to M441 being omitted for clarity. (C) Overlay of the LeuRS editing domain of M. tuberculosis and E. coli in complex with methionine (in yellow) (PDB accession number 2AJH). The 3-aminomethyl group of compound 2 mimics the amino group of methionine, including the interaction with the bacterium-specific residue D447.

FIG 5

In vivo efficacy of compounds 11, 12, 13, and 14 in models of acute and chronic of TB infections. (A) In vivo efficacy in a murine GKO (C57BL/6-Ifngtm1ts) model of acute TB. Compounds were dosed orally daily for 14 days after 10 days of infection (start) with a low-dose aerosol of M. tuberculosis Erdman. Mean lung CFU were determined from five mice at the end. (B) In vivo efficacy in a murine GKO (C57BL/6-Ifngtm1ts) model of acute TB. Oral treatment was started 13 days after infection (start) with a low-dose aerosol of M. tuberculosis Erdman lux and continued for 9 consecutive daily treatments until day 21, when mice were sacrificed on day 22 (end). Mean lung CFU were determined from five mice at the end. (C) In vivo efficacy in a murine BALB/c model of chronic TB infection. Compounds were dosed orally 5 days a week for 4 weeks after infection with M. tuberculosis Erdman with a low-dose aerosol 21 days prior (start). Lung and spleen CFU were determined from six mice at the end. **, P < 0.01; *, P < 0.001 (by pairwise multiple-comparison procedures [Tukey test] compared to controls).

FIG 6

M. tuberculosis H37Rv in vitro kill kinetics. Cells were incubated with compounds at 20 times their MIC values for different times over 14 days in 10 ml of Middlebrook 7H9 medium containing 10% (vol/vol) ADC and 0.05% (vol/vol) Tween 80. The MIC values used in this experiment were 0.013 μg/ml, 0.6 μg/ml, and 0.06 μg/ml for compound 14, linezolid, and moxifloxacin, respectively. The means and the standard deviations of data from triplicate cultures at each point are shown.

FIG 7

Efficacy of compound 14 in a mouse model of chronic TB infection. C57BL/6J mice were infected with M. tuberculosis H37Rv intratracheally (∼10² CFU) and were dosed once daily for 8 weeks starting 6 weeks after infection. Mice were sacrificed 24 h after the last drug administration. Every column represents the mean values ± standard deviations of data from 7 mice per group for untreated and linezolid-treated groups and from 3 mice for compound 14-treated mice. *, P < 0.001 by pairwise multiple-comparison procedures (Tukey test) compared to controls.

FIG 8

Efficacy of compound 14 in a mouse model of acute TB infection under different dosing regimes of once a day (QD), twice a day (BID), or every other day (q48h). C57BL/6J mice were infected with M. tuberculosis H37Rv intratracheally (∼10⁵ CFU) and were dosed starting on the following day after infection for 8 days. Only one dose was administered on day 8 under the BID schedule. Mice were sacrificed at least 24 h after the last drug administration. Every dot represents one mouse data point except for linezolid (mean of data for 5 mice ± standard deviation).