Identification and mitigation of a reactive metabolite liability associated with aminoimidazoles.

Reactive metabolites (RMs) have been implicated as causal factors in many drug-associated idiosyncratic toxicities. This study aims at identification and mitigation of an RM liability associated with aminoimidazole and amino(aza)benzimidazole structural motifs from an antimalarial project. Nineteen compounds with different structural modifications were studied in rat and human liver microsomes using glutathione (GSH) and N-acetyl cysteine (NAC) as trapping agents for RM. Metabolite profiling of aminoimidazole compounds in initial studies revealed the presence of dihydrodiol metabolites suggestive of reactive epoxide precursors, confirmed by the identification of a dihydrohydroxy GSH conjugate in GSH supplemented incubations. Substitution of methyl group at a potential site of metabolism blocked the epoxidation; however, formation of an imine-methide RM was suspected. Masking the site of metabolism via benzimidazole and 4/7-azabenzimidazole resulted in the possible formation of quinone-imine intermediates as a product of bioactivation. Further, substitutions with electron withdrawing groups and steric crowding did not address this liability. Mitigation of bioactivation was achieved with 5/6-azabenzimidazole and with CF3 substitution at the 6-position of the 7-azabenzimidazole ring. Moreover, compounds devoid of imidazole -NH2 do not undergo bioactivation. This study, therefore, establishes aminoimidazole and amino(aza)benzimidazoles as potential toxicophores and describes ways to mitigate this bioactivation liability by chemical modification.

Thiazolopyridone ureas as DNA gyrase B inhibitors: optimization of antitubercular activity and efficacy.

Scaffold hopping from the thiazolopyridine ureas led to thiazolopyridone ureas with potent antitubercular activity acting through inhibition of DNA GyrB ATPase activity. Structural diversity was introduced, by extension of substituents from the thiazolopyridone N-4 position, to access hydrophobic interactions in the ribose pocket of the ATP binding region of GyrB. Further optimization of hydrogen bond interactions with arginines in site-2 of GyrB active site pocket led to potent inhibition of the enzyme (IC50 2 nM) along with potent cellular activity (MIC=0.1 µM) against Mycobacterium tuberculosis (Mtb). Efficacy was demonstrated in an acute mouse model of tuberculosis on oral administration.

Azaindoles: noncovalent DprE1 inhibitors from scaffold morphing efforts, kill Mycobacterium tuberculosis and are efficacious in vivo.

We report 1,4-azaindoles as a new inhibitor class that kills Mycobacterium tuberculosis in vitro and demonstrates efficacy in mouse tuberculosis models. The series emerged from scaffold morphing efforts and was demonstrated to noncovalently inhibit decaprenylphosphoryl-ß-D-ribose2'-epimerase (DprE1). With "drug-like" properties and no expectation of pre-existing resistance in the clinic, this chemical class has the potential to be developed as a therapy for drug-sensitive and drug-resistant tuberculosis.

Thiazolopyridine ureas as novel antitubercular agents acting through inhibition of DNA Gyrase B.

A pharmacophore-based search led to the identification of thiazolopyridine ureas as a novel scaffold with antitubercular activity acting through inhibition of DNA Gyrase B (GyrB) ATPase. Evaluation of the binding mode of thiazolopyridines in a Mycobacterium tuberculosis (Mtb) GyrB homology model prompted exploration of the side chains at the thiazolopyridine ring C-5 position to access the ribose/solvent pocket. Potent compounds with GyrB IC50 ≤ 1 nM and Mtb MIC ≤ 0.1 µM were obtained with certain combinations of side chains at the C-5 position and heterocycles at the C-6 position of the thiazolopyridine core. Substitutions at C-5 also enabled optimization of the physicochemical properties. Representative compounds were cocrystallized with Streptococcus pneumoniae (Spn) ParE; these confirmed the binding modes predicted by the homology model. The target link to GyrB was confirmed by genetic mapping of the mutations conferring resistance to thiazolopyridine ureas. The compounds are bactericidal in vitro and efficacious in vivo in an acute murine model of tuberculosis.

In vitro and in vivo efficacy of ß-lactams against replicating and slowly growing/nonreplicating Mycobacterium tuberculosis.

Beta-lactams, in combination with beta-lactamase inhibitors, are reported to have activity against Mycobacterium tuberculosis bacteria growing in broth, as well as inside the human macrophage. We tested representative beta-lactams belonging to 3 different classes for activity against replicating M. tuberculosis in broth and nonreplicating M. tuberculosis under hypoxia, as well as against streptomycin-starved M. tuberculosis strain 18b (ss18b) in the presence or absence of clavulanate. Most of the combinations showed bactericidal activity against replicating M. tuberculosis, with up to 200-fold improvement in potency in the presence of clavulanate. None of the combinations, including those containing meropenem, imipenem, and faropenem, killed M. tuberculosis under hypoxia. However, faropenem- and meropenem-containing combinations killed strain ss18b moderately. We tested the bactericidal activities of meropenem-clavulanate and amoxicillin-clavulanate combinations in the acute and chronic aerosol infection models of tuberculosis in BALB/c mice. Based on pharmacokinetic/pharmacodynamic indexes reported for beta-lactams against other bacterial pathogens, a cumulative percentage of a 24-h period that the drug concentration exceeds the MIC under steady-state pharmacokinetic conditions (%TMIC) of 20 to 40% was achieved in mice using a suitable dosing regimen. Both combinations showed marginal reduction in lung CFU compared to the late controls in the acute model, whereas both were inactive in the chronic model.

Effect of repeated dosing on rifampin exposure in BALB/c mice.

The discovery of novel therapeutics for the treatment of tuberculosis involves routine testing in a mouse model over four weeks of daily dosing with test compounds. In this model, daily oral administration of rifampin (10 mg/kg) showed significantly lower plasma exposure on day 5 compared to day 1. The absence of PXR-mediated induction of mouse Cyp3a isoforms was confirmed in the present study by incubating liver microsomes prepared from control and rifampin treated mice with probe substrates of CYP3A. To test whether the reduction in exposure was due to Pgp-mediated efflux, verapamil, a known Pgp inhibitor, was dosed to the rifampin pre-treated mice which led to an increase in exposure to that obtained after a single dose of rifampin, suggesting the role of Pgp induction in reducing exposure to rifampin. To further confirm Pgp induction in rifampin treated mice, digoxin, a known substrate of Pgp, was administered to the rifampin pre-treated mice, and a significant drop in the digoxin exposure was observed compared to the control group. Collectively, our results show that repeated administration of rifampin in mice leads to a reduction in oral exposure due to induction of Pgp-mediated efflux of rifampin, and not via induction of CYP3A isoforms.