Targeting Mycobacterial Dormancy Survival Regulator (DosR) With Generative Artificial Intelligence and Omics Methods

Area of Science:

• Biochemistry and Molecular Biology
• Computational Chemistry
• Drug Discovery

Background:

• Tuberculosis (TB) remains a significant global health threat, primarily due to Mycobacterium tuberculosis's ability to enter a dormant state.
• This dormancy allows the pathogen to evade immune responses and resist conventional antibiotic treatments.
• The dormancy survival regulator (DosR) protein is crucial for this survival mechanism, making it a key therapeutic target.

Purpose of the Study:

• To identify novel inhibitors of the DosR protein using a comprehensive in silico approach.
• To develop a robust computational framework for discovering new drugs against dormant tuberculosis.

Main Methods:

• In silico analysis including domain and motif analysis, multiple sequence alignment (MSA), and consensus sequence generation.
• Screening of FDA-approved compounds via molecular docking against the DosR active site.
• De novo molecule generation using REINVENT4, followed by ADMET analysis, molecular dynamics (MD) simulations, and MMGBSA.

Main Results:

• Identified conserved regions within the DosR protein across various Mycobacterium species.
• Screened FDA-approved compounds and generated a novel library of potential inhibitors.
• Selected top 5 compounds, including RI081, RI089, and RI107, demonstrating stability and strong binding affinity to DosR.

Conclusions:

• The multi-tiered computational approach successfully identified promising novel inhibitors of the DosR protein.
• These findings provide a strong foundation for developing new therapeutic strategies against dormant tuberculosis.
• The developed framework can be applied to the discovery of inhibitors for other essential bacterial proteins.