Fighting tuberculosis by drugs targeting nonreplicating Mycobacterium tuberculosis bacilli.

Abstract

Current tuberculosis (TB) treatment requires 6 months of combination therapy with isoniazid (INH), rifampin (RIF), pyrazinamide (PZA), and ethambutol for active TB and 9 months of INH or 3 months of rifapentine (RFP) + INH for latent TB. The lungs of patients with active and latent TB contain heterogeneous mixtures of cellular and caseous granulomas harboring Mycobacterium tuberculosis bacilli ranging from actively replicating (AR) to nonreplicating (NR), phenotypically drug-resistant stages. Several in vitro models to obtain NR cells were reported, including exposure to hypoxia, nutrient starvation, acid + nitric oxide, and stationary phase. Overall, these models showed that RIF, RFP, PA-824 (PA), metronidazole (MZ), bedaquiline (BQ), and fluoroquinolones were the most active drugs against NR M. tuberculosis. In hypoxia at pH 5.8, some combinations killed AR plus NR cells, as shown by lack of regrowth in liquid media, whereas in hypoxia at pH 7.3 (the pH of the caseum), only RIF and RFP efficiently killed NR bacilli while several other drugs showed little effect. In conventional mouse models, combinations containing RFP, BQ, PA, PZA, moxifloxacin, sutezolid, linezolid, and clofazimine sterilized animals in ≤2 months, as shown by lack of viable bacilli in lung homogenates after 3 months without therapy. Drugs were less effective in C3HeB/FeJ mice forming caseous granulomas. Overall, in vitro observations and in vivo studies suggest that the search for new TB drugs could be addressed to low lipophilic molecules (e.g., new rpoB inhibitors with clogP < 3) killing NR M. tuberculosis in hypoxia at neutral pH and reaching high rates of unbound drug in the caseum.