Inhibition of Human Platelet Aggregation by Gangliosides

Thrombosis Research 98 (2000) 51–57

REGULAR ARTICLE

Inhibition of Human Platelet Aggregation by Gangliosides Hugo A. Guglielmone1, Jose´ J. Daniele2, Ismael D. Bianco2,3, Eduardo J. Fernandez1 and Gerardo D. Fidelio2 Laboratorio de Ana´lisis Clı´nicos Especializados (L.A.C.E.); 2 Departamento de Quı´mica Biolo´gica—CIQUIBIC—Facultad de Ciencias Quı´micas, Universidad Nacional de Co´rdoba, Ciudad Universitaria; and 3CEPROCOR, 5000 Co´rdoba, Argentina.

1

(Received 25 February 1999 by Editor J. Aznar; revised/accepted 21 October 1999)

Abstract The content and composition of gangliosides is modified upon platelet stimulation, suggesting that these lipids may play functional roles in platelet physiology. Therefore, the effect of exogenously added gangliosides on human platelet aggregation was evaluated. The pretreatment of platelets with a mixture of total gangliosides from bovine brain and a series of purified mono-, di- and tri-sialogangliosides partially inhibit the collagen-induced aggregation process and ATP release and completely block the generation of the second aggregation wave when ADP is used as agonist. The inhibition was exerted at around 100 ␮M by GTOT as well as purified GM1, GM3, GD1a, and GT1b gangliosides, whereas asialoGM1 and sulphatide did not show a significant influence on platelet aggregation. Thrombin, Ca2⫹ ionophores (A23187 and Ionomycin), arachidonic acid, and U46619 were unable to bypass the inhibitory effect exerted by gangliosides, suggesting that gangliosides inhibit platelet aggregaAbbreviations: AA, Arachidonic Acid; PLC, phospholipase C; PLA2, phospholipase A2; GTOT, total bovine brain gangliosides; ATP, adenosine triphosphate; PRP, platelet rich plasma; PPP, platelet poor plasma; TXA2, thromboxane A2; PMA, phorbol myristate acetate; PKC, protein kinase C; the designation of individual gangliosides follows the recommendations of the IUPACIUB Commision on Biochemical Nomenclature. Corresponding author: Dr. Gerardo D. Fidelio, Departamento de Quı´mica Biolo´gica, Facultad de Ciencias Quı´micas, Universidad Nacional de Co´rdoba, Ciudad Universitaria C.C. 61, A.P. 4, 5000 Co´rdoba, Argentina. Tel: ⫹54 (351) 433 4168 /433 4171; Fax: ⫹54 (351) 433 4074; E-mail: ⬍[email protected]⬎.

tion by inhibiting the synthesis or action of prostaglandins. Gangliosides inhibited U46619-induced aggregation, thus suggesting that they block the action of thromboxane A2. Epinephrine induces a partial aggregation on gangliosides-treated platelets, similar to fluoroaluminate and phorbol myristate acetate, indicating that these platelets are still functional. To summarize, these results indicate that the major pathway(s), but not all, driving to the aggregation process following the interaction of ligand-receptor may be blocked by pretreatment of human platelets with gangliosides.  2000 Elsevier Science Ltd. All rights reserved. Key Words: Gangliosides; Platelet aggregation; Phospholipases; Signal transduction

G

angliosides are complex glycosphingolipids with sialic acid as a component of the sugar backbone of their polar head groups. These lipids are found in the outer layer of cell membranes, being particularly abundant in the nervous system. Different roles have been assigned to gangliosides in differentiation, adhesion, modulation of immunological response, neuron synaptic transmission, and induction of neuritogenesis (see [1–3] for reviews). These lipids may modulate the ability of the cell to modify its response to signals from the microenvironment. Based on differences in ganglioside composition associated with changes in the activation state of platelets, these lipids have been suggested to play functional roles in platelet

0049-3848/00 $–see front matter  2000 Elsevier Science Ltd. All rights reserved. PII S0049-3848(99)00208-X

52

H.A. Guglielmone et al./Thrombosis Research 98 (2000) 51–57

physiology [4]. However, the role of gangliosides on platelet function is not clear. GD3 ganglioside adsorbed to plastic stimulates platelet adhesion and activation [5]. It has also been reported that gangliosides isolated from tumours enhance platelet activation, GD2 being the main individual specie responsible for this effect [6]. In addition, GM1 and GM3 gangliosides have been shown to induce platelet shape changes by themselves and elicited aggregation in combination with epinephrine [7]. However, platelet adhesion to fibronectin, fibrinogen, and von Willebrand Factor substrates has been inhibited using complex gangliosides [8]. Similar inhibitory effect on platelet aggregation was described using a modified lyso ganglioside (de-Nacetyl-lyso-GM1) [9]. It has also been shown that gangliosides can inhibit the adhesion of thrombinactivated platelets to polystyrene surfaces modulating the platelet integrin receptor glycoprotein IIb-IIIa [10]. When platelets are activated, several phospholipase activities are triggered and the intracellular Ca2⫹ level increases previous to aggregation and secretion of granules [11]. In this connection, gangliosides have shown a potent inhibitory effect on the enzymatic activities of both phospholipase C (PLC) and phospholipase A2 (PLA2) on different artificial model membrane systems as substrates [12–15]. Additionally, it has been reported that gangliosides can inhibit agonist-induced arachidonic acid release and show an anti-inflammatory effect on in vitro and in vivo models [16–18]. Studies on the effect of gangliosides as Ca2⫹ flux modulators showed that gangliosides increase the intracellular Ca2⫹ level in several cell types and model membrane systems, including platelets [7,19,20]. In the present report, we studied and compared the effect of a series of chemically related glycosphingolipids on platelet aggregation and ATP release induced by several agonists. The present study provides evidence that exogenously incorporated gangliosides may inhibit human platelet aggregation by interfering with the major ligandreceptor-mediated activation processes.

Argentina) and correspond to 21% GM1, 42% GD1a, 18% GD1b, and 19% GT1b (average molecular weight 1800). Glycosphingolipids were purified as previously described [12]. All the ganglioside preparations used were over 98% on the base of dry weight and sialic acid content. Firefly luciferin/luciferase, ADP, collagen, epinephrine, arachidonic acid, and thrombin were obtained from ChronoLog Corp. (Havertown, PA, USA). U46619 (a stable analogue of prostaglandin H2) was obtained from The Upjohn Co. (Kalamazoo, MI, USA). A23187, Ionomycin, and adenosine triphosphate (ATP) standard were purchased from Sigma Chem. Co. (St. Louis, MO, USA); all the other reagents were analytical grade. Tyrode’s buffer (pH 7.4) containing 136 mM NaCl, 11.9 mM NaHCO3, 2.65 mM KCl, 0.42 mM NaH2PO4, 5 mM dextrose, and 2 mM MgCl2 was supplemented with 3.5 mg/ml bovine serum albumin.

1.2. Platelet Preparation

1. Materials and Methods

Blood from healthy volunteers, who denied having taken any medication in the previous 2 weeks, was drawn into a plastic syringe and mixed with 3.8%buffered sodium citrate anticoagulant, pH 6.5 (blood:anticoagulant, 9:1 v/v). Platelet rich plasma (PRP) was obtained by centrifugation at 135 g for 20 minutes at room temperature. Platelet poor plasma (PPP) was prepared by centrifugation of the residual blood at 1500 g for 20 minutes at room temperature. PRP was always diluted to give 300,000 platelet/mm3 using autologous PPP for the assays performed. Washed human platelets were prepared as previously described [21]. Briefly, PRP was centrifuged at 800 g for 10 minutes at room temperature. Platelet pellet was resuspended in Tyrode’s solution plus 1 mM EDTA, 5 mM HEPES, pH 6.5. This washing procedure was repeated three times. Platelets were finally resuspended in Tyrode’s solution pH 7.4, counted in a Cell-Dyn Counter Model 1700 (Abbott, Chicago, IL, USA) and adjusted to 300,000 platelet/mm3 for aggregation studies. Platelets were always used within 2 to 3 hours after PRP preparation.

1.1. Materials

1.3. Platelet Aggregation and ATP Release

Total gangliosides from bovine brain (GTOT) were purchased from Laboratorios Beta (Buenos Aires,

Aggregation was measured with the turbidimetric method described by Born [22]. PRP (0.475 ml)

H.A. Guglielmone et al./Thrombosis Research 98 (2000) 51–57

53

was pre-incubated in silanized glass cuvettes with the indicated concentration of gangliosides at 37⬚C with stirring (900 rpm). After 15 minutes of preincubation, 25 ␮l of firefly luciferin/luciferase was added, and 1 minute later, the indicated agonist. When arachidonic acid, A23187, Ionomycin, or PMA were used as agonists, PRP pre-incubated with gangliosides was washed twice with Tyrode’s buffer, and the platelets count was adjusted to 300,000/mm3 with autologous PPP. Aggregation was monitored as an increase in light transmittance and release of platelet ATP measured in a Lumiaggregometer Model 540 (Chrono-Log, Havertown, PA, USA). At the peak of ATP secretion of luminescent intensity, 2 nM ATP was added to calibrate the amount of released ATP. Each experiment performed in triplicate was repeated at least twice and SE values of the averages were below 10%.

2. Results 2.1. Gangliosides Inhibit the Aggregation Process on Human Platelets A mixture of GTOT was used to evaluate the effect of glycosphingolipids on platelet aggregation. GTOT inhibited the aggregation process induced either by collagen or ADP, selected as strong and weak agonists, respectively (Figure 1). GTOT did not induce morphological changes in platelets by themselves, as detected by a change in light transmittance. The pre-incubation of platelets with GTOT partially inhibited the aggregation induced by collagen and avoided the generation of the second wave on ADP-induced aggregation process. The ATP release was also affected in gangliosidestreated platelets (see Figure 1). The effect was observed around 100 ␮M ganglioside concentration (Figure 2). On this basis, 100 ␮M was selected for the comparative analysis of the effects of individual glycosphingolipids. Results obtained with the different lipids on both ADP- and collagen-induced platelet aggregation are summarised in Table 1. All of the gangliosides tested were able to avoid the generation of the second wave of aggregation when ADP was used as agonist and partially inhibit the collagen-induced aggregation. The uncharged glycosphingolipid Gg4Cer (asialo GM1), which has

Fig. 1. Data recorded directly from the aggregometer showing the inhibitory effect of GTOT on aggregation and ATP release induced by ADP and collagen. Platelets were preincubated for 15 minutes with 100 ␮M GTOT and then exposed to the agonist. (Upper panel) Aggregation (a, b) and ADP release (c, d) induced by 1 ␮g/ml collagen, in the presence (b, d) and absence (a, c) of GTOT. (Lower panel) Aggregation (e, f) and ADP release (c, d) induced by 2.5 ␮M ADP in the presence (f, h) and absence (e, g) of GTOT.

a similar sugar backbone than does GM1 but lacks the sialic acid residue, had no effect on platelet aggregation. The fact that the negatively charged sphingolipid sulphatide had no effect on the aggregation process rules out a simple inhibitory effect exerted by the incorporation of negative charges onto the surface of platelets.

2.2. Effect of Gangliosides on the Aggregation Induced by Different Agonists In order to obtain some evidence about the mechanism involved in the inhibition observed, different

54

H.A. Guglielmone et al./Thrombosis Research 98 (2000) 51–57

Fig. 2. Dosis-response curve of GTOT. Human platelets were pre-incubated with the indicated concentration of GTOT by 15 minutes and subsequently stimulated with 1 ␮g/ml collagen (䊊) or 2.5 ␮M ADP (䊉) as indicated in Figure 1.

agonists were used to study the aggregation process of ganglioside-pretreated platelets (Table 2). In order to avoid the unspecific partition of hydrophobic agonists into ganglioside micelles, these assays were performed with washed human pretreated platelets. In these conditions, PRP was mixed with 100 ␮M GTOT for 15 minutes; then the platelets were washed and exposed to the indicated agonist in a gangliosides-free medium. Neither arachidonic acid (activating by TXA2-dependent mechanism), U46619 (a prostaglandin analogue), calcium ionophores (A23187 and ionomycin) (activating by TXA2-independent mechanism), nor thrombin

Table 1. Effect of individual gangliosides on platelet aggregation Ganglioside (100 ␮M) Saline solution Total gangliosides GM1 GD1a GT1b GM3 Gg4Cer Sulphatide

Platelet aggregation (%) ADP (second wave, 2.5 ␮M)

Collagen (1 ␮g/ml)

100 0 0 0 0 0 83 93

100 65 62 69 58 58 93 93

PRP was preincubated with 100 ␮M of the indicated ganglioside and 15 minutes later stimulated with ADP or collagen. The percentage of ADP-induced aggregation corresponds to the second wave. The results represent the mean of at least two determinations performed in triplicate.

was able to bypass the inhibitory effect exerted by gangliosides (see Table 2). Collagen and epinephrine induced a partial aggregation, similar to fluoroaluminate (a trimmeric G-protein activator) and PMA (a phorbol ester used as a diacylglicerol analogue) (see Table 2). The results observed with ATP release from platelets, used to measure the fusion of dense granules and therefore the releasing of the content to the media, were in correspondence with the effect of gangliosides on the aggregation process (see Table 2). Agonists that induce a high ATP release, as calcium ionophores, were completely blocked by gangliosides.

3. Discussion In this study, it is shown that the pretreatment of platelets with a mixture of bovine brain gangliosides inhibits the major pathways driving to the aggregation process. Gangliosides have been shown to have neuritogenic and neuronotrophic activities, and to facilitate repair of neuronal tissue after mechanical, biochemical, or toxic injuries. The content and composition of gangliosides significantly change during platelet differentiation and maturation [23,24]. Even more, exogenous gangliosides have been shown to induce megakaryocyte differentiation and maturation [23]. In an apparent contradiction, exogenous gangliosides have been shown to enhance and inhibit platelet activation and adhesion [5–8,25,26]. The results of the different groups could be reconciled, at least in part, if the experimental designs are carefully analysed. So, gangliosides have been reported to play a role in platelet shape change and aggregation based on the effect of a co-incubation with the agonist or their addition a short time after the stimulation [6,7,25]. It is well known that gangliosides induce a rapid and transient increase in the intracellular Ca2⫹ level in platelets [19]. It is likely that this increment could act sinergically with the agonist, overcoming other effects that gangliosides could potentially have on platelet function. In our experimental design, gangliosides were added several minutes before the agonist stimulation in order to simulate a change in platelet membrane ganglioside content and composition. When the ganglioside-pretreated platelets were exposed to the ago-

55

H.A. Guglielmone et al./Thrombosis Research 98 (2000) 51–57

Table 2. Effect of total gangliosides on platelet aggregation induced by differents agonists Platelet aggregation (%)

ATP release (nM)

Agonist

Concentration

Control

GTOT (100 ␮M)

Control

GTOT (100 ␮M)

ADP Epinephrine Collagen Arachidonic acid* U-46619* Ionomycine* A 23187* Thrombin* PMA* KF/A1⫺4

2.5 ␮M 50 ␮M 1 ␮g/ml 0.5 mM 1 ␮M 1 ␮M 5 ␮M 0.1 UI/ml 100 nM 10 mM/30 ␮M

75 69 92 86 96 87 84 84 67 85

0 15 61 0 0 0 0 0 26 27

0.80 2.28 1.33 5.80 4.00 6.20 7.65 0.61 n.d. n.d.

0 0.31 0.42 0 0 0 0 0 n.d. n.d.

The results represent the mean of at least two determinations performed in triplicate. n.d., not determined. * Assays performed with washed human platelets.

nists, the intracellular Ca2⫹ level may had already been restored to resting values [19]. Under this situation, we observed that the increase in the membrane ganglioside content inhibits platelet aggregation. Using a similar experimental design, alkaline hydrolysed GM1 has been shown to have an anti-aggregating effect either in vitro or in vivo [9]. In our assay system, all the evaluated gangliosides inhibited the aggregation process to a similar extent. The neutral Gg4Cer (Asialo-GM1) had no effect on the aggregation, suggesting that the negative charge or the sialic acid might be important for the observed effects. However, sulphatide, a molecule with similar hydrophobic core but with a negative charge in the polar region, had no effect on aggregation, indicating that the net charge of the lipid is not relevant for the observed effects (see Table 1). The enhancement of collagen-mediated platelet activation by gangliosides shed by neuroblastoma tumour cells suggests that the only presence of sialic acid in the sugar backbone is not necessarily related to an inhibitory effect on platelet aggregation. In this connection, individual gangliosides have opposite effects on serotonin uptake, suggesting that the specificity of the effect could be dependent on the spatial distribution of the sialic acid residues in the sugar backbone [27]. Indirect evidence about the mechanism involved in the inhibition was obtained using different agonists to stimulate the aggregation. The first wave of ADP-induced aggregation is not altered upon pretreatment of platelets with gangliosides, sug-

gesting that ligand-receptor interaction that determines the first morphological changes in platelets is not affected by gangliosides. Myosin light chain phosphorylation is associated with initiation of platelet shape changes [first change observed in light transmission (see Figure 1)] in the absence of platelet secretion after agonist stimulation [28]. The first aggregation wave was not affected by ganglioside-pretreatment, suggesting that the initial phosphorylation events in platelet activation are not affected by gangliosides. One of the first events in platelet activation is the breakdown of phosphoinositides by phospholipase C, which produces inositol trisphosphate (IP3) and diacylglycerol (DAG) that, in turn, increase the intracellular Ca2⫹ level and activate PKC, respectively [11,29]. In this connection, we have shown that gangliosides can inhibit the enzymatic activities of phospholipases C and A2 in several artificial membrane model system [12–15]. Thus, the inhibition of platelet phospholipases could be, at least in part, responsible for the inhibitory effects of gangliosides on platelet function. The aggregation and ATP release of platelets stimulated with Ca2⫹ agonists was completely blocked by gangliosides. It has been previously shown that gangliosides increase the intracellular Ca2⫹ level in platelets and other cell types [7,19,20]. This result is indicative that events downstream intracellular Ca2⫹ increment are blocked in platelets previously treated with gangliosides. Arachidonic acid and U46619 were unable to bypass the inhibitory effect exerted by gangliosides,

56

H.A. Guglielmone et al./Thrombosis Research 98 (2000) 51–57

suggesting that gangliosides inhibit platelet aggregation by inhibiting the synthesis or action of prostaglandins. Gangliosides inhibited U46619-induced aggregation, thus suggesting that gangliosides block the action of TXA2. However, gangliosidetreated platelets are still functional because other receptor-mediated agonists, such as collagen and epinephrine, are able to induce partial aggregation. The difference should be found in the different pathways of activation triggered by these agonists, compared to arachidonic acid and U46619. Observations that epinephrine does not stimulate phospholipase C in human platelets are in agreement with the possibility that gangliosides could be inhibiting platelet aggregation through an inhibition of phospholipases [30]. This study also revealed that fluoroaluminate, a direct activator of trimmeric G-proteins, triggered a partial aggregation. The result is indicative that some pathways of activation are still functional whereas others are not. The phosphorylation in the initial steps of response is not affected, because ganglioside-treated platelets are still able to give a first response after ADP stimulation. PMA, a phorbol ester mimicking the action of diacylglicerol, was also used to evaluate the functional response of ganglioside-treated platelets. The partial inhibition obtained may suggest a dysfunction in the PKC activity caused by gangliosides, as previously found in other cell types [31,32]. However, an inhibitory effect of gangliosides on PKC should be bypassed by Ca2⫹ ionophores, suggesting that the major pathways inhibited are downstream PKC activation (e.g., the enzymatic activity of phosphorylated cytoplasmic PLA2 may be inhibited by the presence of gangliosides). It is widely known that the increase of platelet cyclic AMP (cAMP) inhibits all platelet responses to activating factors [21,33]. A stimulation of inhibitory pathways in human platelets by gangliosides cannot be completely discarded, but in this case the partial response obtained with some agonists and the morphological change detected after ADP-stimulation (first wave) of gangliosidestreated platelets should not be observed. In summary, the remarkable inhibition in platelet function induced by exogenous gangliosides interacting with platelet membranes support the hypothesis that changes in ganglioside composition may be one of the causes of abnormal platelet functionality. The opposite effects of associated

vs. non-associated exogenous gangliosides suggest that these lipids may act as bimodal regulators of either positive or negative signals for platelet function. This work was supported by CONICET, SECyT-UNC, and CONICOR. J.J. Daniele is a fellow, and I.D. Bianco and G.D. Fidelio are career investigators from CONICET.

References 1. Maggio B. The surface behavior of glycosphingolipids in biomembranes: A new frontier of molecular ecology. Prog Biophys Molec Biol 1994;62:55–117. 2. Yates AJ, Rampesaud A. Sphingolipids as receptor modulators. An overview. Ann NY Acad Sci 1998;845:57–71. 3. Yu RK, Ariga T. The role of glycosphingolipids in neurological disorders. Mechanisms of immune action. Ann NY Acad Sci 1998;845: 285–306. 4. Ferroni P, Lenti L, Martini F, Ciatti F, Pontieri GM, Gazzaniga PP. Ganglioside content of human platelets-differences in resting and activated platelets. Thromb Haemost 1997;77: 548–54. 5. Mazurov AV, Prokazova NV, Mikhailenko IA, Mukjin DN, Repin VS, Bergelson LD. Stimulation of platelet adhesion and activation by ganglioside GD3 adsorbed to plastic. Biochim Biophys Acta 1988;968:167–71. 6. Valentino LA, Ladisch S. circulating tumor gangliosides enhance platelet activation. Blood 1994;83:2872–7. 7. Yatomi Y, Igarashi Y, Hakomori S. Effects of exogenous gangliosides on intracellular Ca2⫹ mobilization and functional responses in human platelets. Glycobiology 1996;6:347–53. 8. Santoro A. Inhibition of platelet adhesion to fibronectin, fibrinogen and Willebrand factor substrates by complex gangliosides. Blood 1989; 73:1645–54. 9. Tubaro E, Belogi L, Croce C, Cavallo G, Guida G, Borelli GP. Antiplatelet effects of a new de-N-acetyl-lyso-glycosphingolipid. Eur J Pharmacol 1993;248:175–83. 10. Cunningham MT, Olson JD, Koerner TA. Glycosphingolipid inhibition of the adhesion of

H.A. Guglielmone et al./Thrombosis Research 98 (2000) 51–57

11.

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

thrombin-activated platelets to surfaces is potentiated by albumin. Glycobiology 1993;3: 331–7. Lapetina EG. The signal transduction induced by thrombin in human platelets. FEBS Lett 1990;268:400–4. Bianco ID, Fidelio GD, Maggio B. Modulation of phospholipase A2 activity by neutral and anionic glycosphingolipids in monolayers. Biochem J 1989;258:95–9. Bianco ID, Fidelio GD, Maggio B. Effect of sulfatide and gangliosides on phospholipase C and phospholipase A2 activity. A monolayer study. Biochim Biophys Acta 1990;1026:179–85. Maggio B, Bianco ID, Montich GG, Fidelio GD, Yu RK. Regulation by gangliosides and sulfatides of phospholipase A2 activity against dipalmitoyl- and dilauroylphosphatidyl-choline in small unilamellar bilayer vesicles and mixed monolayers. Biochim Biophys Acta 1994;1190:137–48. Daniele JJ, Maggio B, Bianco ID, Gon˜i FM, Alonso A, Fidelio GD. Inhibition by gangliosides of Bacillus cereus phospholipase C activity against monolayers, micelles and bilayer vesicles. Eur J Biochem 1996;239:105–10. Correa SG, Bianco ID, Riera CM, Fidelio GD. Anti-inflammatory effect of gangliosides in the rat hindpaw edema test. Eur J Pharm 1991; 199:93–8. Lazarewicz JW, Salinska E, Wroblewski JT. NMDA receptor-mediated arachidonic acid release in neurons: Role in signal transduction and pathological aspects. In: Bazan NG, Murphy MG, Toffano G, editors. Neurobiology of Essential Fatty Acids. New York, Plenum Press; 1992. p. 73–89. Bressler JP, Belloni-Olivi L, Forman S. Effect of ganglioside GM1 on arachidonic acid release in bovine aortic endothelial cells. Life Sci 1994; 54:49–60. Vasylevskaya VV, Bochkov VC, Prokazova NV, Tkachuk VA. Characteristics and regulation of ganglioside-induced elevation of free cytoplasmic Ca2⫹ in human blood platelets. Biochim Biophys Acta 1992;1127:221–5. Cossy Isasi S, Bianco ID, Fidelio GD. Gangliosides raise the intracellular Ca2⫹ level in different cell types. Life Sci 1995;57:449–56. Pumiglia KM, Huang CK, Feinstein MB. Ele-

22.

23.

24.

25.

26.

27.

28.

29. 30.

31.

32.

33.

57

vation of cAMP, but not cGMP, inhibits thrombin-stimulated tyrosine phosphorylation in human platelets. Biochem Biophys Res Commun 1990;171:738–45. Born GVR. Aggregation of blood platelets by adenosine diphosphate and its reversal. Nature 1962;194:927–9. Watanabe Y, Taniguchi M, Fukamachi N, Sakuma M, Kobayashi B. Effects of gangliosides on cell maturation of murine megakaryocytes in a liquid culture system. Cell Struct Funct 1990;15:79–84. Nakamura M, Kirito K, Yamanoi J, Wainai T, Nojiri H, Saito M. Ganglioside GM3 can induce megakaryocytoid differentiation of human leukemia cell line K562 cells. Cancer Res 1991; 51:1940–5. Valentino LA, Ladisch S. Tumor gangliosides enhance alpha2 beta1 integrin-dependent platelet activation. Biochim Biophys Acta 1996; 1316:19–28. Fang LH, Lucero M, Kazarian T, Wei Q, Luo FY, Valentino LA. Effects of neuroblastoma tumor gangliosides on platelets adhesion to collagen. Clin Exp Metastasis 1997;15:33–40. Vasylevskaya VV, Zvezdina ND, Korotaeva AA, Prokazova NV. The influence of gangliosides on serotonin binding and uptake by human platelets. Platelets 1995;6:37–42. Daniel JL, Molish IR, Rigmaiden M., Steward G. Evidence for a role of myosin phosphorylation in the initiation of the platelet shape change response. J Biol Chem 1984;259:9826–31. Blockmans D, Deckmyn H, Vermylen J. Platelet activation. Blood Reviews 1995;9:143–56. Lapetina EG. Inositol phospholipids and phospholipase C in platelet receptor transduction. Prog Clin Biol Res 1988;283:499–511. Kanda T, Ariga T, Yamawaki M, Yu RK. GM3 regulates protein kinase systems in cultured brain microvascular endothelial cells. J Neurochem 1993;61:1969–72. Yates AJ, Saqr HE, Van Brocklyn J. Ganglioside modulation of the PDGF receptor. A model for ganglioside functions. J Neurooncol 1995;24:65–73. Herman AG. Rationale for the combination of anti-aggregating drugs. Thromb Res 1998;92: (1 Suppl 1):S17–21.