Investigating the mode of action of tuberculosis drugs using hypersensitive mutants of Mycobacterium smegmatis
Mycobacterium tuberculosis, the etiological agent of tuberculosis (TB), is the leading cause of death and disease by a bacterial pathogen worldwide. The growing incidence of drug resistant TB, especially multi-drug resistant TB highlights the need for new drugs with novel modes of action. Current treatment of TB involves a multi-drug regimen of four drugs including isoniazid and rifampicin, both of which were discovered over 40 years ago. Bedaquiline is one of the first novel TB drugs to enter clinical trials since the discovery of rifampicin, and has shown excellent activity against drug resistant TB. Although isoniazid and rifampicin are well established anti-TB drugs, significant gaps in knowledge regarding their modes of action exist. Furthermore, little information on the mode of action of the novel drug bedaquiline is known beyond its primary target. Characterisation of drug mode of action facilitates rational modifications of drugs to improve the treatment of TB.
The aim of this study was to identify novel aspects of the modes of action of isoniazid, rifampicin, and bedaquiline by characterising drug hypersensitive mutants of M. smegmatis mc²155. A sub-saturated M. smegmatis mc²155 transposon mutant collection with 1.1-fold genome coverage (7680 mutants) was constructed, with this collection estimated to contain mutations in 73.2% of all genes capable of maintaining a transposon insertion (non-essential genes). A high-throughput assay was developed for screening the collection, and mutants related to known drug mode of action were identified for isoniazid (ahpC and eccCa₁) and bedaquiline (atpB). Additionally, known mechanisms of drug inactivation were identified for isoniazid (nudC), rifampicin (arr and lspA), and bedaquiline (mmpL5). The finding that transposon mutants of nudC are hypersensitive to isoniazid independently validated the recent discovery of the role of NudC in basal isoniazid resistance by Wang et al. (2011). The remaining genes identified in this thesis represent potentially novel aspects of the modes of action or resistance mechanisms of these drugs.
Cross-sensitivity to other drugs indicated that the mechanism of sensitivity was drug specific for the mutants examined. Differential-sensitivity testing against drug analogues revealed that Arr is involved in resistance to the rifampicin analogue rifapentine as well, indicating that Arr can detoxify rifapentine similar to rifampicin. The nudC mutant showed increased sensitivity to a range of isoniazid analogues, indicating that it can detoxify these analogues similar to the parent compound. Interestingly six analogues were found to be less active against the nudC mutant than expected. A number of overexpression strains were tested against these six analogues; a nudC overexpression strain, and a strain overexpressing inhA, the primary target for isoniazid. Overexpression of nudC as well inhA increased the resistance of WT to isoniazid, but failed to increase resistance to three of the analogues, NSC27607, NSC33759, and NSC40350. Isoniazid is a prodrug and is activated by the peroxidase/catalase enzyme KatG. Overexpression of katG resulted in increased isoniazid sensitivity, as well as increased sensitivity to NSC27607, NSC33759, and NSC40350. Together these results suggest that NSC27607, NSC33759, and NSC40350 are activated by KatG, but that InhA is not the primary target. Additionally the inability of NudC overexpression to confer resistance suggests these analogues are not acting via a NAD adduct, the mechanism by which isoniazid inhibits InhA. These results suggest that there are other toxic metabolites being produced by KatG activation of these three analogues.
In conclusion, characterisation of mutants identified in a high-throughput assay for drug hypersensitivity identified genes involved in the modes of action or resistance mechanisms for isoniazid, rifampicin, and bedaquiline. Additionally, a number of novel genes were identified that have no known connections to the known modes of action or resistance mechanisms for these drugs. Further testing of a nudC mutant revealed three isoniazid analogues that appear to inhibit growth of M. smegmatis mc²155 independent of InhA, the primary target of isoniazid. This study has successfully demonstrated that screening for drug hypersensitivity can generate novel information on drug mode of action and resistance mechanisms. This information can ultimately be used to help drive the development of new drugs, and improve treatment of TB.