- Email: [email protected]
Large-scale separation of hydroxyanthraquinones from Rheum palmatum L. by pH-zone-refining counter-current chromatography
Available online at www.sciencedirect.com
Journal of Chromatography A, 1176 (2007) 163–168
Large-scale separation of hydroxyanthraquinones from Rheum palmatum L. by pH-zone-refining counter-current chromatography Shengqiang Tong, Jizhong Yan ∗ College of Pharmaceutical Science, Zhejiang University of Technology, Hangzhou 310032, China Received 20 September 2007; received in revised form 24 October 2007; accepted 30 October 2007 Available online 5 November 2007
Abstract pH-zone-refining counter-current chromatography was successfully applied to purify four hydroxyanthraquinones, rhein, emodin, aloeemodin and chrysophanol, from three crude extracts of Rheum palmatum L.. After the two-phase solvent system methyl tert-butyl ether–tetrahydrofuran–water at an optimized volume ratio of 2:2:3 (v/v) was equilibrated, trifluoroacetic acid (10 mM) was added to the organic phase as a retainer and ammonia (10 mM), sodium carbonate (15 mM) and sodium hydroxide (15 mM) were added to the aqueous phase as the eluter, respectively, for three individual runs. Three separation runs of 1.25, 1.53 and 1.41 g of the three crude samples yielded four hydroxyanthraquinones: 0.70 g rhein, 0.81 g emodin, 0.41 g aloe-emodin and 0.94 g chrysophanol at a high purity of over 99.0, 98.5, 98.2 and 97.8% (determined by HPLC), respectively. The structures were identified by electrospray ionization MS–MS and 1 H NMR. © 2007 Elsevier B.V. All rights reserved. Keywords: Counter-current chromatography; Preparative chromatography; Hydroxyanthraquinones; Purification
1. Introduction Rheum palmatum L. is a traditional Chinese medicinal herb, and is officially listed in the Chinese Pharmacopoeia [1]. Its species are widely distributed in China, Kirgheeze desert and Europe. Chinese rhubarb is the only one kind recognized by the United States Pharmacopeia. Pharmacological test revealed that R. palmatum L. had pharmacological activities such as cathartic, anti-psychotic, anti-inflammatory, antimicrobial, hemostasis and so on, so it has been used for the treatment of dysentery, cholera, uraemia, leukaemia, diabetes, lung cancer and widely used in combination with other crude drugs for the treatment of wound healing from ancient times in China [2,3]. The major active components of the herb are hydroxyanthraquinones (HAQs), including rhein, emodin, aloe-emodin and chrysophanol (see Fig. 1), which are often used as criteria in the quality control of R. palmatum L.. HAQs are also active components in a large number of plant-derived drugs such as laxatives from Rheum, Cassia, Aloe and Polygonum species [4]. Pharmacological tests
∗
Corresponding author. Tel.: +86 571 8832 0613; fax: +86 571 8832 0913. E-mail address: [email protected] (J. Yan).
0021-9673/$ – see front matter © 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.chroma.2007.10.095
revealed that the four HAQs present various biological activities [3,5–6]. pH-zone-refining counter-current chromatography (CCC) was first introduced by Ito and co-workers [7–9]. It is generally employed as a large-scale preparative technique for separating ionizable analytes. The method elutes highly concentrated rectangular peaks fused together with minimum overlapping while impurities are concentrated and eluted between the outsides the major peaks according to their pKa and hydrophobicity. In addition, the method provides various special features such as yielding highly concentrated fractions, concentrating minor impurities for detection, and allowing the separation to be monitored by the pH of the effluent when there are no chromophores [10,11]. In the previous studies, this technique was applied to the separation of many natural products [12–14], peptide derivatives [15,16], synthetic products [17–21] and isomeric compounds [18,22,23]. Isolation of the above HAQs by the standard high-speed (HS) CCC technique had been reported in the literature [5,24,25]. However, only small amount of HAQs with high purity was obtained using the standard HSCCC technique. In other words, only about 0.01–0.02 g of rhein, emodin, aloe-emodin and chrysophanol with high purity were obtained during one separation run by standard HSCCC. The need for a large quantity
164
S. Tong, J. Yan / J. Chromatogr. A 1176 (2007) 163–168
BSZ-160 fraction collector (Shanghai Huxi Tech, Shanghai, China). The high-performance liquid chromatography (HPLC) system used was a CLASS-VP Ver.6.1 system (Shimadzu, Japan) comprised a Shimadzu SPD10Avp UV detector, a Shimadzu LC10ATvp Multisolvent Delivery System, a Shimadzu SCL-10Avp controller, a Shimadzu LC pump, and a CLASS-VP Ver.6.1 workstation. 2.2. Reagents and materials
Fig. 1. Chemical structures of hydroxyanthraquinones.
of the material required for clinical trial prompted the development of a new pH-zone-refining CCC method which provides over a 10-fold increase in sample loading capacity compared to the standard HSCCC technique. The present paper reports largescale preparative separation of four HAQs including 0.70 g rhein, 0.81 g emodin, 0.41 g aloe-emodin and 0.94 g chrysophanol with high purity from crude extracts by pH-zone-refining CCC. 2. Experimental 2.1. Apparatus The counter-current chromatography apparatus used in the present study is a multilayer coil planet centrifuge for performing HSCCC made in our laboratory (Zhejiang University of Technology, Hangzhou, China). It holds a column holder and a counterweight in the symmetrical positions at a distance of 8.5 cm from the central axis of the centrifuge. The separation column was prepared by winding a single piece of polytetrafluoroethylene (PTFE) tubing of 155 m in length × 1.8 mm I.D. around the column holder hub making three layers between a pair of flanges spaced 10 in. apart (1 in. = 2.54 cm). The total capacity of column is about 310 ml. The β values of this column range from 0.41 to 0.55 (β = r/R, R = 8.5 cm, where r is the distance from the coil to the holder shaft, and R, the revolution radius or the distance between the holder shaft and central axis of the centrifuge). A pair of flow tubes from the column is first led through the hollow column holder shaft downward, and then makes an arc to enter the side hole of the central pipe, finally exiting the centrifuge at the top plate where they are tightly supported by a pair of clamps. The revolution speed of the apparatus was regulated with a speed controller where an optimum speed of 750 rpm was used in the present studies. The solvent was pumped into the column with a Model NS-1007 constant-flow pump (Beijing Shengyitong Technique Co., Beijing, China). Continuous monitoring of the effluent was achieved with a Model UV-II detector Monitor (Shanghai Institute of Biochemistry of Academy of Science, Shanghai, China) at 254 nm. N2000 workstation (Zhejiang University Zhida Information & Technology Co., Hangzhou, China) was employed to record the chromatogram. Eluate was collected with a Model
All organic solvents used for pH-zone-refining CCC were of analytical grade. Trifluoroacetic acid (TFA), ammonia, sodium carbonate and sodium hydroxide were of reagent grade. Methanol used for HPLC analysis was of chromatographic grade. Other reagents were purchased from Hangzhou HuiPu Chemical Factory, Hangzhou, China. Methyl tert-butyl ether (MtBE) was redistilled before use. R. palmatum L. was purchased from Hangzhou Huadong drugstore, Hangzhou, China. 2.3. Extraction of crude samples from R. palmatum L. Preparation of three crude samples was carried out according to the literature [26]. The dried roots of R. palmatum L. were ground to powder (about 50 mesh). About 500 g of the powder was extracted with the mixture of 20% H2 SO4 and chloroform (1:5, v/v) for 1 h three times under reflux. The chloroform extracts were combined and evaporated under reduced pressure to about 500 ml. Then the chloroform extract was successively extracted with 5% NaHCO3 (aqueous solution I), 5% Na2 CO3 (aqueous solution II) and 5% NaOH (aqueous solution III) each for five times. The three aqueous solutions were acidified respectively with 20% H2 SO4 to pH 3 and then the three acidified aqueous solutions were extracted respectively with chloroform for five times. Each of the three chloroform extracts were evaporated to dryness to give three original samples (1.25 g sample I, 1.53 g sample II and 1.41 g sample III). These three crude samples were directly subjected to pH-zone-refining CCC independently. 2.4. Preparation of two-phase solvent system and sample solution The composition of the two-phase solvent system was selected according to the partition coefficient (K) of target compounds [10]. For the present study, we selected a two-phase solvent system composed of MtBE–tetrahydrofuran (THF)–water (2:2:3, v/v). The solvent mixture was thoroughly equilibrated in a separation funnel and the two phases separated shortly before use. Then, TFA (retainer) was added to the upper organic stationary phase to obtain a final concentration of 10 mM. And three different eluter, including NH4 OH for sample I, Na2 CO3 for sample II and NaOH for sample III was added to the lower aqueous mobile phase to obtain a final concentration of 10 mM NH4 OH, 15 mM Na2 CO3 and 15 mM NaOH. The three sample solutions were prepared by dissolving 1.25 g sample I in 40 ml of the stationary phase (10 mM TFA)
S. Tong, J. Yan / J. Chromatogr. A 1176 (2007) 163–168
and 20 ml water, 1.53 g sample II in 30 ml of the stationary phase (10 mM TFA) and 10 ml water, 1.41 g of sample III in 60 ml of the stationary phase (10 mM TFA) and 20 ml water.
165
3. Results and discussion
in these HAQs do not form stable salts with ammonium hydroxide. This problem was overcome by using a stronger base, such as Na2 CO3 and NaOH, as eluter in the aqueous mobile phase. Our experiment proved that 15 mM Na2 CO3 was suitable for emodin and aloe-emodin as an eluter while 15 mM NaOH was suitable for separation of chrysophanol from crude extract. On the other hand, successful separation of HAQs by pH-zone-refining CCC depends upon the selection of a suitable two-phase solvent system that should have suitable partition coefficient (K) values in both basic (Kbase 1) and acidic (Kacid 1) conditions as well as good solubility of the sample in the solvent system [10,11]. The selection of this two-phase solvent system for the target compound(s) is the most important step in CCC where searching for a suitable two-phase solvent system may be estimated as 90% of the entire work in CCC. A series of experiments were performed to determine the optimal two-phase solvent system for the pH-zone-refining CCC separation of rhein from crude extract. The following systems at different volume ratios were tested in turn: ether–water (1:1), MtBE–acetonitrile–water (1:0:1), (4:1:5) and (2:2:3); nhexane–ethyl acetate–methanol–water (3:7:5:5) and (5:5:5:5), and MtBE–THF–water (4:1:5) and (2:2:3). Among those ether–water (1:1), n-hexane–ethyl acetate–methanol–water (3:7:5:5) and MtBE–acetonitrile–water (1:0:1) produced suitable K values, but the maximum solubility of the sample in those solvent systems were about 150 mg. The solvent system MtBE–acetonitrile–water (4:1:5) and (2:2:3) had even poorer solubility for the sample. This could also explain why only small amount of HAQs with high purity could be obtained by standard HSCCC technique since the two-phase solvent system ether–water (1:1) were employed in the literature [23,24]. However, the solubility of the sample was significantly improved by adding THF to the system MtBE–water (1:1) instead of acetonitrile, and the K values were optimized by selecting MtBE–THF–water (2:2:3, v/v). The Kbase was 0.078 and Kacid was over 8.129 for rhein. The solvent system MtBE–THF–water (2:2:3, v/v) was also suitable for separation other three HAQs including emodin, aloe-emodin and chrysophanol by using the same retainer TFA (10 mM) but different eluter: 15 mM Na2 CO3 for emodin and aloe-emodin, and 15 mM NaOH for chrysophanol. The Kbase and Kacid for emodin, aloe-emodin and chrysophanol met the requirements of pH-zone-refining CCC separation.
3.1. Selection of the pH-zone-refining CCC solvent systems
3.2. Separation of HAQs by pH-zone-refining CCC
The method uses pH-zone-refining CCC was based on HAQs’ characteristic acidity which is determined by the position of the phenolic hydroxyl group in the molecule as well as the number of carboxylic and phenolic hydroxyl groups. The characteristic acidity of the HAQ molecules is different from each other. Among those target compounds, rhein is the most acidic, followed by emodin, aloe-emodin and chrysophanol successively. Rhein could be eluted with the ammonium hydroxide mobile phase used in the ordinary pH-zone-refining system. However, other three target compounds could not be eluted with the ammonium hydroxide mobile phase since weakly acidic phenol groups
Fig. 2 shows typical pH-zone-refining CCC of rhein from the crude extract (sample I). The shadow sect 1 shows 0.70 g rhein separated from 1.25 g sample I. Meanwhile, 0.15 g emodin was separated as showed in the shadow sect 2. The retention of the stationary phase was 35.8%. The separation time was 300 min. The recovery for rhein and emodin were 85.6 and 75.8%, respectively. Fig. 3 shows typical pH-zone-refining CCC of emodin along with aloe-emodin from the second crude extract (sample II). The shadow sect 2 shows 0.81 g emodin separated from 1.53 g sample II and the shadow sect 3 shows 0.41 g aloe-emodin separated at the same separation. The reten-
2.5. Separation procedure Separation of rhein along with emodin from sample I: the column was first entirely filled with the organic stationary phase containing TFA at 10 mM. This was followed by sample injection. Then the aqueous phase containing NH4 OH (eluter base) at 10 mM was pumped into the inlet of the column at a flowrate of 2.0 ml/min in the head-to-tail elution mode, while the apparatus was rotated at 750 rpm. The effluent from the outlet of the column was continuously monitored at 254 nm and collected into test tubes at 2-min intervals 4.0 ml/tube using a fraction collector. After the separation was completed, retention of the stationary phase was measured by collecting the column contents by forcing them out of the column with pressurized air. The same experimental procedure was also performed for samples II and III except for different eluter base: 15 mM Na2 CO3 for sample II and 15 mM NaOH for sample III. 2.6. Analyses and identification of CCC fractions The pH value of each fraction was manually determined with a portable Delta 320-s pH meter (Mettler–Toledo, Greifensee, Switzerland). Each peak fraction from CCC was analyzed by HPLC. The analyses were performed with a Shim-Pack CLCODS C18 column (250 mm × 6 mm I.D.). The mobile phase composed of methanol–0.1% H3 PO4 (80:20) was eluted with isocratic elution. The flow-rate was 1.0 ml/min and the effluent was monitored by a Shimadzu SPD10Avp UV detector at 254 nm. Identification of the target components was carried out by electrospray ionization (ESI)–MS–MS and 1 H NMR spectra as well as comparing their HPLC retention times and UV spectra with those of standard samples. ESI–MS–MS was performed with a Therm LCQ Deca XP Plus ion trap mass spectrometry system (Thermo Scientific, USA). 1 H NMR spectra were recorded with a Bruker Avance 500 MHz spectrometer (Bruker, USA) with TMS (tetramethylsilane) as the internal standard.
166
S. Tong, J. Yan / J. Chromatogr. A 1176 (2007) 163–168
Fig. 2. Chromatogram of the sample I by pH-zone-refining CCC showing the purification of rhein and emodin. The shadow sect 1 shows rhein separated from sample I and the shadow sect 2 shows emodin separated from sample I. pH-zone-refining CCC conditions: solvent system: MtBE–THF–water (2:2:3, v/v/v) (TFA (10 mM) was added to the organic phase as a retainer and NH4 OH (10 mM) to the aqueous phase as an eluter); flow rate: 2 ml/min; rotation speed: 750 rpm; temperature: 28 ◦ C; sample loading: 1.25 g; retention of the stationary phase: 35.8%; sample preparation: 1.25 g sample I dissolved in 40 ml of the stationary phase (10 mM TFA) and 20 ml water.
Fig. 3. Chromatogram of the sample II by pH-zone-refining CCC showing the purification of emodin and aloe-emodin. The shadow sect 2 shows emodin separated from sample II and the shadow sect 3 shows aloe-emodin separated from sample II. pH-zone-refining CCC conditions: solvent system: MtBE–THF–water (2:2:3, v/v/v) (TFA (10 mM) was added to the organic phase as a retainer and Na2 CO3 (15 mM) to the aqueous phase as an eluter); flow rate: 2 ml/min; rotation speed: 750 rpm; temperature: 28 ◦ C; sample loading: 1.53 g; retention of the stationary phase: 35.2%; sample preparation: 1.53 g sample II in 30 ml of the stationary phase (10 mM TFA) and 10 ml water.
tion of the stationary phase was 35.2%. The separation time was 350 min. The recovery for emodin and aloe-emodin were 92.5 and 74.3%, respectively. Fig. 4 shows typical pH-zone-refining CCC of chrysophanol from the crude extract (sample III). The shadow sect 4 shows 0.94 g chrysophanol separated from 1.41 g sample III. The separation time was 450 min. The recovery for chrysophanol was 76.3%. Under optimum separation con-
ditions, each target compound formed rectangular peak while impurities or byproducts were highly concentrated at their front and rear boundaries. During the third separation run, a stepwise elution by increasing the NaOH concentration in the mobile phase was required in order to obtain a satisfactory retention of the stationary phase, though this considerably increased the separation time. The sep-
Fig. 4. Chromatogram of the sample III by pH-zone-refining CCC showing the purification of chrysophanol. The shadow sect 4 shows chrysophanol separated from sample III. pH-zone-refining CCC conditions: solvent system: MtBE–THF–water (2:2:3, v/v/v) (TFA (10 mM) was added to the organic phase as a retainer and NaOH (15 mM) to the aqueous phase as an eluter); flow rate: 2 ml/min; rotation speed: 750 rpm; temperature: 28 ◦ C; sample loading: 1.41 g; retention of the stationary phase: 22.9%; sample preparation: 1.41 g of sample III in 60 ml of the stationary phase (10 mM TFA) and 20 ml water.
S. Tong, J. Yan / J. Chromatogr. A 1176 (2007) 163–168
167
Fig. 5. Results of HPLC analyses of three original samples and their pH-zone-refining CCC fractions: (a) sample I; (b) pH-zone-refining CCC fraction containing rhein; (c) pH-zone-refining CCC fraction containing emodin. (a ) Sample II; (b ) pH-zone-refining CCC fraction containing emodin; (c ) pH-zone-refining CCC fraction containing aloe-emodin. (a ) Sample III; (b ) pH-zone-refining CCC fraction containing chrysophanol. HPLC conditions: column: Shim-Pack CLC-ODS C18 column (250 mm × 6 mm I.D.); mobile phase: methanol–0.1% H3 PO4 (80:20) with isocratic elution; flow rate: 1.0 ml/min; UV wavelength: 254 nm; column temperature: 30 ◦ C.
aration was performed as follows: the column was filled with upper organic phase containing TFA at 10 mM followed by injection of the sample solution. The column was eluted, first with 240 ml lower aqueous phase containing no NaOH and then, after 120 min, the column was eluted with 15 mM NaOH. The crude sample may contain hydrophilic impurities which will emulsify the two phases in the column under a basic condition. This can be avoided by eluting the column with neutral solution until these compounds are eluted out from the column. Under this condition, the retention of the stationary phase was over 22.9%. The low retention of the stationary phase may be attributed to hydrophilic impurities eluted during 150–250 min in which large amount of stationary phase was carried out. The stationary phase could be totally carried out if a stepwise elution by increasing the NaOH concentration in the mobile phase was not used. 3.3. HPLC analyses of three samples and CCC fractions All fractions obtained from CCC separation were acidified with 20% H2 SO4 to pH 3 and extracted with the solvent mixture
of MtBE and THF (1:1, v/v). The extracts were evaporated to dryness under reduced pressure. Each of the three original samples and fractions obtained from pH-zone-refining CCC was analyzed by HPLC under the optimum analytical conditions (see Fig. 5). The analyses were performed with a Shim-Pack CLC-ODS C18 column (250 mm × 6 mm I.D.). The column temperature was controlled at 30 ◦ C with a model HT-230A column heater (Tianjin Hengao Technology, Tianjin, China). The mobile phase composed of methanol–0.1% H3 PO4 (80:20) was eluted with isocratic elution. The flow-rate was 1.0 ml/min and the effluent was monitored by a Shimadzu SPD10Avp UV detector at 254 nm. Routine sample calculations were made by comparison of the peak area with that of the standard. Fig. 5a shows HPLC chromatogram of sample I and there were three major constituents in the sample. Rhein (peak 1) represented 65.2% of the total and emodin (peak 2) was 15.2% of the total sample. The unknown compound t = 4.052 min represented 9.03% of the total. Fig. 5b and c shows HPLC analyses of the fractions containing the target compounds from pH-zone-refining CCC of sample I. HPLC results showed
168
S. Tong, J. Yan / J. Chromatogr. A 1176 (2007) 163–168
that rhein was separated from its crude extract with over 99.0% purity and emodin was over 96.0% purity. Fig. 5a shows HPLC chromatogram of sample II and there were two major components. Emodin (peak 2) represented 56.4% of the total and aloe-emodin (peak 3) was 35.6% of the total sample. Fig 5b and c shows HPLC analyses of the fractions containing the target compounds from pH-zone-refining CCC of sample II. HPLC results showed that emodin was separated from its crude extract with over 98.5% purity and aloe-emodin was over 98.2% purity. Fig. 5a shows HPLC chromatogram of sample III and there were two major constituents in the sample. Chrysophanol (peak 4) represented 85.4% of the total and aloe-emodin (peak 3) was 10.7% of the total sample. Fig. 5b shows HPLC analyses of the fractions containing the target compound chrysophanol from pH-zone-refining CCC of sample III. HPLC results showed that chrysophanol was separated from its crude extract with over 97.8% purity. 3.4. Structural identification Identification of CCC peak fractions was carried out by comparing their HPLC retention times and UV spectra with those of standard samples. The structural identification of the four HAQs was determined by ESI–MS–MS and 1 H NMR spectra. Peak 1: negative ESI–MS m/z: 282.7 [M − H]− , MS–MS fragmentation of m/z 282.7: 256.8 [M–H–CO + 2H]+ , 239.1 [M–H–CO2 ]− . 1 H NMR {[2 H6 ]dimethyl sulfoxide (DMSOd6 )}δ ppm: 7.38 (1H, d, J = 8.0 Hz, C7 –H), 7.72 (1H, d, J = 8.0 Hz, C5 –H), 7.81 (1H, t, J = 8.0 Hz, 8.0 Hz, C6 –H), 7.70 (1H, s, C2 –H), 8.09 (1H, s, C4 –H), 11.87 (2H, s, ␣-OH), 13.84 (1H, br s, –COOH). Peak 2: negative ESI–MS m/z: 269.0 [M − H]− , MS–MS fragmentation of m/z 269.0: 240.9 [M–H–CO]− , 224.9 [M–H–CO–O]− . 1 H NMR (DMSO-d6 ) δ ppm: 2.44 (3H, s, –CH3 ), 6.63 (1H, d, J = 2.5 Hz, C2 –H), 7.28 (1H, d, J = 2.5 Hz, C4 –H), 7.05 (1H, d, J = 2.0 Hz, C6 –H), 7.56 (1H, d, J = 2.0 Hz, C5 –H), 12.17 (1H, s, ␣-OH), 12.23 (1H, s, ␣-OH), 10.66 (1H, br s, -OH). Peak 3: negative ESI–MS m/z: 269.1 [M − H]− , MS-MS fragmentation of m/z 269.1: 239.9 [M–H–CO–H]− , 223.1 [M–H–CO–H2 O]− . 1 H NMR (DMSOd6 ) δ ppm: 4.73 (2H, s, –CH2 –OH), 7.30 (1H, d, J = 8.5 Hz, C7 –H), 7.81 (1H, d, J = 7.5 Hz, C5 –H), 7.71 (1H, t, J = 8.0 Hz, 7.5 Hz, C6 –H), 7.80 (1H, s, C4 –H), 7.35 (1H, s, C2 –H), 12.10 (1H, s, ␣-OH), 12.05 (1H, s, ␣-OH). Peak 4: negative ESI–MS m/z: 253.0 [M − H]− , MS–MS fragmentation of m/z 253.0: 224.7 [M–H–CO]− . 1 H NMR (C2 HCl3 ) δ ppm: 2.46 (3H, s, –CH3 ), 7.10 (1H, s, C2 –H), 7.65 (1H, s, C4 –H), 7.29 (1H, d, J = 8.0 Hz, C7 –H), 7.68 (1H, t, J = 8.0 Hz, 8.0 Hz, C6 –H), 7.82 (1H, d, J = 8.0 Hz, C5 –H), 12.02 (1H, s, ␣-OH), 12.12 (1H, s, ␣-OH). After comparing the data with spectral information
from literature [25,27,28], Peaks 1–4 were confirmed as rhein, emodin, aloe-emodin and chrysophanol, respectively. The results of our studies described above clearly demonstrated that pH-zone-refining CCC was very successful in the preparative separation of HAQs from the crude extract of R. palmatum L. Acknowledgement The authors are indebted to Dr. Yoichiro Ito for his valuable suggestions in the research work. References [1] China Pharmacopoeia Committee, Pharmacopoeia of the People’s Republic of China, First Division, 2000 ed., China Chemical Industry Press, Beijing, 1999, p. 18. [2] Z. Wang, G. Wang, H. Xu, P. Wang, Zhongguo Zhong Yao Za Zhi 21 (1996) 364. [3] T. Tang, L.W. Yin, J. Yang, G. Shan, Eur. J. Pharmacol. 567 (2007) 177. [4] Y.W. Wu, W.Y. Gao, X.H. Xiao, Y. Liu, Thermochim. Acta 429 (2005) 167. [5] Y. Wei, T.Y. Zhang, Y. Ito, J. Chromatogr. A 1017 (2003) 125. [6] X.H. Tan, D.S. Zhang, L. Zhang, F. An, H. Jiang, S.H. Wang, Zhongchengyao 28 (2006) 1039. [7] Y. Ito, A.Weisz, US Patent 5,332,504, 1994. [8] A. Weisz, A.L. Scher, K. Shinomiya, H.M. Fales, Y. Ito, J. Am. Chem. Soc. 116 (1994) 704. [9] Y. Ito, W.D. Conway, High-Speed Counter-Current Chromatography, Wiley, New York, 1996. [10] Y. Ito, Y. Ma, J. Chromatogr. A 753 (1996) 1. [11] Y. Ito, J. Chromatogr. A 1065 (2005) 145. [12] Y. Ma, Y. Ito, E. Sokolosky, H.M. Fales, J. Chromatogr. A 685 (1994) 259. [13] F.Q. Yang, Y. Ito, J. Chromatogr. A 923 (2001) 281. [14] X. Wang, Y.L. Geng, F.W. Li, Q.S. Gao, X.G. Shi, J. Chromatogr. A 1103 (2006) 166. [15] Y. Ma, Y. Ito, J. Chromatogr. A 702 (1995) 197. [16] Y. Ma, Y. Ito, J. Chromatogr. A 771 (1997) 81. [17] H. Oka, M. Suzuki, K.I. Harada, M. Iwaya, K. Fujii, T. Goto, Y. Ito, H. Matsumoto, Y. Ito, J. Chromatogr. A 946 (2002) 157. [18] A. Weisz, E.P. Mazzola, J.E. Matusik, Y. Ito, J. Chromatogr. A 923 (2001) 87. [19] A. Weisz, D. Andrzejewski, Y. Ito, J. Chromatogr. A 678 (1994) 77. [20] A. Weisz, D. Andrzejewski, R.J. Highet, Y. Ito, J. Chromatogr. A 658 (1994) 505. [21] S.Q. Tong, J.Z. Yan, J. Li, J.Z. Lou, J. Sep. Sci. 30 (2007) 1899. [22] C. Denekamp, A. Mandelbaum, A. Weisz, Y. Ito, J. Chromatogr. A 685 (1994) 253. [23] A. Weisz, E.P. Mazzola, C.M. Murphy, Y. Ito, J. Chromatogr. A 966 (2002) 111. [24] F.Q. Yang, T.Y. Zhang, G.L. Tian, H.F. Cao, Q.H. Liu, Y. Ito, J. Chromatogr. A 858 (1999) 103. [25] R.M. Liu, A.F. Li, A.L. Sun, J. Chromatogr. A 1052 (2004) 217. [26] R.S. Xu, Z.L. Chen (Eds.), Zhongcaoyao Youxiao Chengfen Tiqu Yu Fenli, Shanghai Science & Technology Press, Shanghai, 1983, p. 331. [27] Y.R. Ma, X.Y. Zhao, Z. Xu, Hecheng Huaxue 15 (2007) 244. [28] X.H. Ma, S.L. Shen, F.M. Han, Y. Chen, Hubei Da Xue Xue Bao (Nat. Sci.) 28 (2006) 403.
Journal of Chromatography A, 1176 (2007) 163–168
Large-scale separation of hydroxyanthraquinones from Rheum palmatum L. by pH-zone-refining counter-current chromatography Shengqiang Tong, Jizhong Yan ∗ College of Pharmaceutical Science, Zhejiang University of Technology, Hangzhou 310032, China Received 20 September 2007; received in revised form 24 October 2007; accepted 30 October 2007 Available online 5 November 2007
Abstract pH-zone-refining counter-current chromatography was successfully applied to purify four hydroxyanthraquinones, rhein, emodin, aloeemodin and chrysophanol, from three crude extracts of Rheum palmatum L.. After the two-phase solvent system methyl tert-butyl ether–tetrahydrofuran–water at an optimized volume ratio of 2:2:3 (v/v) was equilibrated, trifluoroacetic acid (10 mM) was added to the organic phase as a retainer and ammonia (10 mM), sodium carbonate (15 mM) and sodium hydroxide (15 mM) were added to the aqueous phase as the eluter, respectively, for three individual runs. Three separation runs of 1.25, 1.53 and 1.41 g of the three crude samples yielded four hydroxyanthraquinones: 0.70 g rhein, 0.81 g emodin, 0.41 g aloe-emodin and 0.94 g chrysophanol at a high purity of over 99.0, 98.5, 98.2 and 97.8% (determined by HPLC), respectively. The structures were identified by electrospray ionization MS–MS and 1 H NMR. © 2007 Elsevier B.V. All rights reserved. Keywords: Counter-current chromatography; Preparative chromatography; Hydroxyanthraquinones; Purification
1. Introduction Rheum palmatum L. is a traditional Chinese medicinal herb, and is officially listed in the Chinese Pharmacopoeia [1]. Its species are widely distributed in China, Kirgheeze desert and Europe. Chinese rhubarb is the only one kind recognized by the United States Pharmacopeia. Pharmacological test revealed that R. palmatum L. had pharmacological activities such as cathartic, anti-psychotic, anti-inflammatory, antimicrobial, hemostasis and so on, so it has been used for the treatment of dysentery, cholera, uraemia, leukaemia, diabetes, lung cancer and widely used in combination with other crude drugs for the treatment of wound healing from ancient times in China [2,3]. The major active components of the herb are hydroxyanthraquinones (HAQs), including rhein, emodin, aloe-emodin and chrysophanol (see Fig. 1), which are often used as criteria in the quality control of R. palmatum L.. HAQs are also active components in a large number of plant-derived drugs such as laxatives from Rheum, Cassia, Aloe and Polygonum species [4]. Pharmacological tests
∗
Corresponding author. Tel.: +86 571 8832 0613; fax: +86 571 8832 0913. E-mail address: [email protected] (J. Yan).
0021-9673/$ – see front matter © 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.chroma.2007.10.095
revealed that the four HAQs present various biological activities [3,5–6]. pH-zone-refining counter-current chromatography (CCC) was first introduced by Ito and co-workers [7–9]. It is generally employed as a large-scale preparative technique for separating ionizable analytes. The method elutes highly concentrated rectangular peaks fused together with minimum overlapping while impurities are concentrated and eluted between the outsides the major peaks according to their pKa and hydrophobicity. In addition, the method provides various special features such as yielding highly concentrated fractions, concentrating minor impurities for detection, and allowing the separation to be monitored by the pH of the effluent when there are no chromophores [10,11]. In the previous studies, this technique was applied to the separation of many natural products [12–14], peptide derivatives [15,16], synthetic products [17–21] and isomeric compounds [18,22,23]. Isolation of the above HAQs by the standard high-speed (HS) CCC technique had been reported in the literature [5,24,25]. However, only small amount of HAQs with high purity was obtained using the standard HSCCC technique. In other words, only about 0.01–0.02 g of rhein, emodin, aloe-emodin and chrysophanol with high purity were obtained during one separation run by standard HSCCC. The need for a large quantity
164
S. Tong, J. Yan / J. Chromatogr. A 1176 (2007) 163–168
BSZ-160 fraction collector (Shanghai Huxi Tech, Shanghai, China). The high-performance liquid chromatography (HPLC) system used was a CLASS-VP Ver.6.1 system (Shimadzu, Japan) comprised a Shimadzu SPD10Avp UV detector, a Shimadzu LC10ATvp Multisolvent Delivery System, a Shimadzu SCL-10Avp controller, a Shimadzu LC pump, and a CLASS-VP Ver.6.1 workstation. 2.2. Reagents and materials
Fig. 1. Chemical structures of hydroxyanthraquinones.
of the material required for clinical trial prompted the development of a new pH-zone-refining CCC method which provides over a 10-fold increase in sample loading capacity compared to the standard HSCCC technique. The present paper reports largescale preparative separation of four HAQs including 0.70 g rhein, 0.81 g emodin, 0.41 g aloe-emodin and 0.94 g chrysophanol with high purity from crude extracts by pH-zone-refining CCC. 2. Experimental 2.1. Apparatus The counter-current chromatography apparatus used in the present study is a multilayer coil planet centrifuge for performing HSCCC made in our laboratory (Zhejiang University of Technology, Hangzhou, China). It holds a column holder and a counterweight in the symmetrical positions at a distance of 8.5 cm from the central axis of the centrifuge. The separation column was prepared by winding a single piece of polytetrafluoroethylene (PTFE) tubing of 155 m in length × 1.8 mm I.D. around the column holder hub making three layers between a pair of flanges spaced 10 in. apart (1 in. = 2.54 cm). The total capacity of column is about 310 ml. The β values of this column range from 0.41 to 0.55 (β = r/R, R = 8.5 cm, where r is the distance from the coil to the holder shaft, and R, the revolution radius or the distance between the holder shaft and central axis of the centrifuge). A pair of flow tubes from the column is first led through the hollow column holder shaft downward, and then makes an arc to enter the side hole of the central pipe, finally exiting the centrifuge at the top plate where they are tightly supported by a pair of clamps. The revolution speed of the apparatus was regulated with a speed controller where an optimum speed of 750 rpm was used in the present studies. The solvent was pumped into the column with a Model NS-1007 constant-flow pump (Beijing Shengyitong Technique Co., Beijing, China). Continuous monitoring of the effluent was achieved with a Model UV-II detector Monitor (Shanghai Institute of Biochemistry of Academy of Science, Shanghai, China) at 254 nm. N2000 workstation (Zhejiang University Zhida Information & Technology Co., Hangzhou, China) was employed to record the chromatogram. Eluate was collected with a Model
All organic solvents used for pH-zone-refining CCC were of analytical grade. Trifluoroacetic acid (TFA), ammonia, sodium carbonate and sodium hydroxide were of reagent grade. Methanol used for HPLC analysis was of chromatographic grade. Other reagents were purchased from Hangzhou HuiPu Chemical Factory, Hangzhou, China. Methyl tert-butyl ether (MtBE) was redistilled before use. R. palmatum L. was purchased from Hangzhou Huadong drugstore, Hangzhou, China. 2.3. Extraction of crude samples from R. palmatum L. Preparation of three crude samples was carried out according to the literature [26]. The dried roots of R. palmatum L. were ground to powder (about 50 mesh). About 500 g of the powder was extracted with the mixture of 20% H2 SO4 and chloroform (1:5, v/v) for 1 h three times under reflux. The chloroform extracts were combined and evaporated under reduced pressure to about 500 ml. Then the chloroform extract was successively extracted with 5% NaHCO3 (aqueous solution I), 5% Na2 CO3 (aqueous solution II) and 5% NaOH (aqueous solution III) each for five times. The three aqueous solutions were acidified respectively with 20% H2 SO4 to pH 3 and then the three acidified aqueous solutions were extracted respectively with chloroform for five times. Each of the three chloroform extracts were evaporated to dryness to give three original samples (1.25 g sample I, 1.53 g sample II and 1.41 g sample III). These three crude samples were directly subjected to pH-zone-refining CCC independently. 2.4. Preparation of two-phase solvent system and sample solution The composition of the two-phase solvent system was selected according to the partition coefficient (K) of target compounds [10]. For the present study, we selected a two-phase solvent system composed of MtBE–tetrahydrofuran (THF)–water (2:2:3, v/v). The solvent mixture was thoroughly equilibrated in a separation funnel and the two phases separated shortly before use. Then, TFA (retainer) was added to the upper organic stationary phase to obtain a final concentration of 10 mM. And three different eluter, including NH4 OH for sample I, Na2 CO3 for sample II and NaOH for sample III was added to the lower aqueous mobile phase to obtain a final concentration of 10 mM NH4 OH, 15 mM Na2 CO3 and 15 mM NaOH. The three sample solutions were prepared by dissolving 1.25 g sample I in 40 ml of the stationary phase (10 mM TFA)
S. Tong, J. Yan / J. Chromatogr. A 1176 (2007) 163–168
and 20 ml water, 1.53 g sample II in 30 ml of the stationary phase (10 mM TFA) and 10 ml water, 1.41 g of sample III in 60 ml of the stationary phase (10 mM TFA) and 20 ml water.
165
3. Results and discussion
in these HAQs do not form stable salts with ammonium hydroxide. This problem was overcome by using a stronger base, such as Na2 CO3 and NaOH, as eluter in the aqueous mobile phase. Our experiment proved that 15 mM Na2 CO3 was suitable for emodin and aloe-emodin as an eluter while 15 mM NaOH was suitable for separation of chrysophanol from crude extract. On the other hand, successful separation of HAQs by pH-zone-refining CCC depends upon the selection of a suitable two-phase solvent system that should have suitable partition coefficient (K) values in both basic (Kbase 1) and acidic (Kacid 1) conditions as well as good solubility of the sample in the solvent system [10,11]. The selection of this two-phase solvent system for the target compound(s) is the most important step in CCC where searching for a suitable two-phase solvent system may be estimated as 90% of the entire work in CCC. A series of experiments were performed to determine the optimal two-phase solvent system for the pH-zone-refining CCC separation of rhein from crude extract. The following systems at different volume ratios were tested in turn: ether–water (1:1), MtBE–acetonitrile–water (1:0:1), (4:1:5) and (2:2:3); nhexane–ethyl acetate–methanol–water (3:7:5:5) and (5:5:5:5), and MtBE–THF–water (4:1:5) and (2:2:3). Among those ether–water (1:1), n-hexane–ethyl acetate–methanol–water (3:7:5:5) and MtBE–acetonitrile–water (1:0:1) produced suitable K values, but the maximum solubility of the sample in those solvent systems were about 150 mg. The solvent system MtBE–acetonitrile–water (4:1:5) and (2:2:3) had even poorer solubility for the sample. This could also explain why only small amount of HAQs with high purity could be obtained by standard HSCCC technique since the two-phase solvent system ether–water (1:1) were employed in the literature [23,24]. However, the solubility of the sample was significantly improved by adding THF to the system MtBE–water (1:1) instead of acetonitrile, and the K values were optimized by selecting MtBE–THF–water (2:2:3, v/v). The Kbase was 0.078 and Kacid was over 8.129 for rhein. The solvent system MtBE–THF–water (2:2:3, v/v) was also suitable for separation other three HAQs including emodin, aloe-emodin and chrysophanol by using the same retainer TFA (10 mM) but different eluter: 15 mM Na2 CO3 for emodin and aloe-emodin, and 15 mM NaOH for chrysophanol. The Kbase and Kacid for emodin, aloe-emodin and chrysophanol met the requirements of pH-zone-refining CCC separation.
3.1. Selection of the pH-zone-refining CCC solvent systems
3.2. Separation of HAQs by pH-zone-refining CCC
The method uses pH-zone-refining CCC was based on HAQs’ characteristic acidity which is determined by the position of the phenolic hydroxyl group in the molecule as well as the number of carboxylic and phenolic hydroxyl groups. The characteristic acidity of the HAQ molecules is different from each other. Among those target compounds, rhein is the most acidic, followed by emodin, aloe-emodin and chrysophanol successively. Rhein could be eluted with the ammonium hydroxide mobile phase used in the ordinary pH-zone-refining system. However, other three target compounds could not be eluted with the ammonium hydroxide mobile phase since weakly acidic phenol groups
Fig. 2 shows typical pH-zone-refining CCC of rhein from the crude extract (sample I). The shadow sect 1 shows 0.70 g rhein separated from 1.25 g sample I. Meanwhile, 0.15 g emodin was separated as showed in the shadow sect 2. The retention of the stationary phase was 35.8%. The separation time was 300 min. The recovery for rhein and emodin were 85.6 and 75.8%, respectively. Fig. 3 shows typical pH-zone-refining CCC of emodin along with aloe-emodin from the second crude extract (sample II). The shadow sect 2 shows 0.81 g emodin separated from 1.53 g sample II and the shadow sect 3 shows 0.41 g aloe-emodin separated at the same separation. The reten-
2.5. Separation procedure Separation of rhein along with emodin from sample I: the column was first entirely filled with the organic stationary phase containing TFA at 10 mM. This was followed by sample injection. Then the aqueous phase containing NH4 OH (eluter base) at 10 mM was pumped into the inlet of the column at a flowrate of 2.0 ml/min in the head-to-tail elution mode, while the apparatus was rotated at 750 rpm. The effluent from the outlet of the column was continuously monitored at 254 nm and collected into test tubes at 2-min intervals 4.0 ml/tube using a fraction collector. After the separation was completed, retention of the stationary phase was measured by collecting the column contents by forcing them out of the column with pressurized air. The same experimental procedure was also performed for samples II and III except for different eluter base: 15 mM Na2 CO3 for sample II and 15 mM NaOH for sample III. 2.6. Analyses and identification of CCC fractions The pH value of each fraction was manually determined with a portable Delta 320-s pH meter (Mettler–Toledo, Greifensee, Switzerland). Each peak fraction from CCC was analyzed by HPLC. The analyses were performed with a Shim-Pack CLCODS C18 column (250 mm × 6 mm I.D.). The mobile phase composed of methanol–0.1% H3 PO4 (80:20) was eluted with isocratic elution. The flow-rate was 1.0 ml/min and the effluent was monitored by a Shimadzu SPD10Avp UV detector at 254 nm. Identification of the target components was carried out by electrospray ionization (ESI)–MS–MS and 1 H NMR spectra as well as comparing their HPLC retention times and UV spectra with those of standard samples. ESI–MS–MS was performed with a Therm LCQ Deca XP Plus ion trap mass spectrometry system (Thermo Scientific, USA). 1 H NMR spectra were recorded with a Bruker Avance 500 MHz spectrometer (Bruker, USA) with TMS (tetramethylsilane) as the internal standard.
166
S. Tong, J. Yan / J. Chromatogr. A 1176 (2007) 163–168
Fig. 2. Chromatogram of the sample I by pH-zone-refining CCC showing the purification of rhein and emodin. The shadow sect 1 shows rhein separated from sample I and the shadow sect 2 shows emodin separated from sample I. pH-zone-refining CCC conditions: solvent system: MtBE–THF–water (2:2:3, v/v/v) (TFA (10 mM) was added to the organic phase as a retainer and NH4 OH (10 mM) to the aqueous phase as an eluter); flow rate: 2 ml/min; rotation speed: 750 rpm; temperature: 28 ◦ C; sample loading: 1.25 g; retention of the stationary phase: 35.8%; sample preparation: 1.25 g sample I dissolved in 40 ml of the stationary phase (10 mM TFA) and 20 ml water.
Fig. 3. Chromatogram of the sample II by pH-zone-refining CCC showing the purification of emodin and aloe-emodin. The shadow sect 2 shows emodin separated from sample II and the shadow sect 3 shows aloe-emodin separated from sample II. pH-zone-refining CCC conditions: solvent system: MtBE–THF–water (2:2:3, v/v/v) (TFA (10 mM) was added to the organic phase as a retainer and Na2 CO3 (15 mM) to the aqueous phase as an eluter); flow rate: 2 ml/min; rotation speed: 750 rpm; temperature: 28 ◦ C; sample loading: 1.53 g; retention of the stationary phase: 35.2%; sample preparation: 1.53 g sample II in 30 ml of the stationary phase (10 mM TFA) and 10 ml water.
tion of the stationary phase was 35.2%. The separation time was 350 min. The recovery for emodin and aloe-emodin were 92.5 and 74.3%, respectively. Fig. 4 shows typical pH-zone-refining CCC of chrysophanol from the crude extract (sample III). The shadow sect 4 shows 0.94 g chrysophanol separated from 1.41 g sample III. The separation time was 450 min. The recovery for chrysophanol was 76.3%. Under optimum separation con-
ditions, each target compound formed rectangular peak while impurities or byproducts were highly concentrated at their front and rear boundaries. During the third separation run, a stepwise elution by increasing the NaOH concentration in the mobile phase was required in order to obtain a satisfactory retention of the stationary phase, though this considerably increased the separation time. The sep-
Fig. 4. Chromatogram of the sample III by pH-zone-refining CCC showing the purification of chrysophanol. The shadow sect 4 shows chrysophanol separated from sample III. pH-zone-refining CCC conditions: solvent system: MtBE–THF–water (2:2:3, v/v/v) (TFA (10 mM) was added to the organic phase as a retainer and NaOH (15 mM) to the aqueous phase as an eluter); flow rate: 2 ml/min; rotation speed: 750 rpm; temperature: 28 ◦ C; sample loading: 1.41 g; retention of the stationary phase: 22.9%; sample preparation: 1.41 g of sample III in 60 ml of the stationary phase (10 mM TFA) and 20 ml water.
S. Tong, J. Yan / J. Chromatogr. A 1176 (2007) 163–168
167
Fig. 5. Results of HPLC analyses of three original samples and their pH-zone-refining CCC fractions: (a) sample I; (b) pH-zone-refining CCC fraction containing rhein; (c) pH-zone-refining CCC fraction containing emodin. (a ) Sample II; (b ) pH-zone-refining CCC fraction containing emodin; (c ) pH-zone-refining CCC fraction containing aloe-emodin. (a ) Sample III; (b ) pH-zone-refining CCC fraction containing chrysophanol. HPLC conditions: column: Shim-Pack CLC-ODS C18 column (250 mm × 6 mm I.D.); mobile phase: methanol–0.1% H3 PO4 (80:20) with isocratic elution; flow rate: 1.0 ml/min; UV wavelength: 254 nm; column temperature: 30 ◦ C.
aration was performed as follows: the column was filled with upper organic phase containing TFA at 10 mM followed by injection of the sample solution. The column was eluted, first with 240 ml lower aqueous phase containing no NaOH and then, after 120 min, the column was eluted with 15 mM NaOH. The crude sample may contain hydrophilic impurities which will emulsify the two phases in the column under a basic condition. This can be avoided by eluting the column with neutral solution until these compounds are eluted out from the column. Under this condition, the retention of the stationary phase was over 22.9%. The low retention of the stationary phase may be attributed to hydrophilic impurities eluted during 150–250 min in which large amount of stationary phase was carried out. The stationary phase could be totally carried out if a stepwise elution by increasing the NaOH concentration in the mobile phase was not used. 3.3. HPLC analyses of three samples and CCC fractions All fractions obtained from CCC separation were acidified with 20% H2 SO4 to pH 3 and extracted with the solvent mixture
of MtBE and THF (1:1, v/v). The extracts were evaporated to dryness under reduced pressure. Each of the three original samples and fractions obtained from pH-zone-refining CCC was analyzed by HPLC under the optimum analytical conditions (see Fig. 5). The analyses were performed with a Shim-Pack CLC-ODS C18 column (250 mm × 6 mm I.D.). The column temperature was controlled at 30 ◦ C with a model HT-230A column heater (Tianjin Hengao Technology, Tianjin, China). The mobile phase composed of methanol–0.1% H3 PO4 (80:20) was eluted with isocratic elution. The flow-rate was 1.0 ml/min and the effluent was monitored by a Shimadzu SPD10Avp UV detector at 254 nm. Routine sample calculations were made by comparison of the peak area with that of the standard. Fig. 5a shows HPLC chromatogram of sample I and there were three major constituents in the sample. Rhein (peak 1) represented 65.2% of the total and emodin (peak 2) was 15.2% of the total sample. The unknown compound t = 4.052 min represented 9.03% of the total. Fig. 5b and c shows HPLC analyses of the fractions containing the target compounds from pH-zone-refining CCC of sample I. HPLC results showed
168
S. Tong, J. Yan / J. Chromatogr. A 1176 (2007) 163–168
that rhein was separated from its crude extract with over 99.0% purity and emodin was over 96.0% purity. Fig. 5a shows HPLC chromatogram of sample II and there were two major components. Emodin (peak 2) represented 56.4% of the total and aloe-emodin (peak 3) was 35.6% of the total sample. Fig 5b and c shows HPLC analyses of the fractions containing the target compounds from pH-zone-refining CCC of sample II. HPLC results showed that emodin was separated from its crude extract with over 98.5% purity and aloe-emodin was over 98.2% purity. Fig. 5a shows HPLC chromatogram of sample III and there were two major constituents in the sample. Chrysophanol (peak 4) represented 85.4% of the total and aloe-emodin (peak 3) was 10.7% of the total sample. Fig. 5b shows HPLC analyses of the fractions containing the target compound chrysophanol from pH-zone-refining CCC of sample III. HPLC results showed that chrysophanol was separated from its crude extract with over 97.8% purity. 3.4. Structural identification Identification of CCC peak fractions was carried out by comparing their HPLC retention times and UV spectra with those of standard samples. The structural identification of the four HAQs was determined by ESI–MS–MS and 1 H NMR spectra. Peak 1: negative ESI–MS m/z: 282.7 [M − H]− , MS–MS fragmentation of m/z 282.7: 256.8 [M–H–CO + 2H]+ , 239.1 [M–H–CO2 ]− . 1 H NMR {[2 H6 ]dimethyl sulfoxide (DMSOd6 )}δ ppm: 7.38 (1H, d, J = 8.0 Hz, C7 –H), 7.72 (1H, d, J = 8.0 Hz, C5 –H), 7.81 (1H, t, J = 8.0 Hz, 8.0 Hz, C6 –H), 7.70 (1H, s, C2 –H), 8.09 (1H, s, C4 –H), 11.87 (2H, s, ␣-OH), 13.84 (1H, br s, –COOH). Peak 2: negative ESI–MS m/z: 269.0 [M − H]− , MS–MS fragmentation of m/z 269.0: 240.9 [M–H–CO]− , 224.9 [M–H–CO–O]− . 1 H NMR (DMSO-d6 ) δ ppm: 2.44 (3H, s, –CH3 ), 6.63 (1H, d, J = 2.5 Hz, C2 –H), 7.28 (1H, d, J = 2.5 Hz, C4 –H), 7.05 (1H, d, J = 2.0 Hz, C6 –H), 7.56 (1H, d, J = 2.0 Hz, C5 –H), 12.17 (1H, s, ␣-OH), 12.23 (1H, s, ␣-OH), 10.66 (1H, br s, -OH). Peak 3: negative ESI–MS m/z: 269.1 [M − H]− , MS-MS fragmentation of m/z 269.1: 239.9 [M–H–CO–H]− , 223.1 [M–H–CO–H2 O]− . 1 H NMR (DMSOd6 ) δ ppm: 4.73 (2H, s, –CH2 –OH), 7.30 (1H, d, J = 8.5 Hz, C7 –H), 7.81 (1H, d, J = 7.5 Hz, C5 –H), 7.71 (1H, t, J = 8.0 Hz, 7.5 Hz, C6 –H), 7.80 (1H, s, C4 –H), 7.35 (1H, s, C2 –H), 12.10 (1H, s, ␣-OH), 12.05 (1H, s, ␣-OH). Peak 4: negative ESI–MS m/z: 253.0 [M − H]− , MS–MS fragmentation of m/z 253.0: 224.7 [M–H–CO]− . 1 H NMR (C2 HCl3 ) δ ppm: 2.46 (3H, s, –CH3 ), 7.10 (1H, s, C2 –H), 7.65 (1H, s, C4 –H), 7.29 (1H, d, J = 8.0 Hz, C7 –H), 7.68 (1H, t, J = 8.0 Hz, 8.0 Hz, C6 –H), 7.82 (1H, d, J = 8.0 Hz, C5 –H), 12.02 (1H, s, ␣-OH), 12.12 (1H, s, ␣-OH). After comparing the data with spectral information
from literature [25,27,28], Peaks 1–4 were confirmed as rhein, emodin, aloe-emodin and chrysophanol, respectively. The results of our studies described above clearly demonstrated that pH-zone-refining CCC was very successful in the preparative separation of HAQs from the crude extract of R. palmatum L. Acknowledgement The authors are indebted to Dr. Yoichiro Ito for his valuable suggestions in the research work. References [1] China Pharmacopoeia Committee, Pharmacopoeia of the People’s Republic of China, First Division, 2000 ed., China Chemical Industry Press, Beijing, 1999, p. 18. [2] Z. Wang, G. Wang, H. Xu, P. Wang, Zhongguo Zhong Yao Za Zhi 21 (1996) 364. [3] T. Tang, L.W. Yin, J. Yang, G. Shan, Eur. J. Pharmacol. 567 (2007) 177. [4] Y.W. Wu, W.Y. Gao, X.H. Xiao, Y. Liu, Thermochim. Acta 429 (2005) 167. [5] Y. Wei, T.Y. Zhang, Y. Ito, J. Chromatogr. A 1017 (2003) 125. [6] X.H. Tan, D.S. Zhang, L. Zhang, F. An, H. Jiang, S.H. Wang, Zhongchengyao 28 (2006) 1039. [7] Y. Ito, A.Weisz, US Patent 5,332,504, 1994. [8] A. Weisz, A.L. Scher, K. Shinomiya, H.M. Fales, Y. Ito, J. Am. Chem. Soc. 116 (1994) 704. [9] Y. Ito, W.D. Conway, High-Speed Counter-Current Chromatography, Wiley, New York, 1996. [10] Y. Ito, Y. Ma, J. Chromatogr. A 753 (1996) 1. [11] Y. Ito, J. Chromatogr. A 1065 (2005) 145. [12] Y. Ma, Y. Ito, E. Sokolosky, H.M. Fales, J. Chromatogr. A 685 (1994) 259. [13] F.Q. Yang, Y. Ito, J. Chromatogr. A 923 (2001) 281. [14] X. Wang, Y.L. Geng, F.W. Li, Q.S. Gao, X.G. Shi, J. Chromatogr. A 1103 (2006) 166. [15] Y. Ma, Y. Ito, J. Chromatogr. A 702 (1995) 197. [16] Y. Ma, Y. Ito, J. Chromatogr. A 771 (1997) 81. [17] H. Oka, M. Suzuki, K.I. Harada, M. Iwaya, K. Fujii, T. Goto, Y. Ito, H. Matsumoto, Y. Ito, J. Chromatogr. A 946 (2002) 157. [18] A. Weisz, E.P. Mazzola, J.E. Matusik, Y. Ito, J. Chromatogr. A 923 (2001) 87. [19] A. Weisz, D. Andrzejewski, Y. Ito, J. Chromatogr. A 678 (1994) 77. [20] A. Weisz, D. Andrzejewski, R.J. Highet, Y. Ito, J. Chromatogr. A 658 (1994) 505. [21] S.Q. Tong, J.Z. Yan, J. Li, J.Z. Lou, J. Sep. Sci. 30 (2007) 1899. [22] C. Denekamp, A. Mandelbaum, A. Weisz, Y. Ito, J. Chromatogr. A 685 (1994) 253. [23] A. Weisz, E.P. Mazzola, C.M. Murphy, Y. Ito, J. Chromatogr. A 966 (2002) 111. [24] F.Q. Yang, T.Y. Zhang, G.L. Tian, H.F. Cao, Q.H. Liu, Y. Ito, J. Chromatogr. A 858 (1999) 103. [25] R.M. Liu, A.F. Li, A.L. Sun, J. Chromatogr. A 1052 (2004) 217. [26] R.S. Xu, Z.L. Chen (Eds.), Zhongcaoyao Youxiao Chengfen Tiqu Yu Fenli, Shanghai Science & Technology Press, Shanghai, 1983, p. 331. [27] Y.R. Ma, X.Y. Zhao, Z. Xu, Hecheng Huaxue 15 (2007) 244. [28] X.H. Ma, S.L. Shen, F.M. Han, Y. Chen, Hubei Da Xue Xue Bao (Nat. Sci.) 28 (2006) 403.