Integrative analysis of microRNAs and mRNAs reveals the regulatory networks of triterpenoid saponin metabolism in Soapberry (Sapindus mukorossi Gaertn.)

Abstract

Triterpenoid saponin are important secondary metabolites and bioactive constituents of soapberry (Sapindus mukorossi Gaertn.) and are widely used in medicine and toiletry products. However, little is known about the roles of miRNAs in the regulation of triterpenoid saponin biosynthesis in soapberry. In this study, a total of 3036 miRNAs were identified, of which 1372 miRNAs were differentially expressed at different stages of pericarp development. Important KEGG pathways, such as terpenoid backbone biosynthesis, sesquiterpenoid and triterpenoid biosynthesis, and basal transcription factors were highlighted, as well the roles of some key miRNAs, such as ath-miR5021, han-miR3630-3p, and ppe-miR858, which may play important roles in regulating triterpenoid saponin biosynthesis. In addition, 58 miRNAs might participate in saponin biosynthesis pathways by predicting the targets of those miRNAs to 53 saponin biosynthesis structural genes. And 75 miRNAs were identified to potentially play vital role in saponin accumulation by targeting transcript factor genes, bHLH, bZIP, ERF, MYB, and WRKY, respectively, which are candidate regulatory genes in the pathway of saponin biosynthesis. The results of weighted gene coexpression network analysis (WGCNA) suggested that two saponin-specific miRNA modules and 10 hub miRNAs may participate in saponin biosynthesis. Furthermore, multiple miRNA-mRNA regulatory networks potentially involved in saponin biosynthesis were generated, e.g., ath-miR5021-SmIDI2/SmGPS5/SmbAS1/SmCYP71D-3/SmUGT74G-2, han-miR3630-3p-SmCYP71A-14/SmbHLH54/SmMYB135/SmWRKY32, and ppe-miR858-SmMYB5/SmMYB32. qRT-PCR analysis validated the expression patterns of nine miRNAs and 12 corresponding target genes. This study represents the first comprehensive analysis of miRNAs in soapberry and lays the foundation for further understanding of miRNA-based regulation in triterpenoid saponin biosynthesis.

Keywords: Sapindus mukorossi; biosynthesis; coexpression network; miRNA; triterpenoid saponin.

Figure 1

Methodology procedure chart.

Figure 2

Diversity of sRNAs in soapberry. (A) Images of eight pericarp developmental stages from soapberry. (B) The length distribution of reads in the eight stages. (C) Small RNAs annotation in the Rfam databases. Fractions of different annotations are presented in a stacked bar plot.

Figure 3

Target genes analysis of miRNAs. (A) GO enrichment analysis of target genes. (B) KEGG pathways analysis of target genes.

Figure 4

Hypothetical model of miRNAs regulation in triterpenoid biosynthesis pathway. AACT, acetyl-CoA C-acetyltransferase; HMGS, 3-hydroxyl-3-methylglutaryl-CoA synthase; HMG-CoA, 3-hydroxy-3-methylglutaryl-CoA; HMGR, 3-hydroxyl-3-methylglutaryl-CoA reductase; MVA, mevalonate; MVK, mevalolate kinase; MVAP, 5-metholvalic acid phosphoric; PMK, phosphoric acid valate kinase; MVAPP, 5-metholvalic acid pyrophosphoric acid; MVD, 5-hydroxyvalerate decarboxylase; G3P, 3-glyceraldehyde phosphate; DXS, 1-deoxy-D-xylulose-5-phosphate synthetase; DXP, 1-deoxy-D-xylulose-5-phosphate; DXR, 1-deoxy-D-xylulose-5-phosphate reductoisomerase; MEP, 2-C-methyl-D-erythritol-4-phosphATE; MCT, 4-cytidine diphosphate-2-C-methyl-D-erythritol synthetase; CDP-ME, 4-cytidine diphosphate-2-C-methyl-D-erythritol; CMK, 4-cytidine diphosphate-2-C-methyl-D-erythritol kinase; CDP-ME2P, 4-cytidine diphosphate-2-C-methyl-D-erythritol-2-phosphate; MDS, 2-C-methyl-D-erythritol-2,4-cyclic phosphate synthetase; MEcPP, 2-C-methyl-D-erythritol-2,4-cyclic phosphate; HDS, 4-hydroxy-3-methylbut-2-enyl diphosphate synthase; HMBPP, 4-hydroxy-3-methylbut-2-enyl pyrophosphate; HDR, 4-hydroxy-3-methylbut-2-enyl diphosphate reductases; IDI, isopentenyl pyrophosphate isomerase; IPP, isopentenyl pyrophosphate; DMAPP, dimethylallyl pyrophosphate; GPP, geranyl pyrophosphate; GPS, geranylpyrophosphate synthetase; FPS, farnesyl pyrophosphate synthetase; FPP, farnesyl pyrophosphate; SS, squalene epoxidase; β-AS, beta-amyrin synthase; LUS, lupeol synthase; CYP450, cytochrome P450; UGT, uridine diphosphate-dependent glycosyltransferases.

Figure 5

WGCNA of expressed miRNAs. (A) Hierarchical clustering tree (cluster dendrogram) showing four modules of coexpressed miRNAs by WGCNA. Each leaf of tree corresponds to one miRNA. The major tree branches constitute four modules, labeled with different colors. (B) Module–saponin relationship. Each row represents a module. Each column represents a specific saponin. The value in each cell at the row-column intersection represents the correlation coefficient between the module and the saponin and is displayed according to the color scale on the right. The value in parentheses in each cell represents the P-value. (C) Correlation and P-values of miRNA modules eigengenes with mRNAs modules eigengenes from the same pericarp samples in soapberry. The rows represent miRNA modules and the columns represent mRNA modules. The correlation networks of miRNAs in the Mebrown (D) and Meblue (E) modules, in which only edges with weight above a threshold of 0.10 and 0.25 are displayed, respectively.

Figure 6

Regulatory network of negatively–correlated miRNAs and mRNAs in identified modules. miRNAs in modules Mebrown (A) and Meblue (B) that were significantly associated with saponin biosynthesis and were negatively correlated with various mRNAs as indicated by the arrows. Round rectangle, diamond, hexagon, octagon, ellipse, parallelogram, rectangle, triangle, and V represent miRNA, TFs, TRs, PKs, transporters, enzymes, proteins, others and unknow genes, respectively.

Figure 7

Expression correlation of miRNAs and their targets. The bars and lines indicate the relative expression level of miRNAs and their corresponding targets in the eight developmental stages of soapberry pericarps. The Y-axis on the left and right represents the abundance of the miRNAs and their targets, respectively. SnRNA U6 and SmACT were used for normalizing the relative expression of miRNAs and their targets, respectively. The expression level of the miRNAs and their corresponding targets in the stage S1 were set as 1.0. Relative expression level was calculated using the 2^{–ΔΔ C t} method. Data indicate the mean values of three biological replications.