Metabolic profiles of multidrug resistant and extensively drug resistant Mycobacterium tuberculosis unveiled by metabolomics.

Although treatable with antibiotics, tuberculosis is a leading cause of death. Mycobacterium tuberculosis antibiotic resistance is becoming increasingly common and disease control is challenging. Conventional drug susceptibility testing takes weeks to produce results, and treatment is often initiated empirically. Therefore, new methods to determine drug susceptibility profiles are urgent. Here, we used mass-spectrometry-based metabolomics to characterize the metabolic landscape of drug-susceptible (DS), multidrug-resistant (MDR) and extensively drug-resistant (XDR) M. tuberculosis. Direct infusion mass spectrometry data showed that DS, MDR, and XDR strains have distinct metabolic profiles, which can be used to predict drug susceptibility and resistance. This was later confirmed by Ultra-High-Performance Liquid Chromatography and High-Resolution Mass Spectrometry, where we found that levels of ions presumptively identified as isoleucine, proline, hercynine, betaine, and pantothenic acid varied significantly between strains with different drug susceptibility profiles. We then confirmed the identification of proline and isoleucine and determined their absolute concentrations in bacterial extracts, and found significantly higher levels of these amino acids in DS strains, as compared to drug-resistant strains (combined MDR and XDR strains). Our results advance the current understanding of the effect of drug resistance on bacterial metabolism and open avenues for the detection of drug resistance biomarkers.

Integrated analysis of ethionamide resistance loci in Mycobacterium tuberculosis clinical isolates.

Tuberculosis patients taking second line drugs such as ethionamide (ETH) have often experienced previous treatment failure and usually have a complex history of disease and treatment that can span decades. Mutations in the ETH activating enzyme, EthA, confer resistance through undescribed mechanisms. To explore the impact of EthA mutations on ETH resistance, data from a total of 160 ETHR isolates was analysed. The most frequently mutated positions are within regions that display sequence conservation with the active site of OTEMO, another FAD-containing NADH-binding Baeyer-Villiger monooxygenase (BVMO), or with the sugar binding site of galectin-4N. Additionally, to look at a possible role of EthR on ETH resistance we purified an EthR mutant identified in a clinical isolate, F110L, and found it to bind the ethA-ethR intergenic region with higher affinity than the wild type regulator in gel shift assays. The ability of cyclic di-GMP to enhance DNA binding is maintained in the EthR mutant. To our knowledge, this is the first ETH resistance study that combines sequence and resistance data of clinical isolates with functional and structural information.