Introduction

China is the first country to grow and domesticate the soybean(Glycine max(L.) Merr.), with a cultivation history of at least 4,000 years. Nowadays, soybean is gaining popularity in many countries largely as a vegetable protein and oil source, owing to its bean composition of approximately 40% protein and 20% oil1. Soybean is an important nutritional component of diets and is used in many foods, such as soybean oil, soybean sprout, paste, soymilk, and tofu2. Mostly present in plant leaves and fruits, flavonoids primarily exist as glycosides combining with sugars. An important phenolic secondary metabolite in plants, flavonoids can support plant disease resistance, and in the human body have roles such as anti-oxidation, anti-cancer, and anti-aging3,4. Certain major flavonoids, such as flavonols, flavonol glycosides, isoflavones, chalcones, anthocyanins and procyanidins have been previously isolated and identified from soybeans5,6. Additionally, recent research showed that pinto and black beans as excellent dietary sources of natural antioxidants and thus may serve as disease preventative and healthy foods7. Moreover, a recent review analyzing epidemiological studies in the last ten years and the flavonoid content of diets concluded that dietary flavonoids may prevent a range of different types of cancer8. In fact, free phenolic compounds, such as flavonols and isoflavones, have been found to have with immunomodulatory and cellular antiproliferative functions9.

The sprouting process induces a variety of biochemical changes in soybean seeds, leading to the accumulation of primary and secondary metabolites10. Some studies have found that germination and sprout formation of legumes are accompanied by a significant increase in total flavonoids content with antioxidant potential11,12,13. Generally, content of genistein derivatives is about two times higher than that of daidzein derivatives in raw soybeans14,15,16. Germination can induce isoflavone profile changes in soybeans including decreases of β-glycosides and increases of both 6ʹ-O-malonyl-β-glucosides and aglycones17,18,19. During germination, a 13% increase of total isoflavone content was found by Kim, et al20. One comparative study of the chemical profiles of three Egyptian cultivars of fava beans; Sakha 3, Nubaria 3, and Giza 843, focused on the effects of germination on the chemical profiles of phenolic compounds and saponins and characterized 65 metabolites based on UV spectra, accurate MS, and MS/MS data. Germination was found to dramatically increase the quantities of flavonoids and saponins to which biological activities are attributed20. A previous study has suggested that germination could enhance the content of isoflavonesin soybeans21. Accordingly, studies based on metabolomics have applied combined techniques, such as chromatographic separation (UHPLC and/or gas chromatography) and mass spectrometry, for comprehensive metabolic analysis of plant substrates22,23. Recent technological developments in the field of metabolomics, especially as a result of the widespread use of LC-high resolution MS have considerably improved accuracy and sensitivity of metabolite detection24. For example, non-target LC–MS-Orbitrap metabolomics methods were used to compare and analyze the biologically active metabolites of frozen, boiled, and canned chickpeas, lentils, and white beans25. The main purposes of this study were to: (1) compare the flavonoid contents in soybean seeds and sprouts grown under dark conditions for 3 days at 24 °C (2) analyze the metabolism of flavonoids in soybean sprouts.

Materials and methods

Samples preparation

According to our previous detection of isoflavones in 12 main soybean varieties grown in Heilongjiang, China, two varieties with high isoflavone content were selected for experimentation26. The two soybean cultivars Heinong52 (HN52), and Heinong71(HN71) used in this experiment were cultivated at the Harbin Soybean Research Institute (Modern Agriculture Demonstration Zone of Heilongjiang Academy of Agricultural Sciences, Minzhu Township, Daowai District, Harbin; 126.65°E, 45.78°N), Heilongjiang Academy of Agricultural Sciences, Harbin, China(statement as follow). Soybean seeds were harvested from three replicates of each cultivar for a cropping year and stored at room temperature until analysis of flavonoids compounds. Briefly, high-quality soybean grains were selected, rinsed with clean water, soaked in distilled water at 24 °C for 4 h, spread in a single layer on a tray, covered with gauze, and placed in an artificial climate box (HPG-1600H, Harbin Donglian Electronic Technology Development Co., Ltd. Harbin, China) at about 24 °C and 80% humidity to germinate for 3 days in the dark. The growth chamber was set up to spray water for 3 min every 1 h. After germination, approximately 20 g FW(fresh weight) of soybean sprouts (HNS52,HNS71) per cultivar with three replicates every sampling day were randomly collected from containers, and freeze-dried at − 50 °C for 20 h. The freeze-dried sprouts and seeds were crushed using a mixer mill (MM 400, Retsch) with a zirconia bead for 1.5 min at 30 Hz. 100 mg powder was weighed and extracted overnight at 4 °C with 1.0 mL 70% aqueous methanol. Following centrifugation at 10,000 g for 10 min, the extracts were absorbed (CNWBOND Carbon-GCB SPE Cartridge, 250 mg, 3 mL; ANPEL, Shanghai, China, www.anpel.com.cn/cnw) and filtrated (SCAA-104,0.22 μm pore size; ANPEL, Shanghai, China, http://www.anpel.com.cn/) LC–MS analysis(contained “2.2 HPLC Conditions” and “2.3 ESI-Q TRAP-MS/MS”) .

HPLC conditions

The sample extracts were analyzed using a LC–ESI–MS/MS system (HPLC, Shim-pack UFLC SHIMADZU CBM30A system, www.shimadzu.com.cn/; MS, Applied Biosystems 4500 Q TRAP, www.appliedbiosystems.com.cn/). We used the following steps to analyze, HPLC: column, Waters ACQUITY UPLC HSS T3 C18 (1.8 µm, 2.1 mm*100 mm); solvent system, water (0.04% acetic acid): acetonitrile (0.04% acetic acid); gradient program,100:0 V/V at 0 min, 5:95 V/V at 11.0 min, 5:95 V/V at 12.0 min, 95:5 V/V at 12.1 min, 95:5 V/V at 15.0 min; flow rate, 0.40 mL/min; temperature, 40 °C; injection volume: 5μL. The effluent was alternatively connected to an ESI-triple quadrupole-linear ion trap (QTRAP)-MS27.

LIT and triple quadrupole (QQQ) scans were acquired using a triple quadrupole-linear ion trap mass spectrometer (Q TRAP), API 4500 Q TRAP LC/MS/MS System, equipped with an ESI Turbo Ion-Spray interface, operating in a positive ion mode and controlled by Analyst 1.6.3 software (AB Sciex). The parameters of ESI source operation were as follows: ion source, turbo spray; source temperature 550 °C; ion spray voltage (IS)5500 V; ion source gas I (GSI), gas II(GSII), curtain gas (CUR) were set at 55, 60, and 25.0 psi, respectively; the collision gas(CAD) was high. Instrument tuning and mass calibration were performed with 10 and 100 μmol/L polypropylene glycol solutions in QQQ and LIT modes, respectively. QQQ scans were acquired as MRM experiments with collision gas (nitrogen) set to 5 psi. DP and CE for individual MRM transitions was done with further DP and CE optimization. A specific set of MRM transitions were monitored for each period according to the metabolites eluted within this period28,29,30. Similarly, Valente et al. calculated the metabolic profile and antioxidant activity of 7 European broad bean seed varieties31 and their pods (by-products) using HPLC–DAD-MS/MS-based methods and found the existence of 105 phenolic compounds, alkaloids, jasmonic acids, and organic acids32.

Statistics analysis

For all experiments, three independent assays were carried out. One-way analysis of variance was used to study the differences between means, with a significant level at P < 0.05. Data analysis was carried out with SPSS (Windows version 12.0, SPSS Inc., Chicago,IL, USA). All data are presented as mean ± standard error of means.

Results and discussion

Principal component analysis

Principal component analysis (PCA) was performed on metabolic data to visualize the biological variability of soybean seeds and sprouts. These results indicated that the dominating source of variance was the differential flavonoid content in the two soybean varieties across all germination stages (PC1). From Fig. 1A, the PC1 of 4 groups and mix was 58.6%. The PC1 of two soybean seeds was 77.25%. The PC1 of two soybean seeds and sprouts were74.46% and 81.16% respectively. The PC1 of all samples was greater than 50%. These results indicated that the grouping of samples is reasonable. Through PCA of samples (including quality control samples MIX), in order to have a preliminary understanding of the overall metabolic differences between samples in each group and the degree of variability between samples within the group. The PCA results show the trend of metabolome separation between groups, suggesting whether there is a difference in metabolome between sample groups. The X axis represents the first principal component, and the Y axis represents the second principal component. Grouping principal component analysis: before doing difference analysis, first perform principal component analysis on the grouped samples for difference comparison to observe the degree of variability between the difference groups and the samples within the group.

Figure 1
figure 1

PCA score chart of mass spectrum data of each group of samples and quality control samples. Mix: quality control samples, (A): all samples of PCA. (B): Two soybean seeds of PCA; (C):Soybean and sprouts of PCA(HN52 and HNS52); (D): Soybean and sprouts of PCA(HN71 and HNS71).

Full size image

Differential metabolic analysis of flavonoids

From Table 1, a total of 114 flavonoid metabolites were detected in the two soybean varieties and their sprouts. These consisted of 41 isoflavones, 26 flavonoids, 14 flavonols, 10chalcones, 6 dihydroflavones, 5 tannins, 4 dihydroflavonols, 3 carbonosides, 2 flavonoids, 1 flavanol, 1 proanthocyanidin, and 1anthocyanin. These results indicate that the flavonoids in soybeans and sprouts are mainly composed of isoflavones (35.96%). 13 substances detected in this study were the same as those detected in Egyptian cultivars of chickpea in a previous study30. These were C13H16O10 (6-O-Galloyl-glucose), C13H16O10(3-O-Galloyl-glucose),C13H16O10 (2-O-Galloyl-glucose), C27H30O16(Kaempferol-3-O-neohesperidoside), C27H30O16 (5,7-Dihydroxy-4-methoxyflavone-3-O-xylose-(1–6)-glucose), C21H22O11 (Aromadendrin-7-O-glucoside), C27H30O16 (Quercetin-3-O-robinobioside), C27H30O17 (Quercetin-3-O-sophoroside (Baimaside)), C33H40O21 (Kaempferol-3-O-sophoroside -7-O- glucoside) , C27H32O15 (Eriodictyol-7-O-Rutinoside (Eriocitrin)), C27H30O16 (Quercetin-3-O-rutinoside (Rutin)), C28H32O17 (2'-Hydoxy, 5-methoxy Genistein-4',7- O-diglucoside), C27H30O15 (Eriodictyol-7-O-Rutinoside (Eriocitrin)), respectively. This indicated that these flavonoids may be universal in beans.

Table 1 Flavonoids in soybeans seeds and sprouts.
Full size table

Metabolomic data is massive and multidimensional, so it is imperative to combine univariate and multivariate statistical analyses, to analyze the data from a multivariate perspective according to the characteristics of the data, and to accuratelymine differential metabolites. This study has three biological replicates. The method of combining fold change ≥ 2 and OPLS-DA model VIP (The VIP value represents the influence of the difference between the corresponding metabolites in the classification and discrimination of each group of samples in the model. It is generally believed that the metabolites with VIP ≥ 1 are significantly different.) ≥ 1 was used to screen for differential metabolites. Shown in Fig. 2A, there were a total of 18 significantly different metabolites between the two soybean varieties, of which 14 were up-regulated and 4 were down-regulated. After germination, the significantly different metabolites in the two soybean sprouts were 33, of which 19 were up-regulated and 14 were down-regulated (Fig. 2B). Moreover, there were a total of 27 significantly different metabolites between HN52 and HNS52 cultivars, of which 26 were up-regulated and 1 were down-regulated (Fig. 2C). Additionally, there were a total of 25 significantly different metabolites between HN71 and HNS71, of which 21 were up-regulated and 4 were down-regulated (Fig. 2D). These experimental results showed that both differential and up-regulated metabolites increased in soybean sprouts compared to soybean seeds, thus indicating that soybean germination can increase content of flavonoid metabolites. From a qualitative comparison, that the germination process clearly has an impact on the number of characterized metabolites, especially flavonoid compounds. In fact, the biological activity of the plant substrate is mainly due to these metabolites33,34,35,36,37.

Figure 2
figure 2

Volcano Map of Differential Metabolites. (A): HN52 vs HN71; (B): HNS52 vs HNS71; (C): HN52 vs HNS52; (D): HN71 vs HNS71.

Full size image

Each point in the volcano map represents a metabolite. The abscissa represents the logarithm of the quantitative difference of a certain metabolite in the two samples; the ordinate represents the VIP value. The larger the absolute value of the abscissa, the greater the multiple difference in the expression amount between the two samples; the larger the ordinate value, the more significant the differential expression, and the more reliable the differentially expressed metabolites screened. The green dots in the figure represent down-regulated differentially expressed metabolites, the red dots represent up-regulated differentially expressed metabolites, and black represents detected but not significantly different metabolites.

As shown in Fig. 3, the flavonoids of the two varieties have a balanced distribution of the differential metabolite multiples, and the differential metabolites multiples decrease after germination. This indicated that the flavonoid metabolism of different varieties of soybeans while germinating was highly similar. The flavonoid metabolites in the same variety of soybean seeds and sprouts differ greatly in multiples, showing that flavonoid metabolism is especially active during soybean germination, and that germination is a way to accumulate flavonoids. During the growth process, small sprouts may be damaged by microbial and environmental pressures or nutrient deficiency. Therefore it is necessary for plants to synthesize secondary metabolites (such as flavonoid compounds) through different metabolic pathways to develop defense mechanisms38,39,40.

Figure 3
figure 3

Analysis of the fold difference of metabolites between different samples. (A): HN52 vs HN71; (B): HNS52 vs HNS71; (C): HN52 vs HNS52; (D): HN71 vs HNS71.

Full size image

As shown in Fig. 4, there were total 33 differential metabolites, of which 12 identical metabolites in HN52 vs HNS52 and HN52 vs HN71. There were total 45 differential metabolites, of which 13 metabolites were shared in HN71 vs HNS71 and HNS52 vs HNS71. These results indicated that flavonoid metabolites increased in germinated soybeans. For example, Wu et al. discussed the effect of germination on the antioxidant potential and isoflavonoid content of chickpeas and observed increases in both antioxidant activity and isoflavonoids content with germination41. Additionally, Ayet et al. revealed that germination increased the content of soya saponins in lentils42,43. Moreover, Mekky et al. revealed that germination increased the quantities of flavonoids, phenolic acids and saponins in fava beans30.

Figure 4
figure 4

Venn diagram showing the relationships between different groups of metabolites.

Full size image

Functional annotation and enrichment analysis of differential metabolites

In the homepage of the KEGG website, we directly enter the keyword "Flavonoids" and the top 4 pathways are all pathways involving flavonoid metabolites, including:flavonoid synthesis, anthocyanin synthesis, isoflavone synthesis, flavonoids and flavonols synthesis. Using the KEGG database44 to annotate and display the differential metabolites identified in these samples, 6 metabolite were significantly up-regulated in the isoflavone synthesis pathway, 1 metabolite was significantly down-regulated, and 13 metabolites did not change significantly (Fig. 5).

Figure 5
figure 5

Differential metabolite KEGG pathway diagram. Note: Red indicates that the metabolite content is significantly up-regulated in the experimental group, and blue indicates that the metabolite is detected but not significantly changed. Green indicates that the metabolite content is significantly down-regulated in the experimental group.

Full size image

There are six primary metabolic pathways for differential metabolism of flavonoids; isoflavonoid biosynthesis, flavonoid biosynthesis, flavone and flavonol biosynthesis, secondary metabolite biosynthesis, and phenylpropanoid biosynthesis (Fig. 6). Unlike soybean seeds, soybean sprouts have flavone and flavonol biosynthesis. Isoflavonoid biosynthesis accounts for most of the pathways, and the proportions in different samples are 66.67% in HN52 vs HN71 (Fig. 6A), 41.67% in HNS52 vs HNS71 (Fig. 6B), 54.55% in HN52 vs HNS52 (Fig. 6C), and 40% in HN71 vs HNS71 (Fig. 6D), respectively. Another important pathway is the biosynthesis of secondary metabolites, and the proportions in different samples are 33.33% in HN52 vs HN71 (Fig. 6A, 16.67% in HNS52 vs HNS71 (Fig. 6B), 54.55% in HN52 vs HNS52 (Fig. 6C), 70% in HN71 vs HNS71 (Fig. 6D), respectively). Metabolic pathway data explained why isoflavones were higher in different metabolites.

Figure 6
figure 6

Differential metabolite KEGG classification chart. (A): HN52 vs HN71; (B): HNS52 vs HNS71; (C): HN52 vs HNS52; (D): HN71 vs HNS71. Note: The ordinate is the name of the KEGG metabolic pathway, and the abscissa is the number of metabolites annotated to the pathway and its proportion to the total number of metabolites annotated.

Full size image
Figure 7
figure 7

Differential metabolite KEGG enrichment map. (A): HN52 vs HN71; (B): HNS52 vs HNS71; (C): HN52 vs HNS52; (D): HN71 vs HNS71. Note: The abscissa represents the rich factor corresponding to each channel, the ordinate is the channel name, and the color of the point is p value. The more red, the more significant the enrichment. The size of the dot represents the number of different metabolites enriched.

Full size image

As shown in Fig. 7, that the larger size dot indicated isoflavonoid biosynthesis and biosynthesis of secondary metabolites representing the number of these metabolites enriched. The red point represented isoflavonoid biosynthesis in Fig. 7C, indicating that isoflavonoid was more significantly enriched. The redder points indicating isoflavonoid biosynthesis in Fig. 7A, showed that isoflavonoids were more significantly enriched.

Conclusion

In this study, two soybean varieties with high isoflavone content and their soybean sprouts in Heilongjiang Province, China were analyzed with flavonoid metabolome analysis. 114 differential metabolites were detected which contained 41 isoflavones and 26 flavonoids. These experimental results revealed that the differentially expressed and up-regulated metabolites in soybean sprouts increased compared to soybean seeds. The flavonoid metabolites in the same variety of soybean seeds and sprouts differ greatly in multiples, indicating that flavonoid metabolism is very active during soybean germination, and germination may be a way to accumulate flavonoids. Metabolic pathway data explained why isoflavones were higher than different metabolites. This study detected and analyzed the changes of flavonoid metabolites after soybean germination, providing data reference for the research of flavonoids, and providing ideas for future research on plant secondary metabolism.