244
(10)
(11)
(12) (13) (14) (15) (16) (17)
(18)
(19)
(20) (21) (22)
Anal. Chem. 1990, 62,244-248 650; Gupta, D. C., Ed.; American Society for Testing and Materiais: Philadelphia, PA, 1984; pp 409-425. Tanaka, T.; Kurosawa, S. J. Ekctrochem. Soc.1986, 733, 416-420. Lindstrom, R. M. Microdectronic Processing: Inorganic Meterials Characterization; Caspar, L. A., Ed.; ACS Symposium Series 295; American Chemical Society: Washington, DC, 1986; pp 294-307. Schmidt, P. F.; Pearce, C. W. J . €lechochem. SOC. 1981, 728, 630-637. Piepmeier, E. H. Analytical Applications of Lasers, Piepmeier, E. H., Ed.; J. Wiley: New York, 1986; Chapter 19. Kronert, U.; Becker, St.; Hilberath, Th.; Kluge, H.J.; Schultz, C. Appl. P h y ~ 1987, . A44, 339-345. Donohue, D. L.; Christie, W. H.; Goeringer, D. E.; McKown, H. S. Anal. Chem. 1985, 5 7 , 1193-1197. Kimock, F. M.; Baxter, J. P.; Pappas, D. L.; Koprin, P. H.; Winograd, N. Anal. Chem. 1984, 5 6 , 2782-2791. Travis, J. C.; Fassett, J. D.; Lucatorto, T. B. Resonance Ionization Spechoscopy 1986; Hurst, G. S., Morgan, C. G., Eds.; Instiiute of Physics Conference Series 84; Institute of Physics: Bristol, U.K., 1987; pp 91-96. Young, C. E.; Peiiin, M. J.; Calioway, W. F.; Stein, H. J.; Tsao, S. S. Resonance Ionizatlon Spectroscopy 7986; Lucatorto, T. B., Parks, J. E.,Eds.; Institute of Physics Conference Series 94; Institute of Physics: Bristoi, U.K., 1989; pp 205-206. Moore, L. J.; Spaar, M. T.; Taylor, E. H.; Parks, J. E.; Beekman, D. W. Resonance i m k a t b ~ p e c ~ m c o p1088; y Lucatorto, T. E., Parks, J. E., Eds.; Institute of Physics Conference Series 94; Institute of physics: Bristol, U.K., 1989; pp 273-276. Heumann, K. G. Int. J. Mass Spectrom. Ion Phys. 1982, 4 5 , 87-110. Heumann, K. G. Inwpnic Mass Spectrometry; Adams, F., Gijbeis, R., Van Grieken, R., Eds.; J. Wiley: New York. 1987; Chapter 7. Pauisen, P. J.; Beaty, E. s.; Bushee, D. s.; M O ~ J.~ R., Anal. c k m . 1988, 6 0 , 971-975.
(23) Kern, W.; Deckert, C. A. 7M FUm Processes; Vossen, J. L., Kern, W., Eds.; Academic Press: New York, 1978; Chapter V-1. pp 401-496. (24) Studier, M. H.; Sloth. E. N.; Moore, L. P. J. Phys. Chem. 1962, 66, 133-134. (25) Fassett, J. D.; Moore, L. J.; Travis, J. C.; Lytle, F. E. Int. J . Mass Spectrom. Ion Phys. 1983, 5 4 , 201-216. (26) Fassett, J. D.; Walker, R. J.; Travis, J. C.; Ruegg, F. C. Int. J . Mass Spectrom. Ion Phys. 1987, 75, 111-126. (27) Fassett, J. D.; Moore, L. J.; Shideler, R. W.; Travis, J. C. Anal. Chem. 1984, 5 6 , 203-206. (28) Crouch, E. A.; Webster, R. K. J. Chem. Soc., London 1983, Part I , 118-131. (29) Clark, R. J. H. Comprehensive Inwganic Chemistry; Bailar, J. C., Emeleus, H. J., Nyholm, R., Trotman-Dickerson, A. F., Eds.; Pergamon Press: Oxford, U.K., 1973 Chapter 34. (30) Fassett, J. D.; Kingston, H. M. Anal. Chem. 1985, 5 7 , 2474-2478. (31) Fassett, J. D.; Travis, J. C.; Moore, L. J.; Applbtions of Laser Chemistry and Diagnostics; Harvey, A. B., Ed.; Proceedings SPIE 482, 1984; pp 36-43. (32) Jastrebski, L.; Ipri, A. C. I€€€ Electron Device Left. 1988, 9 , 15 1- 153.
RECEIVED for review July 5,1989. Accepted October 18,1989. Certain c m " r c i a l equipment, instruments, or materials are identified in this paper in order to adequately specify the experimental procedure. Such identification does not imply recommendation or endorsement by the National Institute of Standards and Technology, nor- does it imply that the materials Or equipment identified are necessarily the best available for the purpose.
Application of Countercurrent Chromatography/Thermospray Mass Spectrometry for the Identification of Bioactive Lignans from Plant Natural Products Yue Wei Lee,*Robert D. Voyksner, Terry W. Pack, and C. Edgar Cook Research Triangle Institute, P.O. Box 12194, Research Triangle Park, North Carolina 27709
Q . C. Fang Institute of Materia Medica, Chinese Academy of Medical Sciences, Beijing, China
Yoichiro Ito Laboratory of Technical Development, National Heart, Lung, and Blood Institute, Bethesda, Maryland 20892
The versatlllty and resolving power of countercurrent chromatography (CCC) has been demonstrated with a newly developed analytical high-speed planet centrifuge system. Interfacing countercurrent chromatography wlth mass spectrometry (MS)provldes a new analytical methodology that Integrates the advantages of Countercurrent chromatography wlth the low detection llmlt and ldentlfkatlon capabllity of mass spectrometry. I n this paper the capabillty of thermospray CCC/MS Is demonstrated In Identlfylng and valldatlng the bloactive and structurally known lignans from a crude extract of ScWndra rubrMora Rhed et Wlis, a tradltlonal Chinese herbal medlcine for treatment of hepatltls.
INTRODUCTION The development in the 1980s of modern countercurrent chromatography (CCC) instrumentation based upon the
* Author to whom correspondence should be addressed. 0003-2700/90/0362-0244$02.50/0
fundamental principles of liquid-liquid partition has caused a resurgence of interest in this separation science. The advantages of liquid-liquid extraction-the process for separating a multicomponent mixture according to its differential solubility in two immiscible solvents-have long been recognized. In spite of the limitations of the traditional countercurrent distribution methods that prevailed in the 1950s and 19609, liquid-liquid extraction was used successfully to fractionate commercial insulin into two subfractions differing by only one amide group out of a molecular weight of 6000 (1).
In recent years, significant improvements have been made to enhance the performance and efficiency of countercurrent methods (2). The newly developed high-speed CCC technique utilizes a particular combination of coil orientation and planetary motion to produce a unique hydrodynamic phenomenon in the unilateral phase distribution of two immiscible solvents in a coiled column. The hydrodynamic properties can effectively be applied to perform a variety of countercurrent chromatographies, including true countercurrent (3) 0 1990 American Chemical Society
ANALYTICAL CHEMISTRY, VOL. 62. NO. 3, FEBRUARY 1. 1990
245
1
Flgure 1. Schematic diagram of the CCClihermospray MS system.
and foam countercurrent methods (4). More recently, a 2000-rpm high-speed CCC has been developed, which performs, separations with speeds and resolutions approaching those achieved by HPLC (5). High-performance liquid chromatography coupled with mass spectrometry (HPLC/MS) represents one of the most rapidly advancing analytical techniques. Many different approaches have been developed in combining HPLC with MS, with their advantages and disadvantages extensively reviewed (6-9). Thermospray HPLC/MS (10, 1 1 ) is one technique that can handle a high quantity of aqueous solvent at conventional flow rates while providing a soft means of ionization (12,Z3). The versatility of thermospray HPLC/MS is evidenced in numerous applications of environmental and clinical analyses (Z4-Z8). The interface of countercurrent chromatography with thermospray mass spectrometry initially appeared incompatible due primarily to the pressure limitation imposed on the couuntercurrent system. A 6000-psi hack-pressure commonly generated from the thermospray vaporizer often exceeds the pressure limitation of the Teflon fitting connected to CCC and causes solvent leakage. The newly developed high-speed CCC system not only provided improved efficiency but also could withstand the hack-pressure, thus enabling the coupling of CCC with thermospray MS. Interfacing countercurrent chromatography with thermospray mass spectrometry provides a new analytical methodology. This combination enhances the versatility of countercurrent chromatography with the specific detection and sensitivity of mass spectrometry. The two-phase solvent system commonly employed in CCC offers distinct advantages by allowing the interface to be operated with the use of a n aqueous mobile phase. The high-percentage aqueous condition with a volatile buffer provides the best thermospray/MS sensitivity (12) without the losses in resolution typically observed in HPLC when the mobile phase is switched to higher aqueous percentages. The CCC/MS technique has been initially applied in the analysis of plant alkaloids from Vinca minor (19). This paper describes the application of thermospray CCC/MS in the identification of bioactive and structurdy known lignans from a crude extract of Schisandra rubriflora Rhed e t Wils, a
traditional Chinese herbal medicine for treatment of hepatitis.
EXPERIMENTAL SECTION Reagents and Materials. Ethanol and n-hexane used for prepmation of the twephsse solvent system were of ghdistilled chromatographic grade, purchased from Burdick and Jackson Laboratories,Inc., Muskegon, MI. Experiments were performed with a two-phase solvent system composed of n-hexane, ethanol, and water with a volume ratio of 6 5 5 . The two-phase solvent system was prepared hy thoroughly equilibrating the solvent mixture in a separatory funnel at room temperature followed by filtration and degassing with a 5-pm filter. The Schisandra sample was kindly provided by Professor Chen Y. Y. a t the Institute of Materia Medica, Chinese Academy of Medical Sciences, Meijing, China. The ethanolic extract of the kemels of S. rubriflora was filtered and concentrated to provide the crude sample for CCC/MS study. Analytical H g h - S p e d CCC. A recently developed analytical high-speed planet centrifuge equipped with a multilayer coil column of 0.85 mm i.d. poly(tetrailuoroethy1ene) (PTFE) tubing was employed. The system is capable of revolution a t 2000 rpm with a 5-em radius (3). A Waters 6000A HPLC pump (Waters Associates, Milford, MA) was used for the mobile phase. UV detection was achieved with an ISCO Model 1840 (Lincoln, NE) variable wavelength UV-vis absorbance detector. The column was first filled with the stationary phase (upper phase); then the mobile phase (lower phase) was pumped at 0.8 mL/min while the column was spun at 1500 rpm. T h e sample solution was injected when clean mobile phase (lower phase) was eluted. Thermospray CCC/MS. In our system, shown in Figure 1, the effluent from the CCC, operating at 0.8 mL/min, was introduced into a Waters 6oooA pump through n Zero dead volume tee fitted with a reservoir. The waters pump was necessary to achieve the solvent pressure required for thermospray. Since the pressure in CCC fluctuates, the thermospray Waters pump waa operated at 0.1 mL/min with the reservoir providing extra solvent or venting excess solvent from the CCC system. The effluent from the Waters pump was mixed coavially (20)with 0.3 M ammonium acetate added a t 0.3 mL/min to provide the volatile buffer for ion evaporation ionization. This solvent system (total of 1 mL/min) passed through a UV detector (254 nm) and into the thermospray interface. At lower CCC flow rates (0.3-0.6 mL/min) the pressure drop a m the thermospray vaporizer was sufficiently low to permit direct coupling of the CCC effluent (lower phase) to the thermospray interface without the use of the auxiliary pump. Postcolumn addition of huffer and UV detection of the
246
ANALYTICAL CHEMISTRY, VOL. 62, NO. 3, FEBRUARY 1, 1990 B
A
1
3
A
Table I. Lignans from Sehisandra Detected in Thermospray CCC/MS along with Their Tentative Identification peak no.
Analytical High Speed Countercurrent Chromatography Solvent System Hexanes Ethanol Waler 6 5 5 Flow Rate 0 8 mLimin Column Pressure 165 psi Det U V 2 5 4 n m
c w RO
3
0-
R-H
I
30
60
90
~.cn@ TIme (minuts) Figure 2. (A) CCClUV chromatogram of the crude extract of the bioactive lignans of Schisan&a rubrlfiora. (B) CCClUV chromatogram for analysis of standard of schisanhenol and schisanhenol acetate. 5
CCC effluent were maintained as described above. The thermosprayinterface (Vestec, Houston, TX) was installed on a Finnigan 4500 quadrupole mass spectrometer. The interface included a temperature controller and readout. The temperature zones monitored were the vaporizer, source, and aerosol (justpass the ion exit cone). Electrical cartridge heaters were used in the source, and the vaporizer was directly heated. The thermospray interface was operated at a source temperature of 250 "C and a vaporizer temperature to maximize the HPLC solvent clusters (about 170 "C). The solvent cluster has been shown to comaximi with the analyte being analyzed (14). This interface did not require any splitting of the LC effluent. The large volume of solvent was pumped out of the source with a liquid nitrogen cold trap prior to a mechanical rough pump. Both negative and positive ion detection using ion evaporation ionization and filament on chemical ionization (CI) were employed for the analysis of the herb medicine. The filament was operated at lo00 V with a 0.15-mA emission current. The mass spectrometerwas scanned from m / z 180 to 600 in 2 s. The mass calibration of the quadrupole was verified with poly(propy1ene glycol) (average MW 1000).
RESULTS AND DISCUSSION Countercurrent chromatography (CCC) based on the principle of liquid-liquid partition provides a number of distinct advantages, particularly in dealing with crude natural products (21,22). We have successfully applied the high-speed CCC system in the separation of plant alkaloids, plant indole hormones, and herbicides. Recently, the study of Schisandra rubriflora Rhed et Wils has led to the identification of a number of bioactive lignans (23). Because of structural similarities, these bioactive lignans require extensive effort in isolation and identification following either conventional or off-line HPLC mass spectrometry procedures. An efficient thermospray CCC/MS spectrometric system that integrates the versatility of countercurrent chromatography with the specific detection capability of mass spectrometry has been developed in this laboratory. This new analytical technique has been successfully demonstrated in our preliminary study with a mixture of plant alkaloids (19) and further applied in the identification of bioactive lignans of Schisandra.
RT, re1 peak min area, %
1
pregmoisin
27
21
2
13
3
meso-dihydroguai- 37 aretic acid 54 schisanhenol
4
schisanhenol B
52
