Antioxidative Effects Of Phenylethanoids From Cistanche Deserticola

Contact: Audrey Hu Whatsapp/hp: 0086 13880143964 Email: audrey.hu@wecistanche.com

Quanbo Xiong/ Shigetoshi Kadota," Tadato Tani,6 and Tsuneo Namba"

The acetone-H2O (9:1) extract from the stem of Cistanche deserticola showed a strong free radical scavenging activity. Nine major phenylethanoid compounds were isolated from this extract. They were identified by NMR as acteoside, isoacteoside, l^acetylacteoside, tubuloside B, echinacoside, tubuloside A, syringalide A 3z-a-rhamno- pyranoside, cistanoside A and cistanoside F. All of these compounds showed stronger free radical scavenging activities than a-tocopherol on 1,1 -diphenyI-2-picrylhydrazy 1 (DPPH) radical and xanthine/xanthine oxidase (XOD) generated superoxide anion radical (Oj). Among the nine compounds, isoacteoside and tubuloside B, whose caffeoyl moiety is at d'-position of the glucose, showed an inhibitory effect on XOD. We further studied the effects of these phenylethanoids on the lipid peroxidation in rat liver microsomes induced by enzymatic and non-enzymatic methods. As expected, each of them exhibited significant inhibition on both ascorbic acid/Fe2 + and ADP/NADPH/Fe3 + induced lipid peroxidation in rat liver microsomes, which were more potent than a-tocopherol or caffeic acid. The antioxidative eflFect was found to be potentiated by an increase in the number of phenolic hydroxyl groups in the molecule.

Keywords: Cistanche deserticola; phenylethanoids; antioxidant; DPPH (1 ,1 -diphenyl-2-picrylhydrazyl) radical; superoxide anion radical; lipid peroxidation

stem of Cistanche deserticola

The stem of Cistanche spp. (Cistanche Herba, family Orobanchaceae) is a tonic in traditional Chinese medicine used for deficiency of the kidney characterized by impotence, cold sensation in the loins and knees, female sterility, and constipation due to dryness of the bowel in the senile. A number of constituents including phenyl- ethmoids (sometimes called phenylpropanoids),2) iridoids3* and polysaccharides4) have been isolated from these plants. Sato et al5} reported that some phenylethanoids from this crude drug showed a protective effect against decreases of both sexual and learning behaviors in hanging stress-loaded mice. Phenylethanoids are the major constituents of these plants and are considered to contribute to the various actions of this drug.6)

Phenylethanoids are widely distributed in the plant kingdom and some have been found to have the activity of anti-anoxia,7) anti-neoplasm,8) enzyme inhibition,9* anti-inflammation,10) antinephritis,1 n anti-microorganism, and immunomodulation.12) Studies on the antioxidative activity of some phenylethanoids from Pedicularis were also reported,13 -15) but no report was found dealing with the antioxidative activity of phenylethanoids from Cistanche species.

It is well recognized that free radicals are critically involved in various pathogeneses such as cancer, cardiovascular disorder, arthritis, inflammation, as well as in the degenerative processes associated with aging.16™18) In our screening for anti-aging agents from natural resources, we found that each of the CHC13 (C), EtOAc (E), acetone (A), acetone-H2O (9:1) (AH), and water (H) extract of the stem of Cistanche deserticola showed a potent DPPH radical scavenging activity (Fig. 1), among which, the strongest was exhibited by acetone—H?。(9:1) extract (AH). The acetone-H2O (9:1) extract contained a high ratio of phenylethanoids, which were probably active constituents of the free radical scavenging effects. We isolated the major phenylethanoids from this extract and studied the antioxidative effects by their scavenging on DPPH radical and xanthine/XOD generated superoxide anion radical, and their inhibition of ascorbic acid/ Fe2+ and ADP/NADPH/Fe3 + induced lipid peroxidation in rat liver microsomes.

Cistanche deserticola

MATERIALS AND METHODS

Reagents Polyamide C-200 (75一150/zm) and silica gel (Wakogel C-200, 75—150^m) powder for column chromatography, DPPH (1,1 -diphenyl-2-picrylhydrazyl), xanthine, XOD (xanthine oxidase), NBT (nitroblue tetrazolium), ADP, g-NADPH, caffeic acid and a-tocopherol were purchased from Wako Pure Chemicals, Osaka, Japan. Malonaldehyde bis(dimethylacetal) was from Tokyo Kasei, Tokyo, Japan. BSA (bovine serum albumin) and allopurinol were from sigma,St. Louis, U.S.A. Other chemicals were of analytical grade.

Instruments UV absorbance was measured on a Shimadzu UV-2200 recording spectrophotometer, NMR spectra were carried on a JEOL GX-400 or a JEOL JNM-LA 400WB FT NMR system.

Plant Materials A commercial crude drug (produced in Nei Monggol, purchased in Bozhou Crude Drug Market, Anhui province, China, 1995; the voucher specimen, TMPW No. 15479 deposited in the Museum of Materia Medica, Analytical Research Center for Ethnomedicine, Research Institute for Wakan-Yaku, Toyama Medical and Pharmaceutical University) was identified as stems of Cistanche deserticola Y. C. Ma by comparing its morphological and anatomical characters with an authentic sample provided by Professor Dawen Shi, Department of Pharmacognosy, Shanghai Medical University, China.

Extraction and Isolation The powdered crude drug (5.87 kg) was extracted consecutively with CHC13 (12, 9 1x2) and EtOAc (9 1 x 3) at room temperature and refluxed with acetone (9 1 x 3), acetone-玦0 (9:1,91x3) and H2O (7.5 1x4) to give the extracts of CHC13 (C) (28.7g), EtOAc (E) (13.4g), acetone (A) (95g), acetone- H2O (9:1) (AH) (175 g) and H2O (H) (2.51kg), respectively. A portion of lOOg of the acetone-H2O (9:1) extract was subjected to a polyamide C-200 (400 g) column and eluted with H2O (5 1) and then MeOH (5 1). The MeOH eluent (20 g) was charged to a silica gel (600 g) column and eluted with CHCl3-MeOH-H2O (8: 3:0.3) (5 1) to give 10 fractions. Each fraction was subjected to Sephadex LH-20 column and eluted with various ratios of MeOH-H2O (0一50%); nine major phenylethanoids 1 (607 mg), 2 (286 mg), 3 (43 mg), 4 (187 mg), 5 (1.1 g), 6 (100 mg), 7 (208 mg), 8 (168 mg), 9 (75 mg) were obtained.

Animals Male Std: Wistar rats (8 weeks old, 230一 250 g) were used. The animals were purchased from Shizuoka Laboratory Animal Center and were fed with a laboratory pellet chow (Clea Japan) and given water ad libitum.

Preparation of Rat Liver Microsomal Suspension Rat liver microsomal fraction was prepared by the method of Kiso et al.l9) but without pretreatment of phenobarbital. After animals had fasted for 24 h, the livers were perfused with an ice-cold 0.9% NaCl solution in situ, then cut into thin pieces and homogenized in cold 1.15% KC1 solution (1.0 ml/g of wet liver weight). The homogenate was diluted four times with 1.15% KC1 solution and centrifuged at 8000 g for 20 min at 4 °C. The supernatant fraction was then collected and ultracentrifuged at 105000 g for 60 min. The pellet obtained was resuspended in the same volume of 1.15% KC1 solution. Protein content was determined by the method of Lowry et using albumin as a standard.

Sample Preparation The tested samples were dissolved in water or ethanol and diluted with water to various concentrations. The final concentrations of ethanol were below 1% in all the assay systems except DPPH radical scavenging assay. Ethanol at a final concentration below 1 % showed no effect on these assay systems.

DPPH Radical Scavenging Effect The scavenging effect corresponded to the intensity of quenching DPPH radical, as described by Hatano et al.21) Five hundred /A of 60 卩m DPPH in EtOH was added to 500 /zl of sample solution and allowed to react for 30 min at room temperature, then the optical density was measured at 520 nm. For blank, EtOH was used instead of DPPH solution, and for control, H2O was used instead of sample solution. The IC50 values were calculated from regression lines where the abscissa represented the concentration of tested compound and the ordinate the average percent reduction of DPPH radical from four separate tests.

Effects on Superoxide Anion Radical Generation The production of superoxide anions in the xanthine/XOD system was determined using a half size of the method of Imanari et al.22) A mixture composed of 900/zl of 0.05m Na2CO3 (pH 10.2), 50 /il each of 3mM xanthine, 3mM EDTA, 1.5mg/ml BSA, 0.75 mM NBT and tested sample was added with 50 of 0.1 mg/ml XOD to start the reaction. After incubation for 30 min, the reaction was stopped with 50 /il of 6 mM CuCl2 and the absorbance was measured at 560 nm. The control solution was prepared in the same way, but 50 /il of H2O was used instead of the sample solution. In blank solution, 50 /il of H2O was used instead of XOD solution. IC50 values were calculated from regression lines where the abscissa represented the log concentration of tested compound and the ordinate the mean percent inhibition of NBT reduction of four independent tests.

Inhibitory Effects on the Activity of XOD Inhibitory effects on the activity of XOD were estimated by the method of Hatano et al.* with some modifications. A mixture consisting of 600/11 of a buffer (0.1 m phosphate solution, pH 7.5), 50 以 of XOD solution (0.068 U/ml in buffer) and 50 卩 1 of sample solution was pre-incubated for 10 min at 25 °C. Then, 300 pl of xanthine solution (0.1 mM in buffer) was added to the mixture, and the resulting solution was incubated for 30 min at 25 °C. The enzyme reaction was terminated by adding 1 n HC1 (100 /zl), and the absorbance of the reaction mixture was measured at 295 nm. The concentration of uric acid formed in the mixture was calculated from the absorbance, after subtracting the absorbance of the blank solution which was prepared as described above, except that XOD solution was substituted by 50 /zl of the buffer. The inhibitory effects on XOD were expressed by the percent inhibition (%) of formation of uric acid vs. that in control, in which water was used instead of sample solution. The well-known XOD inhibitor, allopurinol, was used as a positive control.

Inhibitory Effects on Lipid Peroxidation in Rat Liver Microsomes Lipid peroxidation in rat liver microsomes non-enzymatically induced by ascorbic/Fe2+ and enzymatically induced by ADP/NADPH/Fe3 + were measured by the method of Kiso et al.l9) with some modifications.

a) Ascorbic Acid/Fe2+ Induced Lipid Peroxidation in Rat Liver Microsomes: The reaction mixture was composed of 390/11 of 100mM KC1/45 mM Tris-HCl (pH 8.0), 50pl of microsomal fraction in 1.15% KC1 (20mg protein/ml) and 50 以 of a sample solution. After adding 10/11 of 10 mM ascorbic acid/0.5mM FeSO4 and then incubating at 37 °C for 20 min, the reaction mixture was cooled in ice to stop the reaction. The lipid peroxidation was measured by the thiobarbituric acid method24) and expressed as MDA (malondialdehyde) production. That

b) ADP/NADPH/Fe3 + Induced Lipid Peroxidation in Rat Liver Microsomes: The reaction mixture contained 290贝 of 100mM KC1/45him Tris-HCl (pH 8.0), 50/il of microsomal suspension in 1.15% KC1 (20mg protein/ml) and 50 以 of sample in water. Fifty of freshly prepared 20 mM ADP, 50//I of 1 mM NADPH and 10 of 2mM FeCl3 were added and then incubated at 37 °C for 30 min. Lipid peroxidation was measured and IC50 was calculated by the method above.

cistanche benefits: protect the liver

RESULTS

Structure Determination By direct comparison of their ^H-NMR and 13C-NMR data with the literature,2* the nine phenylethanoids were identified as 2,-acetylacteoside (1), cistanoside A (2), tubuloside A (3), echinacoside (4), acteoside (5), syringalide A 3'-a-rhamnopyranoside (6), tubuloside B (7), isoacteoside (8) and cistanoside F (9), respectively (Fig. 2). However, for acteoside, a representative compound of phenylethanoids, assignments of the 13C-NMR signals of C-3 and C-4 in aglycone moiety, C-3, C-4, C-5 and C-6 in caffeic moiety in previous reports25) were confusing. By the method of 2D-NMR such as HMBC and HMQC, we assigned unambiguously the 13C-NMR data of acteoside as aglycone moiety C-l: 131.49, C-2: 117.14, C-3: 146.07, C-4: 144.62, C-5: 116.34, C-6: 121.29, C-a: 72.33, C# 36.54; caffeoyl moiety C-l: 127.63, C-2: 115.26, C-3: 146.79, C-4: 149.78, C-5: 116.55, C-6: 123.24, C-a: 168.33, C-g: 114.68, C-y: 148.04; glucose moiety C-l': 104.16, C-2': 75.97, C-3': 81.66, C4: 70.39, C-5': 76.16, C-6': 62.35; and rhamnose moiety: C-l: 102.99, C-2: 72.22, C-3: 72.05, C-4: 73.79, C-5: 70.58, C-6: 18.46.

The structure-activity relationships were studied for these nine phenylethanoids by their DPPH radical and xanthine/XOD generated superoxide anion radical scavenging activities, ascorbic acid/Fe2 + and ADP/NADPH/ Fe3+ induced lipid peroxidation in rat liver microsomes. Caffeic acid and a-tocopherol which are well-known free radical scavengers and antioxidants were employed as positive control substances.

DPPH Radical Scavenging Activity All of the nine phenylethanoids showed stronger DPPH radical scav-enging activity than either a-tocopherol or caffeic acid, except that cistanoside A (2), syringalide A 3'-a-rhamno- pyranoside (6) and cistanoside F (9) showed weaker activity than caffeic acid (Fig. 3). The activity order was tubuloside B ⑺(IC50 = 2.99 ^m) > echinacoside ⑷ (IC50 = 3.29 fiM)M 2z-acetylacteoside ⑴(IC50 = 3.30 /im) M tubuloside A (3) (IC50 = 3.34/im) acteoside (5) (IC50 = 3.36 〃m)> isoacteoside (8) (IC50 = 3.49 ^m) >caffeic acid (IC50 = 4.79 /2M)> cistanoside A (2) (IC5O = 4.87 /im)> syringalide A S'-a-rhamnopyranoside (6) (IC5O = 5.52 /zm) > cistanoside F (9) (IC50 = 6.29 /im) > a-tocopherol (IC50= 10.2//m). Tubuloside B ⑺，echinacoside (4), 2/-acetylacetoside (1), tubuloside A (3), acteoside (5) and isoacteoside (8) (IC5O = 2.99一3.49 /im), each of which has four phenolic hydroxyl groups in the molecule similarly showed stronger activity than cistanoside A (2) whose 3-OH of aglycone moiety was methylated. Syringalide A 3z-a-rhamnopyranoside (6) which has only one OH group in the aglycone moiety showed relatively weak activity. Cistanoside F (9), which has only two hydroxyl groups located at caffeoyl moiety but has no phenylethanol aglycone, exhibited the weakest activity.

Effects on Superoxide Anion Radical Generation All the tested compounds exhibited a stronger inhibitory activity than a-tocopherol but weaker than caffeic acid on the generation of superoxide anion radical (Fig. 4). The activity order was caffeic acid (IC5()= 1.82jUM)>2'- acetylacteoside (1) (IC50 = 2.25 /im) tubuloside B ⑺ (IC50 = 2.34jUM) >echinacoside (4) (IC50 = 2.74juM)^ isoacteoside (8) (IC50 = 2.88 /zm) acteoside (5) (IC50 = 2.89 f/M) >cistanoside F (9) (IC5o = 3.13 rm) tubuloside A (3) (IC50 = 3.17 jUM)syringalide A 3'-a-rhamnopyrano- side (6) (IC50 = 3.20^m)>cistanoside A (2) (IC5O = 3.69 jUm) > a-tocopherol (IC50 > 10 〃m).

Inhibitory Effects on XOD Activity Only isoacteoside (8) and tubuloside B (7), whose caffeoyl moiety is at 6z-position of the glucose, had inhibitory effects on the activity of XOD, and other phenylethanoids, which have caffeoyl moiety at the ^-position of the glucose, showed no effects at the tested concentrations (25一200 m) (Table 1). isolated nine major phenylethanoid compounds from this extract and determined their structures by NMR. system as tubuloside B (7)M2'-acetylacteoside (1) iso-acteoside (8) echinacoside (4) tubuloside A (3) M acteoside (5) > syringalide A 3'-a-rhamnopyranoside (6) > cistanoside A (2) > cistanoside F (9); in ADP/NADPH/Fe3 + system as tubuloside B (7)2,-acetylacteoside (^^isoacteoside (8) acteoside (5) > tubuloside A (3) > echinacoside (4) > syringalide A 3z-a-rhamnopyranoside (6) > cistanoside A (2) > cistanoside F (9). Both these orders were similar to that in the above free radical scavenging tests, especially in the DPPH radical scavenging test. Still, the number of phenolic hydroxyl groups in the molecule played the most important role in the inhibition of lipid peroxidation in rat liver microsomes.

effects of cistanche

DISCUSSION

We tested the free radical scavenging activities of the various solvent extracts of the stem of Cistanche deserticola on DPPH radical and found that the acetone-H2O (9:1) extract exhibited the strongest activity. To find the active constituents of the free radical scavenging activity, we isolated nine major phenylethanoid compounds from this extract and determined their structures by NMR.

The antioxidative effects of these phenylethanoids were evaluated by their free radical scavenging activity and anti-lipid peroxidation activity. The widely known a-tocopherol was used as a positive control substance. As the lipophilicity of a-tocopherol may affect its activities in the present assay systems except in DPPH radical scavenging assay, we took caffeic acid as another positive control substance.

First, we carried out tests on the free radical scavenging activities. All the 9 tested phenylethanoids undoubtedly exhibited stronger DPPH radical and superoxide anion radical scavenging activities than a-tocopherol, but some were stronger and others were weaker than caffeic acid. The radical scavenging activity increased with an increase in the number of phenolic hydroxyl groups. The incomplete parallelism of DPPH radical and superoxide anion scavenging activities was believed to be due to the different characters between the radical species and some other factors involved in these two reaction systems,21,26) for the DPPH radical is a chemically induced radical which reacts with radical scavengers in a simple chemical manner, while the superoxide anion radical is generated by xanthin/XOD, an enzyme reaction.

If the phenylethanoids inhibit the xanthine/XOD generated superoxide anion radical, the inhibition might be regarded as the result of their superoxide anion radical scavenging activity or (and) their inhibition of the enzyme XOD,23) so we carried out a test of their inhibitory effects of xanthine oxidase. It is interesting that, selectively, only isoacteoside (8) and tubuloside B (7), whose caffeoyl moiety is at 6/-position of the glucose, demonstrated inhibition. They exhibited an effect at the concentration of 25—200jtiM on 3.4xlO^3U/ml (10jug/ml) of XOD, although the activity was weaker than allopurinol; other compounds, whose caffeoyl moiety is at《-position of the glucose, showed no inhibitory effect. This result provided the first report of the inhibitory effect of phenylethanoids on XOD enzyme. Chan et al21) reported that caffeic acid and some of its analogs had inhibitory effects on XOD activity, however, we found no such effect of caffeic acid under the present reaction conditions. In this test, the XOD concentration was similar to that in the test of effect on superoxide anion radical generation (4.3 /zg/ml), but the concentration of compounds was much higher than in the latter. Therefore, inhibition of the Oj generation by these phenylethanoids is due to their radical scavenging activity, but not to their inhibitory activity of the XOD enzyme.

That free radical chain reaction finally leads to lipid peroxidation is well known.28,29) We therefore studied these phenylethanoid compounds for their effects on the lipid peroxidation in rat liver microsomes induced by the enzymatic system ADP/NADPH/Fe3 + and that induced by the non-enzymatic system ascorbic acid/Fe2 + . All these compounds showed stronger anti-lipid peroxidation effects than caffeic acid and a-tocopherol in both systems. It is generally accepted that both systems are catalyzed by iron ions, either Fe2+ or Fe3 + , involved with lipid peroxyl radicals,30,31) and that hydroxyl radical OH plays the most important role in lipid peroxidation,32 _34) although the initiation of OH has been questioned in some reports.35祁6) On the other hand, superoxide anion radical OJ, which is the primary product of e_ attack on O2 in the radical chain, is a rather poorly reactive radical and does not result in lipid peroxidation itself. Although these phenylethanoids showed weaker scavenging activity on superoxide anion radical than caffeic acid, they showed much stronger anti-lipid peroxidation action than the latter. Therefore, we believe that they, like some other aromatic, chain-breaking polyphenols, may inhibit the above lipid peroxidation in rat liver microsomes by chelating Fe2+ or Fe3+ ions, and also by scavenging superoxide radical Oj to break down radical chain reaction.37,38) Moreover, they may also scavenge more toxic radical species such as hydroxyl radical and lipid peroxyl radicals to directly interrupt lipid peroxidation at higher potency than caffeic acid.38,39)

Li et al. studied some phenylethanoid glycosides including acteoside (verbascoside) (5), isoacteoside (iso- verbascoside) (1), and echinacoside (4) from Pedicularis for their inhibition of the autoxidation of linoleic acid in micelles, a non-biological system,13) for their protection against oxidative hemolysis in vitro,14''1 and also for their scavenging effects on NBT/PMS/NADH generated superoxide and inhibition of lipid peroxidation induced by ascorbic acid/Fe2+ in mouse liver microsomes.15) A similar activity-structure relationship was observed in the reports of Li et al. and our results. That is, the activities of phenylethanoids for inhibiting lipid peroxidation depend mainly on the number and steric position of phenolic hydroxyl groups.15) On the basis of their stereochemistry, the phenolic hydroxyl groups of caffeoyl moiety on acteoside (5), tubuloside A (3), and echinacoside (4), which have the caffeoyl at 4z-position of the glucose, would more easily to form a hydrogen bond with the hydroxyl groups of rhamnosyl than those on isoacteoside (8) and tubuloside B (7), which have the caffeoyl at 6'- position of the glucose. Thus, the latter reserved more free phenolic hydroxyl groups than the former and therefore showed stronger anti-lipid peroxidation activity. We also noticed that 2z-acetylation of these phenylethanoids, although slightly, enhanced their antioxidative effects. For example, acteoside (5) and isoacteoside (8), when 2'- acetylated into 2/-acetylacteoside (1) and tubuloside B (7), respectively, showed stronger activities in all four assay systems. Because of the superiority of stereochemistry and 2'-acetylation, tubuloside B (7) exhibited the strongest antioxidative activity among the tested samples. Perhaps due to the substitution of a third sugar moiety at 6'~position, this correlation did not fit well in the case of tubuloside A (3) or echinacoside (4); however, in the ADP/NADPH/Fe3 + system, tubuloside A ⑶ still exhibited stronger anti-lipid peroxidation activity than echinacoside (4).

In addition, the order of DPPH radical scavenging activity showed better parallelism with that of anti-lipid peroxidation than did the order of superoxide anion radical scavenging activity. This proved again that DPPH radical scavenging assay is a convenient and reliable method for antioxidant assay,26) although DPPH radical is a chemically induced radical while superoxide anion radical could be a biologically endogenous radical.

In conclusion, all of the nine phenylethanoids exhibited significant free radical scavenging activities and anti-lipid peroxidation effects in the present study. The antioxidative effect was potentiated by an increase in the number of phenolic hydroxyl groups in the molecule. As can be imagined, the antioxidative effects of these phenylethanoids proved here may play an important role in the actions of the drug Cistanchis Herba and may partly explain the mechanisms of the activities of some phenylethanoids against neoplasm,8* inflammation10) and nephritis,1 n in which free radicals are seriously involved.

Cistsanche Echinacoside: Anti-oxidation

REFERENCES

1) Namba T., "The Encyclopedia of Wakan-Yaku (Traditional Sino-Japanese Medicines) with Color Pictures,5, Vol. II, Hoikusha, Tokyo, 1994, pp. 16—17.

2) a) Kobayashi H,, Karasawa H., Miyase T., Fukushima S., Chem. Pham. BulL, 32, 3009—3014 (1984); b) Idem, ibid., 32, 3880—3885 (1984); c) Idem, ibid., 33, 1452—1457 (1985); d) Karasawa H., Kobayashi H., Takizawa N., Miyase T., Fukushima S., Yakugaku Zasshi, 106, 562—566 (1986); e) Idem, ibid., 106, 721—724 (1986); /) Kobayashi H., Oguchi H., Takizawa N., Miyase T., Ueno A., Usmanghani K., Ahmad M., Chem. Pharm. Bull., 35, 3309一3314 (1987); g) Yoshizawa F., Deyama T., Takizawa N., Usmanghani K., Ahmad M., ibid., 38, 1927—1930 (1990).

3) a) Kobayashi H, Komatsu J., Yakugaku Zasshi, 103, 508一511 (1983); b): Kobayashi H., Karasawa H., Miyase T., Fukushima S., Chem. Pharm. Bull., 32, 1729—1734 (1984); c) Idem, ibid., 33, 3645—3650 (1985).

4) Naran R., Ebringerova A., Hromadkova Z., Patoprsty V., Phytochemistry, 40, 709—715 (1995).

5) Sato T., Kozima S., Kobayashi K., Kobayashi H., Yakugaku Zasshi. 105, 1131—1144 (1985).

6) Jin X. L., Zhang Q. R., China J. Chinese Materia Medica, 19, 695—697 (1994).

7) Yamahara J., Kitani T., Kobayashi H., Kawahara Y., Yakugaku Zasshi, 110, 932—935 (1990).

8) a) Pettit G. R・，Numata A., Takemura T., Ode R. H., Narula A, S., Schmidt J. M., Cragg G. M., Pase C. P., J. Nat. Prod., 53, 456— 58 (1990); b) Saracoglu I., Inoue M., Calis I., Ogihara Y., Biol. Pharm, Bull., 18, 1396—1400 (1995).

9) Andary C., '4Caffeic Acid Glycoside Esters and Pharmacology/ ed. by Scalbert A., Polyphenolic Phenomena, INRA Editions, Paris, 1993, pp. 237—245.

10) Murai M., Tamayama Y., Nishibe S., Planta Med.. 61, 479― 80 (1995).

11) Hayashi K., Nagamatsu T., Ito M., Hattori T., Suzuki Y., Jpn. J. Pharmacol., 66, 47一52 (1994).

12) a) Molnar J., Gunics G., Mucsi I., Koltai M., Petri L, Shoyama Y., Matsumoto M., Nishioka L, Acta Microbiol. Hung.. 36, 425― 32 (1989); b) Sasaki H., Nishimura T., Morota T., Chin M., Mitsuhashi H” Komatsu Y., Maruyama H., Tu G. R., He W., Xiong Y. L., Planta Med., 55, 458二462 (1989).

13) Li J., Wang P. F., Zheng R. L., Liu Z. M., Jia 乙 J., Planta Med., 59, 315—317 (1993).

14) Zheng R. L., Wang P. F., Li J., Liu Z. M,, Jia Z. J., Chem. Phys. Lipids. 65, 151—154 (1993).

15) Li J., Zheng R. L., Liu 乙 M., Jia 乙 J., Acta Pharmacol. Sinica, 13, 427T30 (1992).

16) Barber D. A., Harris S. R., American Pharmacy. 34, 26一35 (1993).

17) Elstner E. F., Klin Wochenschr, 69, 949—956 (1991).

18) Martin G. R., Danner D. B., Holbrook N. J., Ann. Rev. Med., 44, 419—29 (1993).

19) Kiso Y., Tohkino H., Hattori M., Sakamoto T., Namba T., Planta Med.. 50, 298—302 (1984).

20) Lowry O. H., Rosebrough N. J., Farr A. L., Randall R. J., J. Biol. Chem.. 193, 265—275 (1951).

21) Hatano T., Edamatsu R., Hiramatsu M., Mori A., Fujita Y., Yasuhara T” Okuda T., Chem. Pharm. Bull., 37, 2016一2021 (1989).

22) Imanari T., Hirota M., Miyazaki M., Igaku No Ayumi, 101, 496—497 (1977).

23) Hatano T., Yasuhara T, Yoshihara R., Agata L, Noro T., Okuda T., Chem. Pharm. Bull., 38, 1224—1229 (1990).

24) Ohkawa H., Ohishi N., Yagi K., Anal. Biochem., 95, 351一358 (1979).

25) a) Lahloub M, F., Zaghloul A. M., El-Khayat S. A., Afifi M. S.， Sticher O., Planta Med.. 57, 481— 85 (1991); b) Nishimura H., Sasaki H., Inagaki N., Chin M. (Zhen 乙 X.), Mitsuhashi H., Phytochemistry. 30, 965—969 (1991); c) Budzianowski J., Skrzypczak L., ibid,, 38, 997—1001 (1995),

26) Marsden S. B., Nature (London), 181, 1199—1200 (1958).

27) Chan W. S., Wen P. C., Chiang H. C., Anticancer Res., 15, 703一 708, (1995).

28) Janero D. R., Burghardt B., Biochem, Pharmacol.. 37, 3335一3342 (1988).

29) Buettner G. R., Arch. Biochem. Biophys., 300, 535一543 (1993).

30) Herbert V., Shaw S., Jayatilleke E., Stopler-Kasdan T., Stem Cells, 12, 289—303 (1994).

31) Aust S. D., Miller D. M., Samokyszyn V. M.，Methods Enzymol., 186, 457—463 (1990).

32) Kubota S., Ikegami Y., Kurokawa T., Sasaki R., Sugioka K., Nakano M., Biochem. Biophys. Res, Commun., 108, 1025一1031 (1982).

33) Yagi K., Ishida N., Komura S., Ohishi N., Kusai M., Kohno M., Biochem. Biophys. Res. Commun., 183, 945一951 (1992).

34) Hockenbery D. M., Oltvai Z. N., Yin X. M., Milliman C. L., Korsmeyer S. J., Cell. 75, 241—251 (1993).

35) Bucher J. R., Tien M., Aust A. D., Biochem. Biophys. Res, Commun., Ill, 777—784 (1983).

36) Yoshida T., Otake H., Aramaki Y., Hara T., Tsuchiya S., Hamada A., Utsumi H., Biol. Pharm. Bull.. 19, 779—782 (1996).

37) Afanas5 E. V. I. B., Dorozhko A. L, Brodskii A. V., Kostyuk V. A., Potapovitch L A., Biochem. Pharmacol., 38, 1763—1769 (1989).

38) Robak J., Gryglewski R. J., Biochem. Pharmacol.. 37, 837—841 (1988).

39) Chimi H., Morel I., Lescoat G., Pasdeloup N・，Cillard P., Cillard J., Toxicology in Vitro, 9, 695一702 (1995)