Busca farmacofórica baseada em grupos químicos identifica moléculas que sugerem atividade contra proteínas de Mycobacterium tuberculosis

Abstract

Tuberculosis remains one of the leading causes of death from infectious diseases worldwide, further aggravated by the emergence of multidrug-resistant strains of Mycobacterium tuberculosis and by the therapeutic limitations of currently available drugs. In this context, the present study aimed to perform a pharmacophore-based search to identify molecules with potential activity against M. tuberculosis proteins through an integrated computational approach combining ligand-based virtual screening (LBVS), structure-based virtual screening (SBVS), molecular docking, protein–ligand interaction analysis, and in sílico pharmacokinetic evaluation. Initially, 228 analogs of first-line anti-tuberculosis drugs were identified in the PubChem database. In parallel, 3,246 protein structures were retrieved from the Protein Data Bank and subsequently refined after exclusion based on structural similarity. Cavity analysis using fpocket resulted in 4,081 pockets with a druggability score ≥ 0.5, which were subjected to molecular docking with the selected analogs, totaling 63,219 evaluated complexes. Interaction characterization using PLIP enabled the identification of relevant chemical groups, which were employed to construct pharmacophore models applied in the PHARMIT platform, leading to the screening of 117,198 molecules. ADMET evaluation demonstrated that a significant portion of the compounds exhibited pharmacokinetic profiles predictively superior to those of reference drugs. Among the best-ranked complexes were proteins involved in essential metabolic pathways, such as thymidylate kinase, enoyl-ACP reductase, and anthranilate phosphoribosyltransferase. In addition, interactions not previously described in the literature suggest that the screened ligands may explore alternative regions of active sites, expanding the possibilities for the rational design of innovative inhibitors. The incorporation of druggability criteria, Lipinski’s rules, and ADMET analysis allowed refinement of compounds with greater potential for progression within the drug development pipeline. The results indicate that the integrated strategy employed is effective in identifying promising candidates, providing a rational basis for future experimental validation and contributing to the development of new therapeutic alternatives against tuberculosis.