Genome-Wide Identification, Characterization, and Expression Analysis of the CYP450 Family Associated with Triterpenoid Saponin in Soapberry (Sapindus mukorossi Gaertn.)

Abstract

Soapberry (Sapindus mukorossi Gaertn.) is a tree species of the family Sapindaceae, the pericarp of which is rich in triterpenoid saponins, which are important in chemical production, biomedicine, and other fields. Cytochrome P450 monooxygenase (CYP450) is involved in the modification of the skeletons of triterpenoid saponins and is linked to their diversity. We previously identified 323 CYP450 genes in the transcriptome of soapberry and screened 40 CYP450 genes related to the synthesis of triterpenoid saponins by gene annotation and conserved structural domain analysis. The genetic structure and phylogeny of the CYP450 genes were analyzed separately. Phylogenetic analysis categorized the CYP450 genes of soapberry into five subfamilies, the members of which had similar conserved cumulative sequences and intron structures. A cis-acting element analysis implicated several genes in the responses to environmental changes and hormones. The expression of several genes during eight periods of fruit development was analyzed by real-time quantitative qRT-PCR; most showed high expression during the first four periods of fruit development, and their expression decreased as the fruits matured. A co-expression network analysis of SmCYP450s and related genes in the triterpenoid saponin synthesis pathway was performed. Correlation analysis showed that 40 SmCYP450s may be involved in saponin synthesis in soapberry. The triterpenoid saponin synthesis-related candidate genes identified in this study provide insight into the synthesis and regulation of triterpenoid saponins in soapberry.

1. Introduction

Soapberry (Sapindus mukorossi Gaertn.) is a deciduous tree in the family Sapindaceae, which is distributed throughout the southern provinces of China. Soapberry is important in landscaping and timber applications, and its pericarp contains a large number of saponins (4.14%–27.04%). Soapberry has a natural and stable decontamination ability, and it is an excellent natural nonionic surfactant that can be used as a raw material for the production of soap, shampoo, and other toiletries [1]. Soapberry saponins have antibacterial, antitumor, and lipid-lowering activities and have protective effects on the cardiovascular and cerebrovascular systems [2,3,4].Triterpenoid saponins are the main active substances in soapberry. As important products of secondary metabolism in plants, triterpenoid saponins are present in widely distributed dicotyledonous plants, among which studies have focused on medicinal plants, such as Panax ginseng [5], Polygala myrtifolia [6], and Scutellaria baicalensis [7].

Triterpenoid saponins are amphiphilic molecules consisting of a hydrophobic triterpenoid skeleton modified by sugar chains and other functional groups. Triterpenoid saponins are structurally complex with a variety of enzymes involved in their biosynthesis. The structural diversity of saponins is a product of their modular biosynthesis, which involves three major steps: the initiation phase, the skeleton building phase, and the modification phase [8]. Cytochrome P450 monooxygenase (CYP450) plays a role in the modification phase by oxidizing the carbon atoms at different positions on the saponin backbone in a functional group to produce a variety of triterpenoid saponins [9,10,11,12,13]. Cytochrome P450 monooxygenase is an oxidoreductase named for the maximum absorbance value at 450 nm of its conjugate with CO in the reduced state. The proteins encoded by CYP450s represent about 1% of all protein-coding genes in the genome of plants [14]. They are involved in almost all metabolic pathways in plants and are important in the biosynthesis of a variety of secondary metabolites such as terpenoids, phytohormones, phenylpropanoids, alkaloids, phenolic compounds, fatty acids, and sterols. These metabolites are crucial players in detoxification, drug metabolism, the assimilation of carbon sources, and secondary metabolite production [15].

The large number of CYP450 genes and their low sequence similarity hamper functional analyses [16,17,18]. In view of the important role of CYP450 in saponins biosynthesis, related research is increasing in prevalence. In recent years, 334, 116, 118, 187, and 150 CYP450 genes were identified in Linum usitatissimum [19], Salvia miltiorrhiza [5], Taxus chinensis [20], Ginkgo biloba [21], and Aralia elata [22], respectively. However, as an important tree species rich in saponins, systematic research into CYP450 in soapberry remains scarce. In this study, we identified the CYP450 family in soapberry for the first time and performed a comprehensive analysis from the perspectives of gene structure, conserved motifs, physicochemical properties, cis-acting elements, and phylogeny. Meanwhile, we investigated the expression pattern of the SmCYP450 genes in different organs and fruit developmental stages in soapberry. Furthermore, a co-expression network between SmCYP450s and other saponin synthesis pathway genes was constructed, and the correlation of SmCYP450s expression with saponin contents was also analyzed. Our study will contribute to understanding the important role of SmCYP450s in saponin biosynthesis and lay a theoretical foundation for further revealing the molecular regulatory mechanisms of saponin biosynthesis in soapberry, which make it possible to improve the saponin yield and further increase the economic values of soapberry by regulating key saponin-related genes using biotechnology. At the same time, it will serve as a reference for relevant studies on other perennial woody plants in the future.

2. Materials and Methods

2.1. Materials

Three soapberry trees (6.5 m average height, 13.5 cm average diameter at breast height [DBH]) were obtained from an orchard in Jianning County, Fujian Province, China (26°49′ N latitude, 116°52′ E longitude, 300 m above sea level) [23]. Fruit samples were collected at eight seed-growth stages between June and November 2018: S1, early ovary development, 15 days after pollination (DAP); S2, 30% of largest fruit size, 45 DAP; S3, 70% of largest fruit size, 75 DAP; S4, 80% of largest fruit size, 90 DAP; S5, 90% of largest fruit size, 105 DAP; S6, beginning of maturity, 120 DAP; S7, marked change in pericarp, 135 DAP; and S8, fully developed and mature, 150 DAP [24]. Three biological replicate samples were obtained at each stage, for a total of 24 samples. Fruits were randomly picked from the east, south, west, and north sides of the middle and upper parts of the crowns of trees at 10 a.m. on sunny days. After picking the fruit, the pericarp was immediately separated from the seed. A portion of each pericarp sample was transferred to liquid nitrogen and stored at −80 °C for RNA extraction.

2.2. Gene Identification and Structural Analysis

The CYP450 genes of soapberry were obtained from the soapberry reference genome annotation file, and the entire genome was compared to Glycine max (https://www.soybase.org/sbt/, accessed on 15 February 2024) and Medicago truncatula (http://www.medicagogenome.org/, accessed on 15 February 2024) in the NCBI database (BLASTX; e-value ≤ 1 × 10^–5). In total, 40 SmCYP450 genes were annotated.

The conserved structural domains of selected genes were analyzed using Conserved Domains in NCBI (http://www.ncbi.nlm.nih.gov/Structure/cdd/wrpsb.cgi, accessed on 16 February 2024). DNAMAN 7.0 software was used for comparisons [25]. The chromosomal locations of SmCYP450 genes were evaluated using TBtools v2.0 software [26], and the 40 genes were sequentially named according to the order of the genes on the chromosome and their phylogenetic relationships.

2.3. Analyses of Conserved Motifs and Physicochemical Properties

The isoelectric point (pI), molecular weight (MW), and average hydrophilicity (gravity) of SmCYP450 proteins were analyzed using ProtParam (https://web.expasy.org/protparam/, accessed on 16 February 2024). The subcellular localization prediction of SmCYP450 proteins was performed on the WOLF PSORT website (https://wolfpsort.hgc.jp/, accessed on 16 February 2024). Motifs in the amino acid sequence of candidate SmCYP450 were predicted using MEME (http://meme-suite.org, accessed on 17 February 2024) with default parameters. Multiple sequence alignments and phylogenetic (NJ) analyses were performed using MEGA7 [27] with bootstrap values set to 1000 replicates to evaluate the phylogenetic relationships between the CYP450 proteins of soapberry and Lotus corniculatus, Glycyrrhiza uralensis, Lycopersicon esculentum, Centella asiatica, Arabidopsis thaliana, and Medicago truncatula.

2.4. Analyses of Cis-Acting Elements and Proteins

Sequences 2000 bp upstream of the coding region were selected, and the cis-acting elements of SmCYP450 genes were predicted using PlantCARE (https://bioinformatics.psb.ugent.be/webtools/plantcare/html/, accessed on 16 February 2024) [28]. The secondary and tertiary structures of SmCYP450 proteins were investigated using the GORIV secondary structure prediction method and Swiss Model (https://swissmodel.expasy.org/, accessed on 17 February 2024) [29], respectively.

2.5. Analysis of the Correlations of SmCYP450 Genes with Other Saponin Synthesis Pathway Genes

The correlations between the expression of SmCYP450s and other genes in triterpenoid saponin synthesis pathways were analyzed. Results with Pearson correlation coefficients ≥0.6 were selected for mapping in Cytoscape v3.10.1 software [30].

2.6. Analysis of the Expression Pattern of the SmCYP450 Genes

We pre-collected nutrient organs (roots and leaves), reproductive organs (flowers and fruits), and pericarp of plants at predetermined developmental stages, from which high-quality FPKM transcriptome data were obtained. The transcriptomic data were normalized using TBtools v2.0 software to create a map of gene expression patterns [26]. We previously used vanillin-perchloric acid colorimetry to determine the total saponin content at eight developmental stages of the fruit, followed by ultra-high-performance liquid chromatography/Q-Orbitrap mass spectrometry to identify and measure 54 monomeric saponins [24]. Using the OmicShare cloud platform (https://www.omicshare.com/tools/Home/Soft/ica2, accessed on 18 February 2024), we analyzed the correlation between the expression levels of SmCYP450s and the contents of total saponins and 54 monomeric saponins at eight developmental stages of the fruit. The saponin content of the fruit at eight developmental stages is shown in Supplementary Table S1.

Plant materials were stored at −80°C. Samples were ground into powder in liquid nitrogen, and total RNA was extracted using an RNA Rapid Extraction Kit and reverse-transcribed into cDNA. RT-qPCR was performed to evaluate the expression levels of nine randomly selected SmCYP450s at predetermined developmental stages using the primers listed in Supplementary Table S2. The PerfectStart™ Green qPCR SuperMix Kit (Quantype Gold, Beijing, China) was used for PCR on a LightCycler 480 II Fluorescence Quantitative PCR Instrument (Roche, Basel, Switzerland). The primers were designed using Primer 3 (https://bioinfo.ut.ee/primer3-0.4.0/, accessed on 19 February 2024). SmACT was the internal reference [31], and expression levels were calculated using the 2^−∆Ct method [32].

3. Results

3.1. Identification and Structural Analysis of the SmCYP450 Genes

A sequence comparison of 40 protein sequences from soapberry was performed using DNAMAN software (Supplementary Figure S1). The 40 SmCYP450s had typical CYP450 conserved structural domains, and all were members of the CYP450 family. The structures of the 40 SmCYP450 genes were analyzed (Figure 1). The members of the CYP71 subfamily had one or two introns, whereas the CYP72A, CYP716A, and CYP716E subfamilies had a larger number of introns and were more complex. Soapberry has 14 chromosomes, 7 of which harbor the 35 members of the CYP450 family, typically at the ends of the chromosomes. Chromosomes 4 and 5 had the largest number of genes, mainly members of the CYP71A and CYP71D subfamilies; all of the members of the CYP716A and CYP716E subfamilies were distributed on chromosomes 2 and 8, respectively; members of the CYP72A subfamily were distributed on chromosomes 4, 10, 12, and 14 (Figure 2).

3.2. Physicochemical Properties of the SmCYP450 Proteins

The physicochemical properties of the 40 SmCYP450 proteins were analyzed using ProtParam (Supplementary Table S3). The molecular weights of the proteins encoded by the 40 SmCYP450 genes ranged from 13,796.81 to 76,950.06 Da, and all were fat-soluble and hydrophilic. The lengths of the coding sequences (CDS) of the SmCYP450 genes varied from 297 to 2043 bp, and the 40 SmCYP450 proteins were predicted to be made up of 98–680 amino acids. The instability indices of the SmCYP450 proteins were 77.17–100.11, with average gravity values of −0.527–0.002. Their theoretical isoelectric points indicated the SmCYP450 proteins to be acidic proteins (4.69–9.35). Most of the SmCYP450 proteins were localized to the cytoplasm, chloroplast, or nucleus, in a manner possibly related to their function.

3.3. Conserved Motifs of the SmCYP450 Proteins

Sequence comparisons were performed using MEME, and the top 30 motifs were selected (Figure 3). The 40 SmCYP450 proteins contained 1–14 motifs; all contained motif 1 except CYP72A2, probably because its short sequence was not displayed correctly. All of the proteins in the CYP72 family, excluding seven, contained motif 21. The three members of the CYP716A subfamily had identical conserved motifs, and almost identical amino acid sequences. The two members of the CYP716E subfamily had identical conserved motifs, suggesting that they have similar or identical functions.

3.4. Analysis of Cis-Acting Elements

Cis-acting elements were predicted by analyzing the upstream 2000 bp region (Figure 4). In addition to a large number of basic components, those that appeared more frequently were selected. There were four components for sensing photoperiod (Box4, G-Box, GT1-motif, and TCT-motif) and five components related to responses to environmental stresses (MBS, MYB, and MYC, drought response; ARE, anaerobic induction; and circadian, circadian control). There were four elements related to hormone regulation: ABRE (abscisic acid-response element), ERE (ethylene), CGTCA-motif and TGACG-motif (methyl jasmonate), and WUN-motif (trauma). These elements in the promoters of the SmCYP450 genes suggest that the expression of SmCYP450 genes is affected by photosystems, environmental stresses, and hormone signaling.

The effectors of SmCYP71D7, SmCYP71D8, SmCYP72A1, SmCYP72A5, SmCYP72A6, and SmCYP72A7 include four related to environmental stresses: ARE, Box4, MYB, and MYC. A total of 35 genes contained the anaerobic-response element ARE, 34 had the Box4 photoperiod-response element, 38 genes contained the MYB drought-response element, and 39 genes had the MYC drought-response element.

3.5. Phylogenetic Analysis of the SmCYP450 Proteins

Based on the phylogenetic tree (Figure 5), the SmCYP450 proteins belonged to three families (CYP71, CYP72, and CYP716). Of the SmCYP450 proteins, 28 were categorized into the CYP71 family, among which 14 clustered into the CYP71A and CYP71D subfamilies, 7 into the CYP72A subfamily, 5 into the CYP716 subfamily, 3 into the CYP716A subfamily, and 2 genes into the CYP716E subfamily. The phylogenetic relationships of different species in the same family suggest that these genes have similar functions, enabling functional analysis and identification. However, clustering in a phylogenetic tree can only characterize the evolutionary correlations of related genes and does not provide information on their functions. The functions of the 40 SmCYP450 genes need to be determined by heterologous expression in Saccharomyces cerevisiae and transient expression in tobacco.

3.6. Structural Characterization of SmCYP450 Proteins

The secondary structure of CYP450 proteins includes mainly α-helices, extended chains, and irregular convolutions (Supplementary Table S4). Proteins of the CYP71A subfamily were of 214–535 aa and had the largest number and percentage of α-helical amino acids, followed by irregularly coiled amino acids. Proteins of the CYP71D subfamily had lengths of 120–705 aa; α-helical amino acids predominated, followed by irregularly coiled amino acids and extended-chain amino acids. Proteins of the CYP72A subfamily had lengths of 98–545 aa, with the exception of CYP72A5, in which the highest number of α-helical amino acids accounted for the highest percentage of amino acid distribution, followed by irregularly coiled amino acid distribution. In the CYP72A5 protein, the proportion of irregularly coiled amino acids was 50.00%. The CYP716 proteins had lengths of 489–680 aa, with the highest percentage of α-helical amino acids, followed by irregularly coiled amino acids and extended-chain amino acids.

Three-dimensional modeling (Figure 6) showed that the CYP71 family proteins SmCYP71A1, SmCYP71A3, SmCYP71A4, SmCYP71A6, SmCYP71A7, SmCYP71A8, SmCYP71A9, SmCYP71D1, SmCYP71D2, SmCYP71D3, and SmCYP71D4 had similar structures, as did SmCYP71A10, SmCYP71A11, and SmCYP71A14 and SmCYP71D5, SmCYP71D6, SmCYP71D7, SmCYP71D8, SmCYP71D11, SmCYP71D12, and SmCYP71D14. Among the CYP72A family members, only SmCYP72A1 and SmCYP72A3 had similar structures; the different structures of the other proteins may be related to their functions. In the CYP716 family, the three members of the CYP716A subfamily had almost identical structures, and the two members of the CYP716E subfamily had slightly different structures. Therefore, members of the same family had similar structures and may have similar or identical functions. The predicted spatial structural consistencies of the 40 SmCYP450 proteins were 23.27%–46.24%, the GMQE values were 0.42–0.81, and the QMEAN values were 0.42–0.79, suggesting a high degree of confidence in the modeling (Supplementary Table S5).

3.7. Expression of SmCYP450 Genes According to Developmental Stages

The photograph in Figure 7a illustrates the morphological changes in fruit development in soapberry over the course of eight stages. We can clearly see the change in size and color of the fruit at different stages. The expression patterns of the SmCYP450 genes were explored using transcriptomic data from soapberry fruit at eight developmental stages. Heatmaps were produced.

We analyzed the expression profiles of 40 SmCYP450 genes in soapberry during eight stages of fruit development (Figure 7b). The expression patterns of SmCYP450 genes at the eight stages of fruit development were categorized as pre-expression (S1–S3), middle expression (S4–S6), and late expression (S7–S8). The expression of most of the genes was in the pre- and mid-stages of fruit development, with the expression of few peaking in the late stages. Three (SmCYP71A2, SmCYP71A4, SmCYP71A9), six (SmCYP71A1, SmCYP71A5, SmCYP71A6, SmCYP71A8, SmCYP71A13, SmCYP71A14), and five (SmCYP71A3, SmCYP71A7, SmCYP71A10, SmCYP71A11, SmCYP71A12) genes of the CYP71A subfamily were expressed in the early, middle, and late stages, respectively. Eight (SmCYP71D1, SmCYP71D2, SmCYP71D3, SmCYP71D6, SmCYP71D7, SmCYP71D8, SmCYP71D10, SmCYP71D14), three (SmCYP71D4, SmCYP71D5, SmCYP71D11), and three (SmCYP71D9, SmCYP71D12, SmCYP71D13) genes of the CYP71D subfamily were expressed in the early, middle, and late stages, respectively. Two (SmCYP72A1, SmCYP72A4), four (SmCYP72A3, SmCYP72A5, SmCYP72A6, SmCYP72A7), and one (SmCYP72A2) member of the CYP72A family were expressed in the early, middle, and late stages, respectively. The five members of the CYP716 family were expressed only in the early and middle stages. The five members (SmCYP716A1, SmCYP716A2, SmCYP716A3, SmCYP716E1, SmCYP716E2) of the CYP716 subfamily were highly expressed only in the first four stages of fruit development.

RT-qPCR showed (Figure 8) that the eight genes, except for SmCYP72A5, had a trend of high expression from S1 to S4 and a gradual decrease in the expression trend from S5 to S8. The two members of the CYP716A subfamily also had similar expression trends. We performed a correlation analysis of the transcriptomic data to the RT-qPCR data with a coefficient of determination r² of 0.74, and the results indicate that our data are reliable (Supplementary Figure S2).

3.8. Correlations of SmCYP450s with Saponin-Synthesis-Pathway Genes

We subjected the SmCYP450s to co-expression network mapping with other genes in the triterpenoid saponin synthesis pathway in soapberry. There were significant differences in the correlations of SmCYP450s with other genes, with SmCYP71A6 having the strongest correlation with saponin pathway genes, whereas SmCYP71A7 had weaker correlations with these genes (Figure 9).

3.9. Correlation of SmCYP450s Expression with Total and Monomeric Saponin Contents

We correlated total saponin content and monomeric saponin content with the expression of SmCYP450s at eight stages of fruit development (Supplementary Table S1; Supplementary Figure S3). A correlation cluster analysis was performed by assessing the correlations of SmCYP450s expression at eight stages of fruit development with the total and monomeric saponin contents. The genes with strong correlations with total saponins showed strong correlations with a large number of monomeric saponins, and genes with weak correlations with total saponins showed strong correlations with a small number of monomeric saponins. For example, SmCYP71A1 showed strong correlations with 32 monomeric saponins, whereas SmCYP71A2 showed strong correlations with 13 monomeric saponins.

4. Discussion

4.1. Structures and Conserved Motifs of SmCYP450 Genes

All of the SmCYP450 aa sequences had a cytochrome P450 cysteine heme-iron ligand signature, PERF, and conserved EXXR motifs in the K- and I-helices, which contain a highly conserved threonine involved in oxygen activation [33,34]. Among the 40 SmCYP450 proteins, 39 had conserved motif 1 (cysteine heme-iron ligand signature motif), and SmCYP72A2 had the conserved home domain FXXGXRXCXG, suggesting it to be in the CYP450 superfamily.

Intron locations, deletions, and acquisitions can provide insight into the evolution of gene families within a phylogenetic group. The 40 SmCYP450 genes had zero to six introns. Among them, 62.5% had intronic insertions, 50% contained one to two introns, and members of the CYP716 and CYP72 families had two to six introns. The structures of CYP450s in soapberry were similar to those of CYP450s in Ginkgo biloba [21], which have 1–13 introns and of which 82% have one or two introns. We hypothesize that introns were gradually lost during the evolution of SmCYP450s.

4.2. Phylogenetic Analysis of the SmCYP450 Genes

According to transcriptome analysis, the number of SmCYP450 genes was similar in soapberry and flax (Linum usitatissimum) [19]. In nature, CYP450 genes are involved in the biosynthesis of a variety of organic substances, such as terpenoids, phytohormones, alkaloids, fatty acids, sterols, etc. Most of the genes involved in triterpene synthesis in plants are members of the CYP51, CYP71, CYP72, and CYP85 families. A phylogenetic tree constructed using 40 SmCYP450 genes related to saponin synthesis (Figure 5) showed that the SmCYP450 genes were categorized into five subfamilies of three families (CYP71A, CYP71A, CYP72A, CYP716A, and CYP716E). Among them, 28 genes were categorized in the CYP71 family, 14 in each of the CYP71A and CYP71D subfamilies, and 7 in the CYP72A family. Three of the five genes in the CYP716 family clustered in the CYP716A subfamily, and the other two clustered in the CYP716E subfamily. The number of SmCYP450 genes we identified (40 members) is lower than CYP450 in flax, because this study focused on key genes related to triterpene saponin synthesis.

The CYP51H subfamily is important in the modification of the triterpene skeleton in monocotyledonous plants [35], and AsCYP51H10 oxidizes β-amyrin to a 12,13-epoxy metabolite. The CYP71 family accounts for >50% of the CYP450 genes in both the plant kingdom and in soapberry [36]. CYP93E (CYP71) is important for the oxidation of the triterpene skeleton in legumes [37,38]. Other members of the CYP71 family (CYP705A, CYP71A, CYP71D, and CYP81Q) are involved in the synthesis of triterpenes [39]. The CYP72 family has few members, about which we know little. The members of the CYP72A family are important in triterpene synthesis [38,40,41]. The product of MtCYP72A67 catalyzes hydroxylation at the C-2 position on the triterpenoid saponin backbone [42]. The members of the CYP716 family in the CYP85 lineage are important in the synthesis of triterpenoid biosynthesis in dicotyledonous plants and are the top contributors to the diversity of triterpenoid saponins [43,44]. The members of the CYP716A/-C/-E/-Y subfamilies are involved in the oxidation of carbon atoms at different positions of the pentacyclic triterpenoid backbone in dicotyledonous plants [43,45]. MtCYP716A12 was in the same branch of the phylogenetic tree as SmCYP716A1, SmCYP716A2, and SmCYP716A3; it is involved in the early stages of saponin synthesis in Medicago truncatula, and MtCYP716A12 catalyzed the oxidation of β-coumarinol at the C-28 position [46], yielding oleanolic acid. Soapberry contains high levels of oleanocarpane-type saponins, suggesting that the functions of the products of SmCYP716A1, SmCYP716A2, and SmCYP716A3 maybe similar to MtCYP716A12.

4.3. Expression Pattern According to Stage of Fruit Development

The expression patterns of SmCYP450 genes at the eight stages of fruit development were divided into pre-expression (S1–S3), middle expression (S4–S6), and late expression (S7–S8) (Figure 7b). The expression of most genes was in the pre-stages and mid-stages of fruit development, with only a few peaking in the late stages. We verified the expression levels of selected genes by RT-qPCR. We hypothesized that the products of most of the SmCYP450 genes catalyzed the oxidation of carbon atoms on the saponin skeleton during the early stages of fruit development and the formation of triterpenoid saponins in pericarp at the middle and late stages. We also hypothesized that most of the genes were expressed in the early stages of fruit development, followed in order by the middle and late stages. As the fruit matured and senesced, the expression of SmCYP450 genes decreased and approached zero. Similarly, the expression of PgCYP450 genes varied in the four developmental stages of ginseng, and six of the eight PgCYP450 genes selected had high expression in the first and second stages, while only two genes had high expression in the third and fourth stages; these results indicated that the expression of PgCYP450s might also gradually decrease with the maturation of ginseng and the change in saponin content, which was very similar to the expression pattern of SmCYP450 genes in our study [47].

A cis-acting element analysis showed that hormones, drought, and light affect the expression of SmCYP450 genes. Seventeen cis-actors were predicted in Ginkgo biloba, most of which were related to hormones and environmental stresses. Methyl jasmonate increases the saponin content of leaves of ginkgo seedlings, and in ginseng, it upregulates genes related to saponin synthesis. These findings indicate that the saponin contents of plants can be increased by modulating environmental conditions related to cis-acting elements. The correlational analysis of SmCYP450s with other genes of the saponin synthesis pathway (Figure 9) revealed correlations of different strengths, possibly because of differences in when the products of these genes perform their functions or in the functions of their products. SmCYP71A6 exhibited strong correlations with other genes related to saponin synthesis, suggesting that it is likely to be involved in the biosynthesis of saponins.

The expression levels of SmCYP71A1, SmCYP71A8, SmCYP71A13, and SmCYP71D2 were strongly and positively correlated with the contents of multiple monomeric saponins (Supplementary Figure S3). We speculate that these genes may be involved in saponin synthesis. The synthesis of a given saponin was typically positively correlated with the expression of multiple genes; some genes may be related to the synthesis of intermediates of saponin synthesis. Indeed, the synthesis of glycyrrhizin requires the products of multiple CYP450 genes and their substrates, and the various SmCYP450 genes may perform their functions at different times [48].

5. Conclusions

CYP450 enzymes exhibit important biosynthetic roles in the synthesis of triterpenoid saponins from soapberry and are linked to the diversity of triterpenoid glycosides. We identified 40 SmCYP450 genes and analyzed their structures, conserved motifs, chromosomal locations, promoter cis-acting elements, secondary and tertiary structures, and phylogenetic relationships. The findings provide insight into triterpene biosynthesis in soapberry. Predicting the functions of SmCYP450s is problematic, so further cloning and functional studies of SmCYP450s are needed to enable the production of different monomer triterpenoids. Our findings not only reveal the importance of the CYP450 gene family in the biosynthesis of triterpenoid saponins in soapberry, but will also facilitate research on triterpenoid saponins in other plant taxa.