Fast quantified counter chemicals of Dan Shen by fingerprints and correlated with antioxidating profiles

Yujing Zhang, Guoxiang Sun＊

School of Pharmacy, Shenyang Pharmaceutical University, Shenyang 110016, China

Fast quantified counter chemicals of Dan Shen by fingerprints and correlated with antioxidating profiles

Yujing Zhang, Guoxiang Sun＊

School of Pharmacy, Shenyang Pharmaceutical University, Shenyang 110016, China

A combination method of fingerprint analysis and antioxidant analysis was developed and validated in order to evaluate the quality consistency of traditional Chinese medicine (TCM) / Dan Shen (DS). In the fingerprint analysis nine major chromatographic peaks were detected in DS sample, which were identified as: Protocatechuic acid (PHA), Protocatechualdehyde (PHD), Rosmarinic acid (RMA), Caffeic acid (CFA), Tanshinone IIA (T2A), Salvianolic acid B (SAB), Lithospermic acid (LSA), Propanoic acid (PPA), Salvianolic acid A (SAA). In the antioxidant analysis, IC50and antioxidant capacity of the individual markers was successfully measured and predicted coupled with partial least squares regression (PLSR). This study indicated that the antioxidant activity, and the associated fingerprint analysis based on systematically quantified fingerprint method (SQFM) could assess the quality of TCM efficiently and credibly.

Dan Shen; HPLC fingerprint; antioxidant activity; quality control; SQFM

1 Introduction

Dan Shen (DS), has been used for thousands of years in China and now is practised all over the world to cure many diseases, especially treating coronary heart diseases, cerebrovascular disease, bone loss, hepatitis, hepatocirrhosis and chronic renal failure, etc. [1, 2]. There are many components in DS and some of them have amazing antioxidant capacity, and according to the research of Geng Li, etc. [3], the therapeutic effects of DS are based on synergistic effects of many kinds of antioxidant active ingredients. Therefore, it is insufficient to select single or multiple chemical markers or bioactive constituents to assess the quality of complex TCM [4, 5].

Analysis of traditional Chinese medicines (TCMs) plays important roles in quality control of TCMs and understanding their pharmacological effects [6, 7] and HPLC fingerprint has been widely accepted as an important means for the quality control of complex analytes [8-10]. Meanwhile, PLSR is applied to establish a relationship between fingerprint and bioactivity, and to provide the contribution information of each constituent of the spectrum-effect relationship [11-13].

The purpose of this study was to separate the chemical compositions of DS by HPLC-DPPH method, which is capable of assessing the quality consistency of DS by SQFM. In addition, spectrum-

?

3 Materials and Methods

3.1 Chemicals and Reagents

A total of 30 batches of DS samples were from Tasly (Tianjin) named S1～S30, whose origin was Henan Province, China. RMA (111871-201102), PHA (809-200102), PHD (110810-200205), CFA (110885-200102), Tanshinone IIA (T2A, 110766-200518), LSA, SAA, SAB, PPA were obtained from National Institute for Food and Drug Control (NIFDC, Beijing, China). All the nine standard reference compounds have over 99% purity, and their chemical structures were shown in Fig. 1. HPLC grade acetonitrile and methanol were purchased from Yuwang Chemical Industry Co., Ltd (China, Shandong). Deionized water and other reagents were of analytical grade used throughout the experiment.

Fig. 1 The structure of nine marker compounds

3.2 Instruments

HPLC analyses were performed with an Agilent 1100 liquid chromatography instrument, equipped with a diode array detector (DAD), a low pressure mix quaternary pump, online degasser, auto sampler and Chem Station workstation (Agilent Technologies, Ltd.); 722S visible spectrophotometer (quartz color plate 1 cm × 1 cm, Shanghai precision instrument co., Ltd.); KQ-50B ultrasonic cleaners (kunshan ultrasonic instruments co., Ltd.); KDM temperature control electric jacket (Shandong zhencheng China instrument company); Sarturius -BS110S analytical balance (Beijing dolly scales co., Ltd.).

3.3 Preparation of sample solutions and reference solutions

Dried herbs DS samples were ground to a particle size up to standard. 2.50 g of powder accurately weighed was extracted with 25 mL of pure methanol (100%) twice, heated to reflux for 40 min and 30 min, respectively. Hereafter, the extracts of two times were mixed into 50 mL volumetric flasks. Standard solution of 9 standards PHA, PHD, RMA, CFA, T2A, SAB, LSA, SAA, and SAS (It could be converted to Propanoic acid to the reaction system, so it could be also replaced by PPA) were accurately weighed and then dissolved in methanol, respectively. Fresh DPPH stock solutionswere prepared at 88.0 µg/mL and 48.0 µg/mL by dissolving in methanol, sealed and stored in the refrigerator at 4 °C, and diluted to the appropriate concentration before using in online analysis and offline analysis. All the solution should be filtered through 0.45 µm Millipore filters before injected into the HPLC and both of them stored in the dark at 4 °C.

3.4 HPLC-DAD fingerprint testing conditions

The separation system carried out on an Arcus EP-C18 column (250 mm×4.6 mm, 5.0 µm) from Exformma Technologies (Shanghai, China), maintained at 35 °C. The mobile phase was consisted of 5 mmol/L citric acid and 10 mmol/L sodium phosphate monobasic dehydrate in water (B) and 1% acetic acid in Acetontrile (C). The gradient elution sequence for the HPLC fingerprints was used as follows: 0～10 min, linear gradient 10% C; 10～16 min, linear gradient 10%～13% C; 16～25 min, 13%～20% C; 25～30 min, 20%～22% C; 30～40 min, 22%～25% C; 40～60 min, 25%～60% C; 60～75 min, 60%～100% C; 75～85 min, 100% C at a flow rate of 0.8 mL /min. The mobile phase of antioxidant system was 88.0 µg/mL DPPH stock solutions at a flow rate of 0.2 mL/min. The total flow rate was 1 mL/min. The injection volume of samples and standard solutions was all 10 µL. The UV absorbance was monitored at 290 nm for fingerprints and 517 for online antioxidant determination. The mobile phase constituents were degassed and filtered through 0.45 µm Millipore filters prior to HPLC analysis.

3.5 Offline analysis of HPLC-DAD

The principle of this method: DPPH is a relatively stable with lipid free radical, which has a free electron on the nitrogen atoms and when it is dissolved in methyl alcohol, the solution displays as purple and the maximum absorption peaks display at 517 nm. After joining the anti-oxidants, DPPH captures an electron to the free electron pair. After a period of time, purple fades and becomes colorless substance, with the result from the disappearance of the absorption at 517 nm [15].

0.1 mL of methanol extracts of DS was set to 10 mL volumetric flask, methanol added to volume and shook. 0.2, 0.4, 0.6, 0.8, 1.0 mL of the solution was drawn precision respectively, then added 1.8, 1.6, 1.4, 1.2, 1.0 mL of methanol solution corresponding. Finally, 2 mL DPPH solutions were added in those brown volumetric flasks. These solutions were mixed and allowed to react for 40 min at 25 °C. The solutions were transferred to a 1 cm colorimeter cell, and the absorbance was measured at 517 nm. Radical scavenging capacity equation was as follows:

Free radical scavenging capacity (%) = (1-Ai/Ac)×100%

WhereAiis the absorbance of the sample (sample or Vc, methanol and DPPH), andAcis the absorbance of the negative control (methanol and DPPH). The clearance rate (y) of the drug concentration (x) was mapped in a coordinate. The linear equation was established based on it. According to this equation, drug concentration when the clearance rate reached 50% (IC50) was calculated. When Vc was used as positive control, the antioxidant activity of 100 mg/mL DS was equal to 3.62 mg/mL Vc.

3.6 Statistical data analysis

In this experiment, the HPLC data were evaluated by using laboratory-developed“Digital Evaluation System for super-information Characteristics of TCM Chromatographic Fingerprint 4.0” (software, Certificate no.0407573, China) and the biological activities were analyzed by using SIMCA 13.0 software.

4 Results and discussion

4.1 Method validation

The calibration curves, including its regression equations, correlation coefficients (r), linear ranges, LODs and LOQs of the HPLC method were listed in Table 2. All analytes showed good linear relationship (r＞0.99). The relative standard deviation (RSD) values of the relative retention time, and the relative peak area obtained the precision, reproducibility and stability tests were all less than 1.0% and 3.0%, respectively. LOD and LOQ were in the range of 0.01～0.15 µg/mg and 0.05～1.5 µg/mg, respectively. In summary, the test results were sufficiently accurate and reliable.

Table 2 Calibration curves, correlation coefficients (r), linear ranges, LOD and LOQ for the test compounds

4.2 Simultaneous quantitative analysis of nine ingredients of DS

The contents of the nine compounds were calculated by the standard curve method, and the determination was performed in triplicate for each DS sample. The results in Table 3 indicate that the contents of the nine components were significantly variations between 30 samples. Especially, the content of PHD was in the range from 0.0014 to 0.0250 mg/g, and the RSD was higher than 50%. The possible reason is that the harvest seasons, and manufacturing processes are more likely to play a significant role in making it difficult to maintain quality consistency of DS.

Table 3 Contents of 9 compounds in 30 DS samples

Continued Table 3

4.3 Establishment and evaluation of HPLC fingerprints

There are 64 ‘common peaks' existed in all 30 batches of samples in the fingerprint at 290 nm. The 30 batches of sample fingerprints and the nine marker compounds were shown in Fig. 2. To evaluate the quality of DS, the values ofSm,Pm,αand the final quality grades calculated according to SQFM and the evaluating results were summarized in Table 4, which demonstrated that the quality consistency varied notably among DS samples. For example, S3, S4, S16, S19 and S25 were assigned the good quality of grade 3, while S30 fell into the inferior quality of grade 7. All the DS samples have theSmvalues above 0.95 and αvalues below 0.20, which indicated that all samples have the similar chemicals' composition. The acceptableSmvalue should be no less than 0.90,αno more than 0.20. ThePmvalue was a major index to reflect the contentof the sample, which associated with the medicinal effects in clinics. The acceptablePmvalue fluctuated within the range of 75%～125%. The higher value ofPmthe higher overall component contents, for example, S10 and S30; on the contrary, the lower value ofPmthe lower overall component contents (for example, S18). In general, samples within grade 4 were recommended as qualified ones. Accordingly, a total of 3 batches of DS samples failed to qualify because of either too low contents (Pm＜75%) for S18 or too high contents for S10 and S30 (Pm＞125%). In this content, no samples were found to be unqualified due to the large fluctuations of values depending onαwhen compared with other samples.

Table 4 The quality evaluation results of 30 DS samples based on SQFM

Fig. 2 The fingerprints of DS samples and 9 marker compounds

4.4 Correlation analysis of antioxidant capacity and HPLC fingerprints

During DS contain a lot of antioxidants, a PLSR model was established to predict their antioxidant activity, where peak area of 30 batches of samples was mean centered and were treated as response variableX[16]. Meanwhile, the lower IC50value meant the higher antioxidant activity. The correlation coefficient was shown in Fig. 3. It was evident that many compounds such as peaks 1 (PPA), 3 (PHA), 6 (CFA), 18 (RMA), 19 (LSA), 20 (SAB), 25 (SAA), 59 (T2A); a total of 43 peaks are positively correlated with 1/IC50 at 64 peaks. Antioxidant activity of the active substance was much greater than the activity of antioxidant activity antagonistic substances in DS, showing a strong antioxidant activity overall.

Fig. 3 The histogram of the coefficients for the fingerprint peaks

The derived models were validated by the different procedures: First, the omitted data were divided to two groups after removing outliers (S26, S30), which were the calibration set and the prediction set. Second, the calibration set was used for internal cross validation, and Observed vs Predicted Regression Line was established after the three main components were extracted whenevaluated. Three latent variables to a calibration model were selected, which were achieving an explained variance (R2) of 96.2% forYvariables (1/ IC50), a predictive ability (Q2) of 87.5%, and a root mean square error of estimation value of 0.0004353, shown in Fig. 4. Third, the prediction set consisting of the remaining 6 samples was employed for external validation, which was shown in Table 5, proving no significant difference between observed and predicted values. Therefore, the performance of PLSR model in terms of predictive medicine activity was indeed excellent.

Fig. 4 The linear regression plots of the measured vs predicted value of 1/ IC50used in the calibration model

Table 5 Comparisons of experimental and predicted values of antioxidant activity for both the calibration and prediction models

5 Conclusions

In this study, the HPLC fingerprints were established and antioxidant analysis to evaluate the quality consistency of DS, respectively. In addition, we established a simple and rapid method for identifying a biologically active compound content and biological activity evaluation for the first time, which provided a new method and a comprehensive evaluation for modern medicine and could be potentially applied to DS practical production due to its simplicity and economic advantage.

[1] Li X, Wu LH, Fan XH, et al. Network pharmacology study on major active compounds of Fufang Danshen formula. China Journal of Chinese Materia Medica, 2011, 36: 2911-2915.

[2] Chen L, Lu Y, Zheng SZ. Integration effect on pharmacological component in Salvia milt iorrhiza. Chinese Traditional and Herbal Drugs, 2009, 40: 476-479.

[3] Li G, Meng FY, Yang HJ, et al. Simultaneous Determination of 11 Components in Salvia miltiorrhiza by UPLC. China Pharmacy, 2014, 25:1766-1768.

[4] Qi J, Yu BY. A New Methodology for the Quality Evaluation of Traditional Chinese Medicine -Integrated Spectrum-Effect Fingerprint Research. Chinese Journal of Natural Medicines, 2010, 8: 171-176.

[5] Du GH, Wang YH, Zhang R, et al. Multi-component and multi-target pattern, a surface understanding of the action mechanism of Traditional Chinese Medicine. World Science and Technology/Modernization of Traditional Chinese and Medicine and Materia Medica, 2009, 11: 480-484.

[6] Wong MYM, So PK, Yao ZP, et al. Direct Analysis of Traditional Chinese Medicines by Mass Spectrometry. J Chromatogr B, 2015.

[7] Feng CX. Chemical constituents and pharmacological effects profile of Salvia. Chinese journal of ethnomedicine and ethnopharmacy, 2011, 25-26.

[8] Chen L, Du J, Dai Q, et al. Prediction of anti-tumor chemical probes of a traditional Chinese medicine formula by HPLC fingerprinting combined with molecular docking. Eur J Med Chem, 2014, 83: 294-306.

[9] Wang J, Tong X, Li P, et al. Bioactive components on immuno-enhancement effects in the traditional Chinese medicine Shenqi Fuzheng Injection based on relevance analysis between chemical HPLC fingerprints and in vivo biological effects. J Ethnopharmacol, 2014, 155: 405-415.

[10] Liu M, Xia XH, Zhang ZM, et al. A Comparative Study of antioxidant activity of Danshensu, protocatechuic aldehyde, caffeic acid and Sal B in vitro. Journal of Chinese Medicinal Materials, 2009, 32: 265-267.

[11] Li QH, Li L, Huang LL, et al. Research Progress on Antioxidant Activities and Evaluation of Chinese Medicine. Lishizhen Medicine and Materia Medica Research, 2008, 19: 1257-1258.

[12] Qin KM, Zheng LJ, Shen BJ, et al. Application of spectrum-effect relationship in Chinese medicine research and related thinking. China Journal of Chinese Materia Medica, 2013, 38: 26-31.

[13] Zhang J, Cheng YF. Survey about reference choice of Salvia and its preparation by HPLC fingerprints. Journal of Pharmaceutical Research, 2014, 33: 664-673.

[14] Sun GX, Zhang JX. Identification of the authentic quality of Longdanxiegan pill by systematic quantified fingerprint method based on three wavelength fusion chromatogram. Chinese Journal of Chromatography, 2009, 27: 318-322.

[15] Zhang YT, Li Q, Xing H, et al. Evaluation of antioxidant activity of ten compounds in different tea samples by means of an on-line HPLC-DPPH assay. Food Research International, 2013, 53: 847-856.

[16] Wang Y, Sun GX, Liu ZB, et al. Capillary electrophoresis fingerprinting coupled with chemometrics to evaluate the quality consistency and predict the antioxidant activity of Sanhuang tablet as part of its quality control. Journal of Separation Science, 2014, 37: 3571-3578.

* Author to whom correspondence should be addressed. Address: School of Pharmacy, Shenyang Pharmaceutical University, 103 Wenhua Road, Shenyang, Liaoning 110016, China; Tel.: +86-24-23986286; Email: gxswmwys@163.com

Received: 2015-12-21 Accepted: 2016-02-18