Transcriptome Level Reveals the Triterpenoid Saponin Biosynthesis Pathway of Bupleurum falcatum L

Abstract

Bupleurum falcatum L. is frequently used in traditional herbal medicine in Asia. Saikosaponins (SSs) are the main bioactive ingredients of B. falcatum, but the biosynthetic pathway of SSs is unclear, and the biosynthesis of species-specific phytometabolites is little known. Here we resolved the transcriptome profiles of B. falcatum to identify candidate genes that might be involved in the biosynthesis of SSs. By isoform sequencing (Iso-Seq) analyses of the whole plant, a total of 26.98 Gb of nucleotides were obtained and 124,188 unigenes were identified, and 81,594 unigenes were successfully annotated. A total of 1033 unigenes of 20 families related to the mevalonate (MVA) pathway and methylerythritol phosphate (MEP) pathway of the SS biosynthetic pathway were identified. The WGCNA (weighted gene co-expression network analysis) of these unigenes revealed that only the co-expression module of MEmagenta, which contained 343 unigenes, was highly correlated with the biosynthesis of SSs. Comparing differentially expressed gene analysis and the WGCNA indicated that 130 out of 343 genes of the MEmagenta module exhibited differential expression levels, and genes with the most "hubness" within this module were predicted. Manipulation of these genes might improve the biosynthesis of SSs.

Keywords: Bupleurum falcatum; WGCNA; gene expression; saikosaponin biosynthetic pathway; transcriptome.

Figure 1

Differentially expressed genes in the seeding root of B. falcatum among three samples. (A,B) Venn diagram of all down-regulated genes and all up-regulated ones from the transcriptome. S1, the whole fresh roots of 5-day-old seedlings; S2, a 5 mm portion of root tips without the differentiation region of 15-day-old sprout; S3, differentiated root tissue of 15-day-old seedlings. (C,D) Pathway enrichment analysis of DEGs identified in ”S1 vs. S3” (C), and “S2 vs. S3” (D), based on the KEGG database.

Figure 2

WGCNA. Hierarchical gene clustering was used to sort out the clustering tree of the co-expression network module according to the 1-tom matrix. Each module used a dissimilar color. (A) The hierarchical dendrogram displays co-expression modules revealed by WGCNA. One gene is represented by one leaf. Ten modules were found and assigned with different colors based on the calculation of eigengenes. (B) Correlation between traits (bottom) and modules (left), with numbers in left brackets showing the gene number of the module. Blue and red represent negative and positive correlations, respectively, and coefficient values are shown. The darker the color, the higher the correlation coefficient. The p-values are in brackets.

Figure 3

Differential expression of candidate genes of saikosaponin biosynthesis route among S1, S2, and S3. HMG-CoA: 3-hydroxy-3-methy-lglutaryl-COA; MVA-P: 5-phosphoevalonate; MVA-PP: 5-pyrophosphoevalonate; DXP: 1-deoxyxylulose 5-phosphateuvate; CDP-ME: methylerythritol cytidyl diphosphate; CDP-ME-2P: 4-diphosphocytidyl-2-C-methyl-D-erythritol; MEcDP: 2-C-methyl-D-erythritol-2,4-cyclodiphosphate; HMBPP: 4-hydroxy-3-methyl-butenyl-1-diphosphate.

Figure 4

Gene expression profile and cluster analysis of candidate P450. These genes displayed meaningfully different expressions in S3 of B. falcatum.

Figure 5

Hub genes of putative triterpenoid biosynthesis pathway. The MCC (maximal clique centrality) algorithm identified hub genes in PPI networks. The red node represents the high MCC score gene, and the yellow represents the low score gene.

Figure 6

qRT-PCR verification of key genes of saikosaponin biosynthetic pathway. (A–J) The relative expression patterns of BfAACT21654, BfHMGR35352, BfDXS34808, BfHDS30580, BfBAS49741, BfBAS50159, BfCYP18605, BfCYP58609, BfUGT8329 and BfUGT7414. The bar represents the standard error (SE). Data are presented as mean ± SE. Different letters represent different levels of significant difference.