Biol. Pharm. Bull. 25(10) 1328—1332 (2002) Vol

1328 Biol. Pharm. Bull. 25(10) 1328—1332 (2002) Vol. 25, No. 10

Antithrombotic and Antiallergic Activities of Daidzein, a Metabolite of Puerarin and Daidzin Produced by Human Intestinal Microﬂora

a a b ,a,c Min-Kyung CHOO, Eun-Kyung PARK, Hae-Kyung YOON, and Dong-Hyun KIM* a College of Pharmacy, Kyung Hee University; c East-West Research Institute, Kyung Hee University; 1, Hoegi, Dongdaemun-ku, Seoul 130–701, Korea: and b Department of Human Life Science, Kyungwon College; 65, Bokjong, Soojong-ku, Songnam-city, Kyonggi-do, 461–701 Korea. Received February 21, 2002; accepted July 2, 2002

To evaluate the antithrombotic activities of puerarin and daidzin from the rhizome of Pueraria lobata, in vitro and ex vivo inhibitory activities of these compounds and their metabolite, daidzein, were measured. These compounds inhibited ADP- and collagen-induced platelet aggregation. Daidzein was the most potent. However, when puerarin and daidzin were intraperitoneally administered, their antiaggregation activities were weaker than when these compounds were administered orally. When in vivo antithrombotic activities of these compounds against collagen and epinephrine were measured, these compounds showed significant protection from death due to pulmonary thrombosis in mice. To evaluate the antiallergic activity of puerarin, daidzin, and daidzein, their inhibitory effects on the release of b-hexosaminidase from RBL 2H3 cells and on the passive cutaneous anaphylaxis (PCA) reaction in mice were examined. Daidzein exhibited potent inhibitory activity on the b- hexosaminidase release induced by DNP-BSA and potently inhibited the PCA reaction in rats. Daidzein administered intraperitoneally showed the strongest inhibitory activity and significantly inhibited the PCA reaction at doses of 25 and 50 mg/kg with inhibitory activity of 37 and 73%, respectively. The inhibitory activity of intraperitoneally administered daidzein was stronger than those of intraperitoneally and orally administered puerarin and daidzin. Therefore we believe that puerarin and daidzin in the rhizome of Pueraria lobata are prodrugs, which have antiallergic and antithrombotic activities, produced by intestinal microflora. Key words antithrombosis; antiallergy; puerarin; daidzein; intestinal bacteria

The rhizome of Pueraria lobata (family Leguminosae) is MATERIALS AND METHODS frequently used for the liver diseases, strokes, and allergy as a crude substance taken orally in Asian countries as a tradi- Materials Adenosine 5Ј-diphosphate (ADP), epineph- tional ingredient of polyprescriptions.1) Some isoflavones, rine, collagen, bovine serum albumin, prothrombin, thrombo- puerarin, and daidzin, were isolated from the rhizome of plastin, thrombin, hyaluronidase from bovine testis, p-nitro- 2,3) Pueraria lobata as the main component. Among them, phenyl N-acetyl-b -D-glucosaminide, anti-dinitrophenol daidzin has acetaldehyde dehydrogenase and cAMP phos- (DNP)-IgE, DNP-bovine serum albumin (BSA), Evans blue, phoesterase inhibitory and antispasmatic activities,4—6) and hyaluronic acid potassium salts, azelastine, xanthine oxidase puerarin has hypoglycemic and coronary artery blood flow (XOD), 2,2-diphenyl-1-picrylhydrzyl (DPPH), and disodium increasing activity.7,8) cromoglycate (DSCG) were purchased from Sigma Chemical Most herbal medicines are administered orally and their (U.S.A.). Puerarin and daidzin were isolated from the rhi- components are inevitably brought into contact with intesti- zome of P. lobata according to the method of Kakegawa et nal microflora in the alimentary tract. Most components are al.,12—14) and daidzein was isolated according to our previous metabolized by the intestinal bacteria before absorption from method.15) The other chemicals were of analytical reagent the gastrointestinal tract.9) Related to the metabolism of puer- grade. arin and daidzin, Yasuda et al. reported that daidzein and its Animals Sprague-Dawley rats (male, 180—220 g) and conjugates were found in the urine of rats orally administered ICR mice (male, 20—24 g) were purchased from Sam Yook puerarin and daidzin.10,11) We also reported that puerarin and Animal Co. (Korea) and acclimatized for 1 week at a temper- daidzin are transformed to daidzein and calycosin by human ature of 22Ϯ1 °C and humidity of 55Ϯ5% with free access intestinal microflora. Based on these findings, we thought to a commercial pellet diet (Samyang Co., Korea) and drink- that these compounds could be metabolized to their agly- ing water before the experiments. Animal experiments were cones by intestinal bacteria before absorption into the blood carried out in accordance with international guidelines. and expression of their pharmacological actions. However, Preparation of Platelets Blood from rats was collected the relationship between the pharmacological actions, such by cardiac puncture into a plastic flask containing 2.2% as antithrombotic and antiallergic effects, and the metabolism sodium citrate (1 : 9 v/v). Platelet-rich plasma (PRP) was pre- of these compounds by human intestinal microflora was not pared by centrifugation of the blood at 120ϫg for 15 min and completely clarified. further centrifuged at 850ϫg for 10 min to prepare platelet- Therefore we investigated the antithrombotic and antialler- poor plasma (PPP).16) The supernatant was pooled and cen- gic activities of puerarin, daidzin, and daidzein, which is a trifuged at 600ϫg for 15 min at room temperature. The metabolite of puerarin and daidzin produced by human in- platelet pellets were washed with modified Tyrode-HEPES testinal microflora. buffer (129 mM NaCl, 2.8 mM KCl, 8.9 mM NaHCO3, 0.8 mM MgCl2, 0.8 mM KH2PO4, 2 mM EGTA, 5.6 mM glucose, 10 mM HEPES, 0.35% BSA, pH 7.4) and centrifuged at 600ϫg for 15 min. Then platelet pellets were gently resus-

* To whom correspondence should be addressed. e-mail: [email protected] © 2002 Pharmaceutical Society of Japan October 2002 1329 pended in Tyrode-HEPES buffer and used in the experi- allergic reactions (degranulation). The reaction was stopped ments. by cooling in an ice bath for 10 min. The reaction mixture Assay of in Vitro Antiplatelet Aggregation Platelet ag- was centrifuged, and 25 ml aliquots of supernatant were gregation was measured by turbimetry using a dual-channel transferred to 96-well plates and incubated with 25 ml of sub- Whole Lumini-Ionized Calcium Aggregometer (Chrono-Log strate (1 mM p-nitrophenyl-N-acetyl-b-D-glucosaminide) for Co., Ltd., Havertown, PA, U.S.A.) according to the method 1 h at 37 °C. The reaction was stopped by adding 0.1 N of Born and Cross.17) Brieﬂy, rat PRP (300 ml) was incubated NaOH. The absorbance was measured using an ELISA at 37 °C for 2 min in the aggregometer with stirring at reader at 405 nm. 1200 rpm and then stimulated with ADP and collagen. The Passive Cutaneous Anaphylaxis (PCA) reaction An samples or aspirin, as the reference agent, were incubated IgE-dependent cutaneous reaction was measured according with PRP for 3 min, followed by the addition of the aggrega- to the method of Katayama et al.21) Five male ICR mice tion agents. Changes in light transmission were recorded for (25—30 g) in each group were used. Mice were injected in- 10 min after stimulation with these agents. Each inhibition tradermally with 10 mg of anti-DNP IgE into each of four rate was obtained from the maximal aggregation induced by dorsal skin sites that had been shaved 48 h earlier. The sites the respective agonist at several concentrations, and then the were outlined with a water-insoluble red marker. Forty-eight

IC50 values were calculated from the data using a Probit hours later, each mouse received an injection of 200 ml of method. 0.25% Evans blue-PBS containing 100 mg of DNP-HSA via In Vitro Coagulation Parameters The plasma clotting the tail vein. Puerarin, daidzin, daidzein, or DSCG was ad- times, activated partial thromboplastin time (APTT), pro- ministered 1 h prior to DNP-HSA injection. Thirty minutes thrombin time (PT), and thrombin time (TT) were measured after DNP-HSA injection, the mice were killed and their dor- according to the manufacturer’s protocol. The PPP was incu- sal skins were removed for measurement of the pigmented bated with the samples for 7 min at 37 °C, and coagulation area. After extraction with 1 ml of 1.0 N KOH and 9 ml of a was started with by adding CaCl2, 100 ml of thromboplastin, mixture of acetone and phosphoric acid (5 : 13), the amount and 100 ml of bovine thrombin to the 100 ml of incubated of dye was determined colorimetrically (absorbance at 620 plasma for the APTT, PT, and TT assays, respectively. nm). Assay of ex Vivo Antiplatelet Aggregation Five male Assay of Hyaluronidase Activity Hyaluronidase activ- rats in each group were used after overnight fasting. Rats ity was determined using the method of Kakegawa et al.13) were orally or intraperitoneally administered the samples (25 Assay of Antioxidant Activity DPPH (1,1-diphenyl-2- or 50 mg/kg) as a vehicle. Blood was collected 3 h after sam- picrylhydrazyl) radical scavenging and XOD-inhibitory ac- ple treatment and PRP as previously described. Platelet ag- tivities of compounds isolated were measured according to 22) gregation was induced by 80 mg/ml of collagen or 8 m M of the method of Xiong et al. Superoxide anion radical gener- ADP. Antiplatelet activities of the sample were investigated ation of compounds isolated was measured according to the according to the method of Kimura et al.18) method of Xiong et al.22) Assay of in Vivo Antithrombotic Activity The antithrombotic effect in the samples was investigated using the RESULTS mouse thromboembolism test according to the method of Di Minno et al.19) Ten male ICR mice in each group were used Antithrombotic Activities of Puerarin, Daidzin, and after overnight fasting. The samples (25, 50 mg/kg) and 0.5% Daidzein In vitro inhibitory activities of puerarin, daidzin, CMC solution were administered orally. A mixture solution and daidzein against ADP- and collagen-induced platelet of collagen (110 mg) and epinephrine (13 mg) was injected aggregation were measured (Table 1). These compounds in- into the mouse tail vein 3 h after oral administration of the hibited ADP- and collagen-induced platelet aggregation. samples. The number of dead or paralyzed mice was Daidzein was the most potent. Particularly, it inhibited recorded up to 15 min, and the percentage of protection was platelet aggregation induced by collagen more potently than Ϫ ϩ ϫ calculated as follows: [1 (dead paralyzed)/total] 100. that induced by ADP. Its IC50 were 0.2 and 0.05 mg/ml, re- Assay of Antiallergic Activity in RBL-2H3 Cell Line spectively. Against ex vivo ADP- and collagen-induced rat The inhibitory activity of tested compounds against the re- platelet aggregation, these compounds also exhibited potent lease of b-hexosamindase from RBL-2H3 cells was evalu- 20) ated according to the method of Choi et al. RBL-2H3 cells Table 1. Effect of Puerarin, Daidzin, and Daidzein on in Vitro Platelet Ag- were grown in DMEM supplemented with 10% fetal bovine gregation serum and L-glutamine. Before the experiments, cells were 5 a) dispensed into 24-well plates at the concentration of 5ϫ10 IC50 (mg/ml) cells per well using medium containing 0.5 mg/ml of mouse Sample ADPb) Collagenc) monoclonal IgE and were incubated overnight at 37 °C in 5% Ͼ Ͼ CO2 for sensitization of cells. The cells were washed with P. lobata (rhizome) 1 1 500 ml of siraganian buffer (pH 7.2, 119 mM NaCl, 5 mM Puerarin 0.34 0.5 Daidzin 0.38 0.63 KCl, 0.4 mM MgCl , 25 mM PIPES, 40 mM NaOH) and incu- 2 Daidzein 0.20 0.05 bated in 160 ml of siraganian buffer containing 5.6 mM glu- Aspirin 0.04 0.17 cose, 1 mM CaCl2, and 0.1% BSA for an additional 10 min at 37 °C. Then cells were exposed to 40 ml of test materials for a) 50% inhibitory concentration (IC50) was calculated as follows: (control aggregation (%)Ϫherbal medicine-treated aggregation (%))/control aggregation (%)ϫ100ϭin- 20 min, followed by treatment with 20 ml of antigen (DNP- hibition (%). b) Final concentration was 0.008 mM. c) Final concentration was BSA, 1 mg/ml) for 10 min at 37 °C to activate cells to evoke 0.08 mg/ml. 1330 Vol. 25, No. 10

Table 2. Effect of Puerarin, Daidzin, and Daidzein on ex Vivo Antiplatelet Table 4. Antithrombotic Activity of Puerarin, Daidzin, and Daidzein Aggregation Activity Sample Dose (mg/kg) Route Protection (%) Platelet aggregation (%) Dose Samplea) Route Control Vehicle — 20 (mg/kg) ADPb) Collagenc) P. lobata (rhizome) 1000 p.o. 60 Puerarin 25 p.o. 20 Control Vehicle — 60.5Ϯ0.7 56.0Ϯ1.4 50 p.o. 40 Pueraria lobata 1000 p.o. 43.0Ϯ1.4* 53.5Ϯ0.7* 25 i.p. 20 (rhizome) Daidzin 25 p.o. 40 Puerarin 25 p.o. 43.0Ϯ1.4* 45.0Ϯ1.4* 50 p.o. 40 50 p.o. 17.5Ϯ9.2* 3.0Ϯ4.2* 25 i.p. 20 50 i.p. 47.0Ϯ5.7* 48.0Ϯ24.0* Daidzein 25 i.p. 60 Daidzin 25 p.o. 61.5Ϯ4.9 53.5Ϯ4.9 Aspirin 25 p.o. 30 50 p.o. 34.5Ϯ6.4* 4.0Ϯ2.8* 50 p.o. 60 50 i.p. 42.5Ϯ16.3* 44.0Ϯ0.1* Daidzein 25 i.p 37.5Ϯ10.6* 6.5Ϯ0.7* The samples were administered 3 h before tail vein injection of epinephrine and col- Aspirin 50 p.o. 47.5Ϯ3.5* 40.5Ϯ3.5* lagen, and the assay of antithrombotic activity was performed according to Materials and Methods. a) Samples were administered 3 h before the test, and the assay of ex vivo antiplatelet aggregation activity was performed according to Materials and Methods. b) Final concentration was 0.008 mM. c) Final concentration was 0.08 mg/ml. The results were expressed as meanϮS.D. (nϭ5). * Signiﬁcantly inhibited compared with control group ( pϽ0.05).

Table 3. Effect of Puerarin, Daidzin, and Daidzein on Human Plasma Co- agulation Time

Coagulation time (s) Concentration Sample (mg/ml) APTT PT TT

Control 37.7Ϯ0.35 33.1Ϯ1.77 35.3Ϯ1.48 Puerarin 1.5 40.8Ϯ1.0 39.1Ϯ4.6 34.9Ϯ2.2 Daidzin 1.5 37.1Ϯ2.4 36.6Ϯ0.4 35.4Ϯ2.8 Fig. 1. Inhibitory Effect of Puerarin, Daidzin, and Dadizein on b-Hexos- Daidzein 1.5 36.6Ϯ3.9 39.4Ϯ1.5* 36.6Ϯ1.1 amindase Release from RBL-2H3 Cells by DNP-BSA Heparin 3ϫ10Ϫ3 174.8Ϯ22.4* 57.0Ϯ7.7* Ͼ500* ᭹ ᭜ ᭿ ᭡ ϫ 3ϫ10Ϫ4 40.6Ϯ1.0 48.2Ϯ4.9* 45.9Ϯ0.6* , Puerariae Radix; , puerarin, , daidzin; , daidzein; , disodium cromglycate (DSCG). All values are the means of three independent experiments. The results were expressed as meanϮS.D. (nϭ3). * Signiﬁcantly inhibited compared with control group ( pϽ0.05). Table 5. Inhibitory Effect of Puerarin, Daidzin, Daidzein, and DSCG against the PCA Reaction inhibitory activity (Table 2). When puerarin and daidzin were Inhibition (%) administered intraperitoneally, their antiaggregation activities Dose (mg/kg) were weaker than when these compounds were administered p.o. i.p. orally. The rhizome of P. lobata did not inhibit in vitro ADP- P. lobata (rhizome) 500 32Ϯ8.2 — and collagen-induced platelet aggregation. However, it inhib- Puerarin 100 34Ϯ10.3 — ited ex vivo ADP- and collagen-induced platelet aggregation 50 26Ϯ16.9 46Ϯ1.0 Ϯ even if orally administered at a high dose. Aspirin, a refer- Daidzin 100 46 17.3 — 50 25Ϯ4.4 23Ϯ6.2 ence drug widely used as an antiplatelet aggregation agent in Daidzein 50 — 73Ϯ5.7 clinical practice, potently inhibited in vitro and ex vivo 25 — 37Ϯ11.9 platelet aggregation. The antiplatelet aggregation activities of DSCGa) 100 37Ϯ0.2 — these compounds were more potent than that of aspirin. Azelastine 10 81Ϯ15.4 80Ϯ9.0

In vitro inhibitory activities of these compounds against a) Disodium chromoglycate. All agents were administered p.o. or i.p. prior to chal- APTT, PT, and TT-induced human plasma coagulation were lege with antigen, and the assay of PCA reaction-inhibitory activity was performed ac- measured (Table 3). These compounds did not show potent cording to Materials and Methods. The results were expressed as meanϮS.D. (nϭ5). inhibitory activity. In vivo antithrombotic activities of these compounds against collagen and epinephrine were measured tested compounds exhibited inhibitory activity. Daidzein and (Table 4). These compounds and the rhizome of P. lobata daidzin inhibited more potently the release of b-hexos- gave signiﬁcant protection against death due to pulmonary aminidase. Their inhibitory activities were comparable to that thrombosis in mice. Among them, intraperitoneally adminis- of DSCG. To determine their inhibitory effects on the PCA tered daidzein exhibited the most potent protective activity. reaction in mice, these compounds were administered orally Its activity was comparable to that of aspirin at the dose of or intraperitoneally 60 min prior to challenge with antigen 50 mg/kg. (Table 5). These compounds inhibited the PCA reaction on Antiallergic Action of Puerarin, Daidzin, and Daidzein rats. Among the substances tested, daidzein administered in- To evaluate the antiallergic activity of puerarin, daidzin, and traperitoneally showed the strongest inhibitory activity and daidzein, we then examined their inhibitory effects on the re- signiﬁcantly inhibited the PCA reaction at doses of 25 and lease of b-hexosaminidase from RBL-2H3 cells (Fig. 1). All 50 mg/kg with inhibitory activity of 37 and 73%, respec- October 2002 1331

Table 6. Antioxidant and Hyaluronidase-Inhibitory Activities of Puerarin, Table 7. Transforming Activity of Puerarin and Daidzin by Human Fecal Daidzin, and Daidzein Suspension and Mouse Liver

a) a) IC50 (mM) Activity (nmol/min/mg) Enzyme source Agent Superoxide Puerarin Daidzin Hyaluronidase DPPHb) XODc) anion Mouse liverb) 0 0.8Ϯ0.22 P. lobata Ͼ1 Ͼ1 Ͼ0.2 Ͼ0.2 Human fecesc) 1.2Ϯ0.41 44.0Ϯ12.3 (rhizome) Puerarin Ͼ1 Ͼ1 Ͼ0.2 Ͼ0.2 a) The reaction mixture containing 0.3 ml of 1 mM puerarin (or daidzin), 0.1 ml of an Daidzin Ͼ1 Ͼ1 Ͼ0.2 Ͼ0.2 enzyme source, and 0.6 ml of 10 mM sodium phosphate buffer pH 7.0, was incubated at 37 °C at 2 h, and extracted with 2 ml of ethylacetate, and analyzed by TLC (developing Daidzein Ͼ1 0.94 0.18 0.06 ϭ d) solvent, CHCl3 : MeOH 4 : 1) scanner (Shimadzu CS-9301PC). b) Livers of ICR Caffeic acid — 0.004 0.014 — mice were homogenized in 20 mM sodium phosphate buffer, pH 7.0, and centrifuged at Allopurinol — — — 0.04 5000ϫg for 30 min. The resulting supernatant was used as a crude enzyme. c) Fresh human feces were suspended in 20 mM sodium phosphate buffer, pH 7.0, and cen- a) Inhibitory concentration was calculated according to Materials and Methods. b) trifuged at 100ϫg for 10 min. The supernatant was centrifuged at 5000ϫg for 30 min. DPPH, 2,2-diphenyl-1-picrylhydrzyl. c) XOD, xanthine oxidase. d) Not deter- The precipitate was sonicated by ultrasonic and centrifuged at 5000ϫg for 30 min. The mined. All values are the means of three independent experiment. resulting supernatant was used as a crude enzyme. The results were expressed as meanϮS.D. (nϭ5). tively. The inhibitory activity of intraperitoneally adminis- rosis, heart attacks, stroke, and peripheral vascular dis- tered daidzein was stronger than that of intraperitoneally and ease.26—28) Therefore the inhibition of platelet function repre- orally administered puerarin and daidzin. sents a promising approach for the prevention of thrombosis. To understand the antiallergic mechanism of these com- Puerarin, daidzin, and daidzein exhibited inhibitory activity pounds, we measured their hyaluronidase activation in- against platelet aggregation in vitro and ex vivo and protected hibitory and antioxidant activities (Table 6). No tested com- against thromboembolism in mice. Daidzein, which is a pounds inhibited hyaluronidase activation. Puerarin and metabolite of puerarin and daidzin produced by human in- daidzin did not show antioxidant activity against DPPH and testinal microflora before absorption in the intestine, was the free radical-scavenging activities. However, daidzein showed most potent. The rhizome of P. lobata also exhibited in vitro potent antioxidant activity against DPPH and free radical- and ex vivo antiplatelet aggregation activity even if it was ad- scavenging activities. ministered orally at high dose. The rhizome of P. lobata potently protected against thromboembolism. Based on these DISCUSSION findings, the rhizome of P. lobata, particularly daidzein, if puerarin and daidzein are metabolized to daidzein by human The rhizome of P. lobata, an ingredient of many polypre- intestinal bacteria, could prevent the development of recur- scriptions, has been used to treat liver diseases, stroke, and rence and thrombosis. allergy. Its main components are puerarin and daidzin. These Second, DSCG is an antiallergic drug.29) This drug inhib- glycosides are not easily absorbed from intestine to blood ited the release of b-hexosaminidase from RBL-2H3 cells by due to their hydrophilic properties. Therefore these compo- DNP-BSA, which is an essential step in the pathological nents are metabolized to daidzein by human intestinal mi- process of type I allergy.20) These facts suggest that, if sub- croflora before absorption.15) When puerarin and daidzin stances that strongly inhibit b-hexosaminidase release are were administered orally to rats, daidzein in the urine was the discovered, they could be lead compounds for developing main metabolite.10,11) We also identified daidzein in urine, new antiallergic drugs. In the present study, we investigated when puerarin and daidzin were administered orally (data not the inhibitory activity of puerarin, daidzin, and daidzein on shown). When puerarin and daidzin were incubated with the b-hexosaminidase release from RBL-2H3 cells. These com- liver homogenate of mice, puerarin was not transformed to pounds inhibited b-hexosaminidase release, with daidzein daidzein and daidzin was very weakly transformed to being the most potent. Daidzein showed potent antioxidant daidzein (Table 7). However, human fecal suspension po- activity against free radical-scavenging activities. These re- tently transformed puerarin and daidzin to daidzein. These sults support the reports that antioxidants are useful for treat- results suggest that puerarin and daidzin could be trans- ing allergic diseases.30) Daidzein also had the most potent in- formed to daidzein before absorption into the blood. There- hibitory activity against the PCA reaction. These results sug- fore we investigated the antithrombotic and antiallergic activ- gest that, when puerarin and daidzin are administered orally, ities of puerarin, daidzin, and their metabolite daidzein to it is easily metabolized to daidzein in intestine, and daidzein evaluate the role of the intestinal bacteria on the pharmaco- is absorbed into blood and is responsible for the inhibitory logical actions. activity against PCA reaction. First, platelets play an important role in the pathogenesis In this study, daidzein, the metabolite of puerarin and of thrombosis. The interactions between platelets and blood daidzin produced by human intestinal bacteria, demonstrated vessel walls are important in the development of thrombosis the potent inhibitory activity against thrombosis and allergy. and cardiovascular disease such as myocardial infarction, Finally, we believe that puerarin and daidzin in the rhizome stroke, and arteriosclerosis.23—25) Once blood vessels are of P. lobata are prodrugs that have extensive antiallergic and damaged, platelet aggregation occurs rapidly to form hemo- antithrombotic properties. static plugs or arterial thrombi at the sites of vessel injury or in regions where blood flow is disturbed. These thrombi are Acknowledgment This work was supported by a grant the source of thromboembolic complications of arterioscle- from the Good Health R&D Project (HMP-99-O-01-0002), 1332 Vol. 25, No. 10

Ministry of Health and Welfare, Korea (1999). 14) Kakegawa H., Matsumoto H., Satoh T., Planta Med., 54, 385—389 (1988). 15) Kim D.-H., Yu K.-U., Bae E.-A., Han M. J., Biol. Pharm. Bull., 21, REFERENCES 628—630 (1998). 16) Aronson N. N., Davidson E. A., J. Biol. Chem., 242, 441—444 (1967). 1) Hara M., Mod. Orient. Med., 3, 42—53 (1982). 17) Born G. V. R., Cross M. J., J. Physiol., 168, 178—195 (1963). 2) Shibata S., Murakami T., Nishikawa Y., Yakugaku Zasshi, 79, 757— 18) Kimura Y., Tani T., Kanbe T., Watanabe K., Arzneim. Forsch. (Drug 761 (1959). Res.), 35, 1144—1149 (1985). 3) Murakami T., Nishikawa Y., Ando T., Chem. Pharm. Bull., 8, 688— 19) Di Minno G., Silver M., J. Pharmacol. Exp. Ther., 225, 57—60 691 (1960). (1983). 4) Keung W. M., Valee B. L., Proc. Natl. Acad. Sci. U.S.A., 90, 1247— 20) Choi O. H., Kim J. H., Kinet J. P., Nature (London), 380, 634—636 1251 (1993). (1996). 5) Shen Z.-F., Xie M.-Z., Acta Pharm. Sin., 20, 863—867 (1985). 21) Katayama S., Shionoya H., Ohtake S., Microbiol. Immunol., 22, 89— 6) Fan L. L., O’Keefe D. D., Powell W. J., Acta Pharm. Sin., 19, 801— 101 (1978). 804 (1984). 22) Xiong Q., Kadota S., Tani T., Namba T., Biol. Pharm. Bull., 19, 7) Nikaido T., Ohmoto T., Sankawa U., Hamanaka T., Totsuka K., Planta 1580—1585 (1996). Med., 46, 162—166 (1982). 23) Mustard J. F., Drugs, 9, 19—76 (1975). 8) Nakamoto H., Iwasaki Y., Kiku H., Yakugaku Zasshi, 97, 103—105 24) Mustard J. F., Packham M. A., Kinlough-Rathbone R. L., Circulation, (1977). 81, 124—127 (1990). 9) Kobashi K., Akao T., Biosci. Microﬂora, 16, 1—7 (1997). 25) Dinerman J. L., Mehta J. L., J. Am. Coll. Cardiol., 16, 207—222 10) Yasuda T., Kano Y., Saito K., Ohsawa K., Biol. Pharm. Bull., 17, (1990). 1369—1374 (1994). 26) Packham M. A., Can. J. Physiol. Pharmacol., 72, 2782—2784 (1994). 11) Yasuda T., Kano Y., Saito K., Ohsawa K., Biol. Pharm. Bull., 18, 27) Stein B., Fuster V., Cardiovasc. Drugs Ther., 3, 797—813 (1989). 300—303 (1995). 28) MacMahon S., Sharpe N., Med. J. Aust., 154, 477—480 (1991). 12) Kakegawa H., Matsumoto H., Satoh T., Chem. Pharm. Bull., 33, 642— 29) Tasaka K., Mio M., Okamoto M., Ann. Allergy, 56, 464—469 (1986). 646 (1985). 30) Ito H., Miyazawa T., Ono M., Sakurai H., Bioorg. Med. Chem., 6, 13) Kakegawa H., Matsumoto H., Satoh T., Chem. Pharm. Bull., 40, 1051—1056 (1998). 1439—1442 (1992).