Dynamic interaction network-guided engineering of Bacillus subtilis lipase A for enhanced stability in non-aqueous media

Industrial Enzymes often face the challenge of simultaneously improving thermostability and catalytic activity in non-aqueous catalysis. This study proposes an integrated design strategy combining PROSS computational design, molecular dynamics simulations, and innovative dynamic residue interaction network analysis (focusing on edge-weight variance) to synergistically enhance enzyme thermostability and chemical tolerance. Using Bacillus subtilis Lipase A as a model, the variant 6B^pro obtained by this strategy exhibited a melting temperature 33.34 °C higher than that of the wild-type, and its thermal inactivation half-life at 55 °C was extended by more than 136-fold. Dynamic network analysis revealed a significant decrease in edge-weight variance for 6B^pro, indicating enhanced structural rigidity and suppressed dynamic fluctuations. This variant retained 2.5-fold higher catalytic activity than the wild-type in 80% dimethyl sulfoxide. Further mutagenesis targeting key sites (e.g., D34K, R142D) screened via network analysis led to additional improvements in tolerance, confirming the effectiveness of the strategy. This work demonstrates that quantifying the dynamic fluctuations of residue-residue interactions enables the identification of a "core" interaction network that remains stable during protein motion, providing an effective design tool for the multi-attribute synergistic engineering of proteins.

Edge-weight variance; Organic solvent tolerance; Protein interaction network; Thermal stability.