In Silico Approaches to Analyze a Potency of Mangostin Compounds as Inhibitors of Gardnerella vaginalis

Abstract

Background: Gardnerella vaginalis is the predominant pathogen in bacterial vaginosis (BV), known for its biofilm formation and increasing resistance to standard antibiotics such as metronidazole. Natural compounds such as xanthones from Garcinia mangostana, including α-, β-, and γ-mangostin, have shown antimicrobial properties and are promising as alternative therapeutic candidates. Aims: This study aimed to evaluate the potential of α-mangostin, β-mangostin, and γ-mangostin as inhibitors of G. vaginalis using in silico approaches, focusing on their interactions with the DAHP synthase enzyme. Methods: The study involved ligand preparation with energy minimization using Chem3D, receptor modeling of DAHP synthase through SWISS-MODEL, and validation using Ramachandran plot analysis. Molecular docking was conducted using HDock, while pharmacokinetic properties (ADME) were predicted using SwissADME and pkCSM. Molecular dynamic modeling was performed using the iMODS server, analyzing main-chain deformability, eigenvalues, elastic network, variance, and covariance. Results: γ-Mangostin showed the lowest docking score (–165.89) and highest confidence score (0.57), indicating strong affinity and interaction with active site residues such as Arg135A and Asp368A. ADME analysis revealed that all mangostin compounds had high gastrointestinal absorption with no Lipinski violations. Molecular dynamics simulations confirmed γ-mangostin’s ability to induce structural stability and coordinated residue motion in DAHP synthase, suggesting its favorable bioactive profile. Conclusions: γ-Mangostin is a promising natural inhibitor candidate against G. vaginalis through stable and specific interaction with DAHP synthase, supporting its potential for further in vitro and in vivo studies as a BV therapeutic agent.