Malolactic fermentation (MLF) is responsible for the decarboxylation of l-malic into lactic acid in most red wines and some white wines. It reduces the acidity of wine, improves flavor complexity and microbiological stability. Despite its industrial interest, the MLF mechanism is not fully understood. The objective of this study was to provide new insights into the role of pH on the binding of malic acid to the malolactic enzyme (MLE) of Oenococcus oeni. To this end, sequence similarity networks and phylogenetic analysis were used to generate an MLE homology model, which was further refined by molecular dynamics simulations. The resulting model, together with quantum polarized ligand docking (QPLD), was used to describe the MLE binding pocket and pose of l-malic acid (MAL) and its l-malate (−1) and (−2) protonation states (MAL− and MAL2−, respectively). MAL2− has the lowest ∆Gbinding, followed by MAL− and MAL, with values of −23.8, −19.6, and −14.6 kJ/mol, respectively, consistent with those obtained by isothermal calorimetry thermodynamic (ITC) assays. Furthermore, molecular dynamics and MM/GBSA results suggest that only MAL2− displays an extended open conformation at the binding pocket, satisfying the geometrical requirements for Mn2+ coordination, a critical component of MLE activity. These results are consistent with the intracellular pH conditions of O. oeni cells—ranging from pH 5.8 to 6.1—where the enzymatic decarboxylation of malate occurs.
Combining fragment docking with graph theory to improve ligand docking for homology model structures
