Target Identification of Mycobacterium tuberculosis Phenotypic Hits Using a Concerted Chemogenomic, Biophysical, and Structural Approach

Grace Mugumbate, V. Mendes, M. Błaszczyk, Mohamad Sabbah, George Papadatos, Joël Lelièvre, Lluís Ballell, David Barros, Chris Abell, Tom L. Blundell, John P. OveringtonFrontiers in Pharmacology 2017 · DOI

Abstract

Mycobacterium phenotypic hits are a good reservoir for new chemotypes for the treatment of tuberculosis. However, the absence of defined molecular targets and modes of action could lead to failure in drug development. Therefore, a combination of ligand-based and structure-based chemogenomic approaches followed by biophysical and biochemical validation have been used to identify targets for Mycobacterium tuberculosis phenotypic hits. Our approach identified EthR and InhA as targets for several hits, with some showing dual activity against these proteins. From the 35 predicted EthR inhibitors, eight exhibited an IC50 below 50 μM against M. tuberculosis EthR and three were confirmed to be also simultaneously active against InhA. Further hit validation was performed using X-ray crystallography yielding eight new crystal structures of EthR inhibitors. Although the EthR inhibitors attain their activity against M. tuberculosis by hitting yet undefined targets, these results provide new lead compounds that could be further developed to be used to potentiate the effect of EthA activated pro-drugs, such as ethionamide, thus enhancing their bactericidal effect.

Introduction

Phenotypic screening has historically been a highly successful strategy for identifying effective "hits" against Mycobacterium tuberculosis (Mtb). However, this approach often leaves researchers with a "black box" regarding the molecule’s mode of action (MoA). Without a defined molecular target, optimizing drug potency, predicting side effects, and understanding resistance mechanisms become significant hurdles. This study addresses the challenge of target identification by applying a multi-layered, "concerted" pipeline to a library of 776 anti-tuberculosis hits.

Methodology: A Concerted Pipeline

The researchers employed a multi-layered approach to narrow down potential protein-ligand interactions:

  1. Ligand-Based Prediction: Computational methods, including Naive Bayesian Classifiers (MCNBC) and the Similarity Ensemble Approach (SEA), compared phenotypic hits against known drug-target pairs in the ChEMBL database.
  2. Structure-Based Prediction: Hits were docked into the 3D crystal structures of high-priority Mtb targets, specifically EthR (a transcriptional regulator) and InhA (an enzyme essential for cell wall synthesis), using Internal Coordinate Mechanism (ICM) scoring.
  3. Experimental Validation: Predicted compounds were tested using Surface Plasmon Resonance (SPR) for EthR binding and enzymatic assays for InhA inhibition. High-resolution X-ray crystallography was then used to confirm the binding modes within the EthR protein.

Results and Validation

The concerted approach yielded high prediction accuracy. Of the 35 compounds predicted to hit EthR and subsequently tested, 25 (71%) showed significant inhibition (>50% at 100 µM).

Venn diagrams of method overlaps (A, B) and experimental validation of EthR inhibition for 35 selected compounds (C, D).

Eight compounds were identified as potent EthR inhibitors with $IC_{50}$ values below 50 µM. The following figure illustrates the dose-response relationship for Compound 1, one of the primary leads identified in this study.

Dose-response curve for Compound 1 against EthR, demonstrating potent inhibition.

Furthermore, the team solved eight new crystal structures of EthR-inhibitor complexes. These structures revealed that the EthR binding tunnel is a flexible, hydrophobic cavity capable of accommodating diverse chemical scaffolds. Notably, three compounds were confirmed to inhibit both EthR and InhA, providing rare starting points for dual-target therapy.

Discussion and Therapeutic Implications

EthR acts as a repressor for the activation of the pro-drug ethionamide. By inhibiting EthR, these compounds can "boost" the effectiveness of existing TB treatments, potentially reducing required dosages and associated toxicity. While the identified inhibitors may not fully explain the original phenotypic activity—suggesting additional essential targets or polypharmacology—they represent significant leads. Future work will focus on improving the solubility and "drug-likeness" of these compounds to bridge the gap between biochemical potency and whole-cell activity.

Summary updated

Abstract

Mycobacterium phenotypic hits are a good reservoir for new chemotypes for the treatment of tuberculosis. However, the absence of defined molecular targets and modes of action could lead to failure in drug development. Therefore, a combination of ligand-based and structure-based chemogenomic approaches followed by biophysical and biochemical validation have been used to identify targets for <i>Mycobacterium tuberculosis</i> phenotypic hits. Our approach identified EthR and InhA as targets for several hits, with some showing dual activity against these proteins. From the 35 predicted EthR inhibitors, eight exhibited an IC<sub>50</sub> below 50 μM against <i>M. tuberculosis</i> EthR and three were confirmed to be also simultaneously active against InhA. Further hit validation was performed using X-ray crystallography yielding eight new crystal structures of EthR inhibitors. Although the EthR inhibitors attain their activity against <i>M. tuberculosis</i> by hitting yet undefined targets, these results provide new lead compounds that could be further developed to be used to potentiate the effect of EthA activated pro-drugs, such as ethionamide, thus enhancing their bactericidal effect.
BibTeX
@inproceedings{mugumbate2017target,
  author = {Grace Mugumbate and V. Mendes and M. Błaszczyk and Mohamad Sabbah and George Papadatos and Joël Lelièvre and Lluís Ballell and David Barros and Chris Abell and Tom L. Blundell and John P. Overington},
  title = {{Target Identification of Mycobacterium tuberculosis Phenotypic Hits Using a Concerted Chemogenomic, Biophysical, and Structural Approach}},
  booktitle = {Frontiers in Pharmacology},
  year = {2017},
  doi = {10.3389/fphar.2017.00681},
}
Figure