In silico structural analysis of EthA substitutions for ranking priority mutations leading to ethionamide resistance in Mycobacterium tuberculosis

BackgroundTuberculosis (TB) is the second-leading cause of deaths from infectious agents and remains a global health threat. Ethionamide (ETH) is a prodrug used in regimens for multidrug-resistant TB, and, partly due to side effects that can lead to low treatment adhesion, resistance arises. Changes in EthA, the monooxygenase that activates ETH, are the main mechanism of resistance. Yet, of hundreds of EthA substitutions found in resistant isolates, only a handful have been annotated as resistance determinants. ResultsAn in silico analysis was carried out on a previously described panel of Mycobacterium tuberculosis clinical isolates for which genomes and ETH susceptibility testing results were available. EthA substitutions were mapped, revealing the existence of hotspots in its sequence. Visualization of the hotspots in the EthA structural model shows that they cluster in three regions, including ligand binding pockets. Models were built of twenty-three variants found in resistant isolates and changes in local configuration was mapped to identify investigate impact on ETH activation. Information from these models contributed to establishing five criteria for scoring whether substitutions are most likely to lead to resistance. Using these criteria, EthA D58G was selected and its expression is shown to increase growth in high ETH concentrations. ConclusionFunctionally relevant regions of EthA are revealed and point out priority substitutions for functional studies, enhancing identification and detection of substitutions not been previously associated with resistance.