Identification of anti-virulence small molecule compounds targeting Mycobacterium ESAT6/CFP10 proteins via a structure-based approach

Abstract

Mycobacterium tuberculosis is a pathogen which causes the disease tuberculosis. Due to the combination of high occurrence, significant mortality, long treatment period and antibiotic resistance, tuberculosis creates a heavy disease burden in the world. Antibiotic resistance is a huge problem as it renders current drugs used for treatment for diseases much less effective. For tuberculosis, resistance to first line drugs such as rifampicin is common, forcing the use of much more expensive and toxic drugs with lower treatment success rates. To overcome this issue, new treatment methods are desperately needed. However, merely developing novel antibiotics is unlikely to be adequate. The very mechanism of antibiotics, which is to kill the pathogens, exerts significant selection pressure, encouraging rapid emergence of antibiotic resistance. One promising approach to combat this issue is by targeting virulence. Antivirulence drugs would target virulence compounds, therefore preventing them from causing diseases while not killing them directly, reducing the selection pressure significantly. In this study, a pair of virulence factors known as ESAT6/CFP10 was selected as the target for the development of antivirulence compounds against M. tuberculosis and other virulent Mycobacterium species. This pair of heterodimer is believed to facilitate phagosome-cytosolic translocation which causes macrophage necrosis and subsequent spread of pathogen. Virulence of Mycobacteria could be attenuated by blocking the action of these virulence factors. Since a clear target was established first, a structure-based approach was used. Compounds that are able to bind to ESAT6/CFP10 were identified via a combination of in silico screening, followed by validation of binding using NMR spectroscopy. The antivirulence properties of confirmed binders were then tested via in vitro assays using the closely related M. marinum as a model. Finally, the binding site of these compounds was characterized via a combination of site-directed mutagenesis, further NMR studies, and molecular docking. Three candidates were identified with desirable antivirulence properties, selectively inhibiting proliferation of M. marinum inside macrophages but not of those in growth medium. All three compounds were found to interact with a region near the WXG motif of ESAT6, a highly conserved motif of the WXG100 protein superfamily. Apart from the identification of antivirulence compounds, the present study also validated the practicability of the current drug discovery workflow. Following the utilization of an effective in silico structured-based screening platform by combining the docking programme AutoDock Vina and clustering programme AuPosSOM, 4 out of 76 screened compounds were identified to be binders with ligand-based NMR spectroscopy via the waterLOGSY experiment. This was expanded to 19 binders in total after a secondary screen using analogous compounds. Finally, criteria and methodology were developed to screen for antivirulence compounds based on the M. marinum model. Together, these findings and development of methodologies lay the foundation for future antivirulence drug development against tuberculosis.