Inhibition of platelet aggregation by unsaturated fatty acids through interference with a thromboxane-mediated process

cis- and trans-unsaturated fatty acids with 18 carbon atoms (oleic, linoleic, elaidic and linolelaidic acid) inhibited aggregation of washed rabbit platelets stimulated with collagen, arachidonic acid and U46619 when in the same concentration ranges. Thrombin-induced aggregation was not affected by any of them. Saturated fatty acid (stearic acid) had no effect on this response. The inhibition is independent of the induced change in membrane fluidity, since trans-isomers could not induce the change in fluorescence polarization of 1,6-diphenyl-1,3,5-hexatriene. Unsaturated fatty acids, except linoleic acid, did not interfere with the formation of thromboxane B2 from exogenously added arachidonic acid. All the unsaturated fatty acids only slightly inhibited the arachidonic acid liberation by phospholipase A2 in platelet lysate. This indicates that the unsaturated fatty acids may block a process after formation of thromboxane A2 in response to collagen and arachidonic acid. The increase in phosphatidic acid formation stimulated with U46619 was inhibited dose dependently by each of the unsaturated fatty acids but that stimulated with thrombin was not affected by any of them. Phospholipase C activity measured by diacylglycerol formation in unstimulated platelet lysate was not inhibited by the fatty acids. The elevation of cytosolic free Ca2+ induced by arachidonic acid or U46619 and Ca2+ influx by collagen were inhibited almost completely at the same concentration as that which inhibited their aggregation. These data suggest that the unsaturated fatty acids were intercalated into the membrane and inhibited collagen- and arachidonic acid-induced platelet aggregation by causing a significant suppression of the thromboxane A2-mediated increase in cytosolic free Ca2+, probably due to interference with the receptor-operated Ca2+ channel.     

Shape change induced in human platelets by platelet-activating factor. Correlation with the formation of phosphatidic acid and phosphorylation of a 40,000-dalton protein

Washed human platelets that have been separated from plasma in the presence of prostacyclin are activated by the addition of platelet activating factor (PAF). Activation (shape change, serotonin release, and aggregation) correlates closely with the formation of phosphatidic acid and the phosphorylation of a 40,000-dalton protein. Platelet shape change, formation of phosphatidic acid, and protein phosphorylation precede aggregation and are induced at lower concentrations of PAF than those required to induce release of serotonin and platelet aggregation. Platelet shape change, formation of phosphatidic acid, and protein phosphorylation induced by PAF are not affected by trifluoperazine or indomethacin. This indicates that these responses are independent of the liberation of arachidonic acid from platelet phospholipids and the metabolism of arachidonic acid via cyclooxygenase and lipoxygenase. These responses are, however, inhibited by prostacyclin. Platelet shape change is the first measurable physiologic response to platelet agonists and may be associated with the stimulation of phospholipase C, inducing formation of 1,2-diacylglycerol and its phosphorylated product, phosphatidic acid. Transient formation of 1,2-diacylglycerol may also induce the specific activation of the protein kinase C that phosphorylates a 40,000-dalton protein.     

Platelet desensitization by arachidonic acid is associated with the suppression of endoperoxide/thromboxane A2 binding to the membrane receptor

The inhibitory mechanism of high levels of exogenously added arachidonic acid on activation of washed human platelets was investigated. While low levels of arachidonic acid (5-10 microM) induced aggregation, ATP secretion and increase in cytoplasmic free Ca2+ concentration (first phase of activation), these platelet responses did not occur significantly at high concentrations (30-50 microM). However, much higher concentrations than 80 microM again elicited these responses (second phase). The first phase of platelet activation was inhibited by cyclooxygenase inhibitor, indomethacin, whereas the second one was independent of such treatment. Thromboxane B2 was produced dose-dependently until reaching a plateau at arachidonic acid concentrations higher than 20 microM, irrespective of the lack of aggregation and secretion at high concentrations. After that the amount of free arachidonic acid which remained unmetabolized in platelets gradually increased. High concentrations of arachidonic acid as well as other polyunsaturated fatty acids caused desensitization of platelets in response to U46619, and also depressed the specific [3H]U46619-binding to the receptor as well as other polyunsaturated fatty acids. The amount free arachidonic acid needed in platelets to suppress [3H]U46619 binding corresponded to that needed to inhibit platelet aggregation. Furthermore, arachidonic acid dose-dependently induced fluidization of lipid phase of platelet membranes as detected by 1,6-diphenyl-1,3,5-hexatriene. These results suggest that the inhibition of platelet response by high levels of arachidonic acid can be attributed to interference with endoperoxide/thromboxane A2 binding to the receptor, probably due to perturbation of the membrane lipid phase due to excess amounts of free arachidonic acid remaining in the membranes.     

Irreversible platelet aggregation does not depend on lipoxygenase metabolites

Previous investigations in our laboratory demonstrated the existence of an intrinsic mechanism, termed membrane modulation, capable of restoring sensitivity to aspirin treated platelets, resulting in irreversible aggregation in response to arachidonic acid (AA). The mechanism underlying correction of aspirin induced inhibition of platelet function, however, was not clear. In the present study we have evaluated the role of lipoxygenase (LO) metabolites of AA in securing irreversible aggregation of drug induced cyclooxygenase (CO) deficient platelets. Platelets treated with aspirin or Ibuprofen did not convert radiolabeled AA to thromboxane, but generated significant quantities of hydroxy acids via the LO pathway. However, drug exposed platelets, when stirred with epinephrine first and then challenged with AA, aggregated irreversibly. Eicosatetraynoic acid (ETYA 1, U53119) inhibited AA conversion by the LO pathway, whereas 5,8,11,14-eicosatetraynoic acid (ETYA 2) inhibited AA conversion by both CO and LO enzymes. Yet, at the inhibitory concentration these fatty acids failed to prevent AA induced irreversible aggregation of CO deficient, alpha adrenergic receptor stimulated platelets. Results of four studies show that the generation of LO metabolites of AA are not essential for securing irreversible aggregation of platelets.     

The role of phospholipase C in platelet responses

Degradation of inositides induced by phospholipase C in activated platelets leads to the formation of 1,2-diacylglycerol (1,2-DG) and its phosphorylated product, phosphatidic acid (PA). We have studied the relationship between activation of phospholipase C and the appearance of specific platelet responses, such as phosphorylation of proteins, shape change, release reaction and aggregation induced by different stimuli such as thrombin, platelet-activating factor, collagen, arachidonic acid (AA) and dihomogamma linolenic acid. A low degree of platelet activation induces only shape change which is associated with partial activation of phospholipase C (formation of phosphatidic acid), and phosphorylation of both a 40K molecular weight protein (protein kinase C activation) and a 20K molecular weight protein (myosin light chain). A higher degree of platelet activation induces aggregation, release of serotonin and a higher level of phospholipase C and protein kinase C activities. Metabolism of AA occurs concomitantly to aggregation and serotonin release, but AA metabolites are not related to the shape change of human platelets. Platelet shape change and the initial activation of phospholipase C induced by thrombin or platelet-activating factor is independent of the metabolites derived from cyclo-oxygenase activity. Further activation of phospholipase C which occurs during platelet aggregation and release reaction is, however, partly dependent on cyclo-oxygenase metabolites.     

Effect of unsaturated fatty acids and Ca2+ on phosphatidylinositol synthesis and breakdown

CDP-diglyceride : inositol transferase was inhibited by unsaturated fatty acids. The inhibitory activity decreased in the following order: arachidonic acid greater than linolenic acid greater than linoleic acid greater than oleic acid greater than or equal to palmitoleic acid. Saturated fatty acids such as myristic acid, palmitic acid, and stearic acid had no effect. Calcium ion also inhibited the activity of CDP-diglyceride : inositol transferase. In rat hepatocytes, arachidonic acid inhibited 32P incorporation into phosphatidylinositol and phosphatidic acid without any significant effect on 32P incorporation into phosphatidylcholine, phosphatidylethanolamine and phosphatidylserine. Ca2+ ionophore A23187 also inhibited 32P incorporation into phosphatidylinositol. However, 32P incorporation into phosphatidic acid was stimulated with Ca2+ ionophore A23187. Phosphatidylinositol-specific phospholipase C was activated by unsaturated fatty acids. Polyunsaturated fatty acids such as arachidonic acid and linolenic acid had a stronger effect than di- and monounsaturated fatty acids. Saturated fatty acids had no effect on the phospholipase C activity. The phospholipase C required Ca2+ for activity. Arachidonic acid and Ca2+ had synergistic effects. These results suggest the reciprocal regulation of phosphatidylinositol synthesis and breakdown by unsaturated fatty acids and Ca2+.     

Antiplatelet effect of butylidenephthalide

Butylidenephthalide inhibited, in a dose-dependent manner, the aggregation and release reaction of washed rabbit platelets induced by collagen and arachidonic acid. Butylidenephthalide also inhibited slightly the platelet aggregation induced by PAF and ADP, but not that by thrombin or ionophore A23187. Thromboxane B2 formation caused by collagen, arachidonic acid, thrombin and ionophore A23187 was in each case markedly inhibited by butylidenephthalide. Butylidenephthalide inhibited the aggregation of ADP-refractory platelets, thrombin-degranulated platelets, chymotrypsin-treated platelets and platelets in the presence of creatine phosphate/creatine phosphokinase. Its inhibition of collagen-induced aggregation was more marked at lower Ca2+ concentrations in the medium. The aggregability of platelets inhibited by butylidenephthalide could be recovered after the washing of platelets. In human platelet-rich plasma, butylidenephthalide and indomethacin prevented the secondary aggregation and blocked ATP release from platelets induced by epinephrine. Prostaglandin E2 formed by the incubation of guinea-pig lung homogenate with arachidonic acid could be inhibited by butylidenephthalide, indomethacin and aspirin. It is concluded that the antiplatelet effect of butylidenephthalide is mainly due to an inhibitory effect on cyclo-oxygenase and may be due partly to interference with calcium mobilization.     

Inhibitory mechanism of cis-polyunsaturated fatty acids on platelet aggregation: the relation with their effects on Ca2+ mobilization, cyclic AMP levels and membrane fluidity

The in vitro inhibitory effects of cis-polyunsaturated fatty acids, linolenic (18:2 delta 9,12), alpha-linoleic (18:3 delta 9,12,15) and eicosatrienoic (20:3 delta 11,14,17) acid, on bovine platelet aggregation and their inhibitory mechanism were investigated. These fatty acids inhibited platelet aggregation induced by ADP and thrombin to similar extent. Fluorescence analyses with fura-2-loaded platelets showed that, in the concentration ranges that inhibited aggregation, they also inhibited agonist-induced increase in cytoplasmic Ca2+. According to radioimmunoassay study, addition of these fatty acids increased cyclic AMP contents in the presence of theophylline corresponded with their inhibitory effects on aggregation. These fatty acids induced a 1.6-1.8-fold increase over basal concentration of cyclic AMP in the concentration ranges that fully inhibited aggregation. On the other hand, saturated fatty acid, stearic acid, affected neither aggregation nor cyclic AMP levels. As reported previously [1985) Biochim. Biophys. Acta 818, 391), these unsaturated fatty acids induced increase in membrane fluidity in the same concentration range. These results suggest that inhibition of platelet aggregation by cis-polyunsaturated fatty acids is due to the increase in cyclic AMP levels. This increase seems to be due to stimulation of adenylate cyclase which is mediated by membrane perturbation.     

Activation of phospholipases in platelets by polyclonal antibodies against a surface membrane protein.

In a previous paper we demonstrated using immunochemical techniques that propolypeptide of von Willebrand factor was present on the surface of resting platelets. In the present paper we show that polyclonal antibodies against propolypeptide of von Willebrand factor induce activation of phospholipase(s) in platelets and lead to platelet aggregation. The antibody-stimulation of platelets induced the synthesis of thromboxane A2 (TXA2). Furthermore, the aggregation was inhibited by aspirin and an antagonist of TXA2. Aspirin inhibited not only the aggregation but also the activation of arachidonic acid liberation from phospholipids, but the effect of aspirin on arachidonic acid liberation was cancelled by the combined effect of the antibodies and a TXA2 mimetic agonist, which itself did not activate arachidonic acid liberation. The antibody-induced activation of arachidonic acid liberation and the aggregation were blocked by cytochalasin B. All these results obtained with antibodies were quite similar to the results obtained with collagen.     

Inhibition of platelet aggregation and arachidonate metabolism in platelets by procyanidins

The effects of procyanidins on platelet aggregation and arachidonate metabolism in platelets were studied. Nine procyanidins were used in this investigation. Procyanidins B-2-S, EEC and C-1 significantly induced the inhibition of platelet aggregation, and the potency of inhibition was comparable with aspirin. Procyanidin B-2-S was used as a representative of procyanidins for further studies on the effect on arachidonate metabolism. In arachidonate metabolism by fatty acid cyclooxygenase pathway, B-2-S inhibited TXB2 and HHT formation by intact platelets treated with exogenous arachidonic acid. It also inhibited TXB2 formation measured by a specific radioimmunoassay when the cells were challenged with calcium ionophore A23187. In cell-free system, B-2-S inhibited both TXB2 and 12-HETE bioxynthesis in platelet microsome and cytosol, respectively. The inhibitory effect on thromboxane biosynthesis might explain the inhibitory effect of procyanidins on platelet aggregation.     

Antiplatelet effects of clausine-D isolated from Clausena excavata

Clausine-D inhibited concentration-dependently the aggregation and release of washed rabbit platelets induced by arachidonic acid and collagen, without affecting those induced by U46619, PAF and thrombin. The IC₅₀ values of clausine-D on arachidonic acid-and collagen-induced platelet aggregation were calculated to be 9.0±1.1 and 58.9±0.9 μM, respectively. Thromboxane B₂ and prostaglandin D₂ formation in platelets caused by arachidonic acid were also suppressed. Clausine-D inhibited increased intracellular concentration of calcium in platelets caused by arachidonic acid and collagen, and also abolished the generation of inositol monophosphate caused by arachidonic acid, but not that by collagen U46619, PAF and thrombin. In human citrated platelet-rich plasma, clausine-D inhibited the secondary phase, but not the primary phase, of aggregation induced by epinephrine and ADP. These results indicate that the antiplatelet effect of clausine-D is due to inhibition of the formation of thromboxane A₂.     

Effect of docosahexaenoic acid (DHA) on arachidonic acid metabolism and platelet function

Studies from our laboratory have suggested a role for ferrous iron in the metabolism of arachidonic acid and demonstrated that inhibitors of prostaglandin synthesis exert their effect by complexing with the heme group of cyclooxygenase. Docosahexaenoic acid (DHA) is a potent competitive inhibitor of arachidonic acid metabolism by sheep vesicular gland prostaglandin synthetase. In this study we have evaluated the effect of exogenously added DHA on platelet function and arachidonic acid metabolism. DHA at 150 microM concentration inhibited aggregation of platelets to 450 microM arachidonic acid. At this concentration DHA also inhibited the second wave of the platelet response to the action of agonists such as epinephrine, adenosine diphosphate and thrombin. Inhibition induced by this fatty acid could be overcome by the agonists at higher concentrations. DHA inhibited the conversion of labeled arachidonic acid to thromboxane by intact, washed platelet suspensions. However, platelets in plasma incubated first with DHA then washed and stirred with labeled arachidonate generated as much thromboxane as control platelets. These results suggest that the polyenoic acids, if released in sufficient quantities in the vicinity of cyclooxygenase, could effectively compete for the heme site and inhibit the conversion of arachidonic acid.     

Leupeptin selectively inhibits human platelet responses induced by thrombin and trypsin; a role for proteolytic activation of phospholipase C

Thrombin and trypsin induce serotonin release and aggregation in human platelets. Both proteases induce activation of phospholipase C as reflected by formation of inositol phosphates and phosphorylation of the resultant 1,2-diacylglycerol to phosphatidic acid. Also, thrombin and trypsin activate protein kinase C and myosin light chain kinase as indicated, respectively, by phosphorylation of the 40,000 and 20,000 dalton proteins. Leupeptin, a known inhibitor of serine proteases, blocks all the observed responses of human platelets to trypsin and thrombin. Leupeptin does not inhibit serotonin release and aggregation induced by other platelet stimuli such as collagen, platelet-activating factor, ionophore A23187, and arachidonic acid. The implication of a proteolytic-mediated pathway in the transmembrane signalling involved in platelet activation is discussed.     

Potential involvement of vicinal sulfhydryls in stimulus-induced rabbit platelet activation

The potential involvement of vicinal dithiols in the expression of platelet-activating factor (AGEPC)- and A23187-induced alterations in rabbit platelets was explored through the use of phenylarsine oxide (PhAsO) and certain analogous derivatives. PhAsO (As3+) but not phenylarsonic acid (As5+) inhibited markedly at 1 microM concentration the release of arachidonic acid initiated by AGEPC and the ionophore A23187. In contrast, AGEPC-induced phosphatidic acid formation, phosphorylation of 40- and 20-kDa proteins, and Ca2+ uptake from external medium were not inhibited substantially by 1 microM PhAsO. However, these latter metabolic responses to AGEPC were inhibited by PhAsO at higher doses (10 microM). AGEPC- and thrombin-induced platelet aggregation and serotonin secretion also were prevented by PhAsO. The IC50 value of PhAsO was 2.7 +/- 1.2 microM toward AGEPC (5 X 10(-10) M)-induced serotonin release. Further, ATP and cAMP levels in PhAsO-treated platelets were not changed from controls. Interestingly, addition of Ca2+ to platelet sonicates (prepared in EDTA) caused diacylglycerol production and free arachidonic acid formation, even in the presence of 133 microM PhAsO. This would suggest that in the intact platelets PhAsO acted indirectly on phospholipase A2 and/or phospholipase C activities. Finally, a dithiol compound, 2,3-dimercaptopropanol, reversed the inhibition of platelet aggregation and arachidonic acid release effected by PhAsO. On the other hand, a monothiol compound, 2-mercaptoethanol, was not effective in preventing or in reversing the action of PhAsO. These observations suggest that vicinal sulfhydryl residues may be involved in stimulus-induced platelet activation.     

Activation of bovine platelets induced by long-chain unsaturated fatty acids at just below their lytic concentrations,and its mechanism

The effects of long chain unsaturated fatty acids such as linoleic acid on bovine platelets were examined. Not only linoleic acid, but also oleic and linolenic acid, at just below the concentrations causing marked cell lysis, induced an absorbance decrease of the platelet suspension in the presence of Ca²⁺. Since this absorbance decrease was reversed by the addition of EDTA and moreover aggregate formation was found by macroscopic and microscopic observation, it was concluded that unsaturated fatty acids at just below their lytic concentrations caused platelet aggregation. Unsaturated fatty acids also caused release of adenine nucleotides, but there was a lag time between the release and the aggregation, just as with ADP-induced release, suggesting that the aggregation was independent of the release of ADP. It was revealed that this activation of platelets by unsaturated fatty acids was caused by marked Ca²⁺ uptake into the cytoplasm, resulting from significant membrane perturbation.     

Stimulation of arachidonic acid metabolism via phospholipase A2 by triethyl lead

Human blood platelet aggregation and the formation of icosanoids were studied in response to triethyl lead chloride (Et3PbCl). Concentrations higher than 75 microM stimulate platelets to aggregate, whereas low concentrations (less than or equal to 20 microM) caused platelet hypersensitivity to aggregating agents such as collagen or arachidonic acid. Incubation of suspensions of washed platelets with Et3PbCl resulted in a stimulated liberation and subsequent metabolism of arachidonic acid. This response was dependent on the concentration of Et3PbCl and the incubation time. Using low concentrations of Et3PbCl and up to 3 h of incubation, the lipoxygenase product 12-hydroxy-5,8,10,14-icosatetraenoic acid was the major metabolite. Under normal conditions, however, stimulation of platelets with collagen, thrombin, or arachidonic acid leads to higher amounts of the cyclooxygenase products 12-hydroxy-5,8,10-heptadecatrienoic acid and thromboxane B2. The aggregation of human platelets induced by Et3PbCl was inhibited by three different drugs: acetylsalicylic acid, forskolin and quinacrine; but only quinacrine could prevent the liberation of arachidonic acid and the appearance of its metabolites. These specific effects of the inhibitors on Et3PbCl-stimulated platelets as well as the differences in the pattern of arachidonic acid metabolites and phosphatidic acid suggest a direct stimulatory action of Et3PbCl on platelet phospholipase A2.     

Cyclic nucleotides and platelet aggregation. Effect of aggregating agents on the activity of cyclic nucleotide-metabolizing enzymes.

The activities of adenylate and guanylate cyclase and cyclic nucleotide 3':5'-phosphodiesterase were determined during the aggregation of human blood platelets with thrombin, ADP, arachidonic acid and epinephrine. The activity of guanylate cyclase is altered to a much larger degree than adenylate cyclase, while cyclic nucleotide phosphodiesterease activity remains unchanged. During the early phases of thrombin-and ADP-induced platelet aggregation a marked activation of the guanylate cyclase occurs whereas aggregation induced by arachidonic acid or epinephrine results in a rapid diminution of this activity. In all four cases, the adenylate cyclase activity is only slightly decreased when examined under identical conditions. Platelet aggregation induced by a wide variety of aggregating agents including collagen and platelet isoantibodies results in the "release" of only small amounts (1-3%) of guanylate cyclase and cyclic nucleotide phosphodiesterase and no adenylate cyclase. The guanylate cyclase and cyclic nucleotide phosphodiesterase activities are associated almost entirely with the soluble cytoplasmic fraction of the platelet, while the adenylate cyclase if found exclusively in a membrane bound form. ADP and epinephrine moderately inhibit guanylate and adenylate cyclase in subcellular preparations, while arachidonic and other unsaturated fatty acids moderately stimulate (2-4-fold) the former. It is concluded that (1) the activity of platelet guanylate cyclase during aggregation depends on the nature and mode of action of the inducing agent, (2) the activity of the membrnae adenylate cyclase during aggregation is independent of the aggregating agent and is associated with a reduction of activity and (3) cyclic nucleotide phosphodiesterase remains unchanged during the process of platelet aggregation and release. Furthermore, these observations suggest a role for unsaturated fatty acids in the control of intracellular cyclic GMP levels.     

Fatty acids bound to serum albumin potentiate platelet prostaglandin biosynthesis.

Utilization of arachidonic acid by human platelets is increased when the fatty acid content of serum albumin is increased as well. Platelet aggregation induced by arachidonic acid and by low concentrations of thrombin is thus potentiated, suggesting that platelet responsiveness to aggregating agents is influenced by the plasma content in free fatty acids.     

Relative effects of flurbiprofen on platelet 12-hydroxy-eicosatetraenoic acid and thromboxane A2 production: influence on collagen-induced platelet aggregation and adhesion

Flurbiprofen has been shown to inhibit cyclo-oxygenase metabolism of arachidonic acid to thromboxane A2 (TxA2), resulting in the inhibition of platelet aggregation. Recently, our laboratory reported that the "irreversible" phase of platelet aggregation and adhesion were regulated, in part, by the lipoxygenase metabolism of arachidonic acid to 12-hydroxy-eicosatetraenoic acid (12-HETE) in platelets, and that selective inhibition of one enzyme i.e. either cyclo-oxygenase or lipoxygenase, resulted in paradoxical effects on the metabolism of arachidonic acid and platelet response related to the other pathway. Therefore, we performed experiments to assess the relative effects of flurbiprofen on TxA2 and 12-HETE synthesis, and on collagen-induced platelet aggregation and platelet adhesion to collagen-coated surfaces. "Irreversible" collagen-induced platelet aggregation was only partially inhibited by pre-incubation with 1 x 10(-6) M flurbiprofen, while TxA2 production was elevated and 12-HETE production was maximally inhibited in these platelets. At this concentration of flurbiprofen (1 x 10(-6)M), collagen-induced platelet adhesion was also reduced by 50%. At higher concentrations of flurbiprofen, both platelet aggregation and adhesion were further reduced, with a corresponding inhibition of TxA2 production. Thus it appears that the lipoxygenase pathway of arachidonic acid metabolism in platelets is not only inhibited by flurbiprofen, but is more sensitive to inhibition by flurbiprofen than the cyclo-oxygenase pathway. This differential effect of flurbiprofen on arachidonic acid metabolism in the platelet is related to differential effects on platelet function.     

Inhibitory effect of sulfur-containing compounds in Scorodocarpus borneensis Becc. on the aggregation of rabbit platelets

The inhibitory effects of three pure compounds isolated from wood garlic, 2,4,5-trithiahexane (I), 2,4,5,7-tetrathiaoctane (II), and 2,4,5,7-tetrathiaoctane 2,2-dioxide (III), on rabbit platelet aggregation induced by collagen, arachidonic acid, U46619, ADP (adenosine 5'-diphosphate), PAF (platelet aggregating factor), and thrombin were studied in vitro. The anti-aggregating activity of 2,4,5,7-tetrathiaoctane 4,4-dioxide (IV) was also measured with collagen and arachidonic acid. I, II, III, and IV inhibited the platelet aggregation induced by all tested agonists. I, II, and III exhibited a stronger inhibitory effect against the thrombin-induced aggregation of GFP (gel-filtered platelets) than against the aggregation induced by the other agonists. Notably, the IC50 value for III was 4 microM, which is approximately 2.5 times stronger than MATS (methyl allyl trisulfide), a major anti-platelet compound isolated from garlic. In inhibiting collagen-induced aggregation, II was as potent as MATS and aspirin, with a marked disaggregation effect on the secondary aggregation by arachidonic acid, at the rate of 47.05%/min at a concentration of 10(-4) M. I, II, and III also suppressed U46619-induced aggregation. These results suggest that sulfur-containing compounds in wood garlic not only inhibit arachidonic acid metabolism but also suppress aggregation in association with the function of the platelet plasma membrane.