Sage Journals: Discover world-class research

Abstract

An ultra-performance liquid chromatography-electrospray ionization-mass spectrometry targeted metabolomics strategy was applied to analyze protocatechuate, syringin, eleutheroside E, isofraxidin, hyperoside, kaempferol, and oleanolic acid, the active compounds in 3-year-old, 5-year-old, and 9-year-old Acanthopanax senticosus. Then, targeted metabolomics was conducted with 3 growth year plants to identify 19 phenolic metabolites related to the above-mentioned active compounds, including 9 C₆C₃C₆-type, 6 C₆C₃-type, and 4 C₆C₁-type. Multivariate statistical analysis was applied to the bioactive metabolite data, and targeted metabolic profiling was used for marker compound classification and characterization. The results showed that 7 active compounds in the roots and stems in the 3 growth year plants differed. The principal component “Q” values showed that the total contents of 7 active compounds in 5-year-old roots and stems were higher than in other growth years. Results of targeted metabolomics profiling of 19 phenolic metabolites showed that the C₆C₁-type compounds accumulated in 9-year-old plants, the C₆C₃-type in 3-year-old plants, and the C₆C₃C₆-type in 5-year-old plants. The stems had the greatest accumulations of the phenolic metabolites. C₆C₁ and C₆C₃-type metabolites are the most abundant in both roots and stems. In conclusion, the active compounds and pharmacological effects of A. senticosus in different growth years are different. The best harvest age for A. senticosus roots and stems was 5 years. The accumulation of 19 phenolic metabolites in different growth years also showed significant differences.

Keywords

UPLC/MS metabolic profile phenolic compounds quantitative analysis growth year

Acanthopanax senticosus (Rupr. et Maxim.) Harms (A. senticosus thereafter), family Araliaceae,¹ is a tonic and sedative Chinese herb useful in treating various conditions, including stress-induced pathophysiological changes and inflammation.^2,3 The species has been shown to contain many active constituents, including flavonoids,⁴ simple phenylpropanoids,⁵ organic acids,⁶ coumarins,⁷ triterpenoid saponins,⁸ and amino acids. Previous studies have shown that syringin, eleutheroside E, and isofraxidin are the major pharmacologically active components of A. senticosus.¹ Syringin has anti-fatigue, immunomodulation, increasing insulin secretion, and anti-tumor activities.^9
-11 Eleutheroside E has antidepressant and anti-fatigue effects,^12,13 and isofraxidin has anti-tumor, antibacterial, and anti-inflammatory activities.^14,15 Rutin, hyperoside, and kaempferol are common flavonoids, mainly present in A. senticosus leaves.¹⁶ They have anti-oxidant and anti-infective effects.^17
-19 In addition, protocatechuate is also an active compound found in A. senticosus.²⁰

Over the past 20 years, A. senticosus on the market has been derived from wild plants, which grow only at high altitudes in north-eastern China. At present, wild A. senticosus resources are nearing depletion. Thus, it is necessary to cultivate the species to guarantee supplies. Plants have different levels of bioactivities dependent on the number of years of growth. The optimal harvest time of a plant coincides with when it exhibits its highest level of bioactive compounds. So, determining the best harvest time for A. senticosus is of considerable significance, which can guide the large-scale cultivation of this species.

Liquid chromatography (LC)–mass spectrometry (MS) technology was developed in the 1990s. Because of its high selectivity and low detection limits, LC–MS has become a promising analytical method.^21,22 Thus far, several high-performance liquid chromatography (HPLC)–ultraviolet (UV) methods have been used for the quality control of A. senticosus at different growth years. These studies analyzed only a few active compounds and used a long runtime; the results showed that the content of active compounds in the 5-year-old plants was higher than that in other growth years.^23
-25 Although some ultra-performance liquid chromatography (UPLC)-time-of-flight mass spectrometry (TOF)-MS methods have been developed to determine the bioactive constituents of A. senticosus, the difference in different growth years of the plant has not been reported.^26,27 At present, A. senticosus of different ages is sold in the market, but these plants vary in their contents of active ingredients and pharmacological actions. Thus, it is necessary to establish a multi-index method to control the quality of the different growth years of A. senticosus.

Because plants produce diverse compounds, the metabolomics approach is required to estimate the therapeutic effects of natural products and ensure their quality control. Currently, LC–MS has been widely used for the study of plant metabolites.^28,29 As multiple reaction monitoring (MRM) has the advantages of sensitivity, accuracy, and specificity, UPLC-electrospray ionization (ESI)-MS and MRM are often used for the quantitative analysis of targeted metabolites.³⁰ Our previous studies compared the metabolic profiles of 7 active compounds and 19 phenolic metabolites between A. senticosus and Acanthopanax sessiliflorus. The results showed that A. senticosus possess a higher abundance of compounds, particularly in the roots and stems. The 19 phenolic metabolites could be clearly grouped into 5 clusters based on their tissue-specific accumulation patterns. Roots are the tissue with the most abundant accumulations, and the C₆C₃C₆-type metabolites are the most widely existent type in both plants.³¹

Up to date, there has been little research about the metabolomics profiling of cultivated A. senticosus of different ages. We hypothesize that the content of the 7 active compounds and the accumulation of phenolic metabolites in different growth years is different and that the active compounds in 5-year-old A. senticosus are higher than in other growth years. Therefore, metabolomics strategies and correlation analysis were used to determine the growth year properties of 7 active compounds and 19 phenolic metabolites. We hope that the consequences of this research can provide a reference for the cultivation of A. senticosus.

Materials and Methods

Chemicals and Reagents

UHPLC-grade acetonitrile and methanol were bought from Fisher Scientific (Pittsburgh, PA, USA). All other reagents were of analytical purity. Ultrapure water was obtained using a Milli-Q (Millipore, Bedford, MA, USA) water purification system. The reference compounds needed were all purchased from Chroma Dex Inc (Santa Ana, CA, USA). Eleutheroside E, rutin, kaempferol, isofraxidin, protocatechuate, syringin, and hyperoside were used to determine the contents of the active compounds. Syringic acid, vanillic acid, salicylic acid, cinnamic acid, gallic acid, ferulic acid, chlorogenic acid, caffeic acid, p-hydroxycinnamic acid, p-coumaric acid, petunidin, luteolin, apigenin, naringenin, genistein, quercetin-3-O-rhamnoside, quercetin, myricitrin, and catechin were used for the targeted metabolomics study. The purity of these standards was higher than 98%.

Sample Preparation

For eliminating the interference of biotic and abiotic stress factors on the results, 18 samples of A. senticosus, including the roots and stems from plants (3 years, 5 years, and 9 years), were collected from the same producing areas in Qitaihe of Heilongjiang Province in northeast China (45′95N, 131′05E) in September 2017. Because 3-year-old and 5-year-old A. senticosus plants are currently sold as commodities in China, and the 9-year-old A. senticosus is the oldest sample collected, the above 3 growth years were chosen for the research. Six independent biological samples were collected per growth year, randomly selected in our experimental plot (Table 1). The whole plant was dissected to separate roots and stems (ASR and ASS thereafter).

Table 1

Detailed Information of Acanthopanax senticosus Samples.

Sample	Year	Collection time	Medicinal part	Sample	Year	Collection time	Medicinal part
1	3rd	5 September	Root	10	3rd	5 September	Stem
2	3rd	5 September	Root	11	3rd	5 September	Stem
3	3rd	5 September	Root	12	3rd	5 September	Stem
4	5th	5 September	Root	13	5th	5 Septembe	Stem
f	5th	5 September	Root	14	5th	5 September	Stem
6	5th	5 September	Root	15	5th	5 September	Stem
7	9th	5 September	Root	16	9th	5 September	Stem
8	9th	5 September	Root	17	9th	5 September	Stem
9	9th	5 September	Root	18	9th	5 September	Stem

The raw samples were dried at 60°C for 48 hours. Then, the dried samples were powdered in a mill and sieved through a 20 µm mesh. Powdered samples (0.5 g) were extracted with 20 mL methanol for 30 minutes in an ultrasonic bath (100 kHz, 40°C). After filtration, 20 mL solvent was added to the residue and subjected to supersonic extraction for 30 minutes; the 2 filtrates were merged. All samples were filtered through a 0.45 µm Millipore filter before being analyzed by UPLC. The experiment was performed in triplicate.

LC–MS Conditions

The methods used for the quantitative analysis of the 7 active compounds were the same as in our previous studies.³¹ UPLC-MS analysis was performed with an ACQUITY UPLC system (Waters Corporation, Japan) coupled with an LC-20AD pump, SIL-20A autosampler (Waters Corporation, Japan). The ACQUITY UPLC BEH C18 column (1.7 µm, 2.1 mm × 50 mm) used for UPLC analysis was held at 25°C; injection volume was 10.0 µL, and the flow rate was 0.5 mL/minute. Mobile phase A consisted of methanol, while mobile phase B comprised water. The chromatographic column was eluted with a linear gradient of 25% A for 0-1.5 minutes, 25-50% A for 1.5-2.0 minutes, 50% A for 2.0-4.0 minutes, 50%-90% A for 4.0-4.5 minutes, 90% A for 4.0-4.5 minutes, 90%-25% A for 5.5-6.0 minutes, and 25% A for 6.0-7.0 minutes. Mass spectrometric detection was performed using a QTRAP 5500 (AB SCIEX, USA) equipped with an ESI source. Operation parameters were as follows: cone voltage 3 kV and ion source atomizing gas temperature 500°C; 25 psi atomizing gas and 20 psi air curtain gas. The retention time, ion pair, declustering potential,collision energy, and collision chamber injection voltage of the 7 active compounds are shown in Table 2. Both MS and MS/MS data were determined in the positive mode, and data were used for multiple reaction monitoring (MRM).

Table 2

t_R, MS/MS Fragment Ions, DP, CE, and CXP of the 10 Active Metabolites.

Compound	t_R (min)	MS1 (m/z)	MS2 (m/z)	DP	CE	CXP
Eleutheroside E	2.69	765.3	603.1	70	62	23
Rutin	2.98	611.1	464.8	80	20	13
Kaempferol	5.05	287.1	153	70	51	10
Isofraxidin	2.76	223.1	206.3	60	40	9
Protocatechuate	0.86	155	92.9	70	21	17
Syringin	1.37	394.8	231.8	70	40	17
Hyperoside	2.88	487	324	70	43	17

t_R, retention time; DP, declustering potential; CE, collision energy; MS, mass spectrometry; CXP, Collision chamber injection voltage.

The conditions for targeted metabolomics analysis of the 19 phenolic metabolites were the same as in our previous studies.²¹ Samples were analyzed using an LC system coupled to a QTOF tandem-mass spectrometer via an ESI interface (Agilent 6520) (Agilent Technologies, Santa Clara, CA, USA). Sample extracts were separated on a Shimpack LC column (VP-ODS C18 pore size 5.0 µm, 2 × 150 mm), the mobile phase consisting of solvent A and solvent B. Solvent A contained 0.04% acetic acid in water, and solvent B 0.04% acetic acid in acetonitrile. The following gradient was adopted at a flow rate of 0.5 mL/minute: 0-20 minutes, 5% B-95% B; 20-22.1 minutes: 95% B-5% B; 22.1-28 minutes: 5% B-5% B. A blank measurement with the initial solvent was made after each HPLC run. The injection volume and column temperature were adjusted to 5 µL and 40°C, respectively. Optimal MS conditions for positive ion electrospray were: capillary temperature 350°C; curtain gas pressure 40 psi; capillary voltage 3500 V, and fragmentation voltage 135 V. The instrument was tuned prior to each batch run. A full-scan ranging between 50 and 1000 m/z was conducted with a scan time of 1 second and an interscan delay of 0.1 seconds in centered mode.

Statistical Analysis

Analysis of variance of data was performed using SPSS 22.0 software (SPSS, Inc., Chicago, IL USA). The results for 7 active compounds were shown as mean values (mean ± standard error), and the differences between the samples were compared by using a one-way analysis of variance at P < 0.05 probability level (Duncan’s test). The relative peak intensities of 19 phenolic metabolites were normalized and imported into R Project (https://www.r-project.org/) for cluster analysis and heat map drawing. Metabolic pathway enrichment was performed using MBROLE 2.0, and the metabolic network diagram was drawn using Cytoscape 3.7.

Results

Method Validation

As shown in Figure 1, we obtained the total ion current (TIC) spectra for 7 active compounds detected within 6 minutes. The retention times of protocatechuate, syringin, eleutheroside E, isofraxidin, hyperoside, rutin, and kaempferol were found to be almost 0.86, 1.37, 2.69, 2.76, 2.88, 2.98 and 5.05 minutes, respectively. The correlation coefficients (r) of all the calibration curves were higher than 0.998. The limit of detection (LOD) and the limit of quantification (LOQ) values of the protocatechuate, hyperoside, isofraxidin, and kaempferol were 0.030 ng/mL and 0.1 ng/mL, respectively, and values for LOD and LOQ of rutin, syringin, and eleutheroside E were 0.6 ng/mL and 1 ng/mL, respectively. These results displayed good linear relationships between peak area and concentration. The results gave precision (RSD) variations of 0.44%-2.25% and stability ranged from 0.05% to 1.13%, respectively. The repeatability (RSD) ranged between 0.23% and 2.09%. Thus, the developed method was reproducible and accurate. Based on these results, a reliable and straightforward UPLC-MS method for the quantification of the target compounds in A. senticosus has been developed and validated for its linearity, accuracy, precision, LOD, and LOQ.

Figure 1

Ultra-performance liquid chromatography-mass spectrometry chromatograms of 7 standard target compounds.

Accumulation of Active Compounds in Different Growth Years

All A. senticosus samples from different growth years were collected from the same planting area, and their growing environment was the same, so the influence of environmental factors on their active compounds can be ignored. The principal component “Q” is an indicator of a comprehensive, scientific evaluation of objective phenomenon.³² As shown in Figure 2.H, the Q value reflecting accumulation abundance of the 7 active compounds in ASR and ASS at different growth years was obtained. The most extensive accumulation of the active compounds was in the 5-year-old plants. The stems possessed a higher content of the active compounds, especially those of the 3-year-old and 5-year-old plants.

Figure 2

The contents of 7 bioactive compounds in different parts of different growth year (a-g) and the corresponding Q value (h). The letters a, b, c indicates significant differences between different growth year of the same part (P < 0.05); ** indicates significant differences between different tissues of same growth year (P < 0.01) ; * indicates significant differences between different tissues of same growth years (P < 0.01).

Figure 2 shows that the concentrations of the 7 active compounds are different in the 3-year-old, 5-year-old, and 9- year-old plants, being higher in stems than in roots, except for eleutheroside E, which seems almost constant in both organs. The 3-year-old and 5-year-old plants are a clue for the better production of the active compounds than the 9-year-old plants. In the roots of A. senticosus, the content of isofraxidin was 100% higher in the 3-year-old than 5-year-old plants and 33.3% higher than in the 9-year-old. The content of rutin was 1101% higher in the 5-year-old plants than in the 3-year-old ones and 71.4% higher than in the 9-year-old. The content of hyperoside was 333% higher in the 5-year-old plants than in the 3-year-old and 160% higher than in the 9-year-old. The content of eleutheroside E in the 5-year-old plants was 11.0% higher than that in the 3-year-old and 66.0% higher than in the 9-year-old. The content of syringin was 41.7% higher in the 9-year-old plants than in the 5-year-old and 112.5% higher than in the 3-year-old. The content of kaempferol was 14.3% higher in the 9-year-old plants than in the 5-year-old and 166.7% higher than in the 3-year-old. The content of protocatechuate was 25.0% higher in the 9-year-old plants than in the 5-year-old and 66.7% higher than in the 3-year-old. In the stem of A. senticosus, the content of hyperoside was 28.5% higher in the 3-year-old plants than in the 5-year-old and 283.1% higher than in the 9-year-old. The content of eleutheroside E was 57.9% higher in the 3-year-old plants than in the 5-year-old and 87.5% higher than in the 9-year-old. The content of isofraxidin was 283% higher in the 3-year-old than in the 5-year-old and 143% higher than in the 9-year-old. The content of rutin was 265.7% higher in the 5-year-old plants than in the 3-year-old and 905% higher than in the 9-year-old. The content of kaempferol was 199.6% higher in the 5-year-old plants than in the 3-year-old and 211.6% higher than in the 9-year-old. The content of syringin was 133.3% higher in the 5-year-old plants than in the 3-year-old and 218.2% higher than in the 9-year-old. The content of protocatechuate was 300 times higher in the 5-year-old than in the 3-year-old and 666% higher in the 9-year-old.

Targeted Metabolomics Analysis of Phenolic Metabolites

The results of the hierarchical cluster analysis of 19 phenolic metabolites are shown in Figure 3. Accumulation of phenolic metabolites showed an apparent variation in their abundance affected by the different growth years. Based on their accumulation patterns, the 19 phenolic metabolites detected could be grouped into 3 clusters. Cluster 1 showed the highest accumulation in 3-year-old plants; 2 types of compounds accumulated in this cluster: C₆C₃-type (chlorogenic acid, p-coumaric acid, caffeic acid, p-hydroxycinnamic acid, cinnamic acid, ferulic acid) and C₆C₃C₆-type (naringenin). Two types of compounds also occurred in cluster 2: C₆C₁-type (vanillic acid, syringic acid, salicylic acid) and C₆C₃C₆-type (quercetin), with high accumulation detected in the 9-year-old plants. In cluster 3, there were 2 types of compounds: the C₆C₁-type (gallic acid) and C₆C₃C₆ type (catechin, myricitrin, quercetin-3-O-rhamnoside, luteolin, petunidin, genistein, apigenin). The compounds of cluster 3 were accumulated in the 5-year-old plants. As shown in Figure 3, the accumulation of phenolic metabolites in ASR and ASS displayed significant organ-specific mode. ASS contained the most abundant phenolics grouped in clusters 1, 2, and 3, whereas ASR contained cluster 1 compounds.

Figure 3

Heat map visualization of phenolics in the different growth year of ASR and ASS. The color ranges from red to blue indicates relative abundance from high to low (color key scale right the heat map). Each sample was repeated three times. Y, year-old; r, roots; s, stems.

Differences in Target Metabolic Pathways of Above Ground and Underground Organs

To understand the organ specificity of these metabolites in 5-year-old plants, we enriched the metabolic pathways with 26 metabolites. The results revealed that 15 metabolites were involved in 4 metabolic pathways: flavonoid, flavone, flavonol, and phenylpropanoids. We visualized the network map of the metabolomics pathways based on their cumulative situation in the underground and above-ground tissues (Figure 4). We observed a difference in the 2 organs, with most compounds displaying different trends. Syringin takes part in the phenylpropanoid biosynthesis pathway and is mainly accumulated in the above-ground part. Some phenolic metabolites, such as p-hydroxycoumaric acid, protocatechuate, gallic acid, caffeic acid, salicylic acid, chlorogenic acid, apigenin, genistein, and quercetin, also had the same accumulation pattern. However, in underground tissues, abundantly accumulated compounds mainly involved the active ingredients kaempferol and rutin, and the phenolic metabolites naringenin, ferulic acid, and ferulic acid, and they separately take part in the biosynthesis pathways of flavonoid, flavone, flavanol, and phenylpropanoids.

Figure 4

The associated network map of metabolites in ASR and ASS. The circles and diamonds nodes represent different metabolites and related metabolic pathways. Red indicates high abundance, whereas low relative phenolic compounds are green. Phenylpropanoids biosynthesis, biosynthesis of phenylpropanoids, flavonoid biosynthesis, and flavone and flavanol biosynthesis are mutually different .

Correlation Analysis of the Active Compounds and Phenolic Metabolites

To study further the relationship between the detected metabolites, 26 metabolites were subjected to correlation assay. As shown in Figure 5, syringin showed a positive correlation within the C₆C₃C₆-type compounds (naringenin, genistein, and apigenin; P < 0.05). In ASR, the C₆C₁-type compounds, gallic acid and syringic acid, displayed a negative correlation with the C₆C₃-type compounds, chlorogenic acid, caffeic acid, and p-coumaric acid, but, in ASS, a positive correlation with the C₆C₃-type compounds (P < 0.05). In ASR, isofraxidin displayed a positive correlation with p-coumaric acid and caffeic acid (C₆C₃-type compounds) and a negative correlation with the C₆C₃C₆-type compounds genistein, apigenin, and luteolin; the tendency in ASS was the opposite. Eleutheroside E showed a negative correlation with the C₆C₁-type compounds, salicylic acid (P < 0.05) and syringic acid (P < 0.05), but in ASS, a negative correlation within the C₆C₃C₆-type compounds myricetin, luteolin, and quercetin-3-O-rhamnoside (P < 0.05). Kaempferol displayed a negative correlation with chlorogenic acid (P < 0.05) and a positive correlation with gallic acid. Rutin and hyperoside positively correlated with genistein (P < 0.01), apigenin (P < 0.05), and gallic acid (P < 0.05), but a negative correlation with chlorogenic acid and caffeic acid. Protocatechuate displayed a negative correlation with C₆C₁-type compounds and positively correlated with the C₆C₁-type compounds syringic acid and salicylic acid, and the C₆C₃C₆-type compound quercetin. These results were according to the information available in the network map of metabolites (Figure 5).

Figure 5

Correlation coefficients of the metabolites analyzed with 7 bioactive compounds and 19 phenolic compounds in a different part. Each square indicates r (Pearson’s correlation coefficient value for pairs of metabolites). The red color represents a positive correlation and the blue color represents a negative correlation. (a) ASR and (b) ASS.

Discussion

The influence of the growth year on the secondary metabolites of medicinal plants has always been a concern. The analysis of constituents of A. senticosus by UPLC-QTOF-MS was developed, and 131 compounds were identified in leaf extracts, but the growth year of the plant was not reported.¹⁶ The metabolomics strategy was not only applied to distinguish the sexual type of A. senticosus but also was not used for the distinction between growth year.²⁷ In this study, we used a UPLC-MS method for the quantitative analysis of 7 active compounds in different growth years. The MS ion mode, MS parameters, and mobile phase of all analyses were optimized for higher MS sensitivity and better peak shape. After repetitive testing, we found that the positive ion mode and the methanol–water mobile phase gave the optimal conditions. Using these, we improved the speed and quality of detection (Figure 1).

In the root, the contents of rutin, kaempferol, hyperoside, and protocatechuate are meager, so they may not exert any pharmacological effect in this organ. The content of isofraxidin is higher in the 3-year-old plants than in the other growth years. We conclude that the roots of 3-year-old plants have anti-infection and anti-tumor effects.^14,33 The content of eleutheroside E is higher in 5-year-old roots. Therefore, we conclude that the 5-year-old roots have anti-infection, anti-depressant, anti-fatigue, and anti-diabetic effects.^13,34 The content of syringin is higher in the 9-year-old stems. Therefore, we conclude that the 9-year-old roots have immunoregulation, anti-diabetic, and anti-tumor effects.^9,12,35 In the stems of A. senticosus, the contents of hyperoside, isofraxidin, and eleutheroside E are higher in the 3-year-old plants. We conclude that the 3-year-old stems have anti-infection, anti-tumor, anti-oxidation, antidepressant, anti-fatigue, and anti-diabetic effects.^{13,14,17,34,36} Rutin, kaempferol, syringin, and protocatechuate are the highest in 5-year-old stems, and we conclude that 5-year-old stems have anti-infection, anti-oxidation, immunoregulation, anti-tumor, and anti-diabetic effects.^9,18,19,37 The contents of the 7 active compounds in the 9-year-old stems were low, so from this, we conclude that the 9-year-old stems have some pharmacological effects, but the effects are weak. Our results showed that we could target these active marker compounds in different growth years during cultivation and extract them.

Phenolics are important plant secondary metabolites, displaying a tissue and growth year-specific distribution in the content and composition.^38
-40 Targeted analyses were performed to recognize 19 phenolic metabolites in our study. According to Figure 3, the C₆C₃C₆-type metabolites were the most accumulated compounds, and they are mainly involved in the biosynthesis of flavonoids.⁴¹ The results of the determination of 19 phenolic metabolites showed growth year-specific properties, with the C₆C₃-type metabolites mainly accumulated in 3-year-old plants, the C₆C₁-type in 9-year-old plants, and the C₆C₃C₆-type in 5-year-old plants. p-Coumaric acid is said to have antioxidant,⁴² anti-inflammatory,⁴³ and neuroprotective effects.⁴⁴ The results showed that p-coumaric acid accumulated mainly in the 3-year-old plants. A previous study found lonicerin, apigenin, and quercetin-3-O-rhamnoside to have high anti-cancer properties.^45
-47 Therefore, we speculate that 5-year-old A. senticosus has the above effects. Vanillic acid and syringic acid reduce the risk of stroke and myocardial infarction but are only present in 9-year-old plants.⁴⁸ When treating such diseases, 9-year-old A. senticosus can be selected as pharmaceutical material. However, this study only uses relative MS intensity for quantification, and further determination of the above metabolites is needed to confirm these findings.

During the creation of related network maps of metabolites in ASR and ASS, the accumulation patterns of 15 metabolites involved in the 4 metabolomics pathways in different growth years were obtained (Figure 4). The biosynthetic pathway of the C₆C₃C₆-type metabolites results in the production of flavonoids and isoflavones.^21,49 However, C₆C₁-type metabolites may play essential roles, mainly as signaling molecules, such as benzoic acid derivatives and salicylic acid.⁵⁰ C₆C₃-type metabolites may originate in the phenylpropanoid biosynthesis pathway.⁵¹ All compounds have different patterns of accumulation in ASR and ASS. According to this differentiation, we can provide an alternative way to identify ASR and ASS, by the accumulation of target metabolites between the aboveground and underground organs of 5-year-old plants. In this study, only targeted metabolomics research strategies were used to analyze A. senticosus in different growth years, and the study has certain limitations. In the future, non-targeted metabolomics can be used for further research.

Combined with the metabolomics network map and metabolites correlation analysis, we can conclude that the correlation between compounds may affect the accumulation of them in different medicinal organs.⁵⁰ The C₆C₃-type metabolites, namely chlorogenic acid, caffeic acid, p-coumaric acid, and p-hydroxycinnamic acid, are reported to be grouped with salicylic acid.²¹ Syringin showed a negative correlation with them in the roots, and they had a significant correlation; syringin had a significant positive correlation with the C₆C₃-type metabolites in the aboveground organs (Figure 5). We speculate that the C₆C₃-type metabolites have different regulatory mechanisms for the accumulation of syringin in the aboveground and underground organs in the different growth years, which needs further study. Kaempferol, rutin hyperoside, and protocatechuate showed a positive correlation with C₆C₁ and C₆C₃C₆- type metabolites in 2 medicinal organs in different growth years. These phenolic metabolites, as reactive oxygen scavengers, have important functions in plant development and defense.^52,53 These results show that the crucial components might be different in different plant organs. These results indicate that the target secondary metabolic pathways affecting the accumulation of active components are regulated differently in different plant organs.

Conclusion

In summary, the active compounds in A. senticosus have higher accumulation in the roots and stems of 5-year-old plants than in other growth years. The accumulation of phenolic metabolites was significantly different in different growth years and organs. Five-year-old stems are the most abundant resource for the accumulation of phenolic metabolites. We propose that ASR and ASS can be more efficiently used separately for different objectives considering the difference in abundance of their specific active compounds.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: We are thankful for the financial support of the National Key Research and Development Program of China (No. 2016YFC0500303); Heilongjiang Province Foundation for the National Key Research and Development Program of China (No. GX17C006); Post-Doctoral Foundation of Heilongjiang Province of China (No. LBH-Z17208); Key Project of Heilongjiang Provincial Administration of Traditional Chinese Medicine (No. 2018-009).

ORCID iDs

Mingyuan Xu

Zhenyue Wang

References

Huang

Zhao

Huang

Zheng

Peng

Qin

. Acanthopanax senticosus: review of botany, chemistry and pharmacology. Pharmazie. 2011;66(2):83-97.http://www.ncbi.nlm.nih.gov/pubmed/21434569

Jung

HJ.

Park

HJ.

Kim

et al. In vivo anti-inflammatory and antinociceptive effects of liriodendrin isolated from the stem bark of Acanthopanax senticosus . Planta Med. 2003;69(7):610-616.doi:10.1055/s-2003-41127

http://www.ncbi.nlm.nih.gov/pubmed/12898415

Fujikawa

Yamaguchi

Morita

Takeda

Nishibe

. Protective effects of Acanthopanax senticosus Harms from Hokkaido and its components on gastric ulcer in restrained cold water stressed rats. Biol Pharm Bull. 1996;19(9):1227-1230.doi:10.1248/bpb.19.1227

http://www.ncbi.nlm.nih.gov/pubmed/8889047

Chen

Song

Guo

Liu

. Analysis of flavonoid constituents from leaves of Acanthopanax senticosus Harms by electrospray tandem mass spectrometry. Rapid Commun Mass Sp. 2002;16(4):264-271.doi:10.1002/rcm.574

http://www.ncbi.nlm.nih.gov/pubmed/11816040

Zhang

X-Q.

Lai

Y-C.

Wang

et al. Phenylpropanoid constituents from Acanthopanax senticosus . Biochem Syst Ecol. 2011;39(4-6):861-863.doi:10.1016/j.bse.2011.04.008

Zhang

BX.

Jiang

Zhang

. A development of leaves study on effective compositions, pharmacologic action and exploitation in Acanthopanax senticosus (Rupr. et Maxim) Harms [J]. Speci Wild Econ Anim Plant Res. 2009;4:024.

Ryu

JY.

Son

DW.

Kang

et al. Phytochemical constituents of Acanthopanax senticosus (Rupr & Maxim.). Harms stem. Korean J Crop Sci. 2003;11(4):306-310.

Jiang

Han

et al. Biologically active triterpenoid saponins from Acanthopanax senticosus . J Nat Prod. 2006;69(11):1577-1581.doi:10.1021/np060195+

http://www.ncbi.nlm.nih.gov/pubmed/17125224

Cho

JY.

Nam

KH.

Kim

et al. In-vitro and in-vivo immunomodulatory effects of syringin. J Pharm Pharmacol. 2001;53(9):1287-1294.doi:10.1211/0022357011776577

http://www.ncbi.nlm.nih.gov/pubmed/11578112

10.

Liu

KY.

Y-C.

Liu

I-M.

WC.

Cheng

J-T

. Release of acetylcholine by syringin, an active principle of Eleutherococcus senticosus, to raise insulin secretion in Wistar rats. Neurosci Lett. 2008;434(2):195-199.doi:10.1016/j.neulet.2008.01.054

http://www.ncbi.nlm.nih.gov/pubmed/18304730

11.

Panossian

Wagner

. Stimulating effect of adaptogens: an overview with particular reference to their efficacy following single dose administration. Phytother Res. 2005;19(10):819-838.doi:10.1002/ptr.1751

http://www.ncbi.nlm.nih.gov/pubmed/16261511

12.

Kim

MS.

Hyun

. Syringin attenuates insulin resistance via adiponectin-mediated suppression of low-grade chronic inflammation and ER stress in high-fat diet-fed mice. Biochem Bioph Res Co. 2017;488(1):40-45.doi:10.1016/j.bbrc.2017.05.003

http://www.ncbi.nlm.nih.gov/pubmed/28476623

13.

Kimura

Sumiyoshi

. Effects of various Eleutherococcus senticosus cortex on swimming time, natural killer activity and corticosterone level in forced swimming stressed mice. J Ethnopharmacol. 2004;95(2-3):447-453.doi:10.1016/j.jep.2004.08.027

http://www.ncbi.nlm.nih.gov/pubmed/15507373

14.

Yamazaki

Shimosaka

Sakurai

Matsumura

Tsukiyama

Tokiwa

. Anti-inflammatory effects of a major component of Acanthopanax senticosus Harms, isofraxidin. Seibu Butsi Kagaku. 2004;48(2):55-58.doi:10.2198/sbk.48.55

15.

Liu

Tian

Chen

. Interaction of isofraxidin with human serum albumin. Bioorg Med Chem. 2004;12(2):469-474.doi:10.1016/j.bmc.2003.10.030

http://www.ncbi.nlm.nih.gov/pubmed/14723965

16.

Zhang

et al. Application of ultra-performance liquid chromatography with time-of-flight mass spectrometry for the rapid analysis of constituents and metabolites from the extracts of Acanthopanax senticosus Harms leaf. Pharmacogn Mag. 2016;12(46):145-152.doi:10.4103/0973-1296.177902

http://www.ncbi.nlm.nih.gov/pubmed/27076752

17.

Yang

. Hyperoside protects against cerebral ischemia-reperfusion injury by alleviating oxidative stress, inflammation and apoptosis in rats. Biotechnol Biotec Eq. 2019;33(1):798-806.

18.

Stojkovic

Petrovic

Sokovic

Glamoclija

Kukic-Markovic

Petrovic

. In situ antioxidant and antimicrobial activities of naturally occurring caffeic acid, p-coumaric acid and rutin, using food systems. J Sci Food Agric. 2013;93(13):3205-3208.doi:10.1002/jsfa.6156

http://www.ncbi.nlm.nih.gov/pubmed/23553578

19.

García-Mediavilla

Crespo

Collado

et al. The anti-inflammatory flavones quercetin and kaempferol cause inhibition of inducible nitric oxide synthase, cyclooxygenase-2 and reactive C-protein, and down-regulation of the nuclear factor kappaB pathway in Chang liver cells. Eur J Pharmacol. 2007;557(2-3):221-229.doi:10.1016/j.ejphar.2006.11.014

http://www.ncbi.nlm.nih.gov/pubmed/17184768

20.

Jia

et al. Simultaneous determination of protocatechuic acid, syringin, chlorogenic acid, caffeic acid, liriodendrin and isofraxidin in Acanthopanax senticosus Harms by HPLC-DAD. Biol Pharm Bull. 2006;29(3):532-534.doi:10.1248/bpb.29.532

http://www.ncbi.nlm.nih.gov/pubmed/16508160

21.

Liu

Wang

et al. The combined effects of ethylene and MeJA on metabolic profiling of phenolic compounds in Catharanthus roseus revealed by metabolomics analysis. Front Physiol. 2016;7(939):1-11.doi:10.3389/fphys.2016.00217

22.

Liu

Zhao

Zhang

ZH.

Tang

. Profiling of ginsenosides in the two medicinal Panax herbs based on ultra-performance liquid chromatography-electrospray ionization–mass spectrometry. Springerplus. 2016;5(1):1770.doi:10.1186/s40064-016-3427-3

http://www.ncbi.nlm.nih.gov/pubmed/27795912

23.

Zhou

Zhao

et al. Simultaneous qualitative and quantitative evaluation of Ilex kudingcha C. J. Tseng by using UPLC and UHPLC-qTOF-MS/MS. J Pharm Biomed Anal. 2018;155:15-26.doi:10.1016/j.jpba.2018.02.037

http://www.ncbi.nlm.nih.gov/pubmed/29605682

24.

Wang

ZQ.

Wang

ZY.

Chen

LC.

Tang

XM.

Men

. Studies on different harvest periods and different processing methods of Acanthopanax senticosus from different producing areas. Res Infor Trad Chi Med. 2015;7(9):13-15.

25.

Zhang

. HPLC determination of isofraxidin in Acanthopanax senticosus (Rupr. et Maxim.) Harms of different sources and growth time limits. Chi J Pharml Anal. 2010;30(12):2433-2435.

26.

Liu

SP.

J-T.

Wang

. Simultaneous quantification of five bioactive components of Acanthopanax senticosus and its extract by ultra performance liquid chromatography with electrospray ionization time-of-flight mass spectrometry. Molecules. 2012;17(7):7903-7913.doi:10.3390/molecules17077903

http://www.ncbi.nlm.nih.gov/pubmed/22751226

27.

Sun

Han

Zhang

A-H

et al. UPLC-MS based metabolic profiling of the phenotypes of Acanthopanax senticosus reveals the changes in active metabolites distinguishing the diversities of the plant grown in northeast area of China. Chin J Nat Med. 2012;10(3):196-206.

28.

Kiefer

Portais

JC.

Vorholt

. Quantitative metabolome analysis using liquid chromatography-high-resolution mass spectrometry. Anal Biochem. 2008;382(2):94-100.doi:10.1016/j.ab.2008.07.010

http://www.ncbi.nlm.nih.gov/pubmed/18694716

29.

Kim

MS.

Jin

JS.

Kwak

YS.

Hwang

. Metabolic response of strawberry (Fragaria x ananassa) leaves exposed to the angular leaf spot bacterium (Xanthomonas fragariae). J Agric Food Chem. 2016;64(9):1889-1898.doi:10.1021/acs.jafc.5b05201

http://www.ncbi.nlm.nih.gov/pubmed/26890088

30.

Kuzniewski

Zaluski

Olech

Banaszczak

Nowak

. LC-ESI-MS/MS profiling of phenolics in the leaves of Eleutherococcus senticosus cultivated in the West Europe and anti-hyaluronidase and anti-acetylcholinestarase activities. Nat Prod Res. 2018;32(4):448-452.doi:10.1080/14786419.2017.1308369

http://www.ncbi.nlm.nih.gov/pubmed/28349704

31.

K-X.

Liu

et al. A comparative metabolomics analysis reveals the tissue-specific phenolic profiling in two Acanthopanax Species. Molecules. 2018;23(8):E2078. 20 Aug 2018.doi:10.3390/molecules23082078

http://www.ncbi.nlm.nih.gov/pubmed/30127238

32.

Pan

YJ.

Liu

Guo

XR.

Y-G.

Tang

Z-H

. Gene transcript profiles of the TIA biosynthetic pathway in response to ethylene and copper reveal their interactive role in modulating TIA biosynthesis in Catharanthus roseus . Protoplasma. 2015;252(3):813-824.doi:10.1007/s00709-014-0718-9

http://www.ncbi.nlm.nih.gov/pubmed/25344654

33.

Yamazaki

Tokiwa

. Isofraxidin, a coumarin component from Acanthopanax senticosus, inhibits matrix metalloproteinase-7 expression and cell invasion of human hepatoma cells. Biol Pharm Bull. 2010;33(10):1716-1722.doi:10.1248/bpb.33.1716

http://www.ncbi.nlm.nih.gov/pubmed/20930381

34.

Ahn

MY.

Lee

Jung

CH.

Heo

SH.

. Eleutheroside E, an active component of Eleutherococcus senticosus, ameliorates insulin resistance in type 2 diabetic db/db mice. Evid Based Complement Alternat Med. 2013;2013:1-9.doi:10.1155/2013/934183

http://www.ncbi.nlm.nih.gov/pubmed/23690865

35.

Lee

CH.

Huang

CW.

Chang

et al. Reactive oxygen species mediate the chemopreventive effects of syringin in breast cancer cells. Phytomedicine. 2019;61:152844.doi:10.1016/j.phymed.2019.152844

http://www.ncbi.nlm.nih.gov/pubmed/31029906

36.

Liu

Zhao

YH.

Chen

SH.

Zhang

XG.

Mao

. Hyperoside inhibits proinflammatory cytokines in human lung epithelial cells infected with Mycoplasma pneumoniae . Mol Cell Biochem. 2019;453(1-2):179-186.doi:10.1007/s11010-018-3443-4

http://www.ncbi.nlm.nih.gov/pubmed/30350306

37.

Niu

HS.

Liu

IM.

Cheng

JT.

Lin

CL.

Hsu

. Hypoglycemic effect of syringin from Eleutherococcus senticosus in streptozotocin-induced diabetic rats. Planta Med. 2008;74(2):109-113.doi:10.1055/s-2008-1034275

http://www.ncbi.nlm.nih.gov/pubmed/18203055

38.

Hegde

RS.

Fletcher

. Influence of plant growth stage and season on the release of root phenolics by mulberry as related to development of phytoremediation technology. Chemosphere. 1996;32(12):2471-2479.doi:10.1016/0045-6535(96)00144-0

39.

Jia

CH.

Wang

Zhao

. Change of phenolic acids concentration in soil under wheat straw mulch and the effect of phenolic acids on early growth stage of summer maize. Acta Agr Bor—sin. 2004;19(4):84-87.

40.

Marakaev

OA.

Tselebrovskiĭ

MV.

Nikolaeva

TN.

Zagoskina

. Some aspects of underground organs of spotleaf orchis growth and; phenolic compounds accumulation at the generative stage of ontogenesis. Izv Akad Nauk SSSR Biol. 2013;40(3):281-288.doi:10.1134/S1062359013030060

http://www.ncbi.nlm.nih.gov/pubmed/24171312

41.

Trantas

EA.

Koffas

MA.

Ververidis

. When plants produce not enough or at all: metabolic engineering of flavonoids in microbial hosts. Front Plant Sci. 2015;6:1-16.doi:10.3389/fpls.2015.00007

42.

Benbettaieb

Nyagaya

Seuvre

AM.

Debeaufort

. Antioxidant activity and release kinetics of caffeic and p-coumaric acids from hydrocolloid-based active films for healthy packaged food. J Agric Food Chem. 2018;66(26):6906-6916.doi:10.1021/acs.jafc.8b01846

http://www.ncbi.nlm.nih.gov/pubmed/29852064

43.

Zhu

Liang

QH.

Xiong

et al. Anti-inflammatory effects of p-coumaric acid, a natural compound of Oldenlandia diffusa, on arthritis model rats. Evid Based Complement Alternat Med. 2018;2018:1-9.doi:10.1155/2018/5198594

http://www.ncbi.nlm.nih.gov/pubmed/29681976

44.

Sakamula

Thong-Asa

. Neuroprotective effect of p-coumaric acid in mice with cerebral ischemia reperfusion injuries. Metab Brain Dis. 2018;33(3):765-773.doi:10.1007/s11011-018-0185-7

http://www.ncbi.nlm.nih.gov/pubmed/29344828

45.

Jose Jara-Palacios

Goncalves

Hernanz

Heredia

FJ.

Romano

. Effects of in vitro gastrointestinal digestion on phenolic compounds and antioxidant activity of different white winemaking byproducts extracts. Food Res Int. 2018;109:433-439.doi:10.1016/j.foodres.2018.04.060

http://www.ncbi.nlm.nih.gov/pubmed/29803468

46.

Chen

KC.

Hsu

WH.

J-Y

et al. Flavonoids luteolin and quercetin inhibit RPS19 and contributes to metastasis of cancer cells through c-myc reduction. J Food Drug Anal. 2018;26(3):1180-1191.doi:10.1016/j.jfda.2018.01.012

http://www.ncbi.nlm.nih.gov/pubmed/29976410

47.

Madunic

Madunić

IV.

Gajski

Popić

Garaj-Vrhovac

. Apigenin: a dietary flavonoid with diverse anticancer properties. Cancer Lett. 2018;413:11-22.doi:10.1016/j.canlet.2017.10.041

http://www.ncbi.nlm.nih.gov/pubmed/29097249

48.

Rasheeda

Bharathy

Nishad Fathima

. Vanillic acid and syringic acid: Exceptionally robust aromatic moieties for inhibiting in vitro self-assembly of type I collagen. Int J Biol Macromol. 2018;113:952-960.doi:10.1016/j.ijbiomac.2018.03.015

http://www.ncbi.nlm.nih.gov/pubmed/29522822

49.

Mouradov

Spangenberg

. Flavonoids: a metabolic network mediating plants adaptation to their real estate. Front Plant Sci. 2014;5:1-16.doi:10.3389/fpls.2014.00620

http://www.ncbi.nlm.nih.gov/pubmed/25426130

50.

Kim

JK.

Park

S-Y.

Lim

S-H.

Yeo

Cho

HS.

S-H

. Comparative metabolic profiling of pigmented rice (Oryza sativa L.) cultivars reveals primary metabolites are correlated with secondary metabolites. J Cereal Sci. 2013;57(1):14-20.doi:10.1016/j.jcs.2012.09.012

51.

Seigler

. Phenylpropanoids. In: Seigler

DS.

, ed. Plant Secondary Metabolism. CH: Springer; 1998:106-129.

52.

Murphy

Peer

WA.

Taiz

. Regulation of auxin transport by aminopeptidases and endogenous flavonoids. Planta. 2000;211(3):315-324.doi:10.1007/s004250000300

http://www.ncbi.nlm.nih.gov/pubmed/10987549

53.

Watkins

JM.

Hechler

PJ.

Muday

. Ethylene-induced flavonol accumulation in guard cells suppresses reactive oxygen species and moderates stomatal aperture. Plant Physiol. 2014;164(4):1707-1717.doi:10.1104/pp.113.233528

http://www.ncbi.nlm.nih.gov/pubmed/24596331

Targeted Development-Dependent Metabolomics Profiling of Bioactive Compounds in Acanthopanax senticosus by UPLC-ESI-MS

Abstract

Keywords

Materials and Methods

Chemicals and Reagents

Sample Preparation

LC–MS Conditions

Statistical Analysis

Results

Method Validation

Accumulation of Active Compounds in Different Growth Years

Targeted Metabolomics Analysis of Phenolic Metabolites

Differences in Target Metabolic Pathways of Above Ground and Underground Organs

Correlation Analysis of the Active Compounds and Phenolic Metabolites

Discussion

Conclusion

Footnotes

Declaration of Conflicting Interests

Funding

ORCID iDs

References