Analytica Chimica Acta 585 (2007) 76–80
Fast determination of paeonol in plasma by headspace solid-phase microextraction followed by gas chromatography–mass spectrometry Ling Dong a , Chunhui Deng b , Jiyao Wang a , Xizhong Shen a,∗ a
Zhongshan Hosptial, Medical college of Shanghai, Fudan University, Shanghai 200433, China b Department of Chemistry, Fudan University, Shanghai 200433, China
Received 18 September 2006; received in revised form 5 December 2006; accepted 7 December 2006 Available online 20 December 2006
Abstract Paeonol is the active component in the traditional Chinese medicines (TCMs), such as Cynanchum paniculatum, which has been used to treat many diseases, such as eczema. In this work, a simple, rapid and sensitive method was developed for the determination of paeonol in rabbit plasma, which was based on headspace solid-phase microextraction (HS-SPME) followed by gas chromatography–mass spectrometry (GC–MS). The extraction parameters of fiber coating, sample temperature, extraction time, stirring rate and ion strength were systemically optimized; the method linearity, detection limit and precision were also investigated. It was shown that the proposed method provided a good linearity (0.02–20 g mL−1 , R2 > 0.990), low detection limit (2.0 ng mL−1 ) and good precision (R.S.D. value less than 8%). Finally, GC/MS following HS-SPME was applied to fast determination of paeonol in rabbit plasma at different time point after oral demonstration of Cynanchum paniculatum essential oil. The experimental results suggest that the proposed method provided an alternative and novel approach to the pharmacokinetics study of paeonol in the TCMs. © 2006 Elsevier B.V. All rights reserved. Keywords: Paeonol; Solid-phase microextraction; Gas chromatography–mass spectrometry; Plasma; Traditional Chinese medicine; Cynanchum paniculatum
1. Introduction Cynanchum paniculatum is a common traditional Chinese medicine (TCM), which have been used to treat many diseases, such as eczema for a thousand years . Recently, it was found that the extract of Cynanchum paniculatum could inhibit the growth of human cancer cells . Modern pharmacologic studies have shown that paeonol (Fig. 1a) is an important active component present in Cynanchum paniculatum and other TCMs, which has antiaggregatory, antioxidant and anti-inflammatory activities [3,4]. Many methods for the quantification of paeonol in TCMs have been developed, which included micellar capillary electrophoresis , capillary zone electrophoresis and micellar electrokinetic chromatography (MEKC) , high performance liquid chromatography [7–10], high performance liquid chromatography–electrospray ionization–mass spectrometry , and gas chromatography–mass spectrometry (GC–MS) . Recently, to investigate the drug effect of paeonol from TCM essential oil, some animal modes have been
Corresponding author. Fax: +86 21 65641740. E-mail address: xzshen [email protected]
0003-2670/$ – see front matter © 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.aca.2006.12.024
built. The pharmacokinetic study of paeonol is performed in many labs. Therefore, developing a simple, rapid and sensitive method for the quantification of paeonol in biological fluids is very desirable. GC/MS and liquid chromatography (LC) combined with some sample preparation techniques, such as solid-phase extraction (SPE) liquid–liquid extraction (LLE) were developed for the determination of paeonol in rabbit plasma, and applied to the pharmacokinetic study of paeonol [13–15]. However, the two sample techniques of SPE and LLE are time-consuming, and require large amount of organic solvent. Development of a fast, sensitive, solvent-free sample extraction technique for the analysis of paeonol in plasma is interesting. Solid-phase microextraction (SPME) introduced by Pawliszyn and co-worker  is a simple, rapid, and solventfree sample extraction and concentration technique. It has widely been used to the biological analysis [17–24]. In our previous studies [25–29], SPME followed by GC/MS was developed for the analysis of amino acids, carbonyls and organic acids in human biological fluids. Krogh et al.  successfully developed SPME technique for the pharmacokinetic study of valproic acid in human plasma. LC or LC–MS combined with SPME was developed for the pharmacokinetic
L. Dong et al. / Analytica Chimica Acta 585 (2007) 76–80
2.3. Gas chromatography/mass spectrometry
Fig. 1. The chemical structures of paeonol (a) and menthol (IS, b).
study of mycophenolic acid and verapamil in human plasma [31,32]. Headspace (HS) SPME combined with GC/MS was also applied to the pharmacokinetics studies of amphetamine, midazolam, thymol, and ethanol in human plasma [33–37]. In this study, HS-SPME followed by GC/MS was developed for fast determination of paeonol in rabbit plasma, and then applied to the pharmacokinetics study of paeonol from Cynanchum paniculatum essential oil. The experimental parameters of fiber coating, sample temperature, extraction time, stirring rate, ion strength were studied, and the method linearity, detection limit and precision were also investigated. 2. Materials and methods 2.1. Chemicals, calibration solutions and SPME ﬁber Paeonol was obtained from Drug Monitoring Center (Shanghai, China). The internal standard (IS) of menthol (Fig. 1b) was obtained from Chemical Agent Company (Shanghai, China). Standard stock solutions of 10.0 mg mL−1 for paeonol were prepared in methanol. The IS solution (0.6 mg mL−1 ) was also made in methanol. Calibration solutions ranged from 0.02 to 20 g mL−1 were prepared by adding the standard stocking solution to the drug-free human plasma. The SPME fibers: 100-m polydimethylsiloxane (PDMS), 65-m polydimethylsiloxane/divinylbenzene (PDMS-DVB), 65-m carbowax/divinylbenzene (CW-DVB), 75-m carboxen poly(dimethylsiloxane) (CAR-PDMS), and 85-m polyacrylate (PA) were purchased from Supelco (Bellefonte, PA, USA). 2.2. Extraction of essential oil of Cynanchum paniculatum The TCM of Cynanchum paniculatum was purchased from a Chinese Herb Store of LeiYinShang (Shanghai, China). Eighty grams of Cynanchum paniculatum was transferred into a 1000mL distillation flask. Five hundred millilitres of distilled water was added and volatile oil distillation apparatus was set according to the Chinese pharmacopoeia , and the mixture was distilled for 6 h. Oil was collected from the condenser, dried over anhydrous sodium sulfate, and the recorded yield of the sample was about 1.23%. The essential oil was analyzed by GC/MS , the paeonol concentration in the essential oil is 84% (w/w).
An Hewlett-Packard (HP) 6890 GC system, coupled with a HP MD5973 quadrupole mass spectrometer was used. The fiber bearing the extracted analytes was inserted into the split/splitless injector of the gas chromatograph and heated to 250 ◦ C; the compounds were thermally desorbed in splitless mode for 3 min, and afterwards the splitter was opened and the fiber was removed. An HP-5MS fused-silica capillary column (30 m × 0.25 mm I.D., 0.25 m film) purchased from Agilent Technologies Co. Ltd. (Shanghai, China), was used in this study. The carrier gas was helium at a flow rate of 1.0 mL min−1 . The injector temperature was set at 250 ◦ C. The column temperature program was: initial temperature 50 ◦ C, increase to the final temperature of 300 ◦ C at 10 ◦ C min−1 , hold for 5 min. Electron impact ionization (EI) with a nominal electron energy of 70 eV was used. The ion source temperature of mass spectrometer was 230 ◦ C. The scan range was m/z 41–350. Quadrupole temperature and transfer line temperature were 150 and 280 ◦ C, respectively. Quantitative analysis was done in selected ion monitoring (SIM) mode. The molecular ion peaks at m/z 166 for paeonol and m/z 138 for IS were used for SIM experiment. 2.4. Collection of plasma samples A rabbit (2.5 kg) was obtained from Shanghai Medical University, and it was fasted for 12 h with free access to water before the experiments. Essential oil was administered orally to the rabbit at a single dose of 0.4 g kg−1 , and blood samples were collected in heparinized tube at time points 30 to 7200 min after the administration. Then blood samples were immediately centrifuged at 4000 rpm for 10 min to separate the plasma fractions. Each plasma sample (1.0 mL) was placed into a 2.0 mL headspace vial with 10 L IS solution, respectively. The obtained plasma samples were stored at −20 ◦ C until analysis. 2.5. Optimization of HS-SPME conditions In the headspace extraction process, the parameters of fiber coating, sample temperature, extraction time, stirring rate and ion strength can affect the extraction efficiency of SPME, so, these parameters were systemically studied. At first, the fiber coating was investigated. The calibration solution of paeonol (1.0 mL, 5 g mL−1 ) was introduced into 2 mL headspace vials equipped with 1-cm stirring bar, respectively. Headspace extraction of paeonol in plasma was performed by using different SPME fibers (PDMS, DVB-CW, PDMS-DVB, PA and CARPDMS), with sample temperature of 25 ◦ C, extraction time of 20 min, stirring rate of 500 rpm. Next, sample temperature (25 to 80 ◦ C), extraction time (5 to 40 min), stirring rate (300 to 1300 rpm), and salt effect (NaCl concentration ranged from 0 to 30%) were investigated. 2.6. Linearity, detection limit, precision and recovery The method linearity, detection limit, precision and recovery were studied with the optimal HS-SPME conditions (PDMS-
L. Dong et al. / Analytica Chimica Acta 585 (2007) 76–80
DVB fiber, extraction time of 10 min, extraction temperature of 60 ◦ C, stirring rate of 1100 rpm, NaCl concentration of 30%). The linearity was investigated by replicate three analyses of the interesting concentration range (0.02–20 g mL−1 ). To obtain the detection limit, a calibration solution with the lowest concentration of 0.02 g mL−1 for paeonol was analyzed, and the detection limit value was calculated on the basis of extrapolation to a signal-to-noise (S/N) ratio of 3. Precision of the assay was measured via the replicated three analyses of the plasma samples (paeonol concentration: 4.2 g mL−1 , sample volume: 1.0 mL) by the complete analytical procedure. In order to estimate recovery, 10 L of working solution (10 g mL−1 ) and with 10 L IS solution were added to 1 mL plasma sample with the paeonol concentrations of 4.2 g mL−1 . HS-SPME was performed at the optimized conditions. Recovery was obtained by comparing the added paeonol value with that measured using the present internal standard method. 2.7. Determination of paeonol in plasma The rabbit plasma (1.0 mL) after oral administration of the TCM essential oil (1.0 g) was introduced into 2 mL headspace vials with 10 L IS solution, respectively. Headspace extraction was performed by using a PDMS/DVB fiber at 60 ◦ C for 10 min, with the stirring rate of 1100 rpm and salt concentration of 30% NaCl. The extracted analytes on fiber were desorbed and analyzed by GC/MS. To quantify the paeonol in plasma, the calibration solutions ranged from 0.02 to 20 g mL−1 were introduced into 2 mL headspace vials with 10 L IS solution, respectively. Headspace extraction and GC/MS analysis were performed at the same extraction conditions described above. The paeonol concentration in plasma was quantified by using internal standard method. 3. Results and discussion Headspace (HS) SPME can be avoid of interference with sample maxries, and prevent from contamination of fiber in plasma samples. So, the headspace mode of SPME was used in this work. In the proposed method, the paeonol and IS in plasma samples were headspace extracted by SPME fiber, and then desorbed at GC–MS injector. The desorption conditions including temperature (200–300 ◦ C) and time (1.0–5.0 min) were studied. We found that the complete desorption for the analytes was obtained at 250 ◦ C for 3 min. This is similar to that in our previous work . So, the desorption conditions (250 ◦ C and 3 min) were used in the following work.
Fig. 2. The effect of fiber coating on extraction efficiency of paeonol in plasma.
Replicate three extractions for each fiber was performed. The peak area of paeonol obtained at different fibers is shown in Fig. 2. As seen from Fig. 2, among the five fibers, the PDMSDVB fiber has the best extraction efficiency. The effect of sample temperature and extraction time on the extraction efficiency was also studied. The extraction was carried out at different sample temperature (25 to 80 ◦ C) and extraction time (5 to 40 min). It can be seen from Fig. 3 that the best extraction efficiency was obtained at high sample temperature of 60 ◦ C, and the extraction equilibration was obtained at a short extraction time of 10 min. This can be explained that when the sample temperature increases, more analyte is evaporated into the headspace, so the high sample temperature can improve SPME extraction efficiency and shorten the extraction time. On the other hand, sample temperature increases, the headspace temperature also increase, and decrease the extraction efficiency. Stirring the solution improves mass transfer in the aqueous phase and induces the convection in the headspace. Therefore, the equilibrium between the aqueous and vapor phases can be achieved more rapidly. So, sample stirring reduces the time required to reach the equilibrium and extraction time by
3.1. Optimization of HS-SPME parameters Selection of the optimal coating is essential in the HS-SPME process, five different fibers of PDMS, DVB-CW, PDMS-DVB, PA and CAR-PDMS were used for the headspace extraction of paeonol in 1.0 mL plasma (5.0 g mL−1 for paeonol). Headspace extraction was performed at the sample temperature of 25 ◦ C, extraction time of 20 min, stirring rate of 500 rpm. The extracted analytes were desorbed and analyzed by GC/MS.
Fig. 3. The effect of extraction temperature and time on extraction efficiency of paeonol in plasma.
L. Dong et al. / Analytica Chimica Acta 585 (2007) 76–80
Fig. 6. The SIM chromatogram of rabbit plasma after the oral administration of the essential oil (1.0 g) from Cynanchum paniculatum. Fig. 4. The effect of stirring rate on extraction efficiency of paeonol in plasma.
enhancing the diffusion of the analyte towards the fiber. The experimental results shown in Fig. 4 indicate that the signal increases with increase in stirring speed up to 1100 rpm. At elevated rates at 1300 rpm, the spattering of solution occurs which causes signal decreasing. Thus, 1100 rpm was selected as optimum stirring rate. The influence of ionic strength on the efficiency of HS-SPME was studied. As seen from Fig. 5, the maximum peak area of paeonol can be obtained at the NaCl concentration of 30%. This can be explained that the addition of ionic strength promotes the transport of the analyte to the headspace and improves the extraction efficiency. 3.2. Method linearity, detection limit, precision and recovery The optimal HS-SPME conditions were applied to the study of the method validation. The linearity was investigated by replicate analyses of calibration solutions in the interesting con-
Fig. 5. The effect of NaCl concentration on extraction efficiency of paeonol in plasma.
centration range (0.02–20 g mL−1 ). The calibration curve was found to be very linear (R2 = 0.998) over the concentration range (0.02–20 g mL−1 ). To obtain detection limit, replicate analyses of 1.0 mL calibration solution with the lowest concentration (0.02 g mL−1 ) were performed. On the basis of extrapolation to a signal-to-noise (S/N) ratio of 3, the detection limit value was calculated, and the obtained value was 2.0 ng mL−1 . The detection limit by the proposed method was lower than those by LLE and SPE, respectively [13–15]. Precision was expressed by relative standard deviation (R.S.D.) value, and R.S.D. value measured was 7.3%. Recovery was also calculated, and the value was 89.4%. All results show that HS-SPME followed by GC/MS is a reliable method for the determination of paeonol in plasma. 3.3. Determination of paeonol in plasma by HS-SPME followed by GC/MS The plasma samples (1.0 mL for each plasma sample) collected from the rabbit after the administration of essential oil (1.0 g) from Cynanchum paniculatum, were introduced into 2 mL headspace vials with 10 L internal standard solution. Headspace extraction was performed by the PDMS-DVB fiber at the sample temperature of 60 ◦ C for 10 min, with the NaCl concentration of 30% and the stirring rate of 1100 rpm. Finally, the paeonol on the fiber were desorbed at 250 ◦ C for 3.0 min, and detected by GC/MS. Paeonol and IS can produce the molecular ion peak at m/z 166 and 138, respectively. The two characteristic ions were used in selected ion monitoring (SIM) mode for the quantitative analysis of paeonol in plasma. Fig. 6 is the SIM chromatogram of a rabbit plasma sample after the oral administration of the TCM essential oil, using HS-SPME followed by GC/MS. The retention times of paeonol and IS are 9.82 and 7.34 min, respectively. To quantify paeonol concentration in plasma samples, HS-SPME of calibration solutions (1.0 mL, the paeonol concentration ranged from 0.02 to 20 g mL−1 ) were performed at the optimal HS-SPME parameters. The calibration curve was obtained: Y = 0.186X + 0.004, R2 = 0.998 (Y, the peak area ratio of paeonol to IS; X, the paeonol concentration, g mL−1 ). The
L. Dong et al. / Analytica Chimica Acta 585 (2007) 76–80
Fig. 7. Representative data showing plasma paeonol concentration–time profiles of rabbit after the oral administration of the essential oil (1.0 g) from Cynanchum paniculatum.
paeonol concentrations in plasma samples were calculated by the internal standard method. Fig. 7 is the rabbit plasma paeonol concentration–time profiles after the oral administration of the TCM essential oil (1.0 g). These results are similar to those by LLE and SPE [13–15]. This shows that HS-SPME followed by GC/MS is an alternative method for the quantification of paeonol in biological fluids, can be used as a powerful tool for pharmacokinetic study. 4. Conclusions In the work, we successfully developed GC/MS following HS-SPME for determination of paeonol in plasma. Due to the extraction and concentration in a single step, the proposed method needed very short time (only 10 min) to prepare sample. Moreover, little volume of sample (1.0 mL), and no organic solvent were required. Compared with LLE and SPE, the method needed a simple operation, and provided low detection limit (2.0 ng mL−1 ). It has been shown that GC/MS following HSSPME is a simple, fast, sensitive and solvent-free method for quantitative analysis of paeonol in plasma samples, and also an alternative approach to the pharmacokinetic study of paeonol from TCM essential oil. References  F. Liu, T.B. Ng, Life Sci. 66 (2000) 725.  S.K. Lee, K.A. Nam, Y.H. Heo, Planta Med. 69 (2003) 21.  J.R. Dean, B. Liu, Phytochem. Anal. 11 (2000) 1.
 T. Okubo, F. Nagai, T. Seto, K. Satoh, Biol. Pharm. Bull. 23 (2000) 199.  G.L. Yang, X.R. Song, H.Y. Zhang, D.X. Sun, Chin. J. Anal. Chem. 27 (1999) 1.  H.K. Wu, S.J. Shen, J. Chromatogr. A 753 (1996) 139.  X.N. Wu, H.L. Chen, Z.D. Hu, Biomed. Chromatogr. 17 (2003) 504.  K. Yu, Y.W. Wang, Y.Y. Cheng, Chromatographia 63 (2006) 359.  J.W. Zhao, W.B. Yan, G.P. Dai, Chromatographia 62 (2005) 543.  H.W. Zhao, W.B. Yan, J. Pharm. Biomed. Anal. 38 (2005) 571.  X.F. Zhao, Y.Q. Sun, Anal. Sci. 19 (2003) 1313.  C.H. Deng, N. Yao, B. Wang, X.M. Zhang, J. Chromatogr. A 1103 (2006) 15.  C. Zhang, Chin. Herbs 18 (1996) 47.  Y.H. Jie, X.C. Tan, Chin. J. Chromatogr. 23 (2005) 212.  H.J. Liao, Z.L. Liu, Y.H. Li, Acta Pharm. Sin. 32 (1997) 928.  H. Lord, J. Pawliszyn, J. Chromatogr. A 902 (2000) 17.  G. Theodoridis, E.H.M. Koster, G.J. de Jong, J. Chromatogr. B 745 (2000) 49.  S. Ulrich, J. Chromatogr. A 902 (2000) 167.  N.H. Snow, J. Chromatogr. A 885 (2000) 445.  J. Szpunar, Analyst 130 (2005) 442.  E. Gallardo, M. Barroso, C. Margalho, A. Cruz, D.N. Vieira, M. LopezRivadulla, Rapid Commun. Mass Spectrom. 20 (2006) 865.  E. Gallardo, M. Barroso, C. Margalho, A. Cruz, D.N. Vieira, M. LopezRivadulla, J. Chromatogr. B 832 (2006) 162.  G. Frison, L. Tedeschi, S. Maietti, S.D. Ferrara, Rapid Commun. Mass Spectrom. 14 (2000) 2401.  G. Frison, L. Tedeschi, D. Favretto, A. Reheman, S.D. Ferrara, Rapid Commun. Mass Spectrom. 19 (2005) 919.  C.H. Deng, N. Li, X.M. Zhang, Rapid Commun. Mass Spectrom. 18 (2004) 2558.  C.H. Deng, X.M. Zhang, N. Li, J. Chromatogr. B 808 (2004) 269.  C.H. Deng, S. Lin, X.M. Zhang, Rapid Commun. Mass Spectrom. 20 (2006) 2120.  C.H. Deng, W. Zhang, J. Zhang, X.M. Zhang, J. Chromatogr. B 805 (2004) 235.  N. Li, C.H. Deng, X.Y. Yin, N. Yao, X.Z. Shen, X.M. Zhang, Anal. Biochem. 342 (2005) 318.  M. Krogh, K. Johansen, F. Tonnesen, K.E. Rasmussen, J. Chromatogr. B 673 (1995) 299.  C.G. Zambonin, A. Aresta, F. Palmisano, J. Chromatogr. B 806 (2004) 89.  M. Walles, W.M. Mullett, K. Levsen, J. Borlak, G. Wunsch, J. Pawliszyn, J. Pharm. Biomed. Anal. 30 (2002) 307.  B.S. De Martinis, M.A.M. Ruzzene, C.C.S. Martin, Anal. Chim. Acta 522 (2004) 163.  N. Pizarro, J. Ortuno, M. Farre, M. Mas, J. Cami, R. de la Torre, J. Pharm. Biomed. Anal. 21 (1999) 739.  G. Frison, L. Tedeschi, S. Maietti, S.D. Ferrara, Rapid Commun. Mass Spectrom. 15 (2001) 2497.  C. Kohlert, G. Abel, E. Schmid, M. Veit, J. Chromatogr. B 767 (2002) 11.  C. Kohlert, G. Schindler, R.W. Marz, G. Abel, B. Brinkhaus, H. Derendorf, E.U. Grafe, M. Veit, J. Clin. Pharm. 42 (7) (2002) 31.  Editorial Committee of the Pharmacopoeia of People’s Republic of China, The Pharmacopoeia of People’s Republic of China (Part I), Chemical Industry Press, Beijing, 2000.  B.Q. Wang, J. Xian Med. Univ. 16 (1995) 47.