Journal of Chromatography A, 1216 (2009) 4140–4146

Contents lists available at ScienceDirect

Journal of Chromatography A journal homepage: www.elsevier.com/locate/chroma

Rapid and preparative separation of traditional Chinese medicine Evodia rutaecarpa employing elution-extrusion and back-extrusion counter-current chromatography: Comparative study Yanbin Lu a , Wenyan Ma a , Ruilin Hu a , Alain Berthod b,∗ , Yuanjiang Pan a,∗∗ a b

Department of Chemistry, Zhejiang University, Hangzhou 310027, Zhejiang Province, China Laboratoire des Sciences Analytiques, CNRS, Université de Lyon, 69622 Villeurbanne cedex, France

a r t i c l e

i n f o

Article history: Available online 5 November 2008 Keywords: Counter-current chromatography Elution-extrusion Back-extrusion Traditional Chinese medicines Evodia rutaecarpa

a b s t r a c t Traditional Chinese medicines (TCMs) have attracted much attention in recent years. Elution-extrusion and/or back-extrusion counter-current chromatography (EECCC/BECCC) both take full advantage of the liquid nature of the stationary phase. They effectively extend the solute hydrophobicity window that can be studied and rendered the CCC technique particularly suitable for rapid analysis of complex samples. In this paper, a popular traditional Chinese medicine, Evodia rutaecarpa, was used as the target complex mixture for extrusion CCC separations. With a carefully selected biphasic liquid system (nhexane/ethyl acetate/methanol/water, 3/2/3/2, v/v) and optimized conditions (VCM = VC , mobile phase flow rate: 3 mL/min in descending mode, sample loading: 100 mg), five fractions could be obtained in only 100 min on a 140-mL capacity CCC instrument using both elution- and back-extrusion methods. Each fraction was analyzed and identified compared with the data of major standards using LC/MS. Moreover, the performance of both extrusion protocols was systematically compared and summarized. EECCC could be operated continuously and was found extremely suitable for high-throughput separation; however, post-column addition of a clarifying reagent is recommended to smooth the UV-signal during the extrusion process. Considering BECCC, the practical operation is very simple by just switching a 4-port valve to change the flow direction. The change of flowing direction should be done after a sufficient amount of mobile phase has flushed the column in the classical mode so that solutes with small and medium distribution constants have been eluted. Otherwise, a significant portion of the solutes will stay in the mobile phase inside the column, mix together and produce a broad peak showing in the mobile phase eluting after the stationary phase extrusion. Compared with classical CCC or other preparative separation tools, extrusion CCC approaches exhibit distinguished superiority in the modernization process of traditional Chinese medicines. © 2008 Elsevier B.V. All rights reserved.

1. Introduction Traditional Chinese medicines (TCMs) are playing very important roles in clinical therapy because of the high pharmacological activity, low toxicity and rare complication [1]. TCMs may have the longest medical history in the world dating back several thousands of years. According to such a long history of development, 12,806 medical resources of TCMs, including 11,145 medicinal plants, 1581 medicinal animals and 80 medicinal minerals, and also 3214 other related products have been compiled in the 2005 edition of

∗ Corresponding author. Tel.: +33 472431434; fax: +33 472431078. ∗∗ Corresponding author. Tel.: +86 571 87951264; fax: +86 571 87951629.. E-mail addresses: [email protected] (A. Berthod), [email protected] (Y. Pan). 0021-9673/$ – see front matter © 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.chroma.2008.10.095

the Pharmacopoeia of China [2]. Recently, much interest was reattracted in the TCMs-related subjects. The identification of active components in TCMs is especially investigated. Counter-current chromatography (CCC) was developed by Ito in the late 1960s [3]. This technique is an all-liquid method, without solid phases, which relies on the partition of a sample between two immiscible solvents to achieve separation. The relative proportion of solute passing into each of the two phases is determined by the respective distribution constants [4,5]. Therefore, CCC benefits from great advantages when compared with the traditional liquid–solid separation methods: (1) it eliminates the complications resulting from the solid support matrix, such as irreversible adsorptive sample loss and deactivation, tailing of solute peaks, and contamination; (2) it is a very economical method (the instrument is relatively cheaper than HPLC, no expensive columns are required, low solvent consumption and only common solvents

Y. Lu et al. / J. Chromatogr. A 1216 (2009) 4140–4146

can be consumed). In addition, CCC is a preparative technique with high recovery, acceptable efficiency and easy scaling-up [6]. Due to its distinct feature, CCC has obviously played an important role in analysis and separation of various TCMs [7–12]. However, the composition of any TCM is rather complicated. A TCM is a mixture of an extremely large number of small molecules differing in molecular weight, structural class, and, most important, hydrophobicity. This high hydrophobicity variety presents a significant challenge for the standard CCC technique, where longer separation time is always needed due to the narrow hydrophobicity window of any single biphasic solvent system in the isocratic elution mode. Gradient elution is a practical and effective solution in HPLC with chemically stable solid stationary phases. However, in CCC, the use of a biphasic liquid system to separate solutes requires knowledge of the liquid–liquid equilibrium and solute partitioning theory. Any change of the mobile phase composition may change the stationary phase composition, resulting in the possible loss of hydrodynamic equilibrium, as well as loss of the overall separation efficiency. Taking advantage of the liquid nature of the stationary phase in CCC, the recently developed elution-extrusion (EECCC) [13–17] and back-extrusion CCC (BECCC) [18] methods greatly extended the hydrophobicity window of CCC technique. Based on the fact that band broadening inside chromatographic columns depends only on the band position [13–16], the major advantage of extrusion CCC relies on that the narrow band widths present inside the column are preserved during the extrusion process, resulting in extremely sharp extruded peaks. Additionally, because the EECCC method extends the reachable hydrophobicity windows to all solutes, i.e. having distribution constants, KD , ranging from zero to ∞, special chromatographic representations with an X-axis graduated in KD and/or 1/KD values were recently proposed as the reciprocal symmetry (ReS) and reciprocal shifted symmetry (ReSS) KD plots [19]. Due to the above-described advantages, it is predictable that the extrusion CCC protocol will be increasingly applied to the rapid separation of complex natural and/or biological samples, especially in natural drug discovery programs. Evodia rutaecarpa Hooker f. et Thomas (Rutaceae) is a popular TCM known in Chinese as “Wu-Zhu-Yu”. It has been prescribed for the treatment of gastrointestinal disorders (abdominal pain, dysentery), headache, postpartum hemorrhage, and amenorrhea [2,23]. It has also been reported to possess remarkable central stimulating, transient hypertensive, positive ionotropic and chronotropic effects [24,25]. The major phytochemical molecules present in Wu-ZhuYu were found to be quinazolinocarboline alkaloids [26–28]. Both EECCC and BECCC methods were evaluated and compared for the fast fractionation and separation of the TCM based on E. rutaecarpa. It is demonstrated that the extrusion CCC approaches exhibit great potential in the modernization process of TCM fractionation using the CCC technique. 2. Experimental 2.1. Reagents and materials All organic solvents were of analytical grade and purchased from Huadong Chemicals, Hangzhou, China. Reverse osmosis MilliQ water (18 M) (Millipore, Bedford, MA, USA) was used for all solutions and dilutions. Methanol and acetonitrile used for HPLC analysis were of chromatographic grade and purchased from Merck, Darmstadt, Germany. The dried fruits of E. rutaecarpa were purchased from a local drug store and identified by Professor Hongxiang Sun (Zhejiang University, Hangzhou, China). The standard alkaloids were purchased from the National Institute for the Control of Pharmaceutical and Biological Products, Ministry of

4141

Health, Beijing, China. The 0.5 mg/mL stock solutions were prepared by dissolving 5.0 mg of each alkaloid in 10.0 mL of methanol, and then stored in a refrigerator. The working solutions were prepared by suitable dilution of the stock solutions with methanol. 2.2. Instrumentation The biphasic liquid system selection was performed on an analytical-scale integrated parallel CCC separation system manufactured by the Zhejiang University machine shop (Hangzhou, China) [20]. Extrusion CCC separation was accomplished on a coil planet centrifuge (Ito scheme IV [21]) with one 140-mL coil and a counter-weight. The coil was made winding 26.4 m × 2.6 mm I.D. (3.2 mm O.D.) PTFE tubing in multiple layers. The ˇ-value varied from 0.33 at the internal terminal to 0.60 at the external terminal. The revolution speed of the apparatus can be regulated with a speed controller in the range of 0 and 1000 rpm. The CCC column is equipped with a Model 2W-2B constant-flow pump (Beijing Xingda Equipment, Beijing, China), an injection valve with a 5 mL sample loop, a 4-port switching valve (Cheminert Model C2, Valco Inc.), a HD-9704 UV spectrometer operating at 254 and 280 nm, a BSZ-100 fraction collector, and a N2000 data analysis system (Institute of Automation Engineering, Zhejiang University, Hangzhou, China). The high-performance liquid chromatography (HPLC) used was an Agilent 1100 HPLC system equipped with a G1311A Quatpump, a G1322 Degasser, a G1314A UV detector, a Rheodyne 7725i manual injection valve with a 20 ␮L loop and an Agilent Chemstation for data treatment. HPLC analysis of the crude sample and CCC peak fractions was performed with a Zorbax Eclipse XDB-C8 column (150 mm × 4.6 mm I.D., 5 ␮m). Methanol–acetonitrile–water system was used as mobile phase in gradient mode as follows: 10:18:72 to 15:38:47 in 0–10 min, to 45:38:17 in 10–30 min. The flow rate was 0.8 mL/min, and the effluent was monitored at 239 nm. LC/ESI–MS peak identification was performed using the abovedescribed Agilent HPLC system coupled with a Bruker Esquire 3000 plus ion trap mass spectrometer (Bruker–Franzen Analytik, Bremen, Germany) equipped with an electrospray ionization (ESI). Instrument control and data acquisition were performed using Esquire 5.0 software. The ion source temperature was 250 ◦ C, and needle voltage was always set at 4.0 kV. Nitrogen was used as the drying and nebulizer gases at a flow rate of 10 L min−1 and a backpressure of 30 psi. 2.3. Preparation of crude extract The preparation of crude sample was carried out according to the Chinese Pharmacopoeia [2] as follows. The dried fruits of E. rutaecarpa were ground to powder (about 40 mesh or 0.5 mm). The powder (100 g) was dipped into 500 mL of ethyl acetate for 120 min, and then extracted in an ultrasonic bath for 45 min. The extraction procedure was repeated three times. The three paper filtrated extracts were combined together and evaporated to dryness by rotary vaporization under reduced pressure. About 7.0 g of crude sample was obtained. It was stored in a refrigerator (4 ◦ C) for further use. 2.4. CCC procedures The solvent system used was a four solvent system: n-hexane– ethyl acetate–methanol–water. When the same water/methanol and ethyl acetate/hexane ratios are used, the biphasic liquid system compositions are also called “Arizona” system [22]. The ethyl acetate–water system is the most polar Arizona composition A and the hexane–methanol system is the less polar composition Z. Table 1 lists the compositions of intermediate polarity systems. In

4142

Y. Lu et al. / J. Chromatogr. A 1216 (2009) 4140–4146

Table 1 The Arizona system compositions. Letter

v/v n-Hexane

Increased hydrophobicity (Z to A) A 0 C 1 D 1 F 1 G 1 H 1 J 2 K 1 L 2 M 5 N 1 P 6 Q 3 R 2 S 5 T 3 U 4 V 5 W 6 X 9 Z 1

Ethyl acetate

Methanol

Water

1 9 6 5 4 3 5 2 3 6 1 5 2 1 2 1 1 1 1 1 0

0 1 1 1 1 1 2 1 2 5 1 6 3 2 5 3 4 5 6 9 1

1 9 6 5 4 3 5 2 3 6 1 5 2 1 2 1 1 1 1 1 0

all Arizona systems, the volume of upper organic phase is comparable to the volume of lower aqueous phase. The selected solvent mixture was thoroughly equilibrated in a separatory funnel at room temperature and the two phases were separated shortly before use. The crude extract of E. rutaecarpa was stored in a refrigerator and warmed to room temperature before use. The sample solution for chromatography was prepared by dissolving the desired amount of the extract in a solvent mixture consisting of equal volumes of upper and lower phases. The CCC column is first filled with the upper organic phase as the stationary phase. Then the coils are rotated at the desired speed and the lower aqueous phase (the mobile phase) is pumped at the selected flow rate in the head-to-tail direction (descending mode). When the hydrodynamic equilibrium is established, indicated by that the mobile phase exiting the column outlet instead of the displaced stationary phase, the volume of displaced stationary

phase collected corresponds to the volume of mobile phase, VM , in the equilibrated column. The sample solution could be injected on the column. The effluent is continuously monitored by UV and collected in by the fraction collector. The chromatogram is first developed in a classical way using a predetermined volume (VCM ) of the mobile phase, with CM standing for classical mode. Next, for EECCC separation, the upper stationary phase (instead of the lower mobile phase) is pumped into the column without changing the flow rate, flow direction or rotor rotation speed. This phase change marks the beginning of the extrusion step. For BECCC, only the flow direction of the lower mobile phase is changed by switching the 4-port valve from head-to-tail to tail-to-head [18]. In both cases, the flowing of a liquid phase is the “wrong” (meaning not allowing to maintain the liquid–liquid equilibrium) direction produced the extrusion of the previous stationary phase. All contained solutes could be rapidly eluted out of the column and the collected fractions were analyzed by HPLC and LC/MS to corroborate the UV–vis data. Both methods allowed for full recovery of the injected material [13–18]. 3. Results and discussions 3.1. Biphasic liquid system selection In this study, the screening of biphasic liquid system was performed on an integrated three-column CCC separation system with a newly developed screening strategy. Fig. 2a–f illustrates the solvent selection chromatograms of the crude E. rutaecarpa extract with the parallel protocol [20]. Arizona systems J, N and S (Table 1) were subjected to the first parallel trial. As seen in Fig. 2a, with the relative hydrophilic composition J (i.e. hexane–ethyl acetate–methanol–water, 2/5/2/5, v/v), the major components contained in the crude extract, were eluted as one major peak in the extrusion stage. With composition N (i.e. 5/5/5/5, v/v, Fig. 2b), which has an intermediate polarity, the major analytes were also eluted during the extrusion step but they exhibited a significantly improved peak resolution. However, with the rather hydrophobic composition S (i.e. 5/2/5/2, v/v, Fig. 2c), the situation was just opposite: components were mainly eluted in the classical mode step. From the above analysis, it is evident that the crude extract of E.

Fig. 1. HPLC analysis of the crude extract of E. rutaecarpa as well as chemical structures. Column Zorbax Eclipse XDB-C8 (150 mm × 4.6 mm I.D., 5 ␮m); mobile phase in ternary methanol:acetonitrile:water gradient mode: 0–10 min. 10:18:72 to 15:38:47; 10–30 min, to 45:38:17. Flow rate: 0.8 mL/min, detection UV 239 nm.

Y. Lu et al. / J. Chromatogr. A 1216 (2009) 4140–4146

4143

Fig. 2. Parallel strategy for rapid screening of biphasic liquid systems for E. rutaecarpa extract (a) Arizona liquid system J; (b) Arizona liquid system N; (c) Arizona liquid system S; (d) Arizona liquid system P; (e) Arizona liquid system Q; and (f) Arizona liquid system R. Column capacity of 40 mL for each. Flow rate: 2.0 mL/min of lower aqueous phase in the head-to-tail (descending) direction, up to VCM (20 mL) immediately followed by 2 mL/min of upper phase flowing in the same direction; revolution speed: 650 rpm; detection: 254 nm; injected sample: 10 mg crude extract dissolved in 1 mL upper phase and 1 mL lower phase.

rutaecarpa is a sample with a relatively high hydrophobic polarity and therefore the polar Arizona systems J and N are not adapted for the fractionation. The less polar system S was confirmed for the next parallel screening trial. The chromatograms shown in Fig. 2d–f clearly indicated that when Arizona system Q (i.e. 3/2/3/2, v/v) was employed, the crude extract could achieve optimum peak resolutions and rapidly be separated into five fractions. Thus, the Arizona system Q (i.e. 3/2/3/2, v/v, Fig. 2e) was selected as the optimum biphasic liquid system for the following extrusion CCC fractionation of components from the crude extract of E. rutaecarpa. 3.2. Extrusion CCC separation of E. rutaecarpa extract The two extrusion methods, EECCC and BECCC, were used and compared for the rapid fractionation and separation of the crude extract of the fruits of E. rutaecarpa. Fig. 1 illustrates the HPLC chromatogram of the crude ethyl acetate extract of E. rutaecarpa: it contains eight identified major components with a wide range of polarity. As mentioned above, the Arizona system Q (3/2/3/2, v/v) was selected as biphasic liquid system for rapid fractionation of E. rutaecarpa extract. Next, 100 mg (in 5 mL) of crude extract was injected in a 140-mL capacity hydrodynamic CCC column for reversedphase extrusion CCC separations (aqueous lower mobile phase in descending or head-to-tail mode). The switch volume, VCM , was kept constant at one column volume (VCM = VC = 140 mL). The total duration (volume) for each run was only ∼100 min (280 mL) with the flow rate of 3 mL/min. The Arizona system Q was well retained by the CCC column giving a Sf factor of 76% (VS = 106 mL at 3 mL/min and 650 rpm). Five fractions (numbered I–V) could be separated in less than 100 min using either the EECCC (Fig. 3a) or BECCC (Fig. 3b)

protocol. It is interesting to note that in HPLC analysis peaks 1 and 2 are eluted close together, while in CCC these peaks are well resolved as II and III, and fraction III in CCC separation shows two peaks in HPLC analysis (compounds 2 and 5, Fig. 3c). The X-axes in Fig. 3a and b are graduated in retention time. The value of the distribution ratios (partition coefficients) corresponding to a given retention time was calculated according to the well developed equations [13–18] and indicated on the axes. The isolated fractions were Roman numbered in the order of their increasing polarities. This is also the elution order of the classical reversed-phase elution mode. The EECCC method involves changing of liquids nature but flowing in the same direction. Thus, the solute elution order is the same as the one obtained in the classical elution mode: Fraction I eluted first and Fraction V eluted last in the reversed-phase mode (Fig. 3a). While, with the BECCC method, the elution order is inverted after the switch of flowing direction as shown in Fig. 3b. Additionally, the colored bands placed in Fig. 3 below the chromatograms also indicate the nature of the liquid phase eluting from the CCC column. In EECCC, the nature of the entering phase is changed after the VCM; volume of mobile phase in classical mode. However, the mobile aqueous lower phase is still seen exiting the column for the VCM + VM volume (the “sweeping” step described in Ref. [16]). Next the “stationary” organic upper phase is seen at the column outlet. It is collected carrying all solutes remaining in the column. After a volume VC of “stationary” phase is pumped, it is absolutely sure that no solute remains in the column that is full of “stationary” organic upper phase and ready to be equilibrated with the aqueous mobile phase for the next experiment. In the BECCC method, it is the mobile phase flowing direction that is changed after eluting the same volume VCM . This change in flowing direction

4144

Y. Lu et al. / J. Chromatogr. A 1216 (2009) 4140–4146

Fig. 3. Rapid extrusion CCC separation of E. rutaecarpa extracts using a 140 mL CCC column with Arizona system Q in reversed-phase mode. (a) EECCC. Flow rate: 3.0 mL/min of lower aqueous phase up to VCM (140 mL, 46 min, vertical dotted line) immediately followed by 3 mL/min of upper phase; post-column addition begins at 140 mL. (b) BECCC. Flow rate: 3.0 mL/min of lower aqueous phase. Valve switching at VCM (140 mL, 46 min, vertical dotted line); revolution speed: 650 rpm; detection: 254 nm; injected sample: 100 mg of crude extract in 2 mL upper phase and 2 mL lower phase. (c) HPLC analysis of the extrusion CCC peak fractions I–V.

instantly produces a break in column equilibrium and liquid phase separation [18]. The organic upper “stationary” phase is immediately seen exiting the CCC column after the flow switch. It carries the most retained solutes. All the “stationary” phase contained in the column, VS , is extruded from the column and the aqueous lower mobile phase is seen again at the column outlet carrying the part of the solutes present at the switching time. Obviously, the solute

split between the two liquid phases is clearly a major drawback of the BECCC technique. After each extrusion process, five fractions could be rapidly separated from the 100 mg of the crude extract of E. rutaecarpa in about 100 min. The distribution ratio of Fraction V is estimated to be about 35. With the 106 mL VS volume (Sf = 76%), it would need more than 3.8 L of mobile phase or 21 h of experiment duration to be eluted

Y. Lu et al. / J. Chromatogr. A 1216 (2009) 4140–4146

4145

Table 2 LC/MS identification of major compounds present in E. rutaecarpa. Compound no.

Fraction no.

HPLC retention time (min)

Formula

[M+H]+

Compound identification

1 2 3 4 5 6 7

II III II II III IV IV

16.3 16.5 26.7 28.6 32.2 33.5 34.9

C19 H17 N3 O C18 H13 N3 O C19 H27 NO C2 1H29 NO C21 H31 NO C22 H33 NO C25 H35 NO

304 288 286 312 314 328 366

8

V

37.3

C23 H35 NO

342

Evodiamine Rutuecarpine 1-Methyl-2-nonyl-4(1H)-quinolone 1-Methyl-2-[(Z)-5-undecenyl]-4(1H)-quinolone 1-Methyl-2-undecyl-4(1H)-quinolone 1-Methyl-2-dodecyl-4(1H)-quinolone 1-Methy-2-[(6Z,9Z)]-6,9Pentadecadienyl-4-(1H)-quinolone Dihydroevocarpine

in classical CCC elution. Both extrusion CCC ways allow recovering Fraction V in less than 100 min or 300 mL saving more than 80% of liquid phase and/or time. After concentration, the fractions were analyzed and identified compared with the data of major standards using LC/MS (Table 2). These results indicated that the EECCC and BECCC methods are extremely suitable for rapid separation of complex natural extracts. 3.3. EECCC and BECCC: a comparative study As newly developed techniques, the extrusion CCC protocols share several distinct advantages compared with standard CCC and other preparative separation tools. First of all, taking advantage of the liquid stationary phase, extrusion CCC can easily recover the highly retained solutes, which extensively enhance the separation ability of any single biphasic liquid system and polarity range of CCC separations. Moreover, it is noteworthy that, as a preparative technique, extrusion CCC not only presents relatively sharp peaks, but also exhibits satisfactory resolution factors. The fact that no part of the sample can remain trapped inside the column is another tremendous advantage. These promising features make extrusion CCC protocols extremely suitable for separation of complex samples, which has always been a challenge for analytical chemists. From a practical point of view, the EECCC and BECCC methods both save dramatically separation duration, as well as solvents compared to classical CCC separations. They make the CCC technique a rapid and competitive tool compared with other preparative chromatographic process. Especially, it is essential to point out the facility of sample pre-treatment procedures in extrusion CCC separations. Direct injection of most crude plant extracts is most often possible after a rapid filtration of solid parts. This is another clear advantage of extrusion CCC over other preparative separation techniques, where multiple pre-purification steps are always needed to avoid possible contamination of solid support by sample matrix components. Comparing the two extrusion protocols, they both have similarities as well as their own speciality. The distinct features of EECCC and BECCC protocols are summarized in Table 3. For EECCC, the classical elution mode can be as short as one column volume of mobile phase (VCM = VC ) eluting solutes up to

the distribution ratio of unity (KD = 1). After another column volume of “stationary” phase is further pumped in the CCC column, it is certain that all solutes have been eluted leaving a column filled by clean stationary phase ready to be equilibrated for the next experiment. Thus, EECCC allows for continuous operation and is extremely suitable for high-throughput separations. However, continuous UV detection is always a problem since two very different liquid phases are flowing through the detector cell [13–17]. It was presumed that the poor UV detection was mainly due to micro-droplets of the mobile phase (lower phase) staying trapped in the detector cell and perturbing the light beam and generating noisy baselines. Thus, post-column addition of a clarifying reagent (methanol) was recommended to smooth the UV-signal in the extrusion process [17]. With the BECCC technique, the mobile phase flowing direction is just inverted by a 4-port valve after classical mode elution. This valve is often already installed on modern commercial CCC columns meaning that the BECCC method is very simple to implement. It is often possible to use the BECCC method right away. The change of flow direction will produce a sudden rupture of the two liquid phase equilibrium inside the CCC hydrodynamic column. The “stationary” phase gathers at the column outlet where it is back extruded with all retained solutes in relatively sharp peaks. However, considering the solute fraction located in the mobile phase at the time of the switching of the flow direction, a very significant mixing of the contained solute bands will occur because this mobile phase portion finds itself located behind the “stationary” phase after the phase decantation. It always travels up the whole column length, pushed by the same liquid phase (i.e. no piston effect as with the other phase and maximum Taylor or convective diffusion) before eluting. In this situation, broad “echo” peaks will be seen after the sharper peaks in the extruded “stationary” phase [17]. This is the major drawback of the BECCC method. To reduce this problem, it is better increase the classical mode elution. For example, if at least two column volumes of mobile phase are eluted in the classical elution mode (VCM = 2 VC ), solutes with higher distribution ratios are eluted in the classical mode. With Sf = 76% (our column), solutes with KD < 2.4 are eluted with 2 VC . Only 29% of a solute with KD = 2.4 is located in the mobile phase and 71% will be extruded in the stationary phase. The broad “echo” peak will be reduced. However, the

Table 3 Comparison of EECCC with BECCC. EECCC

BECCC

Advantage

Extensively enhance the polarity range of CCC separations Extremely sharp peaks and satisfactory resolution factors Extremely suitable for separation of complex samples Save six times of separation duration, as well as 80% of liquid phase Facility of sample pretreatment procedures that allowing the injection of most crude plant extracts High-throughput separation Simple operation

Drawback

Poor UV detection Two pumps and a solvent selection valve to minimum the dead volume

Broad “echo” peaks Not continuous for high-throughput separation

4146

Y. Lu et al. / J. Chromatogr. A 1216 (2009) 4140–4146

experiment duration will increase, which is not desirable. In short, EECCC may be the best extrusion technique to recommend. 4. Conclusion Extrusion CCC takes full advantage of the liquid nature of the stationary phase and effectively extends the hydrophobicity window with any biphasic liquid system. In this study, we applied the two recently proposed extrusion protocols, EECCC and BECCC, for rapid and preparative fractionation of a real TCM extract of E. rutaecarpa. With the “Arizona” biphasic liquid system Q (3/2/3/2, v/v) and optimized conditions, five fractions could be obtained in only 100 min on a 140-mL capacity CCC instrument using both extrusion methods. Each fraction was analyzed and identified compared with the data of major standards using LC/MS. Moreover, the performance of both extrusion protocols was systematically compared and summarized. This study provides a promising tool for rapid separation and identification of crude extracts of TCMs, indicating that the extrusion CCC has unique advantages in separation efficiency, preparative capacity, and has a great potential for natural drug discovery programs. Acknowledgements Y.P. thanks the National Science Foundation of China for Grant 20775069, Ministry of Education of China for Grant NCET-06-0520 and Natural Science Foundation of Zhejiang Province for Grant Z206510. A.B. thanks the French Centre National de la Recherche Scientifique for financial support through UMR5180 (P. Lanteri).

References [1] S.Z. Li, Ben Cao Gang Mu, People Health Press, Beijing, 1982. [2] China Pharmacopoeia Committee, Pharmacopoeia of the People’s Republic of China, 2005 edition, China Chemical Industry Press, Beijing, 2005. [3] Y. Ito, R.L. Bowman, Science 167 (1970) 281. [4] Y. Ito, W.D. Conway (Eds.), High-Speed Countercurrent Chromatography, Chemical Analysis, vol. 132, Wiley Interscience, New York, 1996. [5] A. Berthod (Ed.), Countercurrent Chromatography, the Support Free Liquid Stationary Phase, Comprehensive Analytical Chemistry, vol. 38, Elsevier, Amsterdam, 2002. [6] I.A. Sutherland, J. Chromatogr. A 1151 (2007) 6. [7] Y. Ito, J. Chromatogr. A 1065 (2005) 145. [8] A. Marston, K. Hostettmann, J. Chromatogr. A 1112 (2006) 181. [9] Y. Pan, Y. Lu, J. Liq. Chromatogr. Relat. Technol. 30 (2007) 649. [10] O. Sticher, Nat. Prod. Rep. 25 (2008) 517. [11] S. Liu, L.-Z. Yi, Y.-Z. Liang, J. Sep. Sci. 31 (2008) 2113. [12] G.F. Pauli, S.M. Pro, J.B. Friesen, J. Nat. Prod. 71 (2008) 1489. [13] A. Berthod, M.J. Ruiz-Angel, S. Carda-Broch, Anal. Chem. 75 (2003) 5886. [14] A. Berthod, M. Hassoun, G. Harris, J. Liq. Chromatogr. Relat. Technol. 28 (2005) 1851. [15] A. Berthod, J. Chromatogr. A 1126 (2006) 347. [16] A. Berthod, J.B. Friesen, T. Inui, G.F. Pauli, Anal. Chem. 79 (2007) 3371. [17] Y. Lu, R. Liu, A. Berthod, Y. Pan, J. Chromatogr. A 1181 (2008) 33. [18] Y. Lu, Y. Pan, A. Berthod, J. Chromatogr. A 1189 (2008) 10. [19] J.B. Friesen, G.F. Pauli, Anal. Chem. 79 (2007) 2320. [20] Y. Lu, W. Ma, R. Hu, A. Berthod, Y. Pan, Presented at the 5th International Conference on Countercurrent Chromatography, Rio de Janeiro, Brazil, July 24–29, 2008, Abstract book, p. 64. [21] Y. Ito, J. Chromatogr. 214 (1981) 122. [22] A. Berthod, M. Hassoun, M.J. Ruiz-Angel, Anal. Bioanal. Chem. 383 (2005) 327. [23] J.F. Liao, C.F. Chen, S.Y. Chow, J. Formosan Med. Assoc. 80 (1981) 30. [24] S.Y. Chow, S.M. Chen, C.M. Yang, J. Formosan Med. Assoc. 75 (1976) 349. [25] C.F. Chen, S.M. Chen, M.T. Lin, S.Y. Chow, Am. J. Chin. Med. 9 (1981) 39. [26] L.L. Yu, J.F. Liao, C.F. Chen, Planta Med. 60 (1994) 308. [27] Y.Q. Tang, X.Z. Feng, L. Huang, Phytochemistry 43 (1996) 719. [28] X.W. Yang, J. Teng, J. Chin. Pharm. Sci. 16 (2007) 20.