Tandemly duplicated CYP82Ds catalyze 14-hydroxylation in triptolide biosynthesis and precursor production in Saccharomyces cerevisiae

Abstract

Triptolide is a valuable multipotent antitumor diterpenoid in Tripterygium wilfordii, and its C-14 hydroxyl group is often selected for modification to enhance both the bioavailability and antitumor efficacy. However, the mechanism for 14-hydroxylation formation remains unknown. Here, we discover 133 kb of tandem duplicated CYP82Ds encoding 11 genes on chromosome 12 and characterize CYP82D274 and CYP82D263 as 14-hydroxylases that catalyze the metabolic grid in triptolide biosynthesis. The two CYP82Ds catalyze the aromatization of miltiradiene, which has been repeatedly reported to be a spontaneous process. In vivo assays and evaluations of the kinetic parameters of CYP82Ds indicate the most significant affinity to dehydroabietic acid among multiple intermediates. The precursor 14-hydroxy-dehydroabietic acid is successfully produced by engineered Saccharomyces cerevisiae. Our study provides genetic elements for further elucidation of the downstream biosynthetic pathways and heterologous production of triptolide and of the currently intractable biosynthesis of other 14-hydroxyl labdane-type secondary metabolites.

Fig. 1. Bioactive metabolites in Tripterygium wilfordii and biosynthesis of triptolide.

a Structures of major bioactive diterpenoids of T. wilfordii. b Proposed triptolide biosynthetic pathway. A solid arrow represents an identified reaction, and a dotted arrow indicates an unknown pathway. GGPP: Geranylgeranyl diphosphate. c Coexpression profiles of CYPs with genes in the triptolide biosynthetic pathway. TwTPS7(v2), TwTPS27(v2) and CYP728B70 are marked in red. The black triangles indicate CYP82D subfamily genes that cluster with identified functional genes. The gene ID of CYP82D274 is TW12G01155.1. Root periderm (RB), root phloem (RP), root xylem (RX), stem vascular bundle (PS), stem periderm (SB), and leaf (L). Source data are provided as a Source Data file.

Fig. 2. Chromosomal localization and evolutionary analysis of the CYP82D genes.

a Chromosomal localization of the 133-kb tandem duplicated CYP82Ds on Chr12 encoding 11 complete genes. The same color indicates orthologous CYP82D genes and the small triangles represent the incomplete gene residues. b Phylogenetic analysis of the TwCYP82D genes. The phylogenetic tree was constructed based on the maximum likelihood method (1000 bootstraps). The numbers represent the predicted divergence time and WGT indicates the whole-genome triplication event of T. wilfordii.

Fig. 3. Functional characterization of CYP82D274 and CYP82D263.

a GC‒MS analysis of methylated products of CYP82D274 and CYP82D263 catalyzing dehydroabietic acid (7) in vivo or in vitro. Empty vector (EV) denotes yeast transformed with an empty vector without CYP. b Catalytic process in a. c Kinetic profiles of CYP82D274 and CYP82D263 catalyzing 7 in vitro. The quantification of 11 was based on the standard curve y = 0.7336x-0.014 (R² = 0.9994) obtained by LC-TQ-MS/MS. The concentration of CYPs was estimated by measuring the reduced CO-difference spectrum. Kinetic parameters were calculated by nonlinear regression analysis using the Michaelis‒Menten model. Data are presented as mean values ± standard deviation SD from three biological independent replicates, and the black circles represent the individual data points. Source data are provided as a Source Data file.

Fig. 4. Intermediates and genes involved in triptolide biosynthesis.

a MALDI-MSI analysis of the distribution of triptolide intermediates in T. wilfordii root tissue. DHB was used as a matrix for metabolite imaging in the positive ion mode. The colors represent the intensity percentage, and each image is independent. b Relative content of diterpenoids and relative quantification of gene expression in 14-hydroxy-dehydroabietic acid (11)-fed cell lines. Cells that were not fed 11 served as a negative control. P-values of genes with significant differences in expression between groups were 0.0375, 0.0178, 0.0031, 0.0448, and 0.0294 in that order. c Metabolite analysis of ¹⁸O-labeled 11-treated cell samples. ¹⁸O-11 was fed to T. wilfordii suspension cells that inhibited the biosynthesis of precursors, and wild-type (WT) and unlabeled 11-treated cells were used as negative controls. ¹⁸O-11 was converted into ¹⁸O-triptinin B (9) in plant cells, and the mass spectra of labeled versus unlabeled metabolites are provided. d Relative content of diterpenoids and relative gene expression in CYP82D274 and CYP82D263 RNAi cell lines. e Relative content of diterpenoids in CYP82D274- and CYP82D263-overexpressing cell lines. The cell lines bombarded with corresponding empty vectors served as controls in d and e, and significant differences (P-values) in metabolites between the groups are shown in Supplementary Data 2 and 3. b, d, e The relative quantification of each metabolite was calculated by dividing each sample by the average content of the control group. The relative expression of genes was determined by the 2^−△△Ct method. EFLα was designated as the housekeeping gene, and the corresponding control group was assigned as the reference sample. Data are presented as mean values ± SD (n = 3 biologically independent replicates). ***P < 0.001, **P < 0.01 and *P < 0.05 determined by two-sided Student’s t test. Source data are provided as a Source Data file.

Fig. 5. Metabolic grid of C-14 hydroxylation and aromatization.

a GC‒MS analysis of methylated products with triptobenzene D (8) as the substrate. b GC‒MS analysis of the catalytic products of miltiradiene (5) in engineered yeast. The mass spectrum results are shown in Supplementary Figs. 12 and 13. c Relative yield of abietatriene (6) in spontaneous, CYP82D274 and CYP82D263 cultures. Data are presented as mean values ± SD (n = 3 biologically independent replicates). d Catalytic process of the CYP82D274 and CYP82D263 metabolic grid for triptolide (1) biosynthesis. A solid arrow represents an identified reaction, and a dotted arrow indicates an unknown reaction. Source data are provided as a Source Data file.

Fig. 6. De novo biosynthesis of 14-hydroxy-dehydroabietic acid (11) in yeast.

a GC‒MS analysis of methylated products in CYP82D274 and CYP82D263 coexpressed with CYP728B70. b Screening for optimal combinations of CYP and TwPOR. Optimal genes are marked in red. N.D. indicates not detected. c Yield of 11 in the engineered strains. The images show the biomass of haploid and diploid strains under the same culture conditions. The genotype schematic is shown in the figure, and detailed information is provided in Supplementary Table 3. Data are presented as mean values ± SD (n = 3 biologically independent replicates). Source data are provided as a Source Data file.

Fig. 7. Phylogenetic analysis of CYP82Ds.

A total of 61 CYP82Ds from flavonoids, furanocoumarins, lignan, fatty acid, sesquiterpenoids, diterpenoids, etc., were included and indicated by different colors. Phylogenetic analysis was performed using MEGA 6.0 software with maximum likelihood method (1000 bootstraps). The two functional CYP82Ds in this study are marked with red triangles. The GenBank accession numbers are provided in Supplementary Data 6.