A Simple and Reliable Method for Distinguishing Danshen in Salvia: Simultaneous Quantification of Six Active Compositions by HPLC

Journal of Chromatographic Science 2014;52:992 998 doi:10.1093/chromsci/bmt140 Advance Access publication September 19, 2013 Article A Simple and Reliable Method for Distinguishing Danshen in Salvia: Simultaneous Quantification of Six Active Compositions by HPLC Tao Wang 1, Hui Zhang 1, Long Wang 1, Yuanyuan Jiang 1, Li Zhang 1 *, Yonghong Zhou 2, Ruiwu Yang 1, Chunbang Ding 1 and Xiaoli Wang 1 1 College of Biology and Science, Sichuan Agricultural University, Ya an 625014, China, and 2 Triticeae Research Institute, Sichuan Agricultural University, Wenjiang 611130, China *Author to whom correspondence should be addressed. Email: zhang8434@sina.com These authors contributed equally to this work. Received 18 May 2013; revised 17 July 2013 A simple and reliable method for distinguishing Danshen is important to evaluate the quality and clinical efficiency of these species. An HPLC method was developed for the determination of protocatechuic aldehyde, salvianolic acid A, salvianolic acid B, cryptotanshinone, tanshinone I and tanshinone IIA in 23 samples of Salvia. The analytes were separated on an Agilent XDB C 18 reversed-phase column coupled with a Phenomenex C 18 guard column using a gradient elution of acetonitile 0.1% aqueous phosphoric acid as the mobile phase at a flow rate 0.8 ml/min and UV detection at 280 nm. The method allowing the simultaneous quantification of six major active compositions was optimized and validated for linearity, precision, accuracy and limits of detection (LOD) and quantification. The LOD ranged from 0.019 to 0.850 mg/ml (R 2 0.9998). Accuracy, precision and reproducibility were all within the required limits. The average recovery between 96.49 and 102.16% and the relative standard deviations were <3.01%. Based on the six compositions content and clustering result, this research results suggest that these six major active compositions could be distinguishing markers for Danshen and non-danshen. Introduction Salvia, a genus of the Lamiaceae, is known as a prolific source of traditional Chinese medicine (TCM) and national drugs (such as Tibetan, Nakhi, etc.) (1). Radix et Rhizoma Salviae Miltiorrhizae (Danshen in Chinese) is a functional food and medicinal material in China. These roots have been commonly used for the treatment of coronary heart disease, cerebrovascular disease, bone loss, hepatitis, hepatocirrhosis and chronic renal failure (2 5). Based on pharmacological research, the active components of Danshen are classified into two groups: phenolic compounds such as potocatechuic aldehyde, salvianolic acid A and salvianolic acid B and tanshinone compounds such as cryptotanshinone, tanshinone I and tanshinone IIA (Figure 1). Pharmacological experiments have demonstrated that phenolic compounds can effectively restrain lipid peroxidation of the brain, liver and kidney; tanshinone compounds can efficiently resist bacteria and tumors, as well as diminish inflammation (6 8). In addition, the dried radix and rhizoma of other Salvia species containing S. castanea and S. przewalskii are commonly used as substitutes for Danshen in regions of China, particularly Sichuan, Yunnan, Qinghai and Gansu because of high demand (1, 9 11). China is the birthplace and also a major producer and exporter of Danshen. But because of reduced Danshen production and increase in demand in the lower reaches, the Danshen market in the world is now in short supply, Danshen prices have gone high and more and more fake Danshen (for example, S. castanea, S. digitaloides, S. maximowicziana, S. paohsingensis, S. trijuga, etc.) products appeared. A simple and reliable method for distinguishing Danshen is important to evaluate the quality and clinical efficiency of these species. Liu et al. studied phenolic acid from Danshen by HPLC DAD ESI MS n, and 42 kinds of phenolic acids were identified (12). In addition, Sun et al. (13) and Lou et al. (14) established an HPLC method for determination of four active constituents in Danshen tablet and Danshen, respectively. Cheng et al. (15) analyzed eight active constituents of Danshen and S. przewalskii.in sum, the goal of previous studies was to determine the active ingredient of Danshen or Salvia herbs species of Salvia is scarce and could not provide indication of Danshen chemical components as identification markers. Thus, the development of a simple and effective HPLC method is essential for the simultaneous determination of two kinds of active components as identification markers. This study aims (1) to develop a simple and reliable HPLC method for simultaneous determination of potocatechuic aldehyde, salvianolic acid A, salvianolic acid B, cryptotanshinone, tanshinone I and tanshinone IIA in Salvia and (2) to provide a method for identification of Danshen. This study is also the first to report on the HPLC quantitative analysis of certain species of Salvia. Materials and Methods Materials and reagents The 23 samples of Salvia (consists of 6 strains and 9 different place of S. miltiorrhiza, 8Salvia species). The Latin names, locations and percentages of the six major active constituents are listed in Table I. The samples of all these voucher herbarium were deposited in Sichuan Agricultural University Herbarium, Sichuan Agricultural University, Ya an, Sichuan, China. Standard substances of potocatechuic aldehyde, salvianolic acid B, cryptotanshinone, tanshinone I and tanshinone IIA were purchased from National Institute for The Control of Pharmaceutical and Biological Products (Beijing, China). Salvianolic acid B was bought from Mansite (Chengdu, China). Acetonitrile was supplied # The Author [2013]. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oup.com

Figure 1. Chemical structures of the six components. Table I Studied Samples and Content (%) of Six Active Components in Salvia Names Location Remark Potocatechuic aldehyde by Fisher (USA). All the other reagents used in this research were of analytical grade. Deionized water was purified in a RM-220 water purification system (Water Purifier Corporation, China). Apparatus and chromatographic conditions All analyses were performed on a Shimadzu LC-20A HPLC system (Shimadzu Technologies, Japan) equipped with a dual-pump solvent delivery system, an online degasser, an auto-sampler, a column temperature controller and an ultraviolet detector. Salvianolic acid A Salvianolic acid B Cryptotanshinone Tanshinone I S. miltiorrhiza Aigan Broad leaf Deyang, Sichuan Culture 0.168 0.553 4.145 0.120 0.009 0.013 S. miltiorrhiza Anhui White leaf Deyang, Sichuan Culture 0.159 0.712 2.005 0.025 0.005 0.015 S. miltiorrhiza Gaogan Broad leaf Deyang, Sichuan Culture 0.153 0.562 3.096 0.047 0.012 0.076 S. miltiorrhiza Sibeiti Lobular leaf Deyang, Sichuan Culture 0.347 0.623 3.749 0.021 0.005 0.009 S. miltiorrhiza Zupei Broad leaf Deyang, Sichuan Culture 0.163 0.120 3.872 0.044 0.006 0.013 S. miltiorrhiza Zupei Lobular leaf Deyang, Sichuan Culture 0.164 0.266 3.144 0.025 0.008 0.020 S. miltiorrhiza Beijing Beijing Culture 0.215 0.019 3.092 0.077 0.002 0.016 S. miltiorrhiza Zhenjiang, Jiangsu Zhenjiang, Jiangsu Culture 0.243 0.143 0.858 0.208 0.037 0.063 S. miltiorrhiza Xi an, Shaanxi Xi an, Shaanxi Culture 0.230 0.150 3.203 0.051 0.027 0.106 S. miltiorrhiza Shaoxing, Zhejiang Shaoxing, Zhejiang Culture 0.066 0.062 3.119 0.089 0.062 0.126 S. miltiorrhiza Bozhou Anhui Bozhou, Anhui Culture 0.123 0.098 4.216 0.077 0.034 0.085 S. miltiorrhiza Nanyang, Henan (C) Nanyang, Henan(C) Culture 0.120 0.066 2.284 0.004 0.002 0.006 S. miltiorrhiza Deyang, Sichuan Deyang, Sichuan Wild 0.102 0.081 2.432 0.001 0.001 0.003 S. miltiorrhiza Nanyang, Henan Nanyang, Wild 0.159 0.191 3.593 0.019 0.012 0.033 (W) Henan(W) S. miltiorrhiza Huanggang, Hubei Huanggang, Hubei Wild 0.301 0.065 2.594 0.322 0.039 0.046 S. brevilabra Ganzi, Sichuan Ganzi, Sichuan Wild 0.005 0.045 0.027 0.050 0.050 0.064 S. castanea Ganzi,Sichuan Ganzi, Sichuan Wild 0.014 0.003 0.073 0.075 0.035 0.115 S. digitaloides Liangshan,Sichuan Liangshan, Sichuan Wild 0.038 0.016 0.143 0.013 0.001 0.018 S. maximowicziana Ya an,sichuan Ya an, Sichuan Wild 0.016 0.003 0.712 S. paohsingensis Ya an, Sichuan Ya an, Sichuan Wild 0.020 0.005 0.139 0.015 0.025 0.036 S. przewalskii Ganzi,Sichuan Ganzi, Sichuan Wild 0.069 0.170 4.865 0.037 0.024 0.019 S. trijuga Liangshan,Sichuan Liangshan, Sichuan Wild 0.017 0.004 0.126 0.149 0.125 0.270 S. yunnanensis Kunming,Yunnan Kunming, Yunnan Wild 0.003 0.032 0.101 0.035 0.021 0.055 means that the composition was undetected. Tanshinone IIA Chromatographic separation was conducted on an Agilent XDB C 18 reversed-phase column (5 mm, 250 4.6 mm ID, Agilent, USA) coupled with a Phenomenex C 18 guard column (5 mm, 4.0 3.0 mm ID, Phenomenex, USA). The detection was performed at 280 nm, the flow rate was 0.8 ml/min and the column temperature was maintained at 208C. The mobile phase was in gradient elution, which consisted of solvents A (0.1% aqueous phosphoric acid, v/v) and B (acetonitrile). The gradient program was optimized and conducted as follows: 0 20 min, 2 17% B; 20 60 min, 17 68% B; 60 70 min, 68 74% B; 70 88 min, 74 30% B; A Simple and Reliable Method for Distinguishing Danshen in Salvia 993

and 88 100 min, 30 2% B. An aliquot of 10 ml of the filtrate was injected into the HPLC for analysis. The ultraviolet spectra of standard substances of potocatechuic aldehyde, salvianolic acid A, salvianolic acid B, cryptotanshinone, tanshinone I and tanshinone IIA were recorded from 190 to 400 nm (Shimadzu UV-1750, Shimadzu Technologies, Japan). Preparation of standard solutions A primary stock standard stock solution of the six compounds was dissolved in methanol and stored at 48C. Standard stock solutions of the six reference standards ( potocatechuic aldehyde, salvianolic acid A, salvianolic acid B, cryptotanshinone, tanshinone I and tanshinone IIA) were prepared by dissolving them in methanol. They were then diluted to six concentrations for construction of calibration plots in the ranges of 16.31 652.8, 24.84 993.6, 154.46 6178, 4.160 166.4, 5.760 230.4 and 16.96 678.4 mg/ml, respectively. Further dilution with the lowest concentrations in the calibration curves were carried out to provide a series of standard solutions for evaluating the limits of detection (LOD) of the compounds. The stock and working solutions were stored at 48C. Sample preparation All samples were milled and over-dried at 508C until a constant weight was achieved. Approximately 0.500 g of each dried sample was accurately weighed and extracted with deionized water in a Soxhlet extractor. Extraction was continued for 4 h and was followed by ultrasound extraction for 25 min with 80% ethanol. The mixed extract solution was concentrated and metered volume to 50 ml with deionized water. For HPLC analysis, a moiety of solution was filtered through a 0.45-mm membrane filter, and 10 ml of the filtrate was injected into the HPLC system. Method validation Linearity Under optimal chromatographic conditions, each calibration graph was obtained by injecting a standard solution with six different concentrations, and this was performed in triplicate. The calibration graphs for each compound were plotted based on the linear regression analysis of the integrated peak areas (Y) versus concentrations (X) of the standard. The regression equation was calculated in the form Y ¼ ax þ b, where Y and X are the values of the peak area and sample amount, respectively. The standard solution was diluted with methanol to provide appropriate concentrations. The LOD was determined at a signal-to-noise ratio of 3. Precision, stability and recovery S. miltiorrhiza Aigan Broad leaf was selected as a sample to validate the method. The intraday and interday variations for determining the precision of the developed method were evaluated based on the results of six replicate analyses in a single day and duplicating the experiments over a period of 3 days. Variations of the peak area were taken as the measures of precision and were expressed as percent relative standard deviations (RSD). Reproducibility was confirmed by preparing six independent analytical sample solutions prepared from the same sample batch. Variations were expressed as the RSD of the data set. Sample stability test was determined with one sample at 0, 2, 4, 6, 8, 12 and 24 h. The solution was stored at ordinary temperature. The stability of the method was expressed as the RSD of the data set. To determine the accuracy of the developed method, the recovery experiments were performed by adding three varying quantities of authentic standards to the samples as well as extracting and quantifying them according to the established procedure. Statistical analysis Cluster analysis with Euclidean Distance and UPGMA with SPSS 19.0 was conducted based on the quantitative distribution of the six major active compounds. Results Selection of HPLC condition It is an important criterion for a high efficiency HPLC condition that the mark peaks have greatly baseline separation with adjacent peaks within a short analysis time as far as possible (15). In this study, different HPLC parameters were examined and compared, including the following: detection wavelengths of 268, 280 and 288 nm; mobile phases consisting of acetonitrile water system and methanol water system; mobile phase flow rates of 1.2, 1.0 and 0.8 ml/min and column temperatures of 30, 25 and 208C. The binary mixtures of the acetonitrile water system exhibited a more efficient separation ability for the detected compounds than did the methanol water system. The 0.1% (v/v) aqueous phosphoric acid was added to improve the peak shape. As indicated in Figures 2 and 3, phenolic and tanshinone compounds have different chemical structures, properties and absorption spectra, and are difficult to baseline resolve under isocratic elution. Thus, gradient elution was used in HPLC analysis. The ultimately selected mobile phase system consisted of acetonitrile 0.1% aqueous phosphoric acid, which provides lower pressure and greater baseline stability. Under these experimental conditions, the retention times of the six components were 100 min, and the separation was satisfactory. The most suitable flow rate was 0.8 ml/min. The results also suggested that separation was more efficient when the column temperature was maintained at 208C, and 280 nm was selected as the chromatogram detection wavelength. Under these conditions, all marker compounds could be detected and showed adequate absorption. Method validation The method was validated by the linearity, precision, stability and reproducibility of the results. Regression equations were derived from the external standard method. The calculated results are listed in Table II. R 2 in Table II refers to the correlation coefficient of the equation. All the standard compounds 994 Wang et al.

Figure 2. (A) The UV spectra of potocatechuic aldehyde (Pa), salvianolic acid A (Sa), salvianolic acid B (Sb), cryptotanshinone (Ct), tanshinone I (T1) and tanshinone IIA (T2) recorded between 190 and 400 nm. (B) Representative HPLC chromatograms of the six components in standard solution (a) and the sample solution: (b) S. miltiorrhiza Aigan Broad leaf, (c) S. miltiorrhiza Anhui White leaf, (d) S. miltiorrhiza Gaogan Broad leaf, (e) S. miltiorrhiza Sibeiti Lobular leaf, (f) S. miltiorrhiza Zupei Broad leaf and (g) S. miltiorrhiza Zupei Lobular leaf. showed good linearity in the relatively wide concentration range. The LOD ranged from 0.019 to 0.850 mg/ml. The RSD was taken as a measure of precision. The data showed that the RSD of intraday and interday variabilities ranged from 0.30 to 1.3% and 0.14 to 2.0%, respectively. For stability test, the same sample solution was analyzed for 24 h at room temperature, and the results indicated that potocatechuic aldehyde, salvianolic acid A, salvianolic acid B, cryptotanshinone, tanshinone I and tanshinone IIA remained stable (RSD 2.6%, Table II). The average recovery of the six components was 96.49 102.16%, whereas that of the six standards ranged from 0.27 2.9% with an RSD of,3% (Table III). The results of these tests indicated that the established method was accurate for the determination of the six chemical compounds in the Salvia samples. Quantitative analysis of samples The established method has been applied for the simultaneous determination of six major active compounds in the roots of 23 A Simple and Reliable Method for Distinguishing Danshen in Salvia 995

Figure 3. Clustering tree of Salvia based on six major active composition distributions and the UPGMA clustering method. The dotted black line is a dividing line between non-danshen and Danshen. Table II HPLC and Validation Data for the Calibration Graphs and Limit of Quantification of the Six Major Components Compound Standard curve R 2 Linear range (mg/ml) LOD (mg/ml) Precision RSD (%, n ¼ 6) Stability RSD (%, 24 h, n ¼ 6) Intraday Interday Potocatechuic aldehyde Y ¼ 1.6648E2007 1.0000 16.32 652.8 0.039 0.91 1.4 1.5 X 2 3.3490E2004 Salvianolic acid A Y ¼ 3.1896E2007 1.0000 24.84 993.6 0.230 1.3 1.5 2.6 X þ 1.4895E2003 Salvianolic acid B Y ¼ 5.6857E2007 0.9998 154.46 6178 0.850 0.83 2.0 0.08 X þ 4.8385E2002 Cryptotanshinone Y ¼ 3.0579E2007 1.0000 4.160 166.4 0.074 0.30 0.14 0.28 X 2 9.0950E2004 Tanshinone I Y ¼ 2.1717E2008 1.0000 5.760 230.4 0.019 0.86 1.2 0.34 X þ 1.9743E2004 Tanshinone IIA Y ¼ 1.9564E2007 X þ 1.1306E2004 1.0000 16.96 678.4 0.075 0.34 0.82 2.1 Salvia samples. Each sample was analyzed in triplicate to determine the mean content (%).The representative HPLC profiles are shown in Figure 2 and the results are summarized in Table I. Table I indicates a significant difference in the contents of the six active major components in 15 samples of Danshen. The contents (%) of potocatechuic aldehyde, salvianolic acid A, salvianolic 996 Wang et al.

Table III Recovery of the Six Major Components Compound Potocatechuic aldehyde Sample contents (mg/g) Added (mg/g) Reproducibility (%) Average reproducibility (%) 1.68 0.85 96.55 96.49 2.5 1.25 95.67 2.00 97.26 RSD (n ¼ 6) Salvianolic acid A 5.53 2.50 97.66 97.82 0.27 4.15 98.12 6.52 97.68 Salvianolic acid B 41.45 20.42 96.74 99.84 2.9 31.09 100.32 49.73 102.45 Cryptotanshinone 1.22 0.60 98.75 99.57 0.8 0.96 99.69 1.45 100.27 Tanshinone I 0.09 0.05 100.67 99.44 1.2 0.07 99.45 0.11 98.21 Tanshinone IIA 0.13 0.06 103.56 102.16 1.4 0.11 102.14 0.16 100.77 acid B, cryptotanshinone, tanshinone I and tanshinone IIA were in the following ranges: 0.066 0.347%, 0.019 0.712%, 0.858 4.216%, 0.001 0.322%, 0.001 0.062% and 0.003 0.126%, respectively. This observation coincides with the results of the study by Cheng et al. (15). S. brevilabra, S. castanea, S. digitaloides, S. maximowicziana, S. przewalskii, S. paohsingensis, S. trijuga and S. yunnanensis were reported for the first time on the simultaneous determination of six major active compounds. The contents (%) of potocatechuic aldehyde, salvianolic acid A, salvianolic acid B, cryptotanshinone, tanshinone I and tanshinone IIA were in the following ranges: 0.003 0.069%, 0.003 0.170%, 0.027 4.865%, not detected to 0.149%, not detected to 0.125% and not detected to 0.270%, respectively. This study confirmed that the six major active components could be used as markers for distinguishing Danshen. Cluster similarities based on the distributions of six major active compositions categorized the analyzed species into two groups: non-danshen and Danshen (Figure 3). S. miltiorrhiza (15 samples) and S. przewalskii clustered together as a result of similarly high levels of concentration of the six major active components. Discussion In this study, an HPLC method was developed for the simultaneous determination of six major active components of Danshen and specific Salvia herbs. The method was validated by its linearity, precision, accuracy and reproducibility. By studying all the sample of Salvia, they were able to determine that the variation in contents may be due to the different geographic environment, variations cultivation model, hereditary character and other factors. S. przewalskii had the highest phenolic compound count among all Salvia herbs in this study. S. castanea and S. trijuga contained high concentrations of tanshinone compounds. The presence of tanshinone compounds in S. maximowicziana was not indicated. Phenolic acids in S. przewalskii and tanshinones in S. castanea and S. trijuga were abundant, but S. maximowicziana and S. digitaloides were not suitable as Danshen substitutes because of relatively low medicinal composition. Chemical identification of TCM materials is an important and useful means. Studies of identification of TCM materials have focused on HPLC fingerprinting with combined statistical analysis in previous studies (16 19). This research results suggest that six major active compositions could be used as markers for distinguishing Danshen and non-danshen by simultaneous determination of active ingredient with cluster analysis. In addition, the result favors S. przewalskii as a substitute for Danshen. But S. castanea, S. digitaloides, S. maximowicziana, S. trijuga and S. yunnanensis may not be appropriate as substitutes for Danshen. Funding This work was financially supported by the Programs of Agriculture Science Technology Achievement Transformation Fund, China (2012GB2F000385), Innovation Fund for Technology Based on Firms, China (12C26215105863), the Science and Technology Department of Sichuan Province, China (2012NZ0020, 2011ZO0006, 2012ZR0050), and Innovation Fund for Technology Based on Firms, Chengdu (11CXYB958JH-003). References 1. Li, X.W., Hedge, I.C.; Flora of China. Vol. (17). Science Press, Beijing, China, and Missouri Botanical Garden Press, St. Louis, UK, (1994), pp. 196 223. 2. Liu, J., Shen, H.M., Ong, C.N.; Salvia miltiorrhiza inhibits cell growth and induces apoptosis in human hepatoma HepG2 cells; Cancer Letters, (2000); 153: 85 93. 3. Guo, B.L., Feng, Y.X., Zhao, Y.J.; Review of germplasm resources on Salvia miltiorrhiza; China Journal Chinese Materia Medica, (2002); 7: 492 495. 4. Su, L., Liu, Q.D.; Recent researches on Pharmacological Actions of TanshinonejjA; Chinese Journal of Pharmaceutical Technology Economics & Management, (2009); 3: 39 45. 5. Pharmacopoeia Commission of Peoples Republic of China. Pharmacopoeia of the People Republic of China. Chemical Industry Press, Beijing, China, (2010), pp. 70. 6. Zhang, J.L., Cui, M., He, Y., Yu, H.L., Guo, D.A.; Chemical fingerprint and metabolic fingerprint analysis of Danshen injection by HPLC-UV and HPLC-MS methods; Journal of Pharmaceutical and Biomedical Analysis, (2005); 36: 1029 1035. 7. Ma, L.J., Zhang, X.Z., Zhang, H.P., Gan, Y.R.; Development of a fingerprint of Salvia miltiorrhiza Bunge by high-performance liquid chromatography with a coulometric electrode array system; Journal of Chromatography B, (2007); 846: 139 146. 8. Luan, X.L., Chen, J., Zhou, J., Zhang, H.; HPLC simultaneous determination of three components in Danxiangguanxin injection; Chinese Journal of Pharmaceutical Analysis, (2011); 31: 1414 1416. 9. Ma, Z.G., Yang, Y.J., Xie, Y., Feng, X.Y.; Pharmacognostic studies on the roots of genus Salvia from Gansu province; Journal of Lanzhou Medical College, (1993); 19: 212 214. 10. Xiao, X.H., Fang, Q.M., Xia, W.J., Yin, G.P., Su, Z.W., Qiao, C.Z.; Numerical taxonomy of medicinal Salvia L. and the genuineness of Danshen; Journal of Plant Resources and Environment, (1997); 6: 17 21. A Simple and Reliable Method for Distinguishing Danshen in Salvia 997

11. Qian, Z.G., Liang, X.Y., Hou, A.G., Ruan, Z.G.; Preliminary study on the resources of Salvia przewalskii; The Chinese Journal of Modern Applied Pharmacy, (2003); 20: 174 176. 12. Liu, A.H., Guo, H., Ye, M., Lin, Y.H., Sun, J.H., Xu, M., et al.; Detection, characterization and identification of phenolic acids in Danshen using high-performance liquid chromatography with diode array detection and electrospray ionization mass spectrometry; Journal of Chromatography A, (2007); 1161: 170 182. 13. Sun, D., Liu, W.Y., Liang, C.; Simultaneous quantitative determination of four active components in Salvia miltiorrhiza tablets by HPLC; Journal of China Pharmceutical University, (2007); 38: 51 54. 14. Lou, Z.Y., Zhang, H., Li, X., Zhu, Z.Y., Zhao, L., Zhang, G.Q., et al.; HPLC simultaneously determines multiple hydrophilic and lipophilic bioactive components in Radix et Rhizoma Salviae miltiorrhiza; Academic Journal of Second Miltary Medical University, (2009); 30: 190 193. 15. Cheng, H.T., Li, X.L., Li, X.R., Li, Y.H., Wang, L.J., Xue, M.; Simultaneous quantification of selected compounds from Salvia herbs by HPLC method and their application; Food Chemistry, (2012); 130: 1031 1035. 16. Liu, X.Q., Du, L.L., Li, W.W., Guan, H.F., Liu, F.; Simultaneous qualitative and quantitative analysis of commercial Bistorta rhizome and its differentiation from closely related herbs using TLC and HPLC-DAD fingerprinting; Chemical & Pharmaceutical Bulletin, (2008); 56: 75 78. 17. Zhao, Y.Y., Zhang, Y.M., Lin, R.C., Sun, W.J.; An expeditious HPLC method to distinguish Aconitum kusnezoffii from related species; Fitoterapia, (2009); 80: 333 338. 18. Lee, D.Y., Cho, J.G., Lee, M.K., Lee, J.W., Lee, Y.H., Yang, D.C., et al.; Discrimination of Panax ginseng roots cultivated in different areas in Korea using HPLC-ELSD and principal component analysis; Journal of Ginseng Research, (2011); 35: 31 38. 19. Wang, L.B., Wang, X.B., Kong, L.Y.; Automatic authentication and distinction of Epimedium koreanum and Epimedium wushanense with HPLC fingerprint analysis assisted by pattern recognition techniques; Biochemical Systematics and Ecology, (2012); 40: 138 145. 998 Wang et al.