Nepetalactol serves as the scaffold
for most iridoids, which exhibit
a wide range of biological and pharmacological activities. Iridoid
synthase (ISY) plays a crucial role in the in vivo synthesis of nepetalactol from 8-oxogeranial. However, the substrate
promiscuity of ISY could result in a deviation of flux toward off-target
routes. In this work, the substrate preference (SP, the ratio of activity
for 8-oxogeranial to geranial) of ISY for nepetalactol was improved
by directed evolution. First, the strategy of focused polarity-steric
mutagenesis scanning (FPSMS) was performed to construct a small mutant
library with NmISY2 from Nepeta mussinii as an object. Four amino acid residues with varying polarity and
steric hindrance, including alanine, aspartic acid, serine, and arginine,
were incorporated to scan hot spots. Consequently, four sites of W109,
M217, K343, and W345 with a significant impact on the substrate preference
of NmISY2 were found. Then, the four sites were combined
by a combinatorial active-site saturation test/iterative saturation
mutagenesis (CAST/ISM) strategy. As a result, the mutant W345D/K343M/W109Y
(3M+) was obtained with a significantly increased SP value for 6 from 8.5 to 293.1. Molecular dynamics simulations revealed
that the steric hindrance and polarity of the substrate tunnel played
pivotal roles in the SP value of NmISY2. Notably,
upon integration of 3M+ into Pichia pastoris, the de novo titer of 6 increased by 24.9 times,
reaching 15.8 mg/L. This study offers a strategic approach to improving
the substrate preference of enzymes.