Extending a Single Residue Switch for Abbreviating Catalysis in Plant ent-Kaurene Synthases

Abstract

Production of ent-kaurene as a precursor for important signaling molecules such as the gibberellins seems to have arisen early in plant evolution, with corresponding cyclase(s) present in all land plants (i.e., embryophyta). The relevant enzymes seem to represent fusion of the class II diterpene cyclase that produces the intermediate ent-copalyl diphosphate (ent-CPP) and the subsequently acting class I diterpene synthase that produces ent-kaurene, although the bifunctionality of the ancestral gene is only retained in certain early diverging plants, with gene duplication and sub-functionalization leading to distinct ent-CPP synthases and ent-kaurene synthases (KSs) generally observed. This evolutionary scenario implies that plant KSs should have conserved structural features uniquely required for production of ent-kaurene relative to related enzymes that have alternative function. Notably, substitution of threonine for a conserved isoleucine has been shown to "short-circuit" the complex bicyclization and rearrangement reaction catalyzed by KSs after initial cyclization, leading to predominant production of ent-pimaradiene, at least in KSs from angiosperms. Here this effect is shown to extend to KSs from earlier diverging plants (i.e., bryophytes), including a bifunctional/KS. In addition, attribution of the dramatic effect of this single residue "switch" on product outcome to electrostatic stabilization of the ent-pimarenyl carbocation intermediate formed upon initial cyclization by the hydroxyl introduced by threonine substitution has been called into question by the observation of similar effects from substitution of alanine. Here further mutational analysis and detailed product analysis is reported that supports the importance of electrostatic stabilization by a hydroxyl or water.

Keywords: carbocation; catalytic mechanism; diterpene synthases; diterpenoids; metabolic engineering; mutagenesis; product specificity.

SCHEME 1

Cyclization mechanism catalyzed by KSs and the Ile mutants investigated here. These KS catalyzed reactions are initiated by ionization of bicyclic ent-CPP (2), generated from the acyclic precursor GGPP by CPS, with initial cyclization to ent-pimar-15-en-8-yl⁺. In Ile→Thr mutants this intermediate is directly deprotonated, affording largely ent-pimara-8(14),15-diene (4) with small amounts of ent-pimara-7,15-diene (6), while Ile→Ala or Ser mutants also produce significant amounts of 8α-hydroxy-ent-pimara-15-ene (7), generated by addition of water prior to deprotonation. In wild-type (WT) KSs secondary cyclization occurs, followed by ring rearrangement, with deprotonation of the neighboring methyl group yielding ent-kaur-16-ene (1). Some of the wild-type KS(L)s investigated here produce either ent-isokaur-15-ene (5) (i.e., OsKSL5i), or a mixture of 1 and 16α-hydroxy-ent-kaurane (4) (i.e., PpCPS/KS), with the latter generated by addition of water prior to deprotonation. Numbers correspond to the compound numbering defined in text.

FIGURE 1

Partial sequence alignment of representative kaurene synthases spanning plant evolution. The highly conserved Ile probed here is indicated by an asterisk (^∗). Residues are numbered as in the full length enzymes. The phylogenetic terms on the right apply to the divisions branching above that point. The aligned sequences are named as follows (NCBI protein database accession): AtKS, Arabidopsis thaliana KS (AAC39443); OsKS, Oryza sativa KS (BAE72099); PgKS, Picea glauca KS (ADB55711); SmKS, Selaginella moellendorffii KS (BAP19110); PpCPS/KS, Physcomitrella patens KS (BAF61135); MpKS, Marchantia polymorpha KS (OAE22677).

FIGURE 2

Effect of Ile-to-Thr mutation on (hydroxy)kaurene synthase product outcome. Chromatograms from GC–MS analysis of the indicated diterpene synthases (wild-type or indicated mutant). (A) MpKS. (B) PpCPS/KS. Numbers correspond to the compound numbering defined in the text (i.e., 1, ent-kaurene; 3, ent-pimara-8(14),15-diene; 4, 16α-hydroxy-ent-kaurane). Enzymatic products were identified by comparison of both retention time and mass spectra to authentic standards (see Supplementary Figure S1).

FIGURE 3

Effect of substitutions for I664 on OsKSL5i product outcome. Chromatograms from GC–MS analysis of the indicated diterpene synthases (wild-type or indicated mutant). Numbers correspond to the compound numbering defined in the text (i.e., 3, ent-pimara-8(14),15-diene; 5, ent-isokaurene), with 2′ corresponding to ent-copalol, the dephosphorylated derivative of ent-CPP (2). Enzymatic products were identified by comparison of both retention time and mass spectra to authentic standards (see Supplementary Figure S1).

FIGURE 4

Effect of various substitutions for I638 on AtKS product outcome. Chromatograms from GC–MS analysis of the indicated diterpene synthases (wild-type or indicated mutant). Numbers correspond to the compound numbering defined in the text (i.e., 1, ent-kaurene; 3, ent-pimara-8(14),15-diene; 6, ent-pimara-7,15-diene; 7, 8α-hydroxy-ent-pimar-15-ene). Enzymatic products were identified by comparison of both retention time and mass spectra to authentic standards (see Supplementary Figure S2).

SCHEME 2

Schematic depicting the water bound in the active site of the AtKS:I638S/A mutants, and its addition to yield 7.