Comprehensive chemical analysis of Schisandra chinensis by HPLC-DAD-MS combined with chemometrics.
ARTICLE INFOKeywords:
Schisandra chinensis
Chemical analysis
Chemometrics
High performance liquid chromatography coupled with diode array detection and mass spectrometry
Lignans
ABSTRACT
The fruit of Schisandra chinensis, namely "Wuweizi" in China, is a well-known herbal medicine and health food. In this paper, an accurate and reliable high performance liquid chromatography coupled with diode array detection and mass spectrometry was developed for quality evaluation of Wuweizi. Nine lignans, including schisandrol A, schisandrol B, angeloylgomisin H, gomisin C, schisantherin A, schisanhenol, schisandrin A, schisandrin B, and schisandrin C were determined simultaneously in forty-three batches of Wuweizi samples collected from different localities. Thirty-six common peaks were unequivocally identified or tentatively assigned by comparing their mass spectrometric data with reference compounds, self-established compound library and published literatures. And the thirty-six common peaks were selected as characteristic peaks to assess the similarity of chromatographic fingerprinting of these Wuweizi samples. Moreover, hierarchical clustering analysis and principal components analysis were successfully applied to demonstrate the variability of these Wuweizi samples. The results indicated the content of nine investigated lignans varied greatly among the samples, and samples collected from different localities could be discriminated. Furthermore, schisandrol A, schisandrol B, schisandrin B, and schisandrin C were found to chemical marker for evaluating the quality of Wuweizi.
Crown Copyright [c] 2013 Published by Elsevier GmbH. All rights reserved.
Introduction
Schisandra chinensis (Turcz.) Baill, is a well-known tonic and sedative herbal medicine and grows mainly in Northeastern China, North China, D.P.R. Korea, R.O. Korea, Japan, and most Eastern parts of Russia (Lebedev 1971; Hancke et al. 1999). The fruit of S. chinensis, namely "Wuweizi" in China, has been used officially for more than two thousand years in China, and it is always recorded in Chinese Pharmacopoeia. It was used as an astringent tonic for the lungs and kidneys, to astringe, replenish qi, promote production of body fluids, tonify the kidney, and induce sedation (Chinese Pharmacopoeia Commission 2010). Existing results showed lignans were the major bioactive compounds of Wuweizi, which exhibited antihepatotoxic, antioxidant, and antitumor activity, as well as effects on physical performance and central nervous system (Liu and Lesca 1982; Upton 1999; Opletal et al. 2004; Panossian and Wikman 2008). In 2002, Wuweizi was enrolled the list available for health food by the Ministry of Health of the People's Republic of China (http://www.moh.gov.cn/publicfiles/business/htmlfiles/mohwsjdj/s3593/200810/38057.htm), which greatly increased its application. So it becomes more important to control and evaluate its quality. Nowadays, the method of quality evaluation of Wuweizi in Chinese Pharmacopoeia (2010 Version) is not sufficient, so it is significant to develop an effective method to analyze its chemical constituents and evaluate its quality accurately and comprehensively.
An accurate and reliable high performance liquid chromatography coupled with diode array detection and mass spectrometry (HPLC--DAD--MS) of multiple components determination in combination with chromatographic fingerprint analysis was developed for quality evaluation of Wuweizi. Nine lignans (Fig. 1), including schisandrol A (1), schisandrol B (4), angeloylgomisin H (8), gomisin C. (12), schisantherin A (17), schisanhenol (20), schisandrin A (27), schisandrin B (30), and schisandrin C (34) were determined simultaneously, and thirty-six common peaks were unequivocally identified or tentatively assigned by comparing their mass spectrometric data with reference compounds, self-established compound library and published literatures. Meanwhile, the thirty-six common peaks were selected as characteristic peaks to assess the similarity of forty-three batches of Wuweizi samples in chromatographic fingerprinting analysis. Moreover, hierarchical clustering analysis (HCA) and principal components analysis (PCA) were successfully applied to demonstrate the variability of the chromatographic fingerprinting analysis in forty-three batches of Wuweizi samples collected from different localities.
Experimental
Chemicals and reagents
Acetonitrile (HPLC grade) were purchased from Merck KGaA (Darmstadt, Germany). Pure water (18.2 M[OMEGA]) for the HPLC analysis was prepared from a Purelab Plus UV System (ELGA, UK). Methanol, formic acid, acetic acid and phosphoric acid (analytical grade) were purchased from Beijing Chemical Works (Beijing, China).
Reference compounds, schisandrol A, schisandrol B, angeloylgomisin H. gomisin G. schisantherin A, schisanhenol, schisandrin A, schisandrin B, and schisandrin C were isolated by our laboratory (Liu et al. 2012). The purities of all the reference compounds were greater than 95%, as determined by HPLC-DAD.
Preparation of reference solutions and samples
Reference compound solutions of schisandrol A (1), schisandrol B (4), angeloylgomisin H (8), gomisin G (12), schisantherin A (17), schisanhenol (20), schisandrin A (27), schisandrin B (30), and schisandrin C (34) were prepared in methanol at the appropriate concentration.
Forty-three batches of Wuweizi samples were collected from different localities including Chinese mainland, Hong Kong, D.P.R. Korea, and R.O. Korea. All of these specimens, identified by Prof. Bengang Zhang, were kept at our laboratory for future reference.
All air-dried samples were ground and sieved (65-mesh) separately. A sample (1,0000g) was suspended in 30ml methanol in a 50-ml capped conical flask, weighed accurately, and then was extracted under ultrasonic bath (60kHZ, 250W) for three cycles (20 min each) at room temperature. After cooling, methanol was added to the original weight. The sample solution was filtered through a 0.22 [micro]m membrane filter prior to injection into the HPLC system.
Chromatographic analysis
Chromatographic analysis was performed by a Waters 2695 high performance liquid chromatography system (Milford, MA, USA), coupled with a 2996 diode array detector. Chromatographic data were processed by Empower 2 software. Chromatographic separation was performed on a Waters Xbridge C18 column (250 mm x 4.6 mm i.d., 5 [micro]m). The mobile phase consisted of acetonitrile (A) and water (B), and the flow rate was 1.0ml/min. The eluting conditions was optimized as follows: isocratic at 38% A (0-26 min), linear gradient from 38% to 44% A (26-30 min), linear gradient from 44% to 48% A (30-45 min), isocratic at 48% A (48-50 min), linear gradient from 48% to 58% A (50-55 min), isocratic at 58% A (55-63 min), linear gradient from 58% to 56% A (63-64 min), isocratic at 56% A (64-85 min), linear gradient from 56% to 80% A (85-95 min), isocratic at 80% A (95-103 min), linear gradient from 80% to 100% A (103-105 min), isocratic at 100% A (105-118 min), linear gradient from 100% to 38% A (118-120 min) and isocratic at 38% A (120-130 min). The monitoring wavelength was set at 220 nm, and the online ultraviolet absorption spectra were recorded in the range of 200-400 nm. The column and auto-sampler were maintained at 35[degrees]C and 20 [degrees]C, respectively. The injection volume was 10 [micro]l.
HPLC-DAD-MS analysis
HPLC-DAD-MS analysis was carried out with Applied Biosystem 3200 Q-Trap mass spectrometer (Foster City, CA, USA) connected to an Agilent 1200 HPLC system via electro spray ionization interface. The chromatographic conditions were as described above. The mass spectrometer was optimized in positive ion mode with an ion spray voltage of 4500 V, curtain gas of 20 psi, nebulizer gas of 30 psi and auxiliary gas 60 psi. The ion source temperature was set at 375 [degrees]C. Ultrapure nitrogen was used as nebulizer, heater, curtain and collision-activated dissociation (CAD) gas. Data were processed by the Analyst 1.4 software (Applied Biosystems/MDS Sciex).
Method validation
Analytical method was validated for the linearity, limit of detection and quantification (LOD and LOQ), precision (inter-day and intra-day), stability, repeatability and recovery, following the International Conference on Harmonization guidelines (ICH 1996) and related literatures on quantitative determination (Zhang et al. 2008, 2009; Lee and Kim 2010).
Chemometric analysis
Similarity analysis was performed by the Similarity Evaluation System for Chromatographic Fingerprint of Traditional Chinese Medicine (Version 2004A), which was recommended by China's State Food and Drug Administration (SFDA). Hierarchical clustering analysis (HCA) and principal components analysis (PCA) were applied to demonstrate the variability of the chromatographic fingerprinting analysis in forty-three batches of Wuweizi samples collected from different localities by using PASW Statistics (Version 18.0) and the Unscrambler X 10.0 software from Camo AS (Trondheim, Norway).
Results and discussion
Optimization of HPLC conditions
To obtain accurate and valid chromatographic conditions, different HPLC parameters were compared and optimized, including various columns (Waters Xbridge C18 250 mm x 4.6 mm i.d., 5 [micro]m; Agilent Eclipse XDB-C18 250 mm x 4.6 mm i.d., 5 [micro]m; Merck Lichrospher 100 RP-18e 250 mm x 4.0 mm i.d., 5 [micro]m; MZ Perfectchrom 100 C18 250mm x4.6mm i.d., 5 [micro]m; and Kromasil KR 100 C18 250mm x 4.6mm i.d., 5 [micro]m), mobile phases (acetonitrile--water and methanol--water with different modifiers, including formic acid, acetic acid, and phosphoric acid), column temperature (30 [degrees]C, 35 [degrees]C, and 40 [degrees]C), and mobile phase flow rate (0.8, 1.0, and 1.2ml/min). Based on the maximum absorption of lignans in the UV spectra of the three-dimensional chromatograms obtained by DAD detection, the monitoring wavelength was set at 220 nm, where most compounds could be detected and had adequate absorption. As a result, the optimized HPLC condition was established by comparing the resolution, baseline, elution time, and the number of characteristic peaks in each chromatogram after repeated experiments. Typical chromatograms for chemical analysis were shown in Fig. 2.
Optimization of extraction procedures
To obtain satisfactory extraction efficiency, ultrasonic, heat refluxing, and soxhlet extraction were compared. It was found that ultrasonic extraction was simpler and more effective for lignan extraction, and then used in further experiments. The other factors of extraction procedures were optimized by [3.sup.4] orthogonal experiment, including extraction solvents (70%, 90%, and 100% of methanol), sample-solvent ratios (1:20, 1:30, and 1:40, w/v), extraction time (10, 20, and 30 min) and extraction cycles (1, 2, and 3 cycles). Comparing the numbers, areas and resolution of the chromatographic peaks obtained by different extraction procedures, the optimized extraction procedures were established. The samples were extracted by ultrasonic extraction with methanol of sample-solvent ratio 1:30 (w/v), and the process carried out three cycles (20 min each).
HPLC-DAD-MS identity confirmation
In search of databases, for example PubMed, ScienceDirect, SciFinder, Google scholar, and CNKI (Chinese National Knowledge Infrastructure), all compounds reported in the literatures on Wuweizi were summarized to establish a compound library of Wuweizi, which included name, molecular formula, molecular weight, chemical structures and references. HPLC--DAD--MS was employed to analyze the extract solution of Wuweizi sample (S1), and about thirty-six peaks were detected. Most of peaks were tentatively assigned as dibenzocylooctadiene lignans according to their maximal UV absorption wavelength ([[lambda].sub.max] 220 nm, 255 nm, and 280 nm) and mass spectrometric data. The total wavelength chromatogram (TWC) of DAD spectral data (B) and total ion chromatogram (TIC) of +Q1 (C) from sample Wuweizi sample (S1) spiked standard mixtures were presented in Fig. 2. By comparing the retention time, UV absorption and mass spectrometric data with reference compounds, 9 peaks were unequivocally identified, namely schisandrol A (1), schisandrol B (4), angeloylgomisin H (8), gomisin G (12), schisantherin A (17), schisanhenol (20), schisandrin A (27), schisandrin B (30), and schisandrin C (34). And other peaks frequently exhibited their quasi-molecular ions [[M+K].sup.+] [[M+Na].sup.+] [[M+[NH.sub.4]].sup.+] and [[M+H].sup.+], which were tentatively assigned by comparing their mass spectrometric data with established compound library and published literatures (Wang et al. 2012; Wagner et al. 2011; Huang et al. 2011; Zhou et al. 2011; Deng et al. 2008; Lu and Chen 2009; Wang et al. 2008; Huang et al. 2007, 2008), which were shown in Table 1.
Table 1 Identification of lignans in the extract solution of Wuwezi
(SI) by HPLC--DAD--MS.
No. RT Identity MW MS data in positive ion mode
(min) (m/z)
1 19.84 Schisandrol A * 432 471.3[[M+K].sup.+],
455.4[[M+Na].sup.+],
433.5[[M+H].sup.+]
2 25.40 Comisin D 530 569.2[[M+K].sup.+],
553.4[[M+Na].sup.+],
531.5[[M+H].sup.+]
3 27.92 Gomisin J 388 427.4[[M+K].sup.+],
411.4[[M+Na].sup.+],
389.3[[M+H].sup.+]
4 29.16 Schisandrol B * 416 455.4[[M+K].sup.+],
439.3[[M+Na].sup.+],
399.5[[M-OH].sup.+]
5 31.12 Micrantherin A 500 539.4[[M+K].sup.+],
523.4[[M+Na].sup.+],
518.5[[M+[NH.sub.4]].sup.+]
6 36.13 Tigloylgomisin H 500 539.4[[M+K].sup.+],
523.4[[M+Na].sup.+],
483.4[[M-OH].sup.+]
7 38.35 Pregomisin 390 429.1[[M+K].sup.+],
413.5[[M+Na].sup.+],
391,3[[M+H.sup.+]
8 40.16 Angeloylgomisin H 500 523.4[[M+Na].sup.+],
* 501,3[[M+H].sup.+],
483.5[[M-OH].sup.+]
9 42.24 Benzoylgomisin H 522 561.2[[M+K].sup.+],
545.4[[M+Na].sup.+],
505.4[[M-OH].sup.+]
10 44.05 Angeloylgomisin Q 530 569.5[[M+K].sup.+],
553.4[[M+Na].sup.+],
548.4[[M+[NH.sub.4]].sup.+]
11 45.02 Gomisin E 514 553.3[[M+K].sup.+],
537.3[[M+Na].sup.+],
497.5[[M-OH].sup.+]
12 46.41 Gomisin C. * 536 575.3[[M+K].sup.+],
559.3[[M+Na].sup.+],
554.5[[M+[NH.sub.4]].sup.+]
13 47.25 Gomisin F 514 553.4[[M+K].sup.+],
537.4[[M+Na].sup.+],
532.3[[M+[NH.sub.4]].sup.+]
14 48.79 (-)-Gomisin 402 425.3[[M+Na].sup.+],
[K.sub.1] 403.3[[M+H].sup.+]
15 49.63 (+)-Gomisin 402 425.4[[M+Na].sup.+],
[K.sub.2] 403.3[[M+H].sup.+]
16 50.87 Schisantherin B 514 553.2[[M+k].sup.+],
537.4[[M+Na].sup.+],
532.3[[M+[NH.sub.4].sup.+]
17 52.83 Schisantherin A * 536 575.2[[M+K].sup.+],
559.4[[M+Na].sup.+],
554.4[[M+[NH.sub.4]].sup.+]
18 54.07 Tigloylgomisin P 514 553.3[[M+K].sup.+],
537.4[[M+Na].sup.+],
532.5[[M+[NH.sub.4]].sup.+]
19 55.48 Schisantherin D 520 559.4[[M+K].sup.+],
543.4[[M+Na].sup.+],
538.2[[M+[NH.sub.4]].sup.+]
20 56.30 Schsanhenol * 402 441,3[[M+K].sup.+],
425.4[[M+Na].sup.+],
403.3[[M+H].sup.+]
21 58.11 (-)-Gomisin 386 409.4[[M+Na].sup.+],
[L.sub.1] 387.4[[M+H].sup.+]
22 58.81 (-)-Gomisin 414 437.2[[M+Na].sup.+],
[L.sub.2] 415.4[[M+H].sup.+]
23 59.37 Schisantherin C 514 553.5[[M+K].sup.+],
537.5[[M+Na].sup.+],
515.3[[M+H].sup.+]
24 60.76 (+)-Gomisin 386 425.2[[M+K].sup.+],
[M.sub.1] 409.4[[M+Na].sup.+],
387.4[[M+H].sup.+]
25 61.88 (+)-Gomisin 386 425.1[[M+K].sup.+],
[M.sub.2] 409.1[[M+Na].sup.+],
387.2[[M+H].sup.+]
26 64.51 Gomisin O 416 455.3[[M+K].sup.+],
417.4[[M+H].sup.+],
399.3[[M-OH].sup.+]
27 66.05 Schisandrin A * 416 455.4[[M+K].sup.+],
439.3[[M+Na].sup.+],
417.4[[M+H].sup.+]
28 68.28 Epigomisin O 416 437.4[[M+Na].sup.+],
417.3[[M+H].sup.+]
29 74.68 Gomisin N 400 439.4[[M+K].sup.+],
423.3[[M+Na].sup.+],
401.5[[M+H].sup.+]
30 77.31 Schisandrin B * 400 439.2[[M+K].sup.+],
423.3[[M+Na].sup.+].
401.5[[M+H].sup.+]
31 78.85 Tigloylgomisin O 498 537.3[[M+K].sup.+],
521,4[[M+Na].sup.+]
32 83.31 Benzoylisogomisin 520 559.4[[M+K].sup.+],
O 543.3[[M+Na].sup.+]
33 84.83 Angeloylisogomisin 498 537.4[[M+K].sup.+],
O 521.5[[M+Na].sup.+]
34 86.64 Schisandrin C * 384 407.2[[M+Na].sup.+],
385.2[[M+H].sup.+]
35 89.99 Benzoylgomisin O 520 559.1[[M+K].sup.+],
543.4[[M+Na].sup.+]
36 91.23 Angeloylgomisin O 498 537.2[[M+K].sup.+],
521.5[[M+Na].sup.+]
Compounds with "*" were identified with references accurately, and
the others were tentatively assigned.
Validation of quantitative analysis method
Linearity and linear ranges of nine lignans were determined by using the developed method. Their correlation coefficient values (r [greater than or equal to] 0.9999) indicated appropriate correlations between concentrations of the investigated compound and their peak areas within the test ranges (Table 2). The LODs and LOQs were less than 0.06 and 0.14 [micro]g/ml, which were determined at a signal-to-noise ratio (S/N) of about 3 and 10, respectively (Table 2). Intra- and inter-day variations of nine lignans were less than 1.00% and 1.08%, respectively (Table 3). The developed method had good repeatability and stability with RSD < 2.02% (Table 3). In addition, the developed method had good accuracy with the recoveries in the range of 93.02-98.98% (RSD < 2.18%) as shown in Table 4.
Table 2 Linearity, LODs and LOQs for nine lignans.
Compound Calibration r Linear range LOD([micro]g/ml)
curve ([micro]g/ml)
Schisandrol A y = 1.0000 5.20-260.00 0.03
56224578.37x
- 22357.78
Schisandrol B y = 1.0000 5.40-270.00 0.03
60130384.09x
- 32332.77
Angeloylgomisin y = 0.9999 3.60-72.00 0.06
H 39159176.57x
- 13928.47
Comisin C y = 1.0000 0.50-100.00 0.05
48482880.11x
- 18412.37
Schisantherin A y = 1.0000 0.50-100.00 0.06
53568965.50x
- 12423.45
Schisanhenol y = 1.0000 0.50-100.00 0.06
68677492.21x
- 17195.07
Schisandrin A y = 1.0000 0.50-150.00 0.05
65629953.00x
- 17045.41
Schisandrin B y = 1.0000 0.50-150.00 0.05
66835719.44x
- 25528.71
Schisandrin C y = 1.0000 0.52-156.00 0.05
64702238.87x
- 19668.09
Compound LOQ([micro]g/ml)
Schisandrol A 0.06
Schisandrol B 0.07
Angeloylgomisin 0.11
H
Comisin C 0.12
Schisantherin A 0.13
Schisanhenol 0.14
Schisandrin A 0.11
Schisandrin B 0.11
Schisandrin C 0.11
Table 3 Precisions, stability and repeatability of nine lignans.
Compound Precision Repeatability Stability
(n = 5) (n = 6) (n = 6)
RSD(%) RSD (%)
Intra-day Inter-day
RSD (%) RSD (%)
Schisandrol A 0.82 0.89 0.57 0.83
Schisandrol B 0.85 0.92 0.74 0.89
Angeloylgomisin 0.70 0.76 0.79 0.80
H
Gomisin C 0.72 0.78 1.30 0.82
Schisantherin A 0.86 0.94 0.83 0.96
Schisanhenol 0.80 0.88 2.02 1.03
Schisandrin A 0.75 0.82 1.13 0.84
Schisandrin B 0.99 1.06 0.52 1.02
Schisandrin C 1.00 1.08 0.49 1.12
Table 4 Recoveries of nine lignans in Wuweizi (S1)(n = 3).
Compound Original Spiked Found Recovery RSD
(mg/g) (mg/g) (mg/g) (%) (%)
Schisandrol A 7.5377 0.1620 7.6924 95.49 1.26
0.8100 8.3364 98.60 0.97
1.6200 9.1568 99.94 0.85
Schisandrol B 2.8803 0.1574 3.0367 99.36 1.31
0.7868 3.6476 97.52 0.94
1.5737 4.4551 100.07 0.82
Angeloylgomisin 2.5912 0.1228 2.7092 96.09 1.52
H
0.6140 3.1982 98.86 0.78
1.2280 3.8172 99.84 0.76
Gomisin G 0.4290 0.1398 0.5618 94.99 1.48
0.6990 1.0940 95.14 0.91
1.3980 1.7984 97.95 1.03
Schisantherin A 0.2352 0.1442 0.3723 95.08 1.98
0.7210 0.9227 95.35 1.53
1.4421 1.6123 95.49 1.44
Schisanhenol 0.1262 0.1409 0.2576 93.26 2.18
0.7046 0.7826 93.16 1.34
1.4092 1.4315 92.63 1.08
Schisandrin A 1.0909 0.1471 1.2318 95.79 1.84
0.7354 1.8183 98.91 1.03
1.4708 2.8947 102.24 0.91
Schisandrin B 4.0491 0.1447 4.1898 97.24 1.25
0.7236 4.7707 99.72 1.09
1.4472 5.4943 99.86 0.86
Schisandrin C 0.9575 0.1530 1.1014 94.05 1.62
0.7647 1.7067 97.97 0.92
1.5296 2.4701 98.89 0.98
Sample analysis
Quantitative analysis
The newly developed method was subsequently applied to quantitative analysis of nine lignans in forty-three batches of Wuweizi samples collected from different localities. Each sample was analyzed three times to determine the mean content (mg/g), and the results were shown in Table 5. These results indicated that the content of nine lignans varied greatly among the samples collected from different localities, and the total content of nine lignans was higher in the samples collected from Heilongjiang province and Liaoning province than those collected in the other localities in China. The total content of nine lignans in the wild samples (S1, S9, S10, S31, and S32) was much higher than those of samples (p < 0.01). The results further revealed that schisandrol A, schisandrol B, angeloylgomisin H, schisandrin A, schisandrin B and schisandrin C were the main chemical constituents of Wuweizi, which were of great importance to establish a better determination method for its quality control.
Table 5 Content of nine lignans in Wuweizi samples collected from
different localities (n = 3).
No. Origin Collecting Content
time (mg/g)
1 2 3 4
S1 Liaoning 2010.10 7.54 2.88 2.59 0.43
Province
Qingyuan County
(W)
S2 Liaoning 2010.10 4.96 1.26 1.40 0.23
Province
Fengcheng County
(C)
S3 Liaoning 2010.10 6.13 1.41 1.74 0.27
Province Xinbin
County (C)
S4 Liaoning 2010.10 6.48 1.39 1.81 0.25
Province Huanren
County (C)
S5 Liaoning 2011.10 5.26 1.49 1.53 0.31
Province Xinbin
County (C)
S6 Liaoning 2011.10 4.89 1.05 1.32 0.21
Province
Fengcheng County
(C)
S7 Liaoning 2011.10 5.26 1.50 1.69 0.26
Province
Qingyuan County
(W)
S8 Liaoning 2011.10 5.50 1.36 1.62 0.29
Province
Kuandian County
(C)
S9 Heilongjiang 2010.10 9.34 3.52 2.22 0.69
Province
Shangzhi County
(W)
S10 Heilongjiang 2010.10 11.08 3.89 3.81 0.53
Province Lesser
Khingan (W)
S11 Heilongjiang 2011.10 4.67 1.78 1.61 0.21
Province Wuchang
County (C)
S12 Heilongjiang 2011.10 5.34 1.98 1.84 0.25
Province Mulan
County (C)
S13 Heilongjiang 2011.10 5.39 1.20 1.50 0.20
Province Yichun
City (C)
S14 Heilongjiang 2011.11 4.27 1.02 1.22 0.17
Province
Fangzheng County
(C)
S15 Heilongjiang 2011.11 4.30 1.68 1.43 0.19
Province Jiayin
County (C)
S16 Heilongjiang 2011.11 5.16 1.23 1.41 0.12
Province Bin
County (C)
S17 Heilongjiang 2011.11 4.94 0.97 1.37 0.15
Province Suihua
City (C)
S18 Jilin Province 2010.10 4.37 0.87 1.19 0.13
Ji'an City (C)
S19 Jilin Province 2010.10 3.65 0.97 1.11 0.20
Jingyu City (C)
S20 Jilin Province 2010.10 3.86 1.27 1.23 0.24
WangqingCounty
S21 Jilin Province 2010.11 3.83 1.40 1.21 0.25
Fusong city (C)
S22 Jilin Province 2010.11 3.51 1.16 1.16 0.19
Huadian County
(C)
S23 Jilin Province 2011.10 4.52 1.05 1.26 0.18
Linjiang County
(C)
S24 Jilin Province 2011.11 4.21 0.98 1.16 0.15
Human County
(C)
S25 Jilin Province 2011.10 3.57 1.57 1.31 0.17
Changbai County
(C)
S26 Jilin Province 2011.11 4.25 1.03 1.23 0.19
Yanbian County
(C)
S27 Jilin Province 2011.11 3.86 0.93 1.10 0.16
Antu County (C)
S28 Jilin 2011.11 4.36 1.03 1.22 0.21
ProvinceTonghua
City (C)
S29 Inner Mongolia 2010.10 3.93 0.67 1.15 0.34
Bayanhot (D)
S30 Heibei Province 2010.10 5.57 1.95 1.82 0.45
Fengning County
(D)
S31 D.P.R. Korea 2011.11 6.19 4.09 2.71 0.46
Ryanggang
Province (W)
S32 D.P.R. Korea 2010.10 6.62 3.43 2.55 0.37
Ryanggang
Province (W)
S33 Hong Kong 2011.11 4.75 1.13 1.32 0.20
Xinchengmen Drug
Store (D)
S34 Hong Kong 2011.11 5.05 1.12 1.41 0.19
Dahetang Drug
Store (D)
S35 Hong Kong 2011.11 5.50 1.23 1.51 0.21
Yurentang Drug
Store (D)
S36 Hong Kong 2011.11 5.10 1.27 1.45 0.20
Lian'anshenrong
Drug Store (D)
S37 Hong Kong 2011.11 4.88 1.18 1.38 0.20
Lichangshenrong
Drug Store (D)
S38 Hong Kong Da'an 2011.11 3.77 0.97 1.06 0.17
Drug Store (D)
S39 R.O. Korea Kgung 2011.10 5.78 1.21 1.20 0.07
book Moon Kgung
1 (D)
S40 R.O. Korea Kgung 2011.10 5.93 1.42 1.43 0.07
book Moon Kgung
2 (D)
S41 R.O. Korea Kgung 2011.10 6.09 1.24 1.27 0.05
book Moon Kgung
3 (D)
S42 R.O. Korea Kang 2011.10 5.29 1.52 1.42 0.12
wondo (D)
S43 R.O. Korea Chong 2011.10 4.11 0.88 1.15 0.13
book Danyong
(D)
No. Origin
5 6 7 8 9 Total
S1 Liaoning 0.24 0.13 1.09 4.05 0.96 19.90
Province
Qingyuan County
(W)
S2 Liaoning 0.18 0.09 0.71 1.92 0.28 11.03
Province
Fengcheng County
(C)
S3 Liaoning 0.22 0.06 1.12 2.18 0.27 13.41
Province Xinbin
County (C)
S4 Liaoning 0.29 0.11 1.23 2.54 0.31 14.40
Province Huanren
County (C)
S5 Liaoning 0.20 0.07 0.80 2.29 0.37 12.32
Province Xinbin
County (C)
S6 Liaoning 0.20 0.07 0.91 1.75 0.21 10.61
Province
Fengcheng County
(C)
S7 Liaoning 0.26 0.01 0.57 2.74 0.25 12.54
Province
Qingyuan County
(W)
S8 Liaoning 0.27 0.12 0.78 2.26 0.28 12.49
Province
Kuandian County
(C)
S9 Heilongjiang 0.18 0.07 1.37 3.35 1.17 21.91
Province
Shangzhi County
(W)
S10 Heilongjiang 0.49 0.19 1.89 5.52 0.90 28.29
Province Lesser
Khingan (W)
S11 Heilongjiang 0.22 0.07 0.65 2.12 0.33 11.66
Province Wuchang
County (C)
S12 Heilongjiang 0.26 0.08 0.82 2.61 0.41 13.59
Province Mulan
County (C)
S13 Heilongjiang 0.19 0.08 0.74 1.91 0.27 11.48
Province Yichun
City (C)
S14 Heilongjiang 0.22 0.06 0.55 1.53 0.22 9.26
Province
Fangzheng County
(C)
S15 Heilongjiang 0.13 0.07 0.53 2.67 0.68 11.68
Province Jiayin
County (C)
S16 Heilongjiang 0.10 0.10 0.88 2.17 0.53 11.69
Province Bin
County (C)
S17 Heilongjiang 0.16 0.10 1.22 1.96 0.21 11.08
Province Suihua
City (C)
S18 Jilin Province 0.13 0.11 1.10 1.68 0.20 9.80
Ji'an City (C)
S19 Jilin Province 0.15 0.05 0.52 1.48 0.22 8.35
Jingyu City (C)
S20 Jilin Province 0.16 0.06 0.61 2.03 0.41 9.87
WangqingCounty
S21 Jilin Province 0.12 0.10 0.79 1.36 0.34 9.39
Fusong city (C)
S22 Jilin Province 0.09 0.05 0.47 1.51 0.35 8.49
Huadian County
(C)
S23 Jilin Province 0.15 0.08 0.79 1.61 0.20 9.85
Linjiang County
(C)
S24 Jilin Province 0.11 0.07 0.77 1.58 0.23 9.27
Human County
(C)
S25 Jilin Province 0.15 0.08 0.71 2.04 0.26 9.86
Changbai County
(C)
S26 Jilin Province 0.14 0.09 0.87 1.92 0.29 9.99
Yanbian County
(C)
S27 Jilin Province 0.15 0.07 0.70 1.50 0.19 8.67
Antu County (C)
S28 Jilin 0.14 0.07 0.50 1.62 0.23 9.39
ProvinceTonghua
City (C)
S29 Inner Mongolia 0.21 0.09 0.57 1.24 0.19 8.40
Bayanhot (D)
S30 Heibei Province 0.07 0.01 0.51 1.05 0.38 11.81
Fengning County
(D)
S31 D.P.R. Korea 0.17 0.10 0.88 3.87 1.61 20.09
Ryanggang
Province (W)
S32 D.P.R. Korea 0.17 0.12 1.14 4.05 1.29 19.74
Ryanggang
Province (W)
S33 Hong Kong 0.17 0.08 0.81 1.74 0.22 10.43
Xinchengmen Drug
Store (D)
S34 Hong Kong 0.18 0.08 0.90 1.71 0.22 10.87
Dahetang Drug
Store (D)
S35 Hong Kong 0.23 0.08 1.08 2.06 0.22 12.12
Yurentang Drug
Store (D)
S36 Hong Kong 0.23 0.09 1.05 2.07 0.22 11.69
Lian'anshenrong
Drug Store (D)
S37 Hong Kong 0.15 0.08 0.81 1.76 0.24 10.70
Lichangshenrong
Drug Store (D)
S38 Hong Kong Da'an 0.12 0.06 0.57 1.47 0.21 8.41
Drug Store (D)
S39 R.O. Korea Kgung 0.05 0.09 0.89 3.03 0.98 13.30
book Moon Kgung
1 (D)
S40 R.O. Korea Kgung 0.07 0.10 1.08 3.49 0.93 14.52
book Moon Kgung
2 (D)
S41 R.O. Korea Kgung 0.06 0.09 0.89 3.29 1.05 14.02
book Moon Kgung
3 (D)
S42 R.O. Korea Kang 0.08 0.08 0.75 2.96 1.05 13.27
wondo (D)
S43 R.O. Korea Chong 0.13 0.06 0.68 1.70 0.18 9.02
book Danyong
(D)
C. cultivated sample; D, drug sample; W, wild sample.
Similarity analysis
In order to evaluate the similarities and differences in these samples. Similarity Evaluation System for Chromatographic Fingerprint of Traditional Chinese Medicine (Version 2004A) was performed based on their HPLC profiles. Chromatograms of these samples were shown in Fig. 3. Thirty-six peaks that existed in all forty-three batches of Wuweizi samples with reasonable heights and good resolution were assigned as "Characteristic peaks". The similarities of the chromatograms of forty-three batches of Wuweizi samples were compared to the reference fingerprint "R" (Fig. 3). The closer the cosine values approached 1, the more similar the two chromatograms were. If a similarity value over a certain value (for example, 0.97) was regarded as the threshold value for qualification, it was easy to identify the qualified sample based on the chromatographic fingerprint. The similarity values of 43 samples were more than 0.97, except for S30, S31 and S32. These meant that forty batches of Wuweizi samples showed good similarity on chemical constituents.
Hierarchical cluster analysis
Hierarchical cluster analysis (HCA) was performed based on thirty-six peaks in their HPLC profiles. The results of HCA were shown in Fig. 4A. The samples were divided into two clusters obviously. Cluster 1 was formed by the samples S31, S32, S1, S9 and S10. Cluster 11 consisted of the remaining samples. The samples in Cluster I were wild, and samples in Cluster II were cultivated or drugs. The total content of nine investigated lignans of samples in Cluster I was much higher than those of samples in Cluster II. In Cluster I, S31 and S32 collected from D.P.R. Korea with different collecting time were clustered into one group. S1, S31 and S32 were clustered into a higher group because of the close content of principle components. S9 and S10 collected from Heilongjiang province had much higher content of total lignans in Cluster I, and the content of schisandrol A was one of important distinguishing factors. In Cluster II, the thirty-eight samples were divided into six groups clearly. Samples collected from close locality were mostly clustered into one group, for example. S39, S40. S41, and S42 collected from different drug stores in R.O. Korea were clustered into one group, and they had relatively high content in schisandrin C than those of samples, which showed some differences with other sample in HPLC fingerprint. S33, S34, S35, S36, and S37 collected from different drug stores in Hong Kong were clustered into one group with samples collected Liaoning province and Heilongjiang province, which indicated that the samples collected from the drug stores in Hong Kong might come from Liaoning province and Heilongjiang province, which were same to the results of market survey. These results indicated HCA was helpful to distinguish the origin information of samples and evaluate the quality of Wuweizi.
Principle component analysis
To identify the differences among forty-three batches of Wuweizi samples, principle component analysis (PCA) was performed on the thirty-six peaks in their HPLC profiles. To display the points on two principal components, PC 1 and PC 2 (first and second principal components) were chosen to represent the information, and their scores were more than 93%. As shown in Fig. 4B, PCA displayed the results that forty-three batches of Wuweizi samples were classified into two groups, which were very similar to the results of HCA. Group I was formed by the wild samples S31, S32, S1, S9 and S10. Group II was formed by the remaining samples. Moreover, the results of the loading plot of the PCA indicated that schisandrol A (1), gomisin J (3), schisandrol B (4), angeloylgomisin H (8), tigloylgomisin P (18), schisandrin A (27), gomisin N (29), schisandrin B (30), and schisandrin C (34) might have more influence on the discrimination of the samples from different localities than other compounds (Fig. 4C). Therein schisandrol A (1), schisandrol B (4), angeloylgomisin H (8), schisandrin A (27) schisandrin B (30), and schisandrin C (34) could be identified accurately, and gomisin J (3), tigloylgomisin P (18), and gomisin N (29) were tentatively assigned. The results showed schisandrol A (1), schisandrol B (4). schisandrin B (30), and schisandrin C (34) could be chosen as the chemical markers for evaluate the quality of Wuweizi.
Conclusions
A reliable method for comprehensive chemical analysis of Wuweizi by HPLC-DAD-MS combined with chemometrics was developed for discriminating the origin information and evaluating the quality of Wuweizi. Thirty-six peaks in the extract solution of Wuweizi were unequivocally identified or tentatively assigned, and nine lignans were analyzed quantitatively in forty-three batches of Wuweizi samples collected from different localities. Chemometrics were successfully applied to comprehensive chemical analysis of forty-three batches of Wuweizi samples to explain the difference. The results indicated the content of nine investigated lignans varied greatly among the samples collected from different localities. Samples with different localities could be discriminated according to the results of chemometrics. Moreover, schisandrol A, schisandrol B, schisandrin B, schisandrin C were found to chemical marker for evaluating the quality of Wuweizi. The results of chemometrics analysis of various Wuweizi herbal drugs make it necessary to compare also the pharmacological activities of the different batches in some pharmacological models and adapt the application and development of new Schisandra containing formulas accordingly.
Acknowledgements
The authors are grateful for the financial support provided by the National Natural Sciences Foundation of China (No. 81001609) and the Fundamental Research Funds for the Central Scientific Research Institutes for Public Welfare.
References
Chinese Pharmacopoeia Commission. 2010. Pharmacopoeia of the People's Republic of China Version (2010). Chin. Med. Sci. Press, Beijing.
Deng, X.X., Chen, X.H., Cheng, W.M., Shen, Z.D., Bi, K.S., 2008. Simultaneous LC-MS quantification of 15 lignans in Schisandra chinensis (Turcz.) Baill. Fruit. Chromatographia 67, 559-566.
Hancke, J.L., Burgos, R.A., Ahumada, F., 1999. Schisandra chinensis (Turcz.) Baill. Fitoterapia 70, 451-471.
Huang, X., Song, F.R., Liu, Z.Q., Liu, S.Y., Ai, J., 2011. Comprehensive quality evaluation of Fructus Schisandrae using electrospray ionization ion trap multiple-stage tandem mass spectrometry coupled with chemical pattern recognition techniques. Analyst 136, 4308-4315.
Huang, X., Song, F.R., Liu, Z.Q., Liu, S.Y., 2008. Structural characterization and identification of dibenzocyclooctadiene lignans in Fructus Schisandrae using electrospray ionization ion trap multiple-stage tandem mass spectrometry and electrospray ionization Fourier transform ion cyclotron resonance multiple-stage tandem mass spectrometry. Analytica Chimica Acta 615, 124-135.
Huang, X., Song, F.R., Liu, Z.Q., Liu, S.Y., 2007. Studies on lignan constituents from Schisandra chinensis (Turcz.) Baill. fruits using high-performance liquid chromatography/electrospray ionization multiple-stage tandem mass spectrometry. Journal of Mass Spectrometry 42, 1148-1161.
http://www.moh.gov.cn/publicfiles/business/htmlfiles/mohwsidj/s3593/200810/38057.htm
ICH, T.Q.B., 1996. Guidance for Industry. Q2B Validat ion of Analytical Procedures. Methodology. FDA. Rockville.
Lebedev, A.A., 1971. Schizandra. Meditsina Publishing House of Uzbek SSR, Tashkent.
Lee, H.J., Kim, C.Y., 2010. Simultaneous determination of nine lignans using pressurized liquid extraction and HPLC-DAD in the fruits of Schisandra chinensis. Food Chemistry 120, 1224-1228.
Liu, H.T., Li, X.B., Zhang, J., Zhu, Y.X., Qi, Y.D., Peng, Y., Zhang, B.G., Xiao, P.G., 2012. Chemical constituents of petroleum ether extract of fruits of Schisandra sphenan-thera. China Journal of Chinese Materia Medica 37, 1597-1601.
Liu, K.T., Lesca, P., 1982. Pharmacological properties of dibenzo[a.c]cyclooctene derivatives isolated from Fructus Schizandrae chinensis III. Inhibitory effects on carbon tetrachloride-induced lipid peroxidation, metabolism and covalent binding of carbon tetrachloride to lipids. Chemico-Biological Interactions 41, 39-47.
Lu, Y., Chen, D.F., 2009. Analysis of Schisandra chinensis and Schisandra sphenanthera. Journal of Chromatography A 1216, 1980-1990.
Opletal, L., Sovova, H., Bartlova, M., 2004. Dibenzo[a,c]cyclooctadiene lignans of the genus Schisandra: importance, isolation and determination. Journal of Chromatography B 812, 357-371.
Panossian, A., Wikman, G., 2008. Pharmacology of Schisandra chinensis Bail.: an overview of Russian research and uses in medicine. Journal of Ethnopharmacology 118, 183-212.
Upton, R., 1999. American Herbal Pharmacopoeia and Therapeutic Compendium. Schisandra Berry. American Herbal Pharmacopoeia, Santa Cruz.
Wagner, H., Bauer, R., Melchart, D., Xiao, P.G., Staudinger, A., 2011. Chromatographic Fingerprint Analysis of Herbal Medicines. Thin-Layer and High Performance Liquid Chromatography of Chinese Drugs (Vol. I). Fructus Schisandrae--Wuweizi. Springer. Wien, New York, pp. 37-44.
Wang, M.C., Lai, Y.C., Chang, C.L., 2008. High throughput screening and antioxidant assay of dibenzo[a,c]cyclooctadiene lignans in modified ultrasonic and supercritical fluid extracts of Schisandra chinensis Baill. by liquid chromatography-mass spectrometry and a free radical-scavenging method. Journal of Separation Science 31, 1322-1332.
Wang, S.W., Wang, C., Zhao, X., Mao, S.L., Wu, Y.T., Fan, G.R., 2012. Comprehensive two-dimensional high performance liquid chromatography system with immobilized liposome chromatography column and monolithic column for separation of the traditional Chinese medicine Schisandra chinensis. Analytica Chimica Acta 713, 121-129.
Zhang, H., Zhang, G.Q., Zhu, Z.Y., Zhao, L. Fei, Y., Jing, J., Chai, Y.F., 2009. Determination of six lignans in Schisandra chinensis (Turcz.) Baill. fruits and related Chinese multi-herb remedies by HPLC. Food Chemistry 115, 735-739.
Zhang, Y.Y., Dan, M., Wu, J.B., Yang, H.Z., Huang, H., Qi, Y., Wei, S.D., Okuyama, T., Nakajima, K., 2008. Study on the chromatographic fingerprinting of Schisandra chinensis (Turcz.) Baiil. by LC coupled with principal component analysis. Chromatographia 68, 101-104.
Zhou, Y., Huang, S.X., Pu, J.X., Li, J.R., Ding, L.S., Chen, D.F., Sun, H.D., Xu, H.X., 2011. Ultra performance liquid chromatography coupled with quadrupole time-of-flight mass spectrometric procedure for qualitative and quantitative analyses of nortriterpenoids and lignans in the genus Schisandra. Journal of Pharmaceutical and Biomedical Analysis 56, 916-927.
Haitao Liu (a), Hongwu Lai (a), Xinyue Jia (b), Jiushi Liu (a), Zhao Zhang (a), Yaodong Qi (a), Jin Zhang (a), Junbin Song (c), Chongming Wu (a), Bengang Zhang (a), *, Peigen Xiao (a)
(a) Key Laboratory of Bioactive Substances and Resources Utilization of Chinese Herbal Medicine (Peking Union Medical College), Ministry of Education, Institute of Medicinal Plant Development, Chinese Academy of Medical Sciences, Peking Union Medical College, Beijing 100193, China
(b) Xinjiang Institute of Chinese and Ethnic Medicine, Urumqi 830002, China
(c) College of Oriental Medicine. Kyung Hee University, Seoul 130-701, Republic of Korea
* Corresponding author. Tel.: +86 10 62899725; fax: +86 10 57833196. E-mail address: bgzhang@implad.ac.cn (B. Zhang).
 Printer friendly
 Cite/link
 Email
 Feedback
| |
| Author: | Liu, Haitao; Lai, Hongwu; Jia, Xinyue; Liu, Jiushi; Zhang, Zhao; Qi, Yaodong; Zhang, Jin; Song, Junb |
|---|---|
| Publication: | Phytomedicine: International Journal of Phytotherapy & Phytopharmacology |
| Article Type: | Report |
| Geographic Code: | 9SOUT |
| Date: | Sep 15, 2013 |
| Words: | 6801 |
| Previous Article: | Protective activity of Cynara scolymus L. leaf extract against chemically induced complex genomic alterations in CHO cells. |
| Next Article: | Bioactive acetylenic metabolites. |
| Topics: | |