[CANCER RESEARCH 42. 1193-1198. April 1982] 0008-5472/82/0042-OOOOS02.00 Isolation and Identification of Kahweol Palmitate and Cafestol Palmitate äsActive Constituents of Green Coffee Beans That Enhance Glutathione S-Transf erase Activity in the Mouse1 Luke K. T. Lam,2 Velta L. Sparnins, and Lee W. Wattenberg Department of Laboratory Medicine and Pathology, University of Minnesota, Minneapolis, Minnesota 55455 ABSTRACT compounds increased the GSH S-transferase activity of the forestomach between 78 and 182%. They were p-methoxy- Glutathione (GSH) S-transferase is a major detoxification phenol, 2-fe/t-butyl-4-hydroxyanisole, coumarin, a-angelica- enzyme system that catalyzes the binding of a variety of elec- lactone, and benzyl isothiocyanate (18). All 5 compounds trophiles, including reactive forms of chemical carcinogens, to inhibited BP-induced neoplasia of the forestomach (13, 20, GSH. Green coffee beans fed in the diet induce increased GSH 21). These data suggest that the capacity to enhance GSH S- S-transferase activity in the mucosa of the small intestine and transferase activity might be used as a method of identifying in the liver of mice. A potent compound that induces increased compounds or natural products likely to inhibit BP or other GSH S-transferase activity was isolated from green coffee carcinogens detoxified in a similar manner (18). beans and identified as kahweol palmitate. The corresponding In efforts at identifying dietary constituents that might protect free alcohol, kahweol, and its synthetic monoacetate are also against chemical carcinogens, the effects of natural products potent inducers of the activity of GSH S-transferase. A similar on GSH S-transferase activity were studied. During this inves diterpene ester, cafestol palmitate, isolated from green coffee tigation, it was found that consumption of diets containing beans was active but less so than was kahweol palmitate. powdered green coffee beans resulted in a very marked en Likewise, the corresponding alcohol, cafestol, and its monoac hancement of GSH S-transferase activity in the liver and mu etate showed moderate potency as inducers of increased GSH cosa of the small bowel of the mouse. The magnitude of S-transferase activity. Kahweol palmitate and cafestol palmi induction was as high as that obtained with any test compound tate were extracted from green coffee beans into petroleum or natural material previously investigated. The coffee beans ether. The petroleum ether extract was fractionated by prepar used in the original study were from Guatemala. In subsequent ative normal-phase and reverse-phase liquid chromatographies work, coffee beans from Brazil, Colombia, El Salvador, Mexico, successively. Final purification with silver nitrate-impregnated and Peru were all found to have a comparable enhancing effect thin-layer chromatography yielded the pure palmitates of ca on GSH S-transferase activity (17). Roasted coffee and instant festol and kahweol. The structures were determined by exam coffee were found to have a weaker inducing activity than did ination of the spectroscopic data of the esters and their parent the green coffee beans studied, i.e., slightly less than 50% as alcohols and by derivative comparison. much. Decaffeinated instant coffee showed activity similar to that of instant coffee. Investigations were then begun to identify INTRODUCTION the constituents of green coffee beans having the capacity to enhance GSH S-transferase activity. In the present study, it GSH3 S-transferase has been studied extensively as a major was found that the coffee constituent kahweol palmitate is a detoxification enzyme system that catalyzes the binding of a highly potent inducer of increased GSH S-transferase activity. wide variety of electrophiles to GSH (3,10). Since most reactive The palmitate of a closely related diterpene, cafestol, was ultimate carcinogenic forms of chemical carcinogens are elec active as an inducer of GSH S-transferase but less so than was trophiles, GSH S-transferase may play a significant role in kahweol palmitate. carcinogen detoxification. Enhancement of the activity of this system potentially could increase the capacity of the organism MATERIALS AND METHODS to withstand the neoplastic effects of chemical carcinogens. Experiments have been carried out to determine the correlation Extraction of Green Coffee Beans. Powdered green coffee beans between increased GSH S-transferase activity in a target organ (Guatemala) were placed in a large modified Soxhlet extractor and were extracted with various solvents of increasing polarity (Chart 1). of chemical carcinogenesis and its response to the carcinogen. PE (b.p. 60-70°) was used first, which was followed by benzene, ethyl For this purpose, the forestomach of the mouse was used. Members of several classes of inhibitors of BP-induced neopla acetate, methanol, and water. Each solvent extraction was carried out over a period of 7 days. The solvent of each extraction was removed sia of the mouse forestomach were studied for their effects on under reduced pressure. The crude extracts were dried under reduced the GSH S-transferase activity in that structure. Five of the pressure until constant weights were obtained. The activities of the extracts were monitored by the GSH S-transferase assay. 1 Supported by USPHS Contract NOI-CP-85613-70 and Grant CA-09599. Fractionation of the PE Extract. The active PE extract was sepa Presented in part at the 182nd American Chemical Society National Meeting, rated into 7 fractions by preparative LC using a Waters Associates New York, N. Y., August 23 to 28, 1981 (14). prepSOO liquid Chromatograph equipped with prepPak-500/silica col 2 To whom requests for reprints should be addressed. 3 The abbreviations used are: GSH, glutathione: BP, benzo(a)pyrene; PE, umns. The elution solvent was PE;ether (2:1, v/v). The active fraction petroleum ether; LC, liquid chromatography; TLC, thin-layer chromatography; from preparative LC was further fractionated into 6 subfractions by NMR, nuclear magnetic resonance. reverse-phase preparative LC. Two prep PAK-500/C18 columns in Received October 7, 1981 ; accepted January 4, 1982. series were used with 98% methanol as the eluting solvent. Silver APRIL 1982 1193 Downloaded from cancerres.aacrjournals.org on October 2, 2021. © 1982 American Association for Cancer Research. L. K. T. Lam et al. Green Coffee Beans GSH S-transferase activity. The inductive effects on the small bowel mucosa under these conditions are about one-third of the magnitude as in the feeding procedure. Liver is much less responsive. GSH S-transferase activity in the cytosol was determined according PE(8%) 0H EtOAc MeOH H2O to previously published procedures (18). All steps were done at 0-4°. Prep LC-silica The tissues to be studied were homogenized in 0.1 M sodium phosphate sÀ267PrepLC-C,sB3(20%)45 buffer (pH 7.5). The homogenate was centrifuged at 100,000 x g for 1 hr. The supernatant was used for the assay of GSH S-transferase activity. The activity was determined spectrophotometrically at 30° C(38%)D7 l with 1-chloro-2,4-dinitrobenzene as substrate according to the proce |TLC-Sllica Gel-AgNO 3 dure of Habig ef al. (8). 1Palmitic RESULTS Isolation Procedure Acid(1C)isaponificationCafestolPalmiticAcid <2a>4 (2c)1Kahweol The activity of the enzyme GSH S-transferase in the liver and mucosa of the small bowel of the mouse was enhanced by the Chart 1. Extraction and isolation scheme for kanweol and cafestol palmitates 0H, benzene; EtOAc, ethyl acetate. PrepPAK-500/silica columns were used for addition of 20% green coffee beans to the diet. The enhance the normal-phase preparatory (Prep)LC. The eluting solvent was PE:ether (2:1, ment of enzyme activity in the liver was approximately 5 times v/v). PrepPAK-500/C18 columns were used for the reverse-phase preparatory LC. The eluting solvent was 98% methanol (MeOH). The developing solvent for that of the control (Chart 28). A slightly smaller enhancement silver nitrate-impregnated Silica Gel GF thin-layer plates was PEiether (2:1 ). nitrate-impregnated TLC was used for the final purification of individual active ingredients. 1 Saponification of Compounds 1 and 2. A sample of Compound 1 (or Compound 2), isolated from silver nitrate-impregnated TLC plates, X was dissolved in 10% aqueous ethanolic potassium hydroxide at room temperature. The reaction mixture was warmed to 50-60° for 0.5 hr. It was then poured into ice, and the aqueous solution was extracted 3 times with ether. The organic layers were combined and dried over anhydrous magnesium sulfate. The ether was removed under reduced pressure. The neutral product was crystallized from PE:ether. I 2 The aqueous layer after ether extraction was acidified with 6 N hydrochloric acid and then extracted with 3 volumes of ether. The ether was dried over anhydrous magnesium sulfate. The solvent was removed under reduced pressure, and the acid thus obtained was dried under reduced pressure overnight. GCB PE BZ EA ME WA RC RS CON Spectroscopic Studies. IR spectra were recorded on a Beckman Acculab 5 spectrophotometer. UV spectra were recorded on a Beck 3 18 man Model 25 spectrophotometer. Proton NMR spectra were deter B mined on a Briiker WM 250 spectrometer at 250.13 MHz with tetra- methylsilane as internal standard. Gas chromatography-mass spec- „ troscopy was determined on an LKB9000 spectrometer. High-resolu tion mass spectroscopy were recorded on a double-beam AEI-MS-30 spectrometer. Melting points were determined on a Fisher-Johns Mel- Temp apparatus and were uncorrected. I - J, Assay for in Vivo Enhancement of GSH S-Transferase Activity. 1 12 Female ICR/Ha mice from the Harlan-Sprague-Dawley Company (In dianapolis, Ind.) were used in all experiments. Mice were randomized 10 by weight at 7 weeks of age into the groups to be used in a particular protocol. In the initial stages of the fractionation, the extracts were taken to dryness and added to a semipurified diet consisting of 27% vitamin-free casein, 59% starch, 10% corn oil, 4% salt mix (U.S.P.