Journal of Chromatographic Science Advance Access published June 11, 2014 Journal of Chromatographic Science 2014;1– 6 doi:10.1093/chromsci/bmu052
Article
Simultaneous Determination of Five Constituents in Qinpijiegu Capsule by High-Performance Liquid Chromatography Coupled with Tandem Mass Spectrometry Minmin Zhao, Weijing Ding, Shuang Wang, Meng Gao, Shan Fu, Juan Zhang, Tao Li, Yin Wu and Qiao Wang* Department of Pharmaceutical Analysis, School of Pharmacy, Hebei Medical University, 361 East Zhongshan Road, Shijiazhuang 050017, P.R. China *Author to whom correspondence should be addressed. Email: [email protected] Received 9 November 2013; revised 30 March 2014
A rapid high-performance liquid chromatography coupled with tandem mass spectrometry method was developed for the simultaneous determination of five constituents in Qinpijiegu capsule (QJC), a classical Tibetan prescription. The separation of five compounds such as aesculin, aesculetin, fraxin, peimine and peiminine was performed on a Purospher STAR LP RP-C18 (250 3 4.6 mm, 5 mm) column with linear gradient elution of acetonitrile – 0.3‰ formic acid water in 13 min. Detection was carried out by multiple reaction monitoring mode using electrospray ionization in the positive and negative ion switching mode. The sample was prepared with ultrasound extraction with methanol, which could obtain higher extraction efficiency and shorter extraction time comparing to reflux extraction with alkalized chloroform – methanol. The proposed method was applied to analyze three batches of samples with acceptable linearity (r 2 > 0.9977), precision [relative standard deviation (RSD) < 7.40%], repeatability (RSD < 2.49%), stability [relative error (RE) < 9.15%] and recovery (RSD < 10.76%). This is the first development of a multicomponent quantitation method for the quality control of QJC. Furthermore, the new established method was proven to be highly sensitive and effective in evaluating the quality of QJC.
Introduction Qinpijiegu capsule (QJC) is a classical Tibetan medicine composed of 113.0 g of Cortex fraxini, 80.0 g of Fritillaria cirrhosa, 80.0 g of Pyrethrum tatsienense and 57.0 g of Keel (1). It displays good effects in analgesia and callus growth comparing with Shangkejiegu tablet, a traditional medicine for treatment of bone injury (2), and is widely used for the treatment of bruises, sprains bones and bruising swelling in China. No doubt, the quality control of QJC is essential to ensure its efficacy and safety. QJC is formulated by four kinds of medicinal materials. It is obvious that determining multicomponents in various medicinal materials can more accurately evaluate its quality than only determining a certain component in one medicinal material. According to the composition of prescription, Cortex fraxini accounts for 34.2%, the largest proportion of the total content. It is from the dry barks of Oleaceae plant Fraxinus rhynchophylla Hance, F. rhynchophylla Hance or Fraxinus paxiana (3). Coumarins, including aesculin, aesculetin, fraxin, and so on, are the characteristic constituents (4 – 8), which are proven to be the active constituents, and commonly possess various biological activities such as anti-inflammation, antivirus, antiarthritis and anticancer (9, 10). Fritillaria cirrhosa accounts for 24.2% of the prescription next to C. fraxini. It has been used as antitussive and expectorant herbs for thousands of
years in China (11). Steroidal alkaloids, including peimine and peiminine, have been demonstrated as its major ingredients and reported to have activities such as antihypertensive, anticholinergic, antitumor, antiasthmatic and antitussive (12, 13). Although P. tatsienense, a traditional folk medicine, occupies the equal proportion as F. cirrhosa in the prescription, there are no reports on its constituents study. Keel, originating from the bone of ancient mammals, accounts for only 17.3% in the prescription. It mainly contains calcium salts and other mineral salts that are difficult to be simultaneously analyzed with organic components. In summary, selecting main coumarins and steroidal alkaloids in C. fraxini and F. cirrhosa as the marker components would be practical and valuable for the quality evaluation of QJC. To date, only one report on quality detection method about QJC was published (1). However, in this paper, aesculin and aesculetin in C. Fraxini and imperialine in F. cirrhosa were separately determined by HPLC-UV according to Chinese Pharmacopoeia (ChP, Version 2010). Moreover, the development and validation of method were not described. Therefore, it is essential to establish a simple and rapid method to simultaneously determine multiple active components in QJC for quality control of QJC. In recent years, because of high sensitivity and specificity, the high-performance liquid chromatography coupled with tandem mass spectrometry (HPLC –MS-MS) method has found wider and wider application than other detection systems in the field of traditional Chinese medicine prescription research (14). In this study, the HPLC–MS-MS method was developed for the simultaneous quantification of five main constituents in QJC, including aesculin, aesculetin, fraxin, peimine and peiminine, and then applied to the quality evaluation of QJC. This is the first time to establish an HPLC – MS-MS method for the simultaneous determination of multiconstituent in QJC.
Experimental Chemicals and materials Aesculin (110740 – 200506) and aesculetin (110741 – 200506) were obtained from the National Institutes for Food and Drug Control (Beijing, China). Fraxin (20111214), peimine (120913001) and peiminine (120921001) were purchased from Shanghai Sunny Biotech Co., Ltd. (Shanghai, China). The purities of all standards were .98%. The chemical structures of five components are shown in Figure 1. HPLC grade acetonitrile was obtained from J.T. Baker Company (Center Valley, PA, USA). Formic acid was purchased from Dikma
# The Author 2014. Published by Oxford University Press. All rights reserved. For Permissions, please email: [email protected]
Figure 1. Chemical structures of five components: (A) aesculin, (B) aesculetin, (C) fraxin, (D) peimine and (E) peiminine.
Company (Lake Forest, CA, USA). Purified water was purchased from Hangzhou Wahaha Group Co., Ltd. (Hangzhou, Zhejiang, China). Three batches of QJC samples (bat. 20110104, 20120610 and 20120904) were manufactured by Arura Tibetan Medicine Co., Ltd. (Xining, Qinghai, China).
Apparatus and HPLC–MS-MS conditions An Agilent 1200 liquid chromatography system (Agilent technology, Santa Clara, CA, USA) equipped with an online vacuum degasser, a quaternary solvent delivery system, an autosampler and a column compartment was used for HPLC analysis. Chromatographic separation was achieved on a Purospher STAR LP RP-C18 (250 4.6 mm, 5 mm) column at room temperature. The mobile phase consisted of acetonitrile (A) and 0.3‰ aqueous formic acid (B) with gradient elution of 20 –30% A at 0 – 13 min. The flow rate was 0.9 mL/min, and the injection volume was 10mL for analysis. The 3200 QTRAPTM system (Applied Biosystems, Foster City, CA, USA) with a hybrid triple quadrupole linear ion trap mass spectrometer equipped with an electrospray ion source was applied to quantitative analysis. Mass spectrometer was operated in the positive and negative ion switching multiple reaction monitoring mode. The switching program was set as follows: period 1, 0 –7.2 min, negative mode for the determination of aesculin, aesculetin and fraxin and period 2, 7.2 – 13 min, positive mode for the determination of peimine and peiminine. The condition of MS analysis was designed as follows: the spray voltage, 5,500 and 24,500 V; the turbo spray temperature, 6508C; nebulizer gas (gas 1), 60 psi; heater gas (gas 2), 65 psi; collision gas, medium and the curtain gas (CUR), 25 psi. Data collection and analysis were accomplished with the Analyst 1.5.2 software provided by Applied Biosystems/MDS Sciex.
Sample and standard solution preparation Fifty milligrams of QJC powder were weighed accurately, and settled to a conical flask with the cover, 25 mL of 75% methanol aqueous solution was precisely added. The conical flask 2 Zhao et al.
containing sample and solvent was weighed. Subsequently, the sample was ultrasonically extracted at 358C for 60 min and then was cooled to room temperature. The lost weight caused by volatilization of extraction solution during extraction was complemented with 75% methanol aqueous solution. After centrifugation at 14,000 rpm for 10 min, solution was filtered through 0.45 mm membrane. The filtrate was diluted eight times with 20% of acetonitrile aqueous solution to eliminate the solvent effect, and finally determined by HPLC–MS-MS. Mixture stock solution (9.20 mg/mL for aesculin; 6.00 mg/mL for aesculetin; 5.85 mg/mL for fraxin; 1.19 mg/mL for peimine and 0.98 mg/mL for peiminine) was prepared by weighing precisely and dissolving the reference standards with methanol. A series of standard solutions were obtained by further diluting the mixture stock solution. All solutions were stored at 48C before analysis.
Results Method validation Linearity, limits of detection and limits of quantification Mixture standard solution of five compounds was diluted to a series of concentrations to establish calibration curves (Table I). Each calibration curve consisted of at least five concentration levels and was plotted with the peak areas of the standards versus their concentration. All calibration curves showed good linearity across the calibration range (r 2 . 0.9977). Limits of detection (LOD) and limits of quantification (LOQ) were the concentrations of standard solution when the ratios of the signal to noise were 3 and 10, respectively. The data of calibration curves, r 2, linear ranges, LOD and LOQ are presented in Table I. Precision, repeatability and stability The precision was investigated by measuring the intra- and interday variances. Six independent samples were extracted and analyzed within a day to evaluate the intraday precision, while interday precision was determined by repeating above experiment once a day over 3 consecutive days. The RSDs of
Table I Calibration Curves, r 2, Linear Ranges, LOD and LOQ of Five Compounds Compound
Regression equation
r2
Linear range (ng/mL)
LOD (ng/mL)
LOQ (ng/mL)
Aesculin Aesculetin Fraxin Peimine Peiminine
y ¼ 2.17e6x þ 3.94e2 y ¼ 2.15e6x þ 3.66e4 y ¼ 4.78e4x 2 211 y ¼ 3.34e3x þ 8.26e3 y ¼ 6.05e3x þ 1.26e3
0.9985 0.9983 0.9987 0.9977 0.9981
115.00 75.00 73.13 29.75 24.50
0.06 4.00 0.08 0.07 0.08
0.25 11.0 0.25 0.31 0.31
intra- and interday were within the range of 3.30 – 4.80% and 5.10 –7.40%, respectively. Repeatability of the instrument was tested by six parallel measurement of the mixed standard solution of five compounds (aesculin, 0.92 mg/mL; aesculetin, 0.60 mg/mL; fraxin, 0.59 mg/mL; peimine, 119.00 ng/mL and peiminine, 98.00 ng/mL) under the above-mentioned HPLC –MS-MS conditions. The RSD values of five compounds were from 0.74 to 2.49%, which showed high repeatability of the instrument. In order to investigate the stability of the sample solution, three samples were analyzed in parallel at different time points: 0, 2, 4, 7 and 10 h, respectively. Stability was expressed by the value of RE, which was calculated as the following formula: RE ð%Þ ¼
Detection amount ð2; 4; 7; 10 hÞ amount ð0 hÞ Amount ð0 hÞ 100%
ð1Þ
The RE values of five compounds were all ,9.15%, which revealed good stability of the sample solution stored within 10 h. The data of precision, repeatability and stability are listed in Table II. Accuracy The accuracy of the method was evaluated by recovery. The recovery test was performed by spiking different known amounts (high, middle and low) of mixed standard solution composed of five compounds to the samples, then the samples were extracted and determined under the same conditions with other sample, and triplicate experiments were repeated at each level. The recovery was calculated according to the following formula: Detection amount Original amount Recovery ð% Þ ¼ Addition 100%
ð2Þ
As shown in Table III, all of the recoveries of five compounds were in the range of 89.69 – 121.40%, which demonstrated that the method was of good accuracy.
Sample analysis The established analytical method was subsequently applied to determine five compounds in three batches of QJC. The contents of five compounds in three batches of QJC are summarized in Table IV, and the typical chromatography of standards and samples is given in Figure 2. As shown in Table IV, aesculin, aesculetin and fraxin were the dominant compounds. Among them, the content of aesculin was
– – – – –
1840.00 1200.00 5850.00 1190.00 980.00
Table II The Results of Precision, Repeatability and Stability Compound
Precision (RSD, n ¼ 6) Intraday
Interday
Aesculin Aesculetin Fraxin Peimine Peiminine
4.10 4.10 3.30 4.00 4.80
7.40 6.20 7.00 6.70 5.10
Repeatability (RSD, n ¼ 6)
Stability (RE, n ¼ 3)
0.7 2.5 1.8 0.8 1.5
,8.9 ,9.1 ,8.1 ,8.6 ,7.6
Table III The Recoveries of Five Compounds with n ¼ 3.7 Compound
Original (mg)
Addition (mg)
Detection (mg)
Recovery (%)
RSD (%)
Aesculin
206.75
Aesculetin
138.00
Fraxin
137.00
122.50 245.00 367.50 60.00 120.00 180.00 52.00 104.00 156.00 2.98 5.95 11.90 0.98 1.96 4.90
335.3 458.0 542.0 195.7 286.0 356.0 192.7 236.0 326.7 9.5 13.0 20.1 3.3 4.4 9.4
108.3 101.1 89.7 93.3 121.4 119.0 120.5 90.4 99.9 113.2 115.6 117.1 107.6 113.5 106.0
4.0 5.6 3.6 5.0 5.0 2.7 2.3 7.5 7.9 5.9 9.65 5.6 10.8 1.6 4.1
Peimine
6.03
Peiminine
2.13
Table IV Contents of Five Compounds in Three Batches of QJC Compound
20110104 Content (mg/g)
20120610 Content (mg/g)
20120904 Content (mg/g)
Mean value (mg/g)
RSD (%)
Aesculin Aesculetin Fraxin Peimine Peiminine
7.65 4.71 5.96 0.18 0.07
8.79 3.49 7.42 0.20 0.08
7.44 5.25 5.18 0.20 0.08
7.96 4.48 6.18 0.19 0.08
8.6 17.4 16.0 6.6 5.9
the highest, followed by fraxin. The mean values of them were 7.96 and 6.18 mg/g, respectively. Aesculetin was lower than former two coumarins with the content of 4.48 mg/g. On the other hand, the contents of peimine and peiminine were relatively low and the content of peiminine (0.08 mg/g) was the lowest. In addition, there were no remarkable variations in the contents of aesculin, peimine and peiminine in different batches, all the RSDs of batch to batch were lower than 10%. While, there were obvious variations in the contents of aesculetin and fraxin, with the RSDs of 17.39 and 16.01%, respectively, which might be Simultaneous Determination of Five Constituents in QJC 3
Figure 2. The typical chromatography of standards (A) and samples (B): 1 aesculin, 2 fraxin, 3 aesculetin, 4 peimine and 5 peiminine.
caused by the variations of medicinal material and production process in different batches.
Discussion Optimization of the extraction conditions Considering the different properties between steroidal alkaloids and coumarins, two methods, ultrasonic extraction and reflux extraction, were tested to optimize extraction conditions. Ultrasonic extraction was a commonly used sample preparation method for the analysis of C. Fraxini. Therefore, this method was first tried to extract five target analytes. Proportion of methanol solvent and ultrasound time was investigated to obtain the optimal extraction conditions. It was found that 75% methanol was the most efficient extraction solvent among the tested methanol – water solution with different proportions (25, 50, 75 and 100%). Different extraction time (30, 45, 60, 75 and 90 min) was tested. The results demonstrated that five compounds could be entirely extracted when the ultrasound extraction time was set at 60 min. Thus, 75% methanol and 60 min were finally selected as the optimal extraction solvent and time, respectively. In Chinese Pharmacopoeia (2010 version), reflux extraction with alkalinized chloroform –methanol is used for the determination of peimine and peiminine in F. cirrhosa. In order to ensure the complete extraction of steroidal alkaloids in QJC, an orthogonal test that contained four variables and three levels was employed to investigate the extraction conditions. The following four variables, each with three levels, were optimized in this study: the time of alkalization (Factor A), its levels (30, 60 and 120 min); the ratio of methanol and chloroform (Factor B), its levels (1 : 1, 1 : 2 and 1 : 4); the volume of solution (Factor C), 4 Zhao et al.
its levels (25, 50 and 75 mL) and the extraction time (Factor D), its levels (30, 60 and 90 min). The interaction between different variables was ignored in this study. The total amount of components in analytes was selected as the index for the evaluation of method. According to the results, the optimum experimental conditions were A1B2C1D3, namely extraction with 25 mL solution (methanol – chloroform, 1 : 2) for 90 min after alkalization 30 min. Comparing optimal ultrasound extraction with methanol and reflux extraction with chloroform –methanol, it was found that there was no significant difference between two methods in content of peimine (87.34 mg/g for ultrasound extraction and 83.62 mg/g for reflux extraction) and peiminine (85.07 mg/g for ultrasound extraction and 89.58 mg/g for reflux extraction). While the content of aesculin, aesculetin and fraxin was obviously increased to 3371.66, 1699.81 and 2526.53 mg/g, respectively, by ultrasound extraction. Moreover, the method of ultrasound extraction was simple, fast and easy to operate. Eventually, the samples were prepared by ultrasound extraction with 25 mL of 75% methanol for 60 min. The comparing results of two methods are shown in Figure 3.
Optimization of the chromatography and mass spectrometry conditions Two types of reversed-phase chromatography column and various composition of the mobile phase were investigated to optimize the chromatography conditions. It was demonstrated that Purospher STAR LP RP-C18 (250 4.6 mm, 5 mm) column showed better peak shape than Phenomenex C18 (150 2.0 mm, 5 mm) column. Mobile phases (methanol – water and acetonitrile – water) were investigated to obtain good chromatographic behavior. Acetonitrile – water was finally selected
Figure 3. The results of comparing two optimal extraction methods: Method 1 ultrasound extraction with methanol and Method 2 reflux extraction with chloroform–methanol.
Table V The Results of RT, Precursor Ion, Product Ion, DP and CE of Five Compounds Compound
RT (min)
MS1 (m/z)
MS2 (m/z)
DP (V)
CE (eV)
Aesculin Fraxin Aesculetin Peimine Peiminine
4.16 4.64 6.74 8.31 9.53
339.1 369.1 176.9 432.6 430.1
177.0 206.7 132.9 414.5 412.5
234 250 240 71 65
230 252 226 47 48
because of lower background signal and better resolution. Acetonitrile – 0.3‰ formic acid water and acetonitrile – 2 mmol ammonium acetate were further examined. The results showed that 2 mmol ammonium acetate had a significant inhibition to the response of aesculin, aesculetin and fraxin. So, acetonitrile– 0.3‰ formic acid water was selected. To obtain the best mass spectrometric conditions, each compound was studied in both positive and negative modes. The results demonstrated that three coumarins such as aesculin, aesculetin and fraxin showed better ion response in negative mode. In the MS spectra, aesculin showed a precursor ion at m/z 339.1 [M–H]2 and the product ion at m/z 177.0 after losing glucose. In the same way, aesculetin and fraxin exhibited precursor ions [M– H]2 at m/z 176.9 and 369.1, respectively. The product ions of aesculetin at m/z 132.9 and fraxin at m/z 206.7 arose from the loss of CO2 and glucose, respectively. However, peimine and peiminine showed better ion response in positive mode. In the MS spectra, the precursor ions of peimine at m/z 432.6 and peiminine at m/z 430.1 were observed. The product ions of peimine at m/z 414.5 and peiminine at m/z 412.5 were detected after both losing H2O. All the product ions were chosen
Figure 4. The mass spectrum of five compounds: (A) aesculin, (B) aesculetin, (C) fraxin, (D) peimine and (E) peiminine.
Simultaneous Determination of Five Constituents in QJC 5
according to the stability and ion response. Finally, for obtaining the highest ion response for all the five compounds, the positive and negative ion switching modes were employed in this research. Furthermore, declustering potential (DP) and collision energy (CE) for each compound was optimized, respectively, to obtain stable and strong response. The results of retention time (RT), precursor ion, product ion, DP and CE of five compounds are listed in Table V, the mass spectra of five compounds in Figure 4. Conclusion In the present study, an HPLC – MS-MS method was established for the simultaneous determination of multiple active components in QJC for the first time. This novel developed method was applied to evaluate three batches of QJC. As a result, the method was demonstrated to be fast, simple and sensitive. At the same time, the established method accurately reflected the characteristic of the preparation because five constituents that worked synergistically were simultaneously determined. It would be reliable and efficient and had significant importance for the quality control of QJC.
4.
5.
6.
7.
8.
9.
10.
Acknowledgments The authors thank financial supports from the National Natural Science Foundation of China (81102412), the Ministry of Education Key Project of Science and Technology Foundation of China (211021), Hundreds of Innovative Talents Project of Hebei Education Department of China, and the Natural Science Foundation of Hebei Province of China (C2011206158, 08B031).
11.
References
13.
1. Zhang, Y.Z., Wang, H.P., Jiang, Y.J., Wei, Y.Y., Yang, W.Y.; Quality detection method for traditional Chinese medicine composition of Cortex fraxini bone-reunion capsule and its preparation; Chinese patent, (2012), CN 102608252 A 20120725. 2. An, J.E.; Clinical observation of Qinpijiegu capsule for the treatment of bone injury; China National Journal of Medicine, (2012); 2(2): 10. 3. Zhou, L., Kang, J., Fan, L., Ma, X.C., Zhao, H.Y., Han, J., et al.; Simultaneous analysis of coumarins and secoiridoids in Cortex fraxini by high-performance liquid chromatography-diode array detection-electrospray ionization tandem mass spectrometry;
6 Zhao et al.
12.
14.
Journal of Pharmaceutical and Biomedical Analysis, (2008); 47(1): 39–46. Shi, Z.H., Zhu, X.M., Zhang, H.Y.; Micelle-mediated extraction and cloud point preconcentration for the analysis of aesculin and aesculetin in Cortex fraxini by HPLC; Journal of Pharmaceutical and Biomedical Analysis, (2007); 44(4): 867– 873. Bo, T., Liu, H.W., Li, K.; An high speed determination of aesculin and aesculetin in Cortex fraxini by micellar electrokinetic chromatography; Chromatographia, (2002); 55(9-10): 621– 624. Zhang, H., Li, Q., Shi, Z., Hu, Z., Wang, R.; Analysis of aesculin and aesculotin in Cortex fraxini by capillary zone electrophoresis; Talanta, (2000); 52(4): 607–621. Li, C.H., Chen, A.J., Chen, X.F., Ma, X., Chen, X.G., Hu, Z.D.; Non-aqueous capillary electrophoresis for separation and simultaneous determination of fraxin, esculin and esculetin in Cortex fraxini and its medicinal preparations; Biomedical chromatography; BMC, (2005); 19(9): 696–702. Yang, M., Chen, J.L., Shi, X.F., Niu, H.J.; Rapid determination of aesculin, aesculetin and fraxetin in Cortex fraxini extract solutions based on ultraviolet spectroscopy; E-Journal of Chemistry, (2011); 8(Suppl 1): S225– S236. Zhao, J.N., Deng, Z.W., Dai, Y., Song, J., Peng, X.H., Wang, X.D., et al.; Effect of total coumarins of Cortex fraxini on acute gouty arthritis and uric acid metabolism; Acta Pharmaceutica Sinica, (2009); 44(10): 751– 754. Li, J.M., Zhang, X., Wang, X., Xie, Y.C., Kong, L.D.; Protective effects of cortex fraxini coumarins against oxonate-induced hyperuricemia and renal dysfunction in mice; European Journal of Pharmacology, (2011); 666(1–3): 196–204. Zhou, J.L., Li, P., Li, H.J., Jiang, Y., Ren, M.T., Liu, Y.; Development and validation of a liquid chromatography/electrospray ionization time-of-flight mass spectrometry method for relative and absolute quantification of steroidal alkaloids in Fritillaria species; Journal of Chromatography A, (2008); 1177(1): 126– 137. Huang, L.F., Chen, S.L., Liu, H., Wang, L.Z., Zhang, Y., Luo, K.; Determination of the content of three alkaloids in Fritillaria cirrhosa processed by different methods by HPLC-ELSD; Traditional Chinese Medicine Patent Prescription, (2009); 31(10): 1560–1564. Zhou, J.L., Xin, G.Z., Shi, Z.Q., Ren, M.T., Qi, L.W., Li, H.J., et al.; Characterization and identification of steroidal alkaloids in Fritillaria species using liquid chromatography coupled with electrospray ionization quadrupole time-of-flight tandem mass spectrometry; Journal of chromatography A, (2010); 1217(45): 7109–7122. Zhang, Y.C., Xu, H.Y., Chen, X.M., Chen, C., Wang, H.J., Meng, F.Y., et al.; Simultaneous quantification of 17 constituents from Yuanhu Zhitong tablet using rapid resolution liquid chromatography coupled with a triple quadrupole electrospray tandem mass spectrometry; Journal of Pharmaceutical and Biomedical Analysis, (2011); 56(3): 497– 504.
Article
Simultaneous Determination of Five Constituents in Qinpijiegu Capsule by High-Performance Liquid Chromatography Coupled with Tandem Mass Spectrometry Minmin Zhao, Weijing Ding, Shuang Wang, Meng Gao, Shan Fu, Juan Zhang, Tao Li, Yin Wu and Qiao Wang* Department of Pharmaceutical Analysis, School of Pharmacy, Hebei Medical University, 361 East Zhongshan Road, Shijiazhuang 050017, P.R. China *Author to whom correspondence should be addressed. Email: [email protected] Received 9 November 2013; revised 30 March 2014
A rapid high-performance liquid chromatography coupled with tandem mass spectrometry method was developed for the simultaneous determination of five constituents in Qinpijiegu capsule (QJC), a classical Tibetan prescription. The separation of five compounds such as aesculin, aesculetin, fraxin, peimine and peiminine was performed on a Purospher STAR LP RP-C18 (250 3 4.6 mm, 5 mm) column with linear gradient elution of acetonitrile – 0.3‰ formic acid water in 13 min. Detection was carried out by multiple reaction monitoring mode using electrospray ionization in the positive and negative ion switching mode. The sample was prepared with ultrasound extraction with methanol, which could obtain higher extraction efficiency and shorter extraction time comparing to reflux extraction with alkalized chloroform – methanol. The proposed method was applied to analyze three batches of samples with acceptable linearity (r 2 > 0.9977), precision [relative standard deviation (RSD) < 7.40%], repeatability (RSD < 2.49%), stability [relative error (RE) < 9.15%] and recovery (RSD < 10.76%). This is the first development of a multicomponent quantitation method for the quality control of QJC. Furthermore, the new established method was proven to be highly sensitive and effective in evaluating the quality of QJC.
Introduction Qinpijiegu capsule (QJC) is a classical Tibetan medicine composed of 113.0 g of Cortex fraxini, 80.0 g of Fritillaria cirrhosa, 80.0 g of Pyrethrum tatsienense and 57.0 g of Keel (1). It displays good effects in analgesia and callus growth comparing with Shangkejiegu tablet, a traditional medicine for treatment of bone injury (2), and is widely used for the treatment of bruises, sprains bones and bruising swelling in China. No doubt, the quality control of QJC is essential to ensure its efficacy and safety. QJC is formulated by four kinds of medicinal materials. It is obvious that determining multicomponents in various medicinal materials can more accurately evaluate its quality than only determining a certain component in one medicinal material. According to the composition of prescription, Cortex fraxini accounts for 34.2%, the largest proportion of the total content. It is from the dry barks of Oleaceae plant Fraxinus rhynchophylla Hance, F. rhynchophylla Hance or Fraxinus paxiana (3). Coumarins, including aesculin, aesculetin, fraxin, and so on, are the characteristic constituents (4 – 8), which are proven to be the active constituents, and commonly possess various biological activities such as anti-inflammation, antivirus, antiarthritis and anticancer (9, 10). Fritillaria cirrhosa accounts for 24.2% of the prescription next to C. fraxini. It has been used as antitussive and expectorant herbs for thousands of
years in China (11). Steroidal alkaloids, including peimine and peiminine, have been demonstrated as its major ingredients and reported to have activities such as antihypertensive, anticholinergic, antitumor, antiasthmatic and antitussive (12, 13). Although P. tatsienense, a traditional folk medicine, occupies the equal proportion as F. cirrhosa in the prescription, there are no reports on its constituents study. Keel, originating from the bone of ancient mammals, accounts for only 17.3% in the prescription. It mainly contains calcium salts and other mineral salts that are difficult to be simultaneously analyzed with organic components. In summary, selecting main coumarins and steroidal alkaloids in C. fraxini and F. cirrhosa as the marker components would be practical and valuable for the quality evaluation of QJC. To date, only one report on quality detection method about QJC was published (1). However, in this paper, aesculin and aesculetin in C. Fraxini and imperialine in F. cirrhosa were separately determined by HPLC-UV according to Chinese Pharmacopoeia (ChP, Version 2010). Moreover, the development and validation of method were not described. Therefore, it is essential to establish a simple and rapid method to simultaneously determine multiple active components in QJC for quality control of QJC. In recent years, because of high sensitivity and specificity, the high-performance liquid chromatography coupled with tandem mass spectrometry (HPLC –MS-MS) method has found wider and wider application than other detection systems in the field of traditional Chinese medicine prescription research (14). In this study, the HPLC–MS-MS method was developed for the simultaneous quantification of five main constituents in QJC, including aesculin, aesculetin, fraxin, peimine and peiminine, and then applied to the quality evaluation of QJC. This is the first time to establish an HPLC – MS-MS method for the simultaneous determination of multiconstituent in QJC.
Experimental Chemicals and materials Aesculin (110740 – 200506) and aesculetin (110741 – 200506) were obtained from the National Institutes for Food and Drug Control (Beijing, China). Fraxin (20111214), peimine (120913001) and peiminine (120921001) were purchased from Shanghai Sunny Biotech Co., Ltd. (Shanghai, China). The purities of all standards were .98%. The chemical structures of five components are shown in Figure 1. HPLC grade acetonitrile was obtained from J.T. Baker Company (Center Valley, PA, USA). Formic acid was purchased from Dikma
# The Author 2014. Published by Oxford University Press. All rights reserved. For Permissions, please email: [email protected]
Figure 1. Chemical structures of five components: (A) aesculin, (B) aesculetin, (C) fraxin, (D) peimine and (E) peiminine.
Company (Lake Forest, CA, USA). Purified water was purchased from Hangzhou Wahaha Group Co., Ltd. (Hangzhou, Zhejiang, China). Three batches of QJC samples (bat. 20110104, 20120610 and 20120904) were manufactured by Arura Tibetan Medicine Co., Ltd. (Xining, Qinghai, China).
Apparatus and HPLC–MS-MS conditions An Agilent 1200 liquid chromatography system (Agilent technology, Santa Clara, CA, USA) equipped with an online vacuum degasser, a quaternary solvent delivery system, an autosampler and a column compartment was used for HPLC analysis. Chromatographic separation was achieved on a Purospher STAR LP RP-C18 (250 4.6 mm, 5 mm) column at room temperature. The mobile phase consisted of acetonitrile (A) and 0.3‰ aqueous formic acid (B) with gradient elution of 20 –30% A at 0 – 13 min. The flow rate was 0.9 mL/min, and the injection volume was 10mL for analysis. The 3200 QTRAPTM system (Applied Biosystems, Foster City, CA, USA) with a hybrid triple quadrupole linear ion trap mass spectrometer equipped with an electrospray ion source was applied to quantitative analysis. Mass spectrometer was operated in the positive and negative ion switching multiple reaction monitoring mode. The switching program was set as follows: period 1, 0 –7.2 min, negative mode for the determination of aesculin, aesculetin and fraxin and period 2, 7.2 – 13 min, positive mode for the determination of peimine and peiminine. The condition of MS analysis was designed as follows: the spray voltage, 5,500 and 24,500 V; the turbo spray temperature, 6508C; nebulizer gas (gas 1), 60 psi; heater gas (gas 2), 65 psi; collision gas, medium and the curtain gas (CUR), 25 psi. Data collection and analysis were accomplished with the Analyst 1.5.2 software provided by Applied Biosystems/MDS Sciex.
Sample and standard solution preparation Fifty milligrams of QJC powder were weighed accurately, and settled to a conical flask with the cover, 25 mL of 75% methanol aqueous solution was precisely added. The conical flask 2 Zhao et al.
containing sample and solvent was weighed. Subsequently, the sample was ultrasonically extracted at 358C for 60 min and then was cooled to room temperature. The lost weight caused by volatilization of extraction solution during extraction was complemented with 75% methanol aqueous solution. After centrifugation at 14,000 rpm for 10 min, solution was filtered through 0.45 mm membrane. The filtrate was diluted eight times with 20% of acetonitrile aqueous solution to eliminate the solvent effect, and finally determined by HPLC–MS-MS. Mixture stock solution (9.20 mg/mL for aesculin; 6.00 mg/mL for aesculetin; 5.85 mg/mL for fraxin; 1.19 mg/mL for peimine and 0.98 mg/mL for peiminine) was prepared by weighing precisely and dissolving the reference standards with methanol. A series of standard solutions were obtained by further diluting the mixture stock solution. All solutions were stored at 48C before analysis.
Results Method validation Linearity, limits of detection and limits of quantification Mixture standard solution of five compounds was diluted to a series of concentrations to establish calibration curves (Table I). Each calibration curve consisted of at least five concentration levels and was plotted with the peak areas of the standards versus their concentration. All calibration curves showed good linearity across the calibration range (r 2 . 0.9977). Limits of detection (LOD) and limits of quantification (LOQ) were the concentrations of standard solution when the ratios of the signal to noise were 3 and 10, respectively. The data of calibration curves, r 2, linear ranges, LOD and LOQ are presented in Table I. Precision, repeatability and stability The precision was investigated by measuring the intra- and interday variances. Six independent samples were extracted and analyzed within a day to evaluate the intraday precision, while interday precision was determined by repeating above experiment once a day over 3 consecutive days. The RSDs of
Table I Calibration Curves, r 2, Linear Ranges, LOD and LOQ of Five Compounds Compound
Regression equation
r2
Linear range (ng/mL)
LOD (ng/mL)
LOQ (ng/mL)
Aesculin Aesculetin Fraxin Peimine Peiminine
y ¼ 2.17e6x þ 3.94e2 y ¼ 2.15e6x þ 3.66e4 y ¼ 4.78e4x 2 211 y ¼ 3.34e3x þ 8.26e3 y ¼ 6.05e3x þ 1.26e3
0.9985 0.9983 0.9987 0.9977 0.9981
115.00 75.00 73.13 29.75 24.50
0.06 4.00 0.08 0.07 0.08
0.25 11.0 0.25 0.31 0.31
intra- and interday were within the range of 3.30 – 4.80% and 5.10 –7.40%, respectively. Repeatability of the instrument was tested by six parallel measurement of the mixed standard solution of five compounds (aesculin, 0.92 mg/mL; aesculetin, 0.60 mg/mL; fraxin, 0.59 mg/mL; peimine, 119.00 ng/mL and peiminine, 98.00 ng/mL) under the above-mentioned HPLC –MS-MS conditions. The RSD values of five compounds were from 0.74 to 2.49%, which showed high repeatability of the instrument. In order to investigate the stability of the sample solution, three samples were analyzed in parallel at different time points: 0, 2, 4, 7 and 10 h, respectively. Stability was expressed by the value of RE, which was calculated as the following formula: RE ð%Þ ¼
Detection amount ð2; 4; 7; 10 hÞ amount ð0 hÞ Amount ð0 hÞ 100%
ð1Þ
The RE values of five compounds were all ,9.15%, which revealed good stability of the sample solution stored within 10 h. The data of precision, repeatability and stability are listed in Table II. Accuracy The accuracy of the method was evaluated by recovery. The recovery test was performed by spiking different known amounts (high, middle and low) of mixed standard solution composed of five compounds to the samples, then the samples were extracted and determined under the same conditions with other sample, and triplicate experiments were repeated at each level. The recovery was calculated according to the following formula: Detection amount Original amount Recovery ð% Þ ¼ Addition 100%
ð2Þ
As shown in Table III, all of the recoveries of five compounds were in the range of 89.69 – 121.40%, which demonstrated that the method was of good accuracy.
Sample analysis The established analytical method was subsequently applied to determine five compounds in three batches of QJC. The contents of five compounds in three batches of QJC are summarized in Table IV, and the typical chromatography of standards and samples is given in Figure 2. As shown in Table IV, aesculin, aesculetin and fraxin were the dominant compounds. Among them, the content of aesculin was
– – – – –
1840.00 1200.00 5850.00 1190.00 980.00
Table II The Results of Precision, Repeatability and Stability Compound
Precision (RSD, n ¼ 6) Intraday
Interday
Aesculin Aesculetin Fraxin Peimine Peiminine
4.10 4.10 3.30 4.00 4.80
7.40 6.20 7.00 6.70 5.10
Repeatability (RSD, n ¼ 6)
Stability (RE, n ¼ 3)
0.7 2.5 1.8 0.8 1.5
,8.9 ,9.1 ,8.1 ,8.6 ,7.6
Table III The Recoveries of Five Compounds with n ¼ 3.7 Compound
Original (mg)
Addition (mg)
Detection (mg)
Recovery (%)
RSD (%)
Aesculin
206.75
Aesculetin
138.00
Fraxin
137.00
122.50 245.00 367.50 60.00 120.00 180.00 52.00 104.00 156.00 2.98 5.95 11.90 0.98 1.96 4.90
335.3 458.0 542.0 195.7 286.0 356.0 192.7 236.0 326.7 9.5 13.0 20.1 3.3 4.4 9.4
108.3 101.1 89.7 93.3 121.4 119.0 120.5 90.4 99.9 113.2 115.6 117.1 107.6 113.5 106.0
4.0 5.6 3.6 5.0 5.0 2.7 2.3 7.5 7.9 5.9 9.65 5.6 10.8 1.6 4.1
Peimine
6.03
Peiminine
2.13
Table IV Contents of Five Compounds in Three Batches of QJC Compound
20110104 Content (mg/g)
20120610 Content (mg/g)
20120904 Content (mg/g)
Mean value (mg/g)
RSD (%)
Aesculin Aesculetin Fraxin Peimine Peiminine
7.65 4.71 5.96 0.18 0.07
8.79 3.49 7.42 0.20 0.08
7.44 5.25 5.18 0.20 0.08
7.96 4.48 6.18 0.19 0.08
8.6 17.4 16.0 6.6 5.9
the highest, followed by fraxin. The mean values of them were 7.96 and 6.18 mg/g, respectively. Aesculetin was lower than former two coumarins with the content of 4.48 mg/g. On the other hand, the contents of peimine and peiminine were relatively low and the content of peiminine (0.08 mg/g) was the lowest. In addition, there were no remarkable variations in the contents of aesculin, peimine and peiminine in different batches, all the RSDs of batch to batch were lower than 10%. While, there were obvious variations in the contents of aesculetin and fraxin, with the RSDs of 17.39 and 16.01%, respectively, which might be Simultaneous Determination of Five Constituents in QJC 3
Figure 2. The typical chromatography of standards (A) and samples (B): 1 aesculin, 2 fraxin, 3 aesculetin, 4 peimine and 5 peiminine.
caused by the variations of medicinal material and production process in different batches.
Discussion Optimization of the extraction conditions Considering the different properties between steroidal alkaloids and coumarins, two methods, ultrasonic extraction and reflux extraction, were tested to optimize extraction conditions. Ultrasonic extraction was a commonly used sample preparation method for the analysis of C. Fraxini. Therefore, this method was first tried to extract five target analytes. Proportion of methanol solvent and ultrasound time was investigated to obtain the optimal extraction conditions. It was found that 75% methanol was the most efficient extraction solvent among the tested methanol – water solution with different proportions (25, 50, 75 and 100%). Different extraction time (30, 45, 60, 75 and 90 min) was tested. The results demonstrated that five compounds could be entirely extracted when the ultrasound extraction time was set at 60 min. Thus, 75% methanol and 60 min were finally selected as the optimal extraction solvent and time, respectively. In Chinese Pharmacopoeia (2010 version), reflux extraction with alkalinized chloroform –methanol is used for the determination of peimine and peiminine in F. cirrhosa. In order to ensure the complete extraction of steroidal alkaloids in QJC, an orthogonal test that contained four variables and three levels was employed to investigate the extraction conditions. The following four variables, each with three levels, were optimized in this study: the time of alkalization (Factor A), its levels (30, 60 and 120 min); the ratio of methanol and chloroform (Factor B), its levels (1 : 1, 1 : 2 and 1 : 4); the volume of solution (Factor C), 4 Zhao et al.
its levels (25, 50 and 75 mL) and the extraction time (Factor D), its levels (30, 60 and 90 min). The interaction between different variables was ignored in this study. The total amount of components in analytes was selected as the index for the evaluation of method. According to the results, the optimum experimental conditions were A1B2C1D3, namely extraction with 25 mL solution (methanol – chloroform, 1 : 2) for 90 min after alkalization 30 min. Comparing optimal ultrasound extraction with methanol and reflux extraction with chloroform –methanol, it was found that there was no significant difference between two methods in content of peimine (87.34 mg/g for ultrasound extraction and 83.62 mg/g for reflux extraction) and peiminine (85.07 mg/g for ultrasound extraction and 89.58 mg/g for reflux extraction). While the content of aesculin, aesculetin and fraxin was obviously increased to 3371.66, 1699.81 and 2526.53 mg/g, respectively, by ultrasound extraction. Moreover, the method of ultrasound extraction was simple, fast and easy to operate. Eventually, the samples were prepared by ultrasound extraction with 25 mL of 75% methanol for 60 min. The comparing results of two methods are shown in Figure 3.
Optimization of the chromatography and mass spectrometry conditions Two types of reversed-phase chromatography column and various composition of the mobile phase were investigated to optimize the chromatography conditions. It was demonstrated that Purospher STAR LP RP-C18 (250 4.6 mm, 5 mm) column showed better peak shape than Phenomenex C18 (150 2.0 mm, 5 mm) column. Mobile phases (methanol – water and acetonitrile – water) were investigated to obtain good chromatographic behavior. Acetonitrile – water was finally selected
Figure 3. The results of comparing two optimal extraction methods: Method 1 ultrasound extraction with methanol and Method 2 reflux extraction with chloroform–methanol.
Table V The Results of RT, Precursor Ion, Product Ion, DP and CE of Five Compounds Compound
RT (min)
MS1 (m/z)
MS2 (m/z)
DP (V)
CE (eV)
Aesculin Fraxin Aesculetin Peimine Peiminine
4.16 4.64 6.74 8.31 9.53
339.1 369.1 176.9 432.6 430.1
177.0 206.7 132.9 414.5 412.5
234 250 240 71 65
230 252 226 47 48
because of lower background signal and better resolution. Acetonitrile – 0.3‰ formic acid water and acetonitrile – 2 mmol ammonium acetate were further examined. The results showed that 2 mmol ammonium acetate had a significant inhibition to the response of aesculin, aesculetin and fraxin. So, acetonitrile– 0.3‰ formic acid water was selected. To obtain the best mass spectrometric conditions, each compound was studied in both positive and negative modes. The results demonstrated that three coumarins such as aesculin, aesculetin and fraxin showed better ion response in negative mode. In the MS spectra, aesculin showed a precursor ion at m/z 339.1 [M–H]2 and the product ion at m/z 177.0 after losing glucose. In the same way, aesculetin and fraxin exhibited precursor ions [M– H]2 at m/z 176.9 and 369.1, respectively. The product ions of aesculetin at m/z 132.9 and fraxin at m/z 206.7 arose from the loss of CO2 and glucose, respectively. However, peimine and peiminine showed better ion response in positive mode. In the MS spectra, the precursor ions of peimine at m/z 432.6 and peiminine at m/z 430.1 were observed. The product ions of peimine at m/z 414.5 and peiminine at m/z 412.5 were detected after both losing H2O. All the product ions were chosen
Figure 4. The mass spectrum of five compounds: (A) aesculin, (B) aesculetin, (C) fraxin, (D) peimine and (E) peiminine.
Simultaneous Determination of Five Constituents in QJC 5
according to the stability and ion response. Finally, for obtaining the highest ion response for all the five compounds, the positive and negative ion switching modes were employed in this research. Furthermore, declustering potential (DP) and collision energy (CE) for each compound was optimized, respectively, to obtain stable and strong response. The results of retention time (RT), precursor ion, product ion, DP and CE of five compounds are listed in Table V, the mass spectra of five compounds in Figure 4. Conclusion In the present study, an HPLC – MS-MS method was established for the simultaneous determination of multiple active components in QJC for the first time. This novel developed method was applied to evaluate three batches of QJC. As a result, the method was demonstrated to be fast, simple and sensitive. At the same time, the established method accurately reflected the characteristic of the preparation because five constituents that worked synergistically were simultaneously determined. It would be reliable and efficient and had significant importance for the quality control of QJC.
4.
5.
6.
7.
8.
9.
10.
Acknowledgments The authors thank financial supports from the National Natural Science Foundation of China (81102412), the Ministry of Education Key Project of Science and Technology Foundation of China (211021), Hundreds of Innovative Talents Project of Hebei Education Department of China, and the Natural Science Foundation of Hebei Province of China (C2011206158, 08B031).
11.
References
13.
1. Zhang, Y.Z., Wang, H.P., Jiang, Y.J., Wei, Y.Y., Yang, W.Y.; Quality detection method for traditional Chinese medicine composition of Cortex fraxini bone-reunion capsule and its preparation; Chinese patent, (2012), CN 102608252 A 20120725. 2. An, J.E.; Clinical observation of Qinpijiegu capsule for the treatment of bone injury; China National Journal of Medicine, (2012); 2(2): 10. 3. Zhou, L., Kang, J., Fan, L., Ma, X.C., Zhao, H.Y., Han, J., et al.; Simultaneous analysis of coumarins and secoiridoids in Cortex fraxini by high-performance liquid chromatography-diode array detection-electrospray ionization tandem mass spectrometry;
6 Zhao et al.
12.
14.
Journal of Pharmaceutical and Biomedical Analysis, (2008); 47(1): 39–46. Shi, Z.H., Zhu, X.M., Zhang, H.Y.; Micelle-mediated extraction and cloud point preconcentration for the analysis of aesculin and aesculetin in Cortex fraxini by HPLC; Journal of Pharmaceutical and Biomedical Analysis, (2007); 44(4): 867– 873. Bo, T., Liu, H.W., Li, K.; An high speed determination of aesculin and aesculetin in Cortex fraxini by micellar electrokinetic chromatography; Chromatographia, (2002); 55(9-10): 621– 624. Zhang, H., Li, Q., Shi, Z., Hu, Z., Wang, R.; Analysis of aesculin and aesculotin in Cortex fraxini by capillary zone electrophoresis; Talanta, (2000); 52(4): 607–621. Li, C.H., Chen, A.J., Chen, X.F., Ma, X., Chen, X.G., Hu, Z.D.; Non-aqueous capillary electrophoresis for separation and simultaneous determination of fraxin, esculin and esculetin in Cortex fraxini and its medicinal preparations; Biomedical chromatography; BMC, (2005); 19(9): 696–702. Yang, M., Chen, J.L., Shi, X.F., Niu, H.J.; Rapid determination of aesculin, aesculetin and fraxetin in Cortex fraxini extract solutions based on ultraviolet spectroscopy; E-Journal of Chemistry, (2011); 8(Suppl 1): S225– S236. Zhao, J.N., Deng, Z.W., Dai, Y., Song, J., Peng, X.H., Wang, X.D., et al.; Effect of total coumarins of Cortex fraxini on acute gouty arthritis and uric acid metabolism; Acta Pharmaceutica Sinica, (2009); 44(10): 751– 754. Li, J.M., Zhang, X., Wang, X., Xie, Y.C., Kong, L.D.; Protective effects of cortex fraxini coumarins against oxonate-induced hyperuricemia and renal dysfunction in mice; European Journal of Pharmacology, (2011); 666(1–3): 196–204. Zhou, J.L., Li, P., Li, H.J., Jiang, Y., Ren, M.T., Liu, Y.; Development and validation of a liquid chromatography/electrospray ionization time-of-flight mass spectrometry method for relative and absolute quantification of steroidal alkaloids in Fritillaria species; Journal of Chromatography A, (2008); 1177(1): 126– 137. Huang, L.F., Chen, S.L., Liu, H., Wang, L.Z., Zhang, Y., Luo, K.; Determination of the content of three alkaloids in Fritillaria cirrhosa processed by different methods by HPLC-ELSD; Traditional Chinese Medicine Patent Prescription, (2009); 31(10): 1560–1564. Zhou, J.L., Xin, G.Z., Shi, Z.Q., Ren, M.T., Qi, L.W., Li, H.J., et al.; Characterization and identification of steroidal alkaloids in Fritillaria species using liquid chromatography coupled with electrospray ionization quadrupole time-of-flight tandem mass spectrometry; Journal of chromatography A, (2010); 1217(45): 7109–7122. Zhang, Y.C., Xu, H.Y., Chen, X.M., Chen, C., Wang, H.J., Meng, F.Y., et al.; Simultaneous quantification of 17 constituents from Yuanhu Zhitong tablet using rapid resolution liquid chromatography coupled with a triple quadrupole electrospray tandem mass spectrometry; Journal of Pharmaceutical and Biomedical Analysis, (2011); 56(3): 497– 504.
