From functional plasticity of two diterpene synthases (IrTPS2/IrKSL3a) to enzyme evolution

Terpenoids are an intriguing class of natural products with diverse structures and biological activities whose complexity stems, in large part, from terpene synthases (TPSs). These enzymes catalyze carbocationic cascade reactions wherein the groups responsible for quenching the final carbocation is generally not well known. IrKSL3a and IrTPS2 from Isodon rubescens share 98% sequence homology but use distinct quenching strategies, with IrKSL3a catalyzing direct deprotonation to generate the olefin isopimaradiene while IrTPS2 adds water to yield the hydroxylated nezukol. In this work, we discovered a threonine and serine that hydrogen-bond the water to be added in IrTPS2. Site-directed mutagenesis and multiscale QM/MM simulations of modeled structures further reveal that binding of this water is blocked by introduction of a β-methyl containing side-chain in a neighboring residue. From these insights it was then possible to engineer IrKSL3a to generate nezukol, with other new hydroxylated products also observed. Inspired by these mechanistic insights into the functional plasticity of IrKSL3a and IrTPS2, we explored the plausible evolutionary relationship of these kaurene synthase-like (KSL) TPSs, as well as prospective utilization of these plasticity sites discovered in IrTPS2/IrKSL3a. Such experiments with a variety of more phylogenetically distant KSLs demonstrated that these residues are necessary but not sufficient to efficiently introduce such addition of water, emphasizing the selective pressure underlying the extended evolutionary process for production of nezukol by IrTPS2.