Structure-based virtual screening and molecular dynamics simulation of bioactive compounds from endophytic fungi and berry plants targeting blaNDM-1

The emergence of New Delhi metallo-β-lactamase-1 (NDM-1)-positive bacterial pathogens has developed into a critical international health issue as they are resistant to β-lactam antibiotics, including carbapenems. The purpose of this study was to discover potential fungal metabolite inhibitors of blaNDM-1 using molecular docking, molecular dynamics (MD) simulation, hydrogen-bond occupancy, MM/PBSA, Zn²⁺ distance analysis, principal component analysis (PCA), ADMET prediction, and DFT calculations. Strong binding affinities were found to be between −10.0 and −8.7 kcal/mol by molecular docking. Mitorubrinol was found to have the highest binding affinity of −10.0 kcal/mol followed by Terreusinone, 1-hydroxy-3-hydroxyethyl-8-ethoxycarbonylxanthone, and Alternariol. The interaction study showed stable hydrogen bonds and hydrophobic interactions with catalytically important residues close to the Zn 2 + binding site. The 100 ns MD simulations showed that the interactions between proteins and ligands formed stable complexes using RMSD, RMSF, radius of gyration, SASA, and the analysis of hydrogen bonds. The hydrogen bond occupancy analysis showed stable binding of the ligands inside the catalytic pocket during the 100 ns of molecular dynamics simulation. The MM/PBSA calculations also indicated negative binding energies and the PCA indicated limited moves in the conformational energies during the simulation. ADMET analysis indicated acceptable pharmacokinetic characteristics of most of the metabolites with Terreusinone having the most attractive drug-likeness profile. Density functional theory (DFT) analysis revealed distinct electronic properties among the selected fungal metabolites. Mitorubrinol had the lowest HOMO-LUMO gap value, reflecting an increased electronic reactivity, while Terraeurusone demonstrated the highest electronic stability, and the analysis of the MEP revealed some oxygen-rich fragments as potential interaction sites. Overall, the results suggest that Terreusinone is the most balanced lead candidate despite Mitorubrinol and Alternariol exhibiting superior performance in individual computational parameters.