Class I and II NADPH-cytochrome P450 reductases exhibit different roles in triterpenoid biosynthesis in Lotus japonicus

Abstract

Cytochrome P450 monooxygenases (CYPs) are enzymes that play critical roles in the structural diversification of triterpenoids. To perform site-specific oxidations of the triterpene scaffold, CYPs require electrons transferred by NADPH-cytochrome P450 reductase (CPR), which is classified into two main classes, class I and class II, based on their structural difference. Lotus japonicus is a triterpenoids-producing model legume with one CPR class I gene (LjCPR1) and a minimum of two CPR class II genes (LjCPR2-1 and LjCPR2-2). CPR classes I and II from different plants have been reported to be involved in different metabolic pathways. By performing gene expression analyses of L. japonicus hairy root culture treated with methyl jasmonate (MeJA), this study revealed that LjCPR1, CYP716A51, and LUS were down-regulated which resulted in no change in betulinic acid and lupeol content. In contrast, LjCPR2s, bAS, CYP93E1, and CYP72A61 were significantly upregulated by MeJA treatment, followed by a significant increase of the precursors for soyasaponins, i.e. β-amyrin, 24-OH β-amyrin, and sophoradiol content. Triterpenoids profile analysis of LORE1 insertion and hairy root mutants showed that the loss of the Ljcpr2-1 gene significantly reduced soyasaponins precursors but not in Ljcpr1 mutants. However, Ljcpr1 and Ljcpr2-1 mutants showed a significant reduction in lupeol and oleanolic, ursolic, and betulinic acid contents. Furthermore, LjCPR1, but not LjCPR2, was crucial for seed development, supporting the previous notion that CPR class I might support plant basal metabolism. This study suggests that CPR classes I and II play different roles in L. japonicus triterpenoid biosynthesis.

Keywords: CRISPR/Cas9; LORE1; Lotus japonicus; NADPH-cytochrome P450 reductases (CPR); cytochrome P450 monooxygenases (CYP); triterpenoid biosynthesis.

Figure 1

Gene co-expression analysis of L. japonicus using transcriptomic data from 35 different samples from different tissues, chemical treatments, or biological treatments. The transcriptomic data were obtained from the Gifu genome assembly v1.2 (https://lotus.au.dk/ of).

Figure 2

Effect of the addition of 100 μM MeJA sampled at different time points on gene expression and metabolite level on triterpenoids and phytosterol biosynthesis in L. japonicus hairy root. Single and double arrows indicate one and two oxidation steps, respectively. Dashed arrows indicate multiple steps. The number in parentheses next to the metabolite name refers to the chromatogram peak in Supplementary Figure S8 . CYP, cytochrome P450; FPP, farnesyl pyrophosphate; SQS, squalene synthase; SQE, squalene epoxidase; bAS, β-amyrin synthase; aAS, α-amyrin synthase; LUS, lupeol synthase; LAS, lanosterol synthase; CAS, cycloartenol synthase. nd, not detected.

Figure 3

The relative expression of triterpenoid biosynthetic genes in Lotus japonicus hairy roots at different time periods after methyl jasmonate (MeJA) treatment. Transcript levels of LjCPR1, LjCPR2-1, LjCPR2-2, bAS, LUS, CAS, LAS, CYP716A51, CYP72A61, and CYP93E1 were analyzed by qRT-PCR in L. japonicus hairy roots treated with 100 μM of MeJA for 0, 3, 6, and 12 h after the treatment. Relative expression levels were normalized to those of ubiquitin and are presented as fold induction relative to the control. Data represent the mean of three independent replicates ± SD. Single-factor ANOVA with Tukey’s post-hoc test was used for statistical comparison with the control sample (0 h). Values were considered statistically significant at * p < 0.05 and ** p < 0.01. SD, standard deviation.

Figure 4

The relative amount of triterpenoids and phytosterol content of L. japonicus hairy roots treated with methyl jasmonate (MeJA) on different time periods. Production levels of α-amyrin, β-amyrin, lupeol, ursolic acid, oleanolic acid, betulinic acid, sophoradiol, 24-OH β-amyrin, soyasapogenols, and phytosterols were analyzed by GC-MS in L. japonicus hairy roots treated with 100 μM MeJA for 0, 12, 24, and 48 h after the treatment. Relative triterpenoids and phytosterol amounts were normalized to that of asiatic acid as the internal standard and are presented as fold induction relative to the control. Data represent the mean of three biological replicates ± SD. Single-factor ANOVA with Tukey's post-hoc test was used for statistical comparison with the control sample (0h). Values were considered statistically significant at *P < 0.05 and **P < 0.01. SD, standard deviation. nd, not detected.

Figure 5

The LORE1 insertion mutant lines selected for this study. Two homozygous LORE1 insertion mutant lines were chosen as (A) Ljcpr1 and (B) Ljcpr2-1 loss-of-function mutants. The (C) pod numbers and (D) length of each mutant were quantified. The photo of representative mutant pods is shown in panel (E) Data represent the mean of three biological replicates in pod count (C) and N = 26 randomly selected pods from each mutant plant for the pod length measurement (D), and both are presented as ± SD. Single-factor ANOVA with Tukey’s post-hoc test was used for statistical comparisons. Values were considered statistically significant at * p < 0.05 and ** p < 0.01. SD, standard deviation.

Figure 6

The relative amount of triterpenoids and phytosterols of soil-cultured Ljcpr LORE1 insertion mutant roots analyzed by GC-MS. Relative triterpenoid and phytosterol amounts were normalized to those of asiatic acid as internal standard and are presented as fold induction relative to the wild-type control (WT). Data represent the mean of three biological replicates ± SD. Single-factor ANOVA with Tukey’s post-hoc test was used for statistical comparison to control (WT). Values were considered statistically significant at * p < 0.05 and ** p < 0.01. SD, standard deviation; GC-MS, gas chromatography–mass spectrometry.

Figure 7

Disruption of LjCPR1 gene in transgenic Lotus japonicus hairy roots by CRISPR/Cas9 system. Two sets of four gRNAs were designed to target LjCPR1 gene in different domain regions (A). The gRNA set nos. 4A and 4B (B) and 5B (C) successfully cut LjCPR1 gene as confirmed by heteroduplex mobility assay (HMA) and sequencing, resulting in four Ljcpr1-KO hairy root mutant lines (B, C).

Figure 8

The relative triterpenoid and phytosterol contents of hairy root Ljcpr-1 mutants analyzed using GC-MS. Relative triterpenoid and phytosterol amounts were normalized to those of asiatic acid as the internal standard and are presented as fold induction relative to the empty vector control. Data represent the mean of three technical replicates ± SD. Single-factor ANOVA was used for statistical comparisons. The different letters indicate significant differences (p < 0.05, one-way ANOVA followed by Tukey’s test). GC-MS, gas chromatography–mass spectrometry.