Computationally Guided Rational Design of a Resveratrol O-Methyltransferase with Expanded Substrate Specificity

Resveratrol, a non-flavonoid polyphenol, has attracted considerable interest due to its exceptional therapeutic potential. However, its clinical application is hindered by rapid metabolism and poor bioavailability. Methylated derivatives of resveratrol typically exhibit enhanced stability, bioavailability, and bioactivity, making enzymatic methylation a promising strategy for optimizing its pharmacological properties. Resveratrol O-methyltransferase (ROMT) plays a pivotal role in this transformation, yet its substrate specificity remains a key challenge. Here, we present a computationally guided rational design of Vitis ROMT to expand its substrate scope. Using molecular docking and molecular dynamics simulations, we identified the mechanisms of the ROMT's binding modes with resveratrol, oxidized resveratrol, resveratrol glycoside, and piceatannol, revealing three conserved residues - F167, F311, and H261 - critical for substrate recognition. Targeted saturation mutagenesis at these positions generated a library of 60 variants, which were subsequently subjected to iterative computational screening. This approach identified F167H as the most stable and promising mutant, demonstrating an expanded substrate profile. Our findings provide mechanistic insights into ROMT-mediated methylation and establish a framework for the rational engineering of methyltransferases with tailored substrate specificity, offering theoretical support for the biocatalytic synthesis of bioactive polyphenols.