Breaking the Stability-Activity-Selectivity Trilemma in Unspecific Peroxygenase through Computation-Based Cross-Regional Combinatorial Mutagenesis

The dynamic coupling relationship between the distal region and the region adjacent to the active site in enzymes is still poorly understood. The resultant trade-off among stability, catalytic activity, and enantioselectivity remains a big challenge in developing robust enzymes in agrochemistry. Here, FoldX-based global saturation mutagenesis predictions with spatially targeted prioritization of functional regions, including the N-terminal domain, substrate access pocket, and dimer interface of the enzyme, are developed and validated by experiment, which successfully overcomes the classical stability-activity-selectivity trade-off in unspecific peroxygenase (UPO) from Collariella virescens (CviUPO). The mutation sites with high predicted stabilization probability and ultralow computed ΔΔG values were selected. Among the eight candidate sites tested, the combinatorial variant M1_T9P_G161L showed a 36-fold extended half-life and a 6-fold higher activity compared to M1, both maintaining exceptional enantioselectivity (>99% ee). This corresponds to a total 75-fold activity improvement relative to the wild type. Semipreparative-scale synthesis of (R)-6-bromochroman-4-ol with 12 mM substrate is achieved with a conversion of 83% in 8 h. Furthermore, molecular dynamics (MD) simulations revealed how the combinatorial mutations stabilized the global folding of the protein molecule and strategically reshaped the catalytic pocket microenvironment, which accounted for the enhanced catalytic performance. This cross-regional combinatorial mutagenesis computational strategy provides a new paradigm for engineering metalloenzymes.