- Email: [email protected]
Inhibition of platelet activation and endothelial cell injury by flavan-3-ol and saikosaponin compounds
Prostaglandins Leukolrienes and Essential Fatty Acids (1991)‘44.51-56 0 Longman Group UK Ltd 1991
Inhibition of Platelet Activation and Endothelial Cell Injury by Flavan-3-01 and Saikosaponin Compounds W.-C. Chang and F.-L. HSU* Department of Pharmacology, College of Medicine, National Cheng Kung University, Tainan, Taiwan and *School of Pharmacy, Taipei Medical College, Taipei, Republic of China (Reprint requests to WCC) ABSTRACT.
The effects of flavan34 and saikosaponin compounds on platelet aggregation, platelet thromboxane biosynthesis and H&induced endothelial cell injury were studied. Seven flavan3-ol compounds isolated from Camellia sinensis L. var sinensis 0. Kuntze (Theaceae) and three saikosaponin compounds isolated from Bupleurum falcatum L. (Umbelliferae) were used. Among the 10 compounds tested, only epigallocatechin and saikosaponin a significantly inhibited human platelet aggregation induced by ADP, and the potency of inhibition was comparable with aspirin. Both of epigaBocatechin and saikosaponin a dosedependently inhibited the platelet thromboxaue formation from exogenous and endogenous arachidonic acid. In the prevention of H&induced endothelial cell injury in culture, only gdlocatechin-3-O-gal and epicatechin3-O-galtate were effective. The inhibitory effect of epigaltocatechin and saikosapouin a on piatelet activation and the cytoprotective effect of gallocatechin-3-O-gallate and epicatechin-3-O-gallate on HzO+duced endothelial cell injury could give evidence of explaining the possible role of flavan3-01 and saikosaponin compounds in maintaining vascular homeostasis.
from Bupleurum falcatum L. on platelet aggregation, platelet thromboxane biosynthesis and H202induced endothelial cell injury were investigated.
INTRODUCTION The vascular endothelium is extremely sensitive to oxidative damage mediated by reactive oxygen metabolites released from inflammatory cells (1, 2). Of these metabolites hydrogen peroxide (H20Z) appears to be an important mediator of acute cellular injury in a variety of settings (2, 3). Such oxidative damage may play a role in the pathogenesis of atherosclerosis (4). Following the injury or dysfunction of endothelium in the pathogenesis of atherosclerosis, platelets are activated and aggregated. Among the products biosynthesized by platelets that show intimate relevance to vascular homeostasis are metabolites of arachidonic acid. Arachidonic acid is converted to thromboxane A2 (TXA2) which is a very potent platelet aggregation inducer and vasoconstrictor (5). In the present study, the effects of several flavan3-01 compounds isolated from Camellia sinensis L. var sinensis 0. Kuntze and saikosaponins isolated
MATERIAL AND METHODS Chemicals Bovine y-globulin, arachidonic acid and calcium ionophore A23187 were from Sigma Chemical Co, St. Louis, MO. TXB2 standard was kindly provided by Ono Pharmaceutical Co Ltd, Osaka, Japan. Hydrogen peroxide was obtained from E. Merck, FRG. Darmstadt, [“Cr]Sodium chromate (386 mCi/mg) and [5, 6, 8, 9, 11, 12, 14, 15(n)3H]thromboxane Bz (180 Ci/mmol) were purchased from NEN, Du Pont, USA. Dulbecco’s modified Eagle medium (DMEM) and fetal bovine serum (FBS) were purchased from GIBCO, NY, USA. All other reagents used were of the highest purity available. Tested oriental medicinal compounds All of the compounds tested in the present study were isolated in our laboratories. Seven flavan-3-01
Date received 22 January 1991 Date accepted 25 April 1991 51
52
Prostaglandins Leukotrienes
and Essential Fatty Acids
T$zC~=
R2
Wlodechin-3-O-galhte
-G
OH
Epic&chin-3-O-gallate
---G
H
EplgAXatechin-3-O-gallatc
--G
OH
GalOCMifl
-on
ctl
Eptgalloutechin
--. O+j
OH
Epkatechh
___ O).j
H
catechln
-OH
H
Ri*vn
Fig. 2.
Structures of saikosaponin compounds
-OH G: -0oc \ 0 Fig. 1.
on
,
OH
Structures of flavan-3-01 compounds
compounds; gallocatechin-3-O-gallate, epicatechin3-0-gallate, epigallocatechin-3-O-gallate, gallocatechin, epigallocatechin, epicatechin and catechin were isolated from Camellia sinensis L. var sinensis 0. Kuntze (Theaceae). The structures were identified by NMR spectra and compared with those reported by Hashimoto et al (6). Saikosaponin a, b2 and c were isolated Bupleurum falcatum L. (Umbelliferae). The structures were also identified by NMR spectra and compared with those reported by Tani et al (7). The structures of flavan-3-01s and saikosaponins are indicated in Figures 1 and 2, respectively. Platelet aggregation assay Blood from human volunteers was withdrawn into a 3.8% solution of sodium citrate (9:1, v/v). Platelet-rich plasma was prepared by centrifugation of titrated blood at 200 x g for 10 min at room temperature. A sample of 0.25 ml of platelet-rich plasma was aggregated at 37°C in a NKK Hema Tracer with 25 ~1 of ADP. The final concentration of ADP was 1 x lo-‘M. Transformation of exogenous arachidonic acid to thromboxane in platelets Platelet-rich plasma was centrifuged at 1800 x g for 30 min in order to obtain the platelet pellet. Platelets were resuspended in Ca*+-and Mg*+-free phosphate-buffered saline (pH 7.5) containing 5.5 mM glucose. Platelet number was determined with a Coulter Counter. For studying the conversion of exogenous arachidonic acid by platelets, each assay tube contained 1 x lo* platelets in 1 ml of
suspension and 5 lug of arachidonic acid. The incubation was performed at 37°C for 6 min. After the incubation, platelets were spun down by an airfuge centrifuge. TXB2 in the supernatant was measured by a specific radioimmunoassay. Transformation of endogenous arachidonic acid to thromboxane in platelets Human platelets (1 X lo8 cells) suspended in 1 ml of Ca’+-and Mg2+-free phosphate-buffer saline (pH 7.4) containing 5.5 mM glucose and 2 mM CaC12 were challenged with 3 PM calcium ionophore A23187 at 37°C for 5 min. Platelets were spun down by an airfuge centrifuge. TXB;! in the supernatant was measured by a specific radioimmunoassay. Radioimmunoassay of thromboxane B2 A specific radioimmunoassay for thromboxane B2 was performed according to the methods described previously (8). Incubation and subsequently separation of the free from the bound form of labeled antigen were carried out at room temperature. Antiserum, labeled antigen, and standards were diluted in the standard radioimmunoassay buffer, 0.05 M Tris-HCl, pH 7.5, containing 0.1% gelatin. The incubation mixture (0.4 ml) contained: 0.2 ml of TXB2 standards or samples, 0.1 ml of antiserum (final dilution, 1:3000), and 0.1 ml of [3H]TXB2 (approximately 10 000 cpm). The incubation was carried out for 1 h. All samples were run in duplicate in 10 X 75 mm glass tubes. Separation of bound from free antigen was achieved by y-globulin plus dextran-coated charcoal. The method of preparing y-globulin and dextran-coated charcoal was as follows. Bovine y-globulin (0.33 g) was first dissolved in 100 ml of 0.9% dextran in standard assay buffer. Charcoal (3.3 g) was then added, and the solution was stirred for 1 h. Immediately before the addition of charcoal suspension, 1 ml of Hz0
Effect of Flavan3-01 and Saikosaponin on Platelets and Endothelial Cells
was added to each incubation tube for easier pipetting. Each incubation tube was capped with a Luckman LP/3S stopper, and 0.2 ml of the charcoal suspension was withheld in the well. Tubes were inverted five times and allowed to stand for 5 min before centrifugation at 100 X g for 10 min. The supernatant containing the antibody-bound antigen was counted in a LKB Rackbeta liquid scintillation counter. Endothelial cell culture Endothelial cells were isolated from bovine carotid arteries as previously reported (9) and maintained in DMEM containing 10% FBS. Cells of passage 15-25 were used in the present series of experiments.
53
Table 1 Inhibition of human platelet aggregation by flavan-3-01 and saikosaponin compounds Inhibition (%)
Compounds
Epigallocatechin Gallocatechin Gallocatechin-3-0-gallate Epicatechin-3-0-gallate Epicatechin Epicatechin-3-0-gallate Epigallocatechin-3-0-gallate Saikosaponin a Saikosaponin bz Aspirin Saikosaponin c
W6 M
lo-’ M
1O-4 M
18 9 5 9 0 9 9 14
31 8 0 0 0 0 0 47
6
0
58 7 5 19 0 3 3 73 2
09
20 0
3:
Human platelet-rich plasma was incubated with the tested compounds at 37°C for 30 min. The platelet aggregation was then induced by lo-’ M ADT. Each value is the mean of duplicate assays.
Endothelial cell injury assay Endothelial cell injury was estimated by the release of “Chromium as previously reported (10). Confluent monolayer cells in 24 multiwell dishes (Falcone Labware) were labeled with 2 &i of [“CrJsodium chromate for 18 h in the growth medium. Cells were then washed twice with DMEM and treated with 2 mM Hz02 in the presence or absence of tested compounds in 0.5 ml of DMEM for 6 h. Aliquot of the cultured DMEM was removed, and the radioactivity due to 51Cr released out of the injured cells was measured by a LKB 1282 Compugamma scintillation spectrophotometer. Results were expressed as percentage of specific 5*Crrelease calculated as follows: (A-B)/(C-D) x NO%, where A represents 51Cr-release due to the test compound; B represents the spontaneous 51Cr release and C represents the maximal release of 5*Cr. Spontaneous and maximum releases were determined in cells incubated with vehicle solvent and 0.1% Triton X-100, respectively.
Results Inhibition of platelet aggregation by flavan3-01 and saikosaponin compounds The effect of 10 compounds on platelet aggregation was studied by using platelet-rich plasma. For comparison, the effect of aspirin was simultaneously tested. Platelet-rich plasma was incubated with the tested compound at 37°C for 30 min. The platelet aggregation was then induced by 1 x 10e5 M ADP. Among the 10 compounds tested, only epigallocatechin and saikosaponin a significantly inhibited the platelet aggregation (Table 1). Their inhibitory effect was as comparable as that of aspirin. The tracers of platelet aggregation affected by epigallocatechin and saikosaponin a are shown in Figures
Control
l$M
f
ld”m I”
/
i
H
i min Fig. 3. Effect of epigallocatechin on platelet aggregation Human platelet-rich plasma was pretreated with different concentrations of epigallocatechin at 37°C for 30 min. The platelet aggregation was then induced by 10-j M ADP.
3 and 4. Both of epigallocatechin and saikosaponin a dose-dependently inhibited the second phase of platelet aggregation induced by ADP in human platelet-rich plasma. Effect of epigallocatechin and saikosaponin a on thromboxane formationin in platelets Since only epigallocatechin and saikosaponin a significantly inhibited the platelet aggregation induced by ADP, the effect of both compounds on thromboxane formation in platelets was then studied. Tbromboxane formations from both exogenous and
54
Prostnplandins Lcukotrienes and Essential Fatty Acids
cndogenous arachidonate were analyzed. In order to study the thromboxane formation from the exogenous arachidonic acid, platelets were treated with 5 pg of arachidonic acid after the preincubation with tested compounds. As indicated in Table 2, both of epigallocatechin and saikosaponin a at the concentrations of 10e5 and 10m4M significantly inhibited the thromboxane formation in platelets. For studying the thromboxane formation from endogenous arachidonic acid, platelets were treated with calcium ionosphere after the preincubation with tested compounds. Both of epigallocatechin and saikosaponin a dose-dependently inhibited the thromboxane formation (Table 3). In both assays, saikosaponin a showed more potent effect than epigallocatechin.
Control
”
Preventive effect on endothelial cell injury induced by H2@ Fig. 4. Effect of saikosaponin a on platelet aggregation Human platelet-rich plasma was pretreated with different concentrations of saikosaponin a at 37°C for 30 min. The platelet aggregation was then induced by 10e5 M ADP.
For studying the preventive effect of flavan-3-01 and saikosaponin compounds on endothelial cell injury induced by H202, the tested compounds were present in culture during H202 treatment. When the cells were incubated with 2 mM Hz02 for 6 h, approximate 45% cell injury was observed. Among the 10 compounds tested, only two flavan-3-01 compounds; gallocatechin-3-O-gallate and epicatechin3-0-gallate significantly showed the preventive effect on endothelial cell injury (Table 4). The preventive effect was dose-dependent.
Table 2 Effect of epigallocatechin
and saikosaponin a on thromboxane formation from exogenous arachidonic acid in intact platelets. Compounds
TXB, formation (ne/l x 10s cells) Control
1O-5 M
1O-4 M
Epigallocatechin
60 + 3.7 (100%)
52 f 2.8* (87%)
22 If: 0.9** (37%)
Saikosaponin a
51 f 2.5 (100%)
17 * 3.1** (33%)
11 + 0.5** (22%)
Platelets were treated with epigallocatechin or saikosaponin a at 37°C for 30 min. Formation for TXB, from exogenous arachidonic acid in intact platelets was measured as described in Methods. Values of product formation represent mean + S.E.M. from quadraplicate assays. Values in parentheses are the percentages compared with their controls. *P < 0.01, **p < 0.001.
DISCUSSION Two kinds of compounds of flavan-3-01s and saikosaponins were used in this series of studies. The flavan-3-01 compounds occur widely in tea plants, and several catechin compounds have been isolated (6). Catechin (cianidanol) has immunomodulatory actions, and has been tried in the
Table 3 Effect of epigallocatechin
formation from endogenous Compounds
and saikosaponin a on thromboxane arachidonic acid in intact platelets.
TXB, formation (rig/l x 10” cells) Control
1O-6 M
1O-5 M
1O-4 M
8.5 + 0.40* (79%)
6.0 +0.21** (56%)
Epigallocatechin
10.7 + 0.80 11.2 f 0.60 (100%) (105%)
Saikosaponin a
10.7 f 0.80 8.5 f 0.70* 2.2 + 0.24** 2.7 f 0.07* (21%) (25%) (100%) (79%)
Platelets were treated with epigallocatechin or saikosaponin a at 37°C for 30 min. Formation of TXBr from endogenous arachidonate triggered by 3 PM calcium ionophore A23187 was performed as described in Methods. Values of product formation represent mean + S.E.M. from quadraplicate assays. Values in parentheses are the percentages compared with their controls. “P C 0.05, **P < 0.001.
Effect of Flavan-3-01 and Saikosaponin on Platelets and Endothelial Cells Table 4 Inhibitory effect of flavan-3-01s and saikosaponins on H,Oiinduced endothelial cell injury Inhibition (%)
Compounds
Gallocatechin-3-0-gallate Epicatechin-3-0-gallate Epigallocatechin-3-0-galfate Gallocatechin Epigallocatechin Epicatechin Catechin Saikosaponin a Saikosaponin b2 Saikosaponin c
1O-6 M
lo-’ M
1O-4 M
9 9 0 0 0 0 0 0 0 6
27 30 0 0 0 0 1 0 0 9
86 53 0 0 0 0 15 0 0 18
After labelling the monolayer endothelial cells with “chromate, cells were treated with 2 mM H,O, in the presence or absence of flavan-3-01s and saikosaponins in 0.5 ml of DMEM for 6 h according to the Methods. Cell injury induced by HP2 was used as control. The percentage of inhibition was obtained by comparing the cytoprotective effect of tested compounds with the control injury. Each value is the mean of two sets of experiments.
treatment of HBeAg-positive chronic hepatitis (11). Saikosaponins are the major saponins in the dry foot of certain species of the Bupleurum plants (Umbelliferae) (7). Saikosaponins have hepatic cytoprotective effects of stimulating membrane stability, hepatic protein synthesis and hepatic glycogen formation (for review see 11). In the present study, the pharmacological effects of seven flavan-3-01 compounds and three saikosaponin compounds on platelet functions and endothelial cell injury were investigated. Among the 10 compounds tested only two compounds, epigallocatechin and saikosaponin a, were effective in inhibiting human platelet aggregation induced by ADP. Two phase aggregation is induced by ADP in human platelet-rich plasma. The second wave aggregation is recognized to be mediated by thromboxane formation (12). As indicated in Figures 3 and 4, both of epigallocatechin and saikosaponin a inhibited the second wave aggregation of ADP in human platelet-rich plasma. The results suggest that both of epigallocatechin and saikosaponin a might interfere the thromboxane formation in platelets. The thromboxane formation from exogenous and endogenous arachidonic acid was then studied. As indicated in Tables 2 and 3, both of epigallocatechin and saikosaponin a dosedependently inhibited the thromboxane formation from both exogenous and endogenous arachidonic acid. Therefore, the inhibition of thromboxane formation could explain the inhibitory effect of both epigallocatechin and saikosaponin a on the second wave aggregation induced by ADP in human platelet-rich plasma. It has been documented previously that active oxygen specifies including H202 can cause cytotoxic
55
injury to endothelial cells (13). Hz02 can cross membrane almost as readily as can water, the unsaturated fatty acids present (phospholipids, glycolipids, and sterols) and the transmembrane proteins containing oxidizable amino acids are susceptible to H202 damage (for review see 14). Since oxidative damage may play a role in the pathogenesis of atherosclerosis, H20zinduced endothelial cells in culture could be used for studying the intervention of oxidative damage in vascular endothelial cells. Among the compounds tested in the present study, only two flavan-3-01 compounds; gallocatechin-3-O-gallate and epicatechin-3-O-gallate significantly prevented the endothelial cell injury induced by H202. The pharmalogical mechanism of the cytoprotective effect is not studied yet in the present report. It might be due to the antioxidative effect of flavan-3-01 compounds since several catechin compounds have been-reported to have antioxidative effect determined by the active oxygen method at 97XC on lard (15). In conclusion, the inhibitory effect of epigallocatechin and saikosaponin a on platelet activation and the cytoprotective effect of gallocatechin-3-0gallate and epicatechin-3-O-gallate on HZOZ-induced endothelial cell injury could give evidence of explaining the possible role of flavan-3-01 and saikosaponin compounds in maintaining the vascular homeostasis. Acknowledgments We are greatly indebted to Ms. Chi-Yen Lin and Ms. I-Ping Li for their excellent technical assistance. Thanks are due to Ono Pharmaceutical Co for the supply of thromboxane B,. This research was supported in part by the National Science Council of the Republic of China (NSC 79-0420-BOO6-18).
References 1. Sacks T, Moldow C F, Craddock P R, Bowers T K, Jacob H S. Oxygen radicals mediate endothelial cell damage by complement-stimulated granulocytes. J. Clin. Invest. 61, 1161, 1978. 2. Weiss S J, Young J, LoBuglio A F, Slivka A, Nimeh N F. Role of hydrogen peroxide in neutrophil-mediated destruction of cultured endothelial cells. J. Clin. Invest. 68, 714, 1981. 3. Babior B. The respiratory burst of phagocytes. J. Clin. Invest. 73. 599, 1984. 4. Mazzone T, Jensen M, Chait A. Human arterial wall cells secrete factors that are chemotactic for monocytes. Proc. Natl. Acad. Sci. USA 80. 5094. 1983. 5 Hamberg M, Svensson J, Samuelsson B. Thromboxanes; a new group of biologically active compounds derived from prostaglandin endoperoxides. Proc. Natl. Acad. Sci. USA 72. 2994, 1975. 6. Hashimoto F, Nonaka G, Nishioka I. Tannin and related compounds. LVI. Isolation of four new acylated flavan-3-01s from oolong tea. Chem. Pharm. Bull. 35, 611, 1987. 7. Tani T, Katsuki T, Kubo M, Arichi S. Distribution of saikosaponins in Bupleurum falcatum root. J. Chromat. 360, 407, 1986.
56
Prostaglandins Leukotrienes
and Essential Fatty Acids
8. Chang W C, Hsu F L. Inhibition of platelet aggregation and arachidonate metabolism in platelets by procyanidins. Prost. Leuk. Essen. Fatty Acids 38, 181, 1989. 9. Sasaguri Y, Morimatsu M, Kinoshita T, Nakashima T, Inagaki T, Yagi K. Difference in susceptibility to injury by linoleic acid hydroperoxide between endothelial and smooth muscle cells of arteries. J. Appl. B&hem. 7, 70, 1985. 10. Abe M, Morita I, Murota S. A new in vitro method using Fura- for the quantification of endothelial cell injury. Prost. Leuk. Essen. Fatty Acids 34, 69, 1988. 11. Oda T, Oka T. Kampo preparations as biological response modifiers in the treatment of chronic viral hepatitis. In: Recent Advances in the Pharmacology of Kampo, pp. 369-377, 1988. (Ed: Hosoya E.
12.
13.
14.
15.
Yamamura Y.) Elsevier Science Publishers LTD. Amsterdam. Ushikubi F. Okuma M, Kanaii V, Sugiyama T, Ogorochi T, Narumiya S. U&no H. Hemorrhagic thrombocytopathy with platelet thromboxane A, receptor abnormality: Defective signal transduction with normal binding activity. Thromb. Haemost. 57, 158, 1987. Henson P M, Johnson R B. Tissue injury in inflammation. Oxidants, proteinases and cationic proteins. J. Clin. Invest. 79, 669, 1987. Weiss S J, LoBuglio A F. Phagocyte-generated oxygen metabolites and cellular injury. Lab. Invest. 47, 5, 1982. Matsuzaki T, Hara Y. Antioxidative activity of tea leaf catechins. Nippon Nogeikagaku Kaishi 59, 129, 1985.
Inhibition of Platelet Activation and Endothelial Cell Injury by Flavan-3-01 and Saikosaponin Compounds W.-C. Chang and F.-L. HSU* Department of Pharmacology, College of Medicine, National Cheng Kung University, Tainan, Taiwan and *School of Pharmacy, Taipei Medical College, Taipei, Republic of China (Reprint requests to WCC) ABSTRACT.
The effects of flavan34 and saikosaponin compounds on platelet aggregation, platelet thromboxane biosynthesis and H&induced endothelial cell injury were studied. Seven flavan3-ol compounds isolated from Camellia sinensis L. var sinensis 0. Kuntze (Theaceae) and three saikosaponin compounds isolated from Bupleurum falcatum L. (Umbelliferae) were used. Among the 10 compounds tested, only epigallocatechin and saikosaponin a significantly inhibited human platelet aggregation induced by ADP, and the potency of inhibition was comparable with aspirin. Both of epigaBocatechin and saikosaponin a dosedependently inhibited the platelet thromboxaue formation from exogenous and endogenous arachidonic acid. In the prevention of H&induced endothelial cell injury in culture, only gdlocatechin-3-O-gal and epicatechin3-O-galtate were effective. The inhibitory effect of epigaltocatechin and saikosapouin a on piatelet activation and the cytoprotective effect of gallocatechin-3-O-gallate and epicatechin-3-O-gallate on HzO+duced endothelial cell injury could give evidence of explaining the possible role of flavan3-01 and saikosaponin compounds in maintaining vascular homeostasis.
from Bupleurum falcatum L. on platelet aggregation, platelet thromboxane biosynthesis and H202induced endothelial cell injury were investigated.
INTRODUCTION The vascular endothelium is extremely sensitive to oxidative damage mediated by reactive oxygen metabolites released from inflammatory cells (1, 2). Of these metabolites hydrogen peroxide (H20Z) appears to be an important mediator of acute cellular injury in a variety of settings (2, 3). Such oxidative damage may play a role in the pathogenesis of atherosclerosis (4). Following the injury or dysfunction of endothelium in the pathogenesis of atherosclerosis, platelets are activated and aggregated. Among the products biosynthesized by platelets that show intimate relevance to vascular homeostasis are metabolites of arachidonic acid. Arachidonic acid is converted to thromboxane A2 (TXA2) which is a very potent platelet aggregation inducer and vasoconstrictor (5). In the present study, the effects of several flavan3-01 compounds isolated from Camellia sinensis L. var sinensis 0. Kuntze and saikosaponins isolated
MATERIAL AND METHODS Chemicals Bovine y-globulin, arachidonic acid and calcium ionophore A23187 were from Sigma Chemical Co, St. Louis, MO. TXB2 standard was kindly provided by Ono Pharmaceutical Co Ltd, Osaka, Japan. Hydrogen peroxide was obtained from E. Merck, FRG. Darmstadt, [“Cr]Sodium chromate (386 mCi/mg) and [5, 6, 8, 9, 11, 12, 14, 15(n)3H]thromboxane Bz (180 Ci/mmol) were purchased from NEN, Du Pont, USA. Dulbecco’s modified Eagle medium (DMEM) and fetal bovine serum (FBS) were purchased from GIBCO, NY, USA. All other reagents used were of the highest purity available. Tested oriental medicinal compounds All of the compounds tested in the present study were isolated in our laboratories. Seven flavan-3-01
Date received 22 January 1991 Date accepted 25 April 1991 51
52
Prostaglandins Leukotrienes
and Essential Fatty Acids
T$zC~=
R2
Wlodechin-3-O-galhte
-G
OH
Epic&chin-3-O-gallate
---G
H
EplgAXatechin-3-O-gallatc
--G
OH
GalOCMifl
-on
ctl
Eptgalloutechin
--. O+j
OH
Epkatechh
___ O).j
H
catechln
-OH
H
Ri*vn
Fig. 2.
Structures of saikosaponin compounds
-OH G: -0oc \ 0 Fig. 1.
on
,
OH
Structures of flavan-3-01 compounds
compounds; gallocatechin-3-O-gallate, epicatechin3-0-gallate, epigallocatechin-3-O-gallate, gallocatechin, epigallocatechin, epicatechin and catechin were isolated from Camellia sinensis L. var sinensis 0. Kuntze (Theaceae). The structures were identified by NMR spectra and compared with those reported by Hashimoto et al (6). Saikosaponin a, b2 and c were isolated Bupleurum falcatum L. (Umbelliferae). The structures were also identified by NMR spectra and compared with those reported by Tani et al (7). The structures of flavan-3-01s and saikosaponins are indicated in Figures 1 and 2, respectively. Platelet aggregation assay Blood from human volunteers was withdrawn into a 3.8% solution of sodium citrate (9:1, v/v). Platelet-rich plasma was prepared by centrifugation of titrated blood at 200 x g for 10 min at room temperature. A sample of 0.25 ml of platelet-rich plasma was aggregated at 37°C in a NKK Hema Tracer with 25 ~1 of ADP. The final concentration of ADP was 1 x lo-‘M. Transformation of exogenous arachidonic acid to thromboxane in platelets Platelet-rich plasma was centrifuged at 1800 x g for 30 min in order to obtain the platelet pellet. Platelets were resuspended in Ca*+-and Mg*+-free phosphate-buffered saline (pH 7.5) containing 5.5 mM glucose. Platelet number was determined with a Coulter Counter. For studying the conversion of exogenous arachidonic acid by platelets, each assay tube contained 1 x lo* platelets in 1 ml of
suspension and 5 lug of arachidonic acid. The incubation was performed at 37°C for 6 min. After the incubation, platelets were spun down by an airfuge centrifuge. TXB2 in the supernatant was measured by a specific radioimmunoassay. Transformation of endogenous arachidonic acid to thromboxane in platelets Human platelets (1 X lo8 cells) suspended in 1 ml of Ca’+-and Mg2+-free phosphate-buffer saline (pH 7.4) containing 5.5 mM glucose and 2 mM CaC12 were challenged with 3 PM calcium ionophore A23187 at 37°C for 5 min. Platelets were spun down by an airfuge centrifuge. TXB;! in the supernatant was measured by a specific radioimmunoassay. Radioimmunoassay of thromboxane B2 A specific radioimmunoassay for thromboxane B2 was performed according to the methods described previously (8). Incubation and subsequently separation of the free from the bound form of labeled antigen were carried out at room temperature. Antiserum, labeled antigen, and standards were diluted in the standard radioimmunoassay buffer, 0.05 M Tris-HCl, pH 7.5, containing 0.1% gelatin. The incubation mixture (0.4 ml) contained: 0.2 ml of TXB2 standards or samples, 0.1 ml of antiserum (final dilution, 1:3000), and 0.1 ml of [3H]TXB2 (approximately 10 000 cpm). The incubation was carried out for 1 h. All samples were run in duplicate in 10 X 75 mm glass tubes. Separation of bound from free antigen was achieved by y-globulin plus dextran-coated charcoal. The method of preparing y-globulin and dextran-coated charcoal was as follows. Bovine y-globulin (0.33 g) was first dissolved in 100 ml of 0.9% dextran in standard assay buffer. Charcoal (3.3 g) was then added, and the solution was stirred for 1 h. Immediately before the addition of charcoal suspension, 1 ml of Hz0
Effect of Flavan3-01 and Saikosaponin on Platelets and Endothelial Cells
was added to each incubation tube for easier pipetting. Each incubation tube was capped with a Luckman LP/3S stopper, and 0.2 ml of the charcoal suspension was withheld in the well. Tubes were inverted five times and allowed to stand for 5 min before centrifugation at 100 X g for 10 min. The supernatant containing the antibody-bound antigen was counted in a LKB Rackbeta liquid scintillation counter. Endothelial cell culture Endothelial cells were isolated from bovine carotid arteries as previously reported (9) and maintained in DMEM containing 10% FBS. Cells of passage 15-25 were used in the present series of experiments.
53
Table 1 Inhibition of human platelet aggregation by flavan-3-01 and saikosaponin compounds Inhibition (%)
Compounds
Epigallocatechin Gallocatechin Gallocatechin-3-0-gallate Epicatechin-3-0-gallate Epicatechin Epicatechin-3-0-gallate Epigallocatechin-3-0-gallate Saikosaponin a Saikosaponin bz Aspirin Saikosaponin c
W6 M
lo-’ M
1O-4 M
18 9 5 9 0 9 9 14
31 8 0 0 0 0 0 47
6
0
58 7 5 19 0 3 3 73 2
09
20 0
3:
Human platelet-rich plasma was incubated with the tested compounds at 37°C for 30 min. The platelet aggregation was then induced by lo-’ M ADT. Each value is the mean of duplicate assays.
Endothelial cell injury assay Endothelial cell injury was estimated by the release of “Chromium as previously reported (10). Confluent monolayer cells in 24 multiwell dishes (Falcone Labware) were labeled with 2 &i of [“CrJsodium chromate for 18 h in the growth medium. Cells were then washed twice with DMEM and treated with 2 mM Hz02 in the presence or absence of tested compounds in 0.5 ml of DMEM for 6 h. Aliquot of the cultured DMEM was removed, and the radioactivity due to 51Cr released out of the injured cells was measured by a LKB 1282 Compugamma scintillation spectrophotometer. Results were expressed as percentage of specific 5*Crrelease calculated as follows: (A-B)/(C-D) x NO%, where A represents 51Cr-release due to the test compound; B represents the spontaneous 51Cr release and C represents the maximal release of 5*Cr. Spontaneous and maximum releases were determined in cells incubated with vehicle solvent and 0.1% Triton X-100, respectively.
Results Inhibition of platelet aggregation by flavan3-01 and saikosaponin compounds The effect of 10 compounds on platelet aggregation was studied by using platelet-rich plasma. For comparison, the effect of aspirin was simultaneously tested. Platelet-rich plasma was incubated with the tested compound at 37°C for 30 min. The platelet aggregation was then induced by 1 x 10e5 M ADP. Among the 10 compounds tested, only epigallocatechin and saikosaponin a significantly inhibited the platelet aggregation (Table 1). Their inhibitory effect was as comparable as that of aspirin. The tracers of platelet aggregation affected by epigallocatechin and saikosaponin a are shown in Figures
Control
l$M
f
ld”m I”
/
i
H
i min Fig. 3. Effect of epigallocatechin on platelet aggregation Human platelet-rich plasma was pretreated with different concentrations of epigallocatechin at 37°C for 30 min. The platelet aggregation was then induced by 10-j M ADP.
3 and 4. Both of epigallocatechin and saikosaponin a dose-dependently inhibited the second phase of platelet aggregation induced by ADP in human platelet-rich plasma. Effect of epigallocatechin and saikosaponin a on thromboxane formationin in platelets Since only epigallocatechin and saikosaponin a significantly inhibited the platelet aggregation induced by ADP, the effect of both compounds on thromboxane formation in platelets was then studied. Tbromboxane formations from both exogenous and
54
Prostnplandins Lcukotrienes and Essential Fatty Acids
cndogenous arachidonate were analyzed. In order to study the thromboxane formation from the exogenous arachidonic acid, platelets were treated with 5 pg of arachidonic acid after the preincubation with tested compounds. As indicated in Table 2, both of epigallocatechin and saikosaponin a at the concentrations of 10e5 and 10m4M significantly inhibited the thromboxane formation in platelets. For studying the thromboxane formation from endogenous arachidonic acid, platelets were treated with calcium ionosphere after the preincubation with tested compounds. Both of epigallocatechin and saikosaponin a dose-dependently inhibited the thromboxane formation (Table 3). In both assays, saikosaponin a showed more potent effect than epigallocatechin.
Control
”
Preventive effect on endothelial cell injury induced by H2@ Fig. 4. Effect of saikosaponin a on platelet aggregation Human platelet-rich plasma was pretreated with different concentrations of saikosaponin a at 37°C for 30 min. The platelet aggregation was then induced by 10e5 M ADP.
For studying the preventive effect of flavan-3-01 and saikosaponin compounds on endothelial cell injury induced by H202, the tested compounds were present in culture during H202 treatment. When the cells were incubated with 2 mM Hz02 for 6 h, approximate 45% cell injury was observed. Among the 10 compounds tested, only two flavan-3-01 compounds; gallocatechin-3-O-gallate and epicatechin3-0-gallate significantly showed the preventive effect on endothelial cell injury (Table 4). The preventive effect was dose-dependent.
Table 2 Effect of epigallocatechin
and saikosaponin a on thromboxane formation from exogenous arachidonic acid in intact platelets. Compounds
TXB, formation (ne/l x 10s cells) Control
1O-5 M
1O-4 M
Epigallocatechin
60 + 3.7 (100%)
52 f 2.8* (87%)
22 If: 0.9** (37%)
Saikosaponin a
51 f 2.5 (100%)
17 * 3.1** (33%)
11 + 0.5** (22%)
Platelets were treated with epigallocatechin or saikosaponin a at 37°C for 30 min. Formation for TXB, from exogenous arachidonic acid in intact platelets was measured as described in Methods. Values of product formation represent mean + S.E.M. from quadraplicate assays. Values in parentheses are the percentages compared with their controls. *P < 0.01, **p < 0.001.
DISCUSSION Two kinds of compounds of flavan-3-01s and saikosaponins were used in this series of studies. The flavan-3-01 compounds occur widely in tea plants, and several catechin compounds have been isolated (6). Catechin (cianidanol) has immunomodulatory actions, and has been tried in the
Table 3 Effect of epigallocatechin
formation from endogenous Compounds
and saikosaponin a on thromboxane arachidonic acid in intact platelets.
TXB, formation (rig/l x 10” cells) Control
1O-6 M
1O-5 M
1O-4 M
8.5 + 0.40* (79%)
6.0 +0.21** (56%)
Epigallocatechin
10.7 + 0.80 11.2 f 0.60 (100%) (105%)
Saikosaponin a
10.7 f 0.80 8.5 f 0.70* 2.2 + 0.24** 2.7 f 0.07* (21%) (25%) (100%) (79%)
Platelets were treated with epigallocatechin or saikosaponin a at 37°C for 30 min. Formation of TXBr from endogenous arachidonate triggered by 3 PM calcium ionophore A23187 was performed as described in Methods. Values of product formation represent mean + S.E.M. from quadraplicate assays. Values in parentheses are the percentages compared with their controls. “P C 0.05, **P < 0.001.
Effect of Flavan-3-01 and Saikosaponin on Platelets and Endothelial Cells Table 4 Inhibitory effect of flavan-3-01s and saikosaponins on H,Oiinduced endothelial cell injury Inhibition (%)
Compounds
Gallocatechin-3-0-gallate Epicatechin-3-0-gallate Epigallocatechin-3-0-galfate Gallocatechin Epigallocatechin Epicatechin Catechin Saikosaponin a Saikosaponin b2 Saikosaponin c
1O-6 M
lo-’ M
1O-4 M
9 9 0 0 0 0 0 0 0 6
27 30 0 0 0 0 1 0 0 9
86 53 0 0 0 0 15 0 0 18
After labelling the monolayer endothelial cells with “chromate, cells were treated with 2 mM H,O, in the presence or absence of flavan-3-01s and saikosaponins in 0.5 ml of DMEM for 6 h according to the Methods. Cell injury induced by HP2 was used as control. The percentage of inhibition was obtained by comparing the cytoprotective effect of tested compounds with the control injury. Each value is the mean of two sets of experiments.
treatment of HBeAg-positive chronic hepatitis (11). Saikosaponins are the major saponins in the dry foot of certain species of the Bupleurum plants (Umbelliferae) (7). Saikosaponins have hepatic cytoprotective effects of stimulating membrane stability, hepatic protein synthesis and hepatic glycogen formation (for review see 11). In the present study, the pharmacological effects of seven flavan-3-01 compounds and three saikosaponin compounds on platelet functions and endothelial cell injury were investigated. Among the 10 compounds tested only two compounds, epigallocatechin and saikosaponin a, were effective in inhibiting human platelet aggregation induced by ADP. Two phase aggregation is induced by ADP in human platelet-rich plasma. The second wave aggregation is recognized to be mediated by thromboxane formation (12). As indicated in Figures 3 and 4, both of epigallocatechin and saikosaponin a inhibited the second wave aggregation of ADP in human platelet-rich plasma. The results suggest that both of epigallocatechin and saikosaponin a might interfere the thromboxane formation in platelets. The thromboxane formation from exogenous and endogenous arachidonic acid was then studied. As indicated in Tables 2 and 3, both of epigallocatechin and saikosaponin a dosedependently inhibited the thromboxane formation from both exogenous and endogenous arachidonic acid. Therefore, the inhibition of thromboxane formation could explain the inhibitory effect of both epigallocatechin and saikosaponin a on the second wave aggregation induced by ADP in human platelet-rich plasma. It has been documented previously that active oxygen specifies including H202 can cause cytotoxic
55
injury to endothelial cells (13). Hz02 can cross membrane almost as readily as can water, the unsaturated fatty acids present (phospholipids, glycolipids, and sterols) and the transmembrane proteins containing oxidizable amino acids are susceptible to H202 damage (for review see 14). Since oxidative damage may play a role in the pathogenesis of atherosclerosis, H20zinduced endothelial cells in culture could be used for studying the intervention of oxidative damage in vascular endothelial cells. Among the compounds tested in the present study, only two flavan-3-01 compounds; gallocatechin-3-O-gallate and epicatechin-3-O-gallate significantly prevented the endothelial cell injury induced by H202. The pharmalogical mechanism of the cytoprotective effect is not studied yet in the present report. It might be due to the antioxidative effect of flavan-3-01 compounds since several catechin compounds have been-reported to have antioxidative effect determined by the active oxygen method at 97XC on lard (15). In conclusion, the inhibitory effect of epigallocatechin and saikosaponin a on platelet activation and the cytoprotective effect of gallocatechin-3-0gallate and epicatechin-3-O-gallate on HZOZ-induced endothelial cell injury could give evidence of explaining the possible role of flavan-3-01 and saikosaponin compounds in maintaining the vascular homeostasis. Acknowledgments We are greatly indebted to Ms. Chi-Yen Lin and Ms. I-Ping Li for their excellent technical assistance. Thanks are due to Ono Pharmaceutical Co for the supply of thromboxane B,. This research was supported in part by the National Science Council of the Republic of China (NSC 79-0420-BOO6-18).
References 1. Sacks T, Moldow C F, Craddock P R, Bowers T K, Jacob H S. Oxygen radicals mediate endothelial cell damage by complement-stimulated granulocytes. J. Clin. Invest. 61, 1161, 1978. 2. Weiss S J, Young J, LoBuglio A F, Slivka A, Nimeh N F. Role of hydrogen peroxide in neutrophil-mediated destruction of cultured endothelial cells. J. Clin. Invest. 68, 714, 1981. 3. Babior B. The respiratory burst of phagocytes. J. Clin. Invest. 73. 599, 1984. 4. Mazzone T, Jensen M, Chait A. Human arterial wall cells secrete factors that are chemotactic for monocytes. Proc. Natl. Acad. Sci. USA 80. 5094. 1983. 5 Hamberg M, Svensson J, Samuelsson B. Thromboxanes; a new group of biologically active compounds derived from prostaglandin endoperoxides. Proc. Natl. Acad. Sci. USA 72. 2994, 1975. 6. Hashimoto F, Nonaka G, Nishioka I. Tannin and related compounds. LVI. Isolation of four new acylated flavan-3-01s from oolong tea. Chem. Pharm. Bull. 35, 611, 1987. 7. Tani T, Katsuki T, Kubo M, Arichi S. Distribution of saikosaponins in Bupleurum falcatum root. J. Chromat. 360, 407, 1986.
56
Prostaglandins Leukotrienes
and Essential Fatty Acids
8. Chang W C, Hsu F L. Inhibition of platelet aggregation and arachidonate metabolism in platelets by procyanidins. Prost. Leuk. Essen. Fatty Acids 38, 181, 1989. 9. Sasaguri Y, Morimatsu M, Kinoshita T, Nakashima T, Inagaki T, Yagi K. Difference in susceptibility to injury by linoleic acid hydroperoxide between endothelial and smooth muscle cells of arteries. J. Appl. B&hem. 7, 70, 1985. 10. Abe M, Morita I, Murota S. A new in vitro method using Fura- for the quantification of endothelial cell injury. Prost. Leuk. Essen. Fatty Acids 34, 69, 1988. 11. Oda T, Oka T. Kampo preparations as biological response modifiers in the treatment of chronic viral hepatitis. In: Recent Advances in the Pharmacology of Kampo, pp. 369-377, 1988. (Ed: Hosoya E.
12.
13.
14.
15.
Yamamura Y.) Elsevier Science Publishers LTD. Amsterdam. Ushikubi F. Okuma M, Kanaii V, Sugiyama T, Ogorochi T, Narumiya S. U&no H. Hemorrhagic thrombocytopathy with platelet thromboxane A, receptor abnormality: Defective signal transduction with normal binding activity. Thromb. Haemost. 57, 158, 1987. Henson P M, Johnson R B. Tissue injury in inflammation. Oxidants, proteinases and cationic proteins. J. Clin. Invest. 79, 669, 1987. Weiss S J, LoBuglio A F. Phagocyte-generated oxygen metabolites and cellular injury. Lab. Invest. 47, 5, 1982. Matsuzaki T, Hara Y. Antioxidative activity of tea leaf catechins. Nippon Nogeikagaku Kaishi 59, 129, 1985.