- Email: [email protected]
Inhibition of thrombin-and collagen-induced phosphoinositides breakdown in rabbit platelets by a PAF antagonist-denudatin B, an isomer of kadsurenone
THROMBOSIS RESEARCH 59; 121-130,199O 0049-3848/90 $3.00 + .OO Printed in the USA. Copyright (c) 1990 Pergamon Press pk. All rights reserved.
INHIBITION OF THROMBIN- AND COLLAGEN-INDUCED PHOSPHOINOSITIDES BREAKDOWN IN RABBIT PLATELETS BY A PAF ANTAGONIST - DENUDATIN B, AN ISOMER OF KADSURENONE Che-Ming Teng, Sheu-Meei Yu, Chien-Chih Chen*, Yu-Lin Huang* and Tur-Fu Huang Pharmacological Institute, College of Medicine, National Taiwan University and *National Research Institute of Chinese Medicine, Taipei, Taiwan
(Received 1.3.1990; accepted in revised form 11.4.1990 by Editor H.C. Pirkle)
ABSTRACT Denudatin B, an isomer of kadsurenone, was isolated from Magnolia farnesii. It inhibited the aggregation and ATP release of washed rabbit platelets caused by platelet-activating factor (PAF) in a The IC5G on PAF (2 ng/ml)-induced concentration-dependent manner. High concentration of denudatin B aggregation was about 10 pg/ml. (>50 pg/ml) also inhibited the aggregation and ATP release of platelets caused by ADP, collagen, arachidonic acid and thrombin. However, shape change of platelets still existed. Prolongation of the incubation time with platelets could not cause further inhibition, and the aggregability of platelets could be restored after denudatin B was washed out from platelets. Thrombin-induced thromboxane B2 formation was almost completely suppressed. In the absence of extracellular calcium (EGTA 1 mM), ATP release caused by thrombin was inhibited. Thrombin-induced rise of the intracellular calcium concentration was suppressed by denudatin B, but not by BN52021 or kadsurenone. The generation of inositol phosphate in washed platelets caused by collagen, PAF and thrombin was also suppressed. The data indicate that PAF antagonist denudatin B has nonspecific antiplatelet action at high concentration by inhibiting phosphoinositides breakdown induced by collagen and thrombin.
Keywords: Denudatin B, PAF antagonists, Phosphoinositides breakdown, Thromboxane B2 formation, Rabbit platelets, Magnolia farnesii. Correpondence to: C.M. Teng, Pharmacol. Inst., College of Medicine, National Taiwan Univ., No. 1, Jen-Ai Rd., Sect. 1, Taipei 10018, Taiwan.
121
122
PAF ANTAGONIST ON PI BREAKDOWN
Vol. 59, No. 1
INTRODUCTION Platelet-activating factor (PAF) is a phospholipid mediator produced by various cells (1). It possesses potent biological activity on many cells and organs, and may have a potential role in various physiopathological situations such as asthma, thrombosis, cardiopathy and nephropathy (2,3). Many PAF antagonists have been introduced recently, such as kadsurenone (4), BN52021 (5) from natural sources, and many synthetic compounds, such as WEB 2086 (6), SM-10061, YM-461 etc. (7). They are very specific for antagonizing the PAF-induced platelet aggregation without having any effect on that caused by ADP, arachidonic acid, thrombin, collagen or other inducers. Denudatin B, isolated from Magnolia fargesii (8), is an isomer of kadsurenone (Fin. 1). It inhibits the aggregation of washed rabbit platelets In
denudatin B
kadsurenone
Fig. 1. Structures of denudatin B and kadsurenone
MATEXIALSAND METHODS Denudatin B was isolated from Magnolia farnesii (8), and Materials. PAF was ourchased from Calbiochemdissolved in dimethvl sulfoxide (DMSO). . Behring Co. and dissolved in chloroform. Thrombin '(bovine)was obtained from Parke Davis & Co. and dissolved in 50% glycerol to give a stock solution of 100 NIH units/ml. Collagen (type 1, bovine Achilles tendon), obtained from Sigma Chem. Co. was homogenized in 25 mM acetic acid and stored (1 mg/ml) at -7oOc. ADP, arachidonic acid, EGTA, EDTA, bovine serum albumin, quin-2/AM, indomethacin, luciferase and luciferin were purchased from Sjgma Chem. Co., [ H] Myo-inositol Ionophore A23187 was obtained from Calbiochem-Behring Co. was purchased from Amersham. BN52021 was a gift of Dr. P. Braquet of Kadsurenone is kindly given by Merck, Sharp & Institute Henri Beaufour. Dohme. Other chemicals were obtained from Wako Pure Chem. Co. r
Platelet aggregation and release reaction. Platelet suspension was obtained from EDTA-anticoazated, platelet-rich plasma, according to the washing procedures described by us (10). Platelet pellets were finally suspended in Tyrode's solution containing calcium (1 mM) and bovine serum albumin (0.35%). Aggregation was measured by the turbidimetric method (11) and assigned the absorbance of platelet suspension as 0% aggregation and the absorbance of Tyrode's solution as 100% aggregation. ATP released from platelets was detected by the bioluminescence method as described by DeLuca and McElory (12). Both the aggregation and ATP release were simultaneously measured by a Lumi-aggregometer (Model 1020, Payton) connected to two dualchannel recorders. Just 1 min before the addition of the aggregation inducer,
Vol. 59, No. 1
PAF ANTAGONIST ON PI BREAKDOWN
123
platelet suspension was stirred at 900 rpm. In order to eliminate the effect of the solvent on the aggregation or ATP release, the final concentration of DMSO in the platelet suspension was fixed to 0.5%. Thromboxane l2 assay. Six min after the incubation of platelets with the inducer, 2 mM EDTA and 50 uM indomethacin were added. After centrifugation in an Eppendorf Centrifuge'(Mode1 5414) for 2 min, thromboxane B2 in the supernatant was assayed using the RIA kits (Amersham) according to the procedure described by the manufacturer. Measurement of intracellular calcium. The method of Rink et al. (13) was followed. Platelets were incubated with auin-2/AM (20 uM) at room temnerature for 40 min and then the platelets were&washed with 'EDNA (2 mM)-containing Tyrode's solution. After centrifugation, platelets were suspended in Tyrode's Fluorescence was measured solution containing 0.35% bovine serum albumin. with a Hitachi Fluorescence Spectrophotometer (Ex 339 nm, Em 492 nm). Intracellular calcium concentration was calculated according to the equation of Tsien et al. (14). Platelet membrane Measurement of phosphoinositides breakdown. phospholipids were-iabelled with ['Hlinositol phosphate according to the methods of Huang and Detwiler (15) and Neylon and Summers (16). The reaction was then carried out at 37'C for 6 min with 1 ml of [3H]inositol labelledplatelets with a stirring bar driven at 900 rpm. An equal volume of 10% (v/v) trichloroacetic acid was added to stop the reaction. After centrifugation at 1000 x g for 10 min, 1 ml of supernatant was pooled and trichloroacetic acid was removed by extracting with 5 x 2 volumes of diethylether. The aqueous phase was applied to a Dowex-1 ion exchange column for separation of inositol phosphates as described by Neylon and Summers (16). All the experiments were carried out in the presence of 5 mM LiCl to inhibit InsP phosphatase. Because the levels of inositol bisphosphate and inositol trisphosphate were very low, inositoi monophosphate was measured as an index of the total inositol phosphates formation. RESULTS Denudatin B inhibited PAF-induced aggregation of rabbit platelets. This inhibition was concentration-dependent on denudatin B. Release of ATP from platelets induced by PAF was also suppressed (Fig. 2). Rabbit platelets were very sensitive to PAF, and Fig. 3 showed dose-response curve of PAF from 0.01 to 10 ng/ml. Denudatin B caused a parallel shift of this curve to the right. The IC50 of denudatin -B on PAF (2 ng/ml)-induced platelet aggregation was calculated to be about 10 pg/ml. Denudatin B was specific for antagonizing platelet aggregation caused by PAF at concentrations lower than 20 pg/ml. At 50 pg/ml, denudatin B inhibited completely the aggregation caused by PAF, and also partially those by ADP, collagen and arachidonic acid (Table 1). Although thrombin (0.1 II/ml)-induced aggregation was not significantly affected by denudatin B (Table 1 and Fig. the ATP release reaction was markedly suppressed. If washed platelets both $ie challenged with lower concentration of thrombin (0.02 U/ml) Hdwever, platelet aggregation and ATP release were completely inhibited. shape change caused by thrombin was still observed (Fig. 4). After treating the platelets with denudatin B (20 ug/ml) for 15 min at 37OC and then washing them twice with the suspending solution, PAF-induced platelet aggregation was completely restored. Incubation of denudatin B with platelets for 30 min at 37'C did not cause more marked inhibition than that incubated for 1 min (data not shown).
124
PAF ANTAGONIST ON PI BREAKDOWN
Vol. 59, No. 1
Fig. 2. Effect of denudatin B on PAF-induced aggregation and ATP release of washed rabbit platelets. Platelets were incubated with DMSO (0.5%, control) or denudatin B (lo,20 or 30 pg/ml> at 37'C for 1 min, then PAF (2 &ml) was added to trigger the aggregation (upward tracings) and ATP release (downward tracings).
100
80
0 1 ??
control 5&ml denudatin B
A
lo/@/Id
A 204ml
0.10
1.00 Pm
10.00
hdml)
Fig. 3. Inhibitory effect of denudatin B on platelet aggregation induced by PAF. Washed rabbit platelets were incubated with denudatin B (5,lO or 20 pg/ml) or DMSO (0.5%, control) for 1 min, then PAF (0.01-10 ng/ml) was added to trigger the aggregation.
Vol. 59, No. 1
PAF ANTAGONIST ON PI BREAKDOWN
125
Table 1. Effect of denudatin B on the aggregation of washed rabbit platelets induced by arachidonic acid (AA), PAF, collagen, ADP, ionophore A23187 and thrombin.
Aggregation (X) Inducer
Control
AA (100 FM)
78.3 + 2.9
74.7 2 3.9
PAF (2 ng/ml)
79.6 + 5.0
19.6 + 1.2""
Collagen (10 us/ml)
79.9 + 2.6
76.9 + 4.5
53.1 + 1.24
ADP (20 j&l)
64.1 + 2.8
63.8 + 1.8
43.1 + 1.2"
Ionophore A23187(2uM) 77.9 + 2.2
77.5 + 1.8
78.1 + 1..5
Thrombin (O.lU/ml)
78.5 + 1.8
79.7 + 2.4
82.7 + 2.0
Denudatin B __-__-----____---____----~__------__ 50 pg/ml 20 @ml
62.1 + 1.64 0.0 + 0.09""
Platelets were preincubated with denudatin B (20 or 50 pg/ml) or 0.5% DMSO (control) at 37'C for 1 min, then the inducer was added. Values are presented as means+S.E. (n-4-6). *: p < 0.05, **: p < 0.01, ***: p < 0.001 as compared with the respective control.
T
I
AT 10%
L
Min
w
m
b thrombin (O.lU/mt)
Fig. 4. Effect of denudatin B on the thrombin-induced aggregation and ATP release of washed rabbit platelets. Platelets were preincubated with 0.5% DMSO (control) or denudatin B (50 ug/ml) at 37'C for 1 min, then thrombin (0.1 or 0.02 U/ml) was added to trigger the aggregation (upward tracings) and ATP release (downward tracings).
126
PAF ANTAGONIST ON PI BREAKDOWN
rig.
Vol. 59, No. 1
5.
Effect of denudatin B on thrombin-induced aggregation and ATP release. Platelet suspension (without calcium) was' preincubated with 0.5% DMSO (left) or denudatin B (right) in the presence of EGTA (1 mM) for 1 min. Then, thrombin (0.1 U/ml) was added, 4 min later, Ca++ (2 mM) was added to trigger further aggregation (upward tracings) and ATP release (downward tracings).
[Ca++]
:q
7601
i
K
BN
denudatin
52021
kadsurenone
B Fig. 6. Effect of PAF antagonists on the increase of intracellular calcium concentration caused by thrombin in quin-2/AM-treated washed rabbit platelets. Platelets were preincubated with 0.5% DMSO (control), denudatin B (50 pg/ml), BN52021 (20 pg/ml) or kadsurenone (20 pg/ml) at 37'C for 1 min, then thrombin (0.1 U/ml) was added to trigger the intracellular calcium mobilization.
Vol. 59, No. 1
PAF ANTAGONIST ON PI BREAKDOWN
127
ihromboxane B2 formation in washed platelets was measured at 6 min after the aggregation inducer was added. Table 2 showed that denudatin B inhibited almost completely the thromboxane B2 formation caused by thrombin, while suppressed slightly, but significantly, that caused by collagen.
Table 2. Effect of denudatin B on the thromboxane B2 formation of washed rabbit platelets caused by collagen, thrombin and ionophore A23187.
Thromboxane B2 (ng/ml) Inducer
Control
Collagen (10 rg/ml)
287.4 +
9.9
Thrombin (0.1 U/ml)
159.5 +
8.8
8.4 i
Ionophore A23187 (2 PM)
163.7 + 19.1
168.6 +
Denudatin B
216.5 + 25.0" 2.1*3t 4.9
Denudatin B (50 pg/ml) or DMSO (0.5%, control) was preincubated with platelets at 37OC for 1 min, then the inducer was added. Aggregation and thromboxane B2 formation were terminated by EDTA (2 mM) and indomethacin (50 pM) 6 min after the addition of the inducer, Values are presented as means.S.E. (n=5). *: p < 0.05, **: p < 0.01 as compared with the respective control. In the absence of extracellular calcium (1 mM EGTA), thrombin induced a marked release of ATP with only slight aggregation. If 2 mM of calcium was then added, a complete aggregation was obtained with some residual release of ATP (Fig. 5). Pretreatment of platelets with denudatin B for 1 min caused a suppression of ATP release especially in the first phase. The change in the intracellular calcium was then studied using quin-Z/AMloaded platelets. As shown in Fig. 6, intracellular calcium was increased after the platelets were stimulated with thrombin. In the presence of denudatin B the rise of intracellular calcium caused by (50 j&ml), thrombin was markedly suppressed. The isomer of denudatin B, kadsurenone (20 pg/ml) or another PAF antagonist BN52021 (20 Pg/ml) did not affect this calcium increase. The concentrations of both PAF antagonists used were about 10 to 15 times that for complete inhibition of PAF. In the presence of calcium (1 mM), collagen and PAF caused 4.7~1.0 and 4.220.7 fold, respectively, of the increase of inositol phosphate formation compared with the resting level. In the absence of calcium, thrombin caused Denudatin B (50 pg/ml) inhibited completely the 8.6t1.8 fold increase. inositol phosphate formation caused by PAF, and markedly those by collagen and thrombin (Fig. 7).
DISCUSSION Denudatin B is a neolignan compound isolated from Magnolia farnesii which has been commonly used in traditional Chinese medicine for the treatment of some nasal symptoms, e.g. sinusitis, allergic rhinitis and nasal congestion with headache (17). It was also isolated from some other species of Magnolia (5). As an isomer of well known PAF antagonist kadsurenone which was isolated from Piper futokadsura (4), denudatin B also competitively inhibited PAFinduced platelet platelet aggregation. In addition to the PAF-antagonizing
128
PAF ANTAGONIST ON PI BREAKDOWN
Vol. 59. No. 1
12T
With Denudatin B
Collagen
PAF
Thrombin
Fig. 7. Inhibition of denudatin B on the formation of inositol hosphate in washed rabbit platelets caused by collagen, PAF and thrombin. [L? H]inositol-labelled platelets were incubated with thrombin (0.1 U/ml) in the absence of calcium or with PAF (2 ng/ml), collagen (10 pg/ml) in the presence of calcium (1 mM) for 6 min. Folds of increase of inositol phosphate (IP) were presented as meanstS.E. *: p < 0.05, ***: p < 0.001 as compared with the respective control. action, high concentration of denudatin B caused some inhibition on the aggregation induced by ADP, collagen, arachidonic acid and also low concentration of thrombin. PAF is one of the most potent inducers of platelet aggregation known in many animal species (18). PAF-induced aggregation is associated with release reaction and degranulation (19). Although thromboxane B formation was reported to be increased in human platelets (20), our an3 other data have found PAF-induced aggregation to be independent of ADP release or of cyclooxygenase products in rabbit platelets (10,21). Recent evidence has shown PAF activates phospholipase C. This leads to the formation of diacylglycerol and inositol polyphosphates, and finally causes the mobilization of calcium and release reaction (22). Most of the PAF antagonists introduced are very highly specific (except some phospholipid analogs) against PAF-induced platelet aggregation, without affecting aggregation caused by ADP, collagen or thrombin (5). Denudatin B (50 pg/ml) inhibited thrombin-induced aggregation, ATP release, thromboxane B2 formation and rise of intracellular calcium. All of these actions could be explained by its inhibition on the phosphoinositides breakdown as revealed by the suppression of inositol phosphate formation caused by thrombin. Denudatin B did not affect the arachidonate-prostaglandin endoperoxides-thromboxane transformation because it did not affect the thromboxane B2 formation in the mixture of platelet homogenate with arachidonic acid (data not shown). PAF-induced pharmacological actions can be antagonized by some calcium channel blockers. Diltiazem and verapamil inhibited aggregation by PAF and also inhibited PAF binding in a dose-dependent manner (23). This suggests
Vol. 59, No. 1
PAF ANTAGONIST ON PI BREAKDOWN
129
that PAF receptor may be closely associated with calcium channels. Our data in this paper also suggest PAF receptor may be closely associated with thrombinor collagen-induced signal transduction, especially phosphoinositides breakdown in the initial activation stage. ACKNOWLEDGEMENTS This work was supported by research grants of the National Science Council of the Republic of China (NSC79-0412-B002-26) and the National Institute of Chinese Medicine. REFERENCES 1. BRAQUET, P., TOUQUI, L., SHEN, T.Y. and VARGAFTIG, B.B. Perspectives in platelet-activating factor research. Pharmacol. -Rev. 39, 97-145, 1987. 2. VARGAFTIG, B.B., CHIGNARD, M., BENVENISTE, J., LEFORT, J. and WALL, F. Background and present status of research on platelet-activating factor (PAF-acether). --Ann. N.Y. Acad. Sci. 370, 119-137, 1981. 3. PAGE, C.P., ARCHER, C.B., PAUL, W. and MORLEY, J. Paf-acether: a mediator of inflammation and asthma. Trends Pharmacol. -_ Sci. 5, 239-241, 1984. 4. SHEN, T.Y., HWANG, S.B., CHANG, M.N., DOEBBER, T.W., LAM, M.H.T., WU, M.S., WANG, X., HAN, G.Q. and LI, R.Z. Characterization of a plateletactivating factor receptor antagonist isolated from haifenteng (Piper in vivo plateletfutokadsura): specific inhibition of in vitro and -activating factor-induced effects. ----ProcTNxAcad. Sci. USA 82, 672676, 1985. 5. BRAQUET, P. and GODFROID, J.J. PAF-acether specific binding sites: 2. Design of specific antagonists. Trends Pharmacol. Sci. 7, 397-403, 1986. M., LEFORT, J., MALANCHERE, E. and VARGAFTIG, B.B. 6. PRETOLANTI, Interference by the novel PAF-acether antagonist WEB 2086 with the bronchopulmonary responses to PAF-acether and to active and passive anaphylactic shock in guinea-pigs. -Eur. J. Pharmacol. 140, 311-321, 1987. 7. PAGE, C. and ABBOTT, A. PAF: new antagonists, new roles in disease and a major role in reproductive biology. Trends Pharmacol. Sci. 10, 256-257, 1989. 8. CHEN, C.C., HUANG, Y.L., CHEN, Y.P., HSU, H.Y. and KUO, Y.H. Three new neolignans, Fargesones A,B and C, from the flower buds of Magnolia fargesii. -CHEM. PHARM. BULL. 36, 1791-1795, 1988. 9. TENG, C.M., YU, S.M. and CHEN, C.C. Effects of PAF antagonists isolated from Magnolia fargesii on platelet aggregation and contraction of rat thoracic aorta. In: Taipei PAF Sattellite Symposium, Proceedings, N. Hicks (editor). Excerpta Medica Asia Ltd 1989 (in press). lO.TENG, C.M., CHEN, W.Y., KO, W.C. and OUYANG, C. Antiplatelet effect of butylidenephthalide. Biochim. Biophys. Acta 924, 375-382, 1987. ll.O'BRIEN, J.R. Platelet aggregation II, some results from a new method of J. Clin. Pathol. 2, 452-455, 1962. study. --
130
PAF ANTAGONIST ON PI BREAKDOWN
Vol. 59. No. 1
12.DELUCA, M. and McELORY, W.D. Purification and properties of firefly luciferase. Methods Enzymol. 57, 3-15, 1978. 13.RINK, T.J., SANCHEZ, A. and HALLAM, T.J. Diacylglycerol and phorbol ester stimulate secretion without raising cytoplasmic free calcium in human platelets. -Nature 305, 317-319, 1983. 14.TSIEN, R.Y., POZZAN, T. and RINK, T.J. Calcium homeostasis in intact lymphocytes: cytoplasmic free calcium monitored with a new, intracellularly trapped fluorescencent indicator. -J. Cell Biol. 94, 325334, 1982. The effect of lithium on platelet 15.HUANG, E.M. and DETWILER, T.C. phosphoinositide metabolism. Biochem. -J. 236, 895-901, 1986. Stimulation of l-adrenoceptors in rat 16. NEYLQN, C.B. and SUMMERS, R.J. Br. J kidney mediates increased inositol phospholipid hydrolysis. -2 Pharmacol. 91, 367-376, 1987. Pharmacology and applications of Chinese 17. CHANG, H.Y. and BUT, P.P.H. Materia Medica, pp 629-635, World Scient. Pub., Philadelphia, 1986. Species 18. NAMM, D.H., ANJANEYULU, S., TADEPALLI, S. and HIGH, J.A. specificity of the platelet response to l-O-alkyl-2-acetyl-sn-glycerophosphocholine. Thromb. Res. 25, 341-350, 1982. Effect of 19. CHESNEY, C.M., PIFER, D.D., BYERS, L.W. and MUIRHEAD, E.E. platelet-activating factor (PAF) in human platelets. Blood 59, 582-585, 1982. 20 McMANUS, L.M., HANAHAN, D.J. and PINCKARD, R.N. stimulation by acetylglyceryl ether phosphorylcholine. 67, 903-906, 1981.
Human platelet J. Clin. Invest. --
21. CAZENAVE, J.P., BENVENISTE, J. and MUSTARD, J.F. Aggregation of rabbit platelets by platelet activating factor is independent of the release reaction and the arachidonate pathway and inhibited by membrane active drugs. Lab. Invest. 4l, 275-285, 1979. 22. ETIENNE, A. and BAROGGI, N. Effects of ginkgolides on PAF-induced calcium mobilization in platelets. In: Ginkgolides - Chemistry, Biology, Pharmacology and Clinical Perspectives. P. Braquet (Ed.), pp. 115-125, J.R. Prous SC&e Publishers, 1988. 23,VALONE, F.H. Inhibition of platelet-activating factor binding to human platelet by calcium channel blockers. Thromb. -Res 45, 427-435, 1987.
INHIBITION OF THROMBIN- AND COLLAGEN-INDUCED PHOSPHOINOSITIDES BREAKDOWN IN RABBIT PLATELETS BY A PAF ANTAGONIST - DENUDATIN B, AN ISOMER OF KADSURENONE Che-Ming Teng, Sheu-Meei Yu, Chien-Chih Chen*, Yu-Lin Huang* and Tur-Fu Huang Pharmacological Institute, College of Medicine, National Taiwan University and *National Research Institute of Chinese Medicine, Taipei, Taiwan
(Received 1.3.1990; accepted in revised form 11.4.1990 by Editor H.C. Pirkle)
ABSTRACT Denudatin B, an isomer of kadsurenone, was isolated from Magnolia farnesii. It inhibited the aggregation and ATP release of washed rabbit platelets caused by platelet-activating factor (PAF) in a The IC5G on PAF (2 ng/ml)-induced concentration-dependent manner. High concentration of denudatin B aggregation was about 10 pg/ml. (>50 pg/ml) also inhibited the aggregation and ATP release of platelets caused by ADP, collagen, arachidonic acid and thrombin. However, shape change of platelets still existed. Prolongation of the incubation time with platelets could not cause further inhibition, and the aggregability of platelets could be restored after denudatin B was washed out from platelets. Thrombin-induced thromboxane B2 formation was almost completely suppressed. In the absence of extracellular calcium (EGTA 1 mM), ATP release caused by thrombin was inhibited. Thrombin-induced rise of the intracellular calcium concentration was suppressed by denudatin B, but not by BN52021 or kadsurenone. The generation of inositol phosphate in washed platelets caused by collagen, PAF and thrombin was also suppressed. The data indicate that PAF antagonist denudatin B has nonspecific antiplatelet action at high concentration by inhibiting phosphoinositides breakdown induced by collagen and thrombin.
Keywords: Denudatin B, PAF antagonists, Phosphoinositides breakdown, Thromboxane B2 formation, Rabbit platelets, Magnolia farnesii. Correpondence to: C.M. Teng, Pharmacol. Inst., College of Medicine, National Taiwan Univ., No. 1, Jen-Ai Rd., Sect. 1, Taipei 10018, Taiwan.
121
122
PAF ANTAGONIST ON PI BREAKDOWN
Vol. 59, No. 1
INTRODUCTION Platelet-activating factor (PAF) is a phospholipid mediator produced by various cells (1). It possesses potent biological activity on many cells and organs, and may have a potential role in various physiopathological situations such as asthma, thrombosis, cardiopathy and nephropathy (2,3). Many PAF antagonists have been introduced recently, such as kadsurenone (4), BN52021 (5) from natural sources, and many synthetic compounds, such as WEB 2086 (6), SM-10061, YM-461 etc. (7). They are very specific for antagonizing the PAF-induced platelet aggregation without having any effect on that caused by ADP, arachidonic acid, thrombin, collagen or other inducers. Denudatin B, isolated from Magnolia fargesii (8), is an isomer of kadsurenone (Fin. 1). It inhibits the aggregation of washed rabbit platelets In
denudatin B
kadsurenone
Fig. 1. Structures of denudatin B and kadsurenone
MATEXIALSAND METHODS Denudatin B was isolated from Magnolia farnesii (8), and Materials. PAF was ourchased from Calbiochemdissolved in dimethvl sulfoxide (DMSO). . Behring Co. and dissolved in chloroform. Thrombin '(bovine)was obtained from Parke Davis & Co. and dissolved in 50% glycerol to give a stock solution of 100 NIH units/ml. Collagen (type 1, bovine Achilles tendon), obtained from Sigma Chem. Co. was homogenized in 25 mM acetic acid and stored (1 mg/ml) at -7oOc. ADP, arachidonic acid, EGTA, EDTA, bovine serum albumin, quin-2/AM, indomethacin, luciferase and luciferin were purchased from Sjgma Chem. Co., [ H] Myo-inositol Ionophore A23187 was obtained from Calbiochem-Behring Co. was purchased from Amersham. BN52021 was a gift of Dr. P. Braquet of Kadsurenone is kindly given by Merck, Sharp & Institute Henri Beaufour. Dohme. Other chemicals were obtained from Wako Pure Chem. Co. r
Platelet aggregation and release reaction. Platelet suspension was obtained from EDTA-anticoazated, platelet-rich plasma, according to the washing procedures described by us (10). Platelet pellets were finally suspended in Tyrode's solution containing calcium (1 mM) and bovine serum albumin (0.35%). Aggregation was measured by the turbidimetric method (11) and assigned the absorbance of platelet suspension as 0% aggregation and the absorbance of Tyrode's solution as 100% aggregation. ATP released from platelets was detected by the bioluminescence method as described by DeLuca and McElory (12). Both the aggregation and ATP release were simultaneously measured by a Lumi-aggregometer (Model 1020, Payton) connected to two dualchannel recorders. Just 1 min before the addition of the aggregation inducer,
Vol. 59, No. 1
PAF ANTAGONIST ON PI BREAKDOWN
123
platelet suspension was stirred at 900 rpm. In order to eliminate the effect of the solvent on the aggregation or ATP release, the final concentration of DMSO in the platelet suspension was fixed to 0.5%. Thromboxane l2 assay. Six min after the incubation of platelets with the inducer, 2 mM EDTA and 50 uM indomethacin were added. After centrifugation in an Eppendorf Centrifuge'(Mode1 5414) for 2 min, thromboxane B2 in the supernatant was assayed using the RIA kits (Amersham) according to the procedure described by the manufacturer. Measurement of intracellular calcium. The method of Rink et al. (13) was followed. Platelets were incubated with auin-2/AM (20 uM) at room temnerature for 40 min and then the platelets were&washed with 'EDNA (2 mM)-containing Tyrode's solution. After centrifugation, platelets were suspended in Tyrode's Fluorescence was measured solution containing 0.35% bovine serum albumin. with a Hitachi Fluorescence Spectrophotometer (Ex 339 nm, Em 492 nm). Intracellular calcium concentration was calculated according to the equation of Tsien et al. (14). Platelet membrane Measurement of phosphoinositides breakdown. phospholipids were-iabelled with ['Hlinositol phosphate according to the methods of Huang and Detwiler (15) and Neylon and Summers (16). The reaction was then carried out at 37'C for 6 min with 1 ml of [3H]inositol labelledplatelets with a stirring bar driven at 900 rpm. An equal volume of 10% (v/v) trichloroacetic acid was added to stop the reaction. After centrifugation at 1000 x g for 10 min, 1 ml of supernatant was pooled and trichloroacetic acid was removed by extracting with 5 x 2 volumes of diethylether. The aqueous phase was applied to a Dowex-1 ion exchange column for separation of inositol phosphates as described by Neylon and Summers (16). All the experiments were carried out in the presence of 5 mM LiCl to inhibit InsP phosphatase. Because the levels of inositol bisphosphate and inositol trisphosphate were very low, inositoi monophosphate was measured as an index of the total inositol phosphates formation. RESULTS Denudatin B inhibited PAF-induced aggregation of rabbit platelets. This inhibition was concentration-dependent on denudatin B. Release of ATP from platelets induced by PAF was also suppressed (Fig. 2). Rabbit platelets were very sensitive to PAF, and Fig. 3 showed dose-response curve of PAF from 0.01 to 10 ng/ml. Denudatin B caused a parallel shift of this curve to the right. The IC50 of denudatin -B on PAF (2 ng/ml)-induced platelet aggregation was calculated to be about 10 pg/ml. Denudatin B was specific for antagonizing platelet aggregation caused by PAF at concentrations lower than 20 pg/ml. At 50 pg/ml, denudatin B inhibited completely the aggregation caused by PAF, and also partially those by ADP, collagen and arachidonic acid (Table 1). Although thrombin (0.1 II/ml)-induced aggregation was not significantly affected by denudatin B (Table 1 and Fig. the ATP release reaction was markedly suppressed. If washed platelets both $ie challenged with lower concentration of thrombin (0.02 U/ml) Hdwever, platelet aggregation and ATP release were completely inhibited. shape change caused by thrombin was still observed (Fig. 4). After treating the platelets with denudatin B (20 ug/ml) for 15 min at 37OC and then washing them twice with the suspending solution, PAF-induced platelet aggregation was completely restored. Incubation of denudatin B with platelets for 30 min at 37'C did not cause more marked inhibition than that incubated for 1 min (data not shown).
124
PAF ANTAGONIST ON PI BREAKDOWN
Vol. 59, No. 1
Fig. 2. Effect of denudatin B on PAF-induced aggregation and ATP release of washed rabbit platelets. Platelets were incubated with DMSO (0.5%, control) or denudatin B (lo,20 or 30 pg/ml> at 37'C for 1 min, then PAF (2 &ml) was added to trigger the aggregation (upward tracings) and ATP release (downward tracings).
100
80
0 1 ??
control 5&ml denudatin B
A
lo/@/Id
A 204ml
0.10
1.00 Pm
10.00
hdml)
Fig. 3. Inhibitory effect of denudatin B on platelet aggregation induced by PAF. Washed rabbit platelets were incubated with denudatin B (5,lO or 20 pg/ml) or DMSO (0.5%, control) for 1 min, then PAF (0.01-10 ng/ml) was added to trigger the aggregation.
Vol. 59, No. 1
PAF ANTAGONIST ON PI BREAKDOWN
125
Table 1. Effect of denudatin B on the aggregation of washed rabbit platelets induced by arachidonic acid (AA), PAF, collagen, ADP, ionophore A23187 and thrombin.
Aggregation (X) Inducer
Control
AA (100 FM)
78.3 + 2.9
74.7 2 3.9
PAF (2 ng/ml)
79.6 + 5.0
19.6 + 1.2""
Collagen (10 us/ml)
79.9 + 2.6
76.9 + 4.5
53.1 + 1.24
ADP (20 j&l)
64.1 + 2.8
63.8 + 1.8
43.1 + 1.2"
Ionophore A23187(2uM) 77.9 + 2.2
77.5 + 1.8
78.1 + 1..5
Thrombin (O.lU/ml)
78.5 + 1.8
79.7 + 2.4
82.7 + 2.0
Denudatin B __-__-----____---____----~__------__ 50 pg/ml 20 @ml
62.1 + 1.64 0.0 + 0.09""
Platelets were preincubated with denudatin B (20 or 50 pg/ml) or 0.5% DMSO (control) at 37'C for 1 min, then the inducer was added. Values are presented as means+S.E. (n-4-6). *: p < 0.05, **: p < 0.01, ***: p < 0.001 as compared with the respective control.
T
I
AT 10%
L
Min
w
m
b thrombin (O.lU/mt)
Fig. 4. Effect of denudatin B on the thrombin-induced aggregation and ATP release of washed rabbit platelets. Platelets were preincubated with 0.5% DMSO (control) or denudatin B (50 ug/ml) at 37'C for 1 min, then thrombin (0.1 or 0.02 U/ml) was added to trigger the aggregation (upward tracings) and ATP release (downward tracings).
126
PAF ANTAGONIST ON PI BREAKDOWN
rig.
Vol. 59, No. 1
5.
Effect of denudatin B on thrombin-induced aggregation and ATP release. Platelet suspension (without calcium) was' preincubated with 0.5% DMSO (left) or denudatin B (right) in the presence of EGTA (1 mM) for 1 min. Then, thrombin (0.1 U/ml) was added, 4 min later, Ca++ (2 mM) was added to trigger further aggregation (upward tracings) and ATP release (downward tracings).
[Ca++]
:q
7601
i
K
BN
denudatin
52021
kadsurenone
B Fig. 6. Effect of PAF antagonists on the increase of intracellular calcium concentration caused by thrombin in quin-2/AM-treated washed rabbit platelets. Platelets were preincubated with 0.5% DMSO (control), denudatin B (50 pg/ml), BN52021 (20 pg/ml) or kadsurenone (20 pg/ml) at 37'C for 1 min, then thrombin (0.1 U/ml) was added to trigger the intracellular calcium mobilization.
Vol. 59, No. 1
PAF ANTAGONIST ON PI BREAKDOWN
127
ihromboxane B2 formation in washed platelets was measured at 6 min after the aggregation inducer was added. Table 2 showed that denudatin B inhibited almost completely the thromboxane B2 formation caused by thrombin, while suppressed slightly, but significantly, that caused by collagen.
Table 2. Effect of denudatin B on the thromboxane B2 formation of washed rabbit platelets caused by collagen, thrombin and ionophore A23187.
Thromboxane B2 (ng/ml) Inducer
Control
Collagen (10 rg/ml)
287.4 +
9.9
Thrombin (0.1 U/ml)
159.5 +
8.8
8.4 i
Ionophore A23187 (2 PM)
163.7 + 19.1
168.6 +
Denudatin B
216.5 + 25.0" 2.1*3t 4.9
Denudatin B (50 pg/ml) or DMSO (0.5%, control) was preincubated with platelets at 37OC for 1 min, then the inducer was added. Aggregation and thromboxane B2 formation were terminated by EDTA (2 mM) and indomethacin (50 pM) 6 min after the addition of the inducer, Values are presented as means.S.E. (n=5). *: p < 0.05, **: p < 0.01 as compared with the respective control. In the absence of extracellular calcium (1 mM EGTA), thrombin induced a marked release of ATP with only slight aggregation. If 2 mM of calcium was then added, a complete aggregation was obtained with some residual release of ATP (Fig. 5). Pretreatment of platelets with denudatin B for 1 min caused a suppression of ATP release especially in the first phase. The change in the intracellular calcium was then studied using quin-Z/AMloaded platelets. As shown in Fig. 6, intracellular calcium was increased after the platelets were stimulated with thrombin. In the presence of denudatin B the rise of intracellular calcium caused by (50 j&ml), thrombin was markedly suppressed. The isomer of denudatin B, kadsurenone (20 pg/ml) or another PAF antagonist BN52021 (20 Pg/ml) did not affect this calcium increase. The concentrations of both PAF antagonists used were about 10 to 15 times that for complete inhibition of PAF. In the presence of calcium (1 mM), collagen and PAF caused 4.7~1.0 and 4.220.7 fold, respectively, of the increase of inositol phosphate formation compared with the resting level. In the absence of calcium, thrombin caused Denudatin B (50 pg/ml) inhibited completely the 8.6t1.8 fold increase. inositol phosphate formation caused by PAF, and markedly those by collagen and thrombin (Fig. 7).
DISCUSSION Denudatin B is a neolignan compound isolated from Magnolia farnesii which has been commonly used in traditional Chinese medicine for the treatment of some nasal symptoms, e.g. sinusitis, allergic rhinitis and nasal congestion with headache (17). It was also isolated from some other species of Magnolia (5). As an isomer of well known PAF antagonist kadsurenone which was isolated from Piper futokadsura (4), denudatin B also competitively inhibited PAFinduced platelet platelet aggregation. In addition to the PAF-antagonizing
128
PAF ANTAGONIST ON PI BREAKDOWN
Vol. 59. No. 1
12T
With Denudatin B
Collagen
PAF
Thrombin
Fig. 7. Inhibition of denudatin B on the formation of inositol hosphate in washed rabbit platelets caused by collagen, PAF and thrombin. [L? H]inositol-labelled platelets were incubated with thrombin (0.1 U/ml) in the absence of calcium or with PAF (2 ng/ml), collagen (10 pg/ml) in the presence of calcium (1 mM) for 6 min. Folds of increase of inositol phosphate (IP) were presented as meanstS.E. *: p < 0.05, ***: p < 0.001 as compared with the respective control. action, high concentration of denudatin B caused some inhibition on the aggregation induced by ADP, collagen, arachidonic acid and also low concentration of thrombin. PAF is one of the most potent inducers of platelet aggregation known in many animal species (18). PAF-induced aggregation is associated with release reaction and degranulation (19). Although thromboxane B formation was reported to be increased in human platelets (20), our an3 other data have found PAF-induced aggregation to be independent of ADP release or of cyclooxygenase products in rabbit platelets (10,21). Recent evidence has shown PAF activates phospholipase C. This leads to the formation of diacylglycerol and inositol polyphosphates, and finally causes the mobilization of calcium and release reaction (22). Most of the PAF antagonists introduced are very highly specific (except some phospholipid analogs) against PAF-induced platelet aggregation, without affecting aggregation caused by ADP, collagen or thrombin (5). Denudatin B (50 pg/ml) inhibited thrombin-induced aggregation, ATP release, thromboxane B2 formation and rise of intracellular calcium. All of these actions could be explained by its inhibition on the phosphoinositides breakdown as revealed by the suppression of inositol phosphate formation caused by thrombin. Denudatin B did not affect the arachidonate-prostaglandin endoperoxides-thromboxane transformation because it did not affect the thromboxane B2 formation in the mixture of platelet homogenate with arachidonic acid (data not shown). PAF-induced pharmacological actions can be antagonized by some calcium channel blockers. Diltiazem and verapamil inhibited aggregation by PAF and also inhibited PAF binding in a dose-dependent manner (23). This suggests
Vol. 59, No. 1
PAF ANTAGONIST ON PI BREAKDOWN
129
that PAF receptor may be closely associated with calcium channels. Our data in this paper also suggest PAF receptor may be closely associated with thrombinor collagen-induced signal transduction, especially phosphoinositides breakdown in the initial activation stage. ACKNOWLEDGEMENTS This work was supported by research grants of the National Science Council of the Republic of China (NSC79-0412-B002-26) and the National Institute of Chinese Medicine. REFERENCES 1. BRAQUET, P., TOUQUI, L., SHEN, T.Y. and VARGAFTIG, B.B. Perspectives in platelet-activating factor research. Pharmacol. -Rev. 39, 97-145, 1987. 2. VARGAFTIG, B.B., CHIGNARD, M., BENVENISTE, J., LEFORT, J. and WALL, F. Background and present status of research on platelet-activating factor (PAF-acether). --Ann. N.Y. Acad. Sci. 370, 119-137, 1981. 3. PAGE, C.P., ARCHER, C.B., PAUL, W. and MORLEY, J. Paf-acether: a mediator of inflammation and asthma. Trends Pharmacol. -_ Sci. 5, 239-241, 1984. 4. SHEN, T.Y., HWANG, S.B., CHANG, M.N., DOEBBER, T.W., LAM, M.H.T., WU, M.S., WANG, X., HAN, G.Q. and LI, R.Z. Characterization of a plateletactivating factor receptor antagonist isolated from haifenteng (Piper in vivo plateletfutokadsura): specific inhibition of in vitro and -activating factor-induced effects. ----ProcTNxAcad. Sci. USA 82, 672676, 1985. 5. BRAQUET, P. and GODFROID, J.J. PAF-acether specific binding sites: 2. Design of specific antagonists. Trends Pharmacol. Sci. 7, 397-403, 1986. M., LEFORT, J., MALANCHERE, E. and VARGAFTIG, B.B. 6. PRETOLANTI, Interference by the novel PAF-acether antagonist WEB 2086 with the bronchopulmonary responses to PAF-acether and to active and passive anaphylactic shock in guinea-pigs. -Eur. J. Pharmacol. 140, 311-321, 1987. 7. PAGE, C. and ABBOTT, A. PAF: new antagonists, new roles in disease and a major role in reproductive biology. Trends Pharmacol. Sci. 10, 256-257, 1989. 8. CHEN, C.C., HUANG, Y.L., CHEN, Y.P., HSU, H.Y. and KUO, Y.H. Three new neolignans, Fargesones A,B and C, from the flower buds of Magnolia fargesii. -CHEM. PHARM. BULL. 36, 1791-1795, 1988. 9. TENG, C.M., YU, S.M. and CHEN, C.C. Effects of PAF antagonists isolated from Magnolia fargesii on platelet aggregation and contraction of rat thoracic aorta. In: Taipei PAF Sattellite Symposium, Proceedings, N. Hicks (editor). Excerpta Medica Asia Ltd 1989 (in press). lO.TENG, C.M., CHEN, W.Y., KO, W.C. and OUYANG, C. Antiplatelet effect of butylidenephthalide. Biochim. Biophys. Acta 924, 375-382, 1987. ll.O'BRIEN, J.R. Platelet aggregation II, some results from a new method of J. Clin. Pathol. 2, 452-455, 1962. study. --
130
PAF ANTAGONIST ON PI BREAKDOWN
Vol. 59. No. 1
12.DELUCA, M. and McELORY, W.D. Purification and properties of firefly luciferase. Methods Enzymol. 57, 3-15, 1978. 13.RINK, T.J., SANCHEZ, A. and HALLAM, T.J. Diacylglycerol and phorbol ester stimulate secretion without raising cytoplasmic free calcium in human platelets. -Nature 305, 317-319, 1983. 14.TSIEN, R.Y., POZZAN, T. and RINK, T.J. Calcium homeostasis in intact lymphocytes: cytoplasmic free calcium monitored with a new, intracellularly trapped fluorescencent indicator. -J. Cell Biol. 94, 325334, 1982. The effect of lithium on platelet 15.HUANG, E.M. and DETWILER, T.C. phosphoinositide metabolism. Biochem. -J. 236, 895-901, 1986. Stimulation of l-adrenoceptors in rat 16. NEYLQN, C.B. and SUMMERS, R.J. Br. J kidney mediates increased inositol phospholipid hydrolysis. -2 Pharmacol. 91, 367-376, 1987. Pharmacology and applications of Chinese 17. CHANG, H.Y. and BUT, P.P.H. Materia Medica, pp 629-635, World Scient. Pub., Philadelphia, 1986. Species 18. NAMM, D.H., ANJANEYULU, S., TADEPALLI, S. and HIGH, J.A. specificity of the platelet response to l-O-alkyl-2-acetyl-sn-glycerophosphocholine. Thromb. Res. 25, 341-350, 1982. Effect of 19. CHESNEY, C.M., PIFER, D.D., BYERS, L.W. and MUIRHEAD, E.E. platelet-activating factor (PAF) in human platelets. Blood 59, 582-585, 1982. 20 McMANUS, L.M., HANAHAN, D.J. and PINCKARD, R.N. stimulation by acetylglyceryl ether phosphorylcholine. 67, 903-906, 1981.
Human platelet J. Clin. Invest. --
21. CAZENAVE, J.P., BENVENISTE, J. and MUSTARD, J.F. Aggregation of rabbit platelets by platelet activating factor is independent of the release reaction and the arachidonate pathway and inhibited by membrane active drugs. Lab. Invest. 4l, 275-285, 1979. 22. ETIENNE, A. and BAROGGI, N. Effects of ginkgolides on PAF-induced calcium mobilization in platelets. In: Ginkgolides - Chemistry, Biology, Pharmacology and Clinical Perspectives. P. Braquet (Ed.), pp. 115-125, J.R. Prous SC&e Publishers, 1988. 23,VALONE, F.H. Inhibition of platelet-activating factor binding to human platelet by calcium channel blockers. Thromb. -Res 45, 427-435, 1987.