Rap1 is a small, Ras‐like GTPase whose function and regulation are still largely unknown. We have developed a novel assay to monitor the active, GTP‐bound form of Rap1 based on the differential affinity of Rap1GTP and Rap1GDP for the Rap binding domain of RalGDS (RBD). Stimulation of blood platelets with α‐thrombin or other platelet activators caused a rapid and strong induction of Rap1 that associated with RBD in vitro. Binding to RBD increased from undetectable levels in resting platelets to >50% of total Rap1 within 30 s after stimulation. An increase in the intracellular Ca2+ concentration is both necessary and sufficient for Rap1 activation since it was induced by agents that increase intracellular Ca2+ and inhibited by a Ca2+‐chelating agent. Neither inhibition of translocation of Rap1 to the cytoskeleton nor inhibition of platelet aggregation affected thrombin‐induced activation of Rap1. In contrast, prostaglandin I2 (PGI2), a strong negative regulator of platelet function, inhibited agonist‐induced as well as Ca2+‐induced activation of Rap1. From our results, we conclude that Rap1 activation in platelets is an important common event in early agonist‐induced signalling, and that this activation is mediated by an increased intracellular Ca2+ concentration.
Rap1A and Rap1B are two members of the Ras family of small GTPases. Rap1A was first identified both by its homology to Ras and as the product of a cDNA that induces flat revertants of K‐ras‐transformed cells (Krev‐1). Rap1B is a very close relative of Rap1A, differing in only nine amino acids predominantly in the C‐terminal part of the protein (Noda, 1993). The functional difference between the two proteins is unclear and in most studies no discrimination between the two has been made. Various functions of Rap1 have been described, some of which are related to the ability of Rap1 to counteract the function of Ras. For instance, introduction of the active, GTP‐bound form of Rap1 into fibroblasts inhibits Ras‐mediated activation of MAP kinase (Cook et al., 1993). In addition, constitutively active Rap1 can inhibit Ras‐mediated induction of germinal vesicle breakdown in Xenopus oocytes (Campa et al.,1991). In vitro Rap1 binds regulators (p120RasGAP; Frech et al., 1990; Hata et al., 1990) and effectors (Raf1, RalGDS; Spaargaren and Bischoff, 1994; Nassar et al., 1995; Wittinghofer and Herrmann, 1995) of Ras, suggesting that the inhibition of Ras function is due to sequestration of Ras effectors by Rap1. Interestingly, cAMP, an inhibitor of Ras signalling in various cell types (Burgering et al., 1993; Cook et al., 1993; Wu et al., 1993), has been reported to induce the activation of exogenously expressed epitope‐tagged Rap1 in a Ras‐transformed fibroblast (Altschuler et al., 1995). However, not all effects of Rap1 point to an inhibition of Ras signalling and indicate that Rap1 has a separate function in various cell types. For instance, in Swiss 3T3 fibroblasts, constitutively active Rap1 induces DNA synthesis (Yoshida et al., 1992). In neutrophils, Rap1 is found in association with the NADPH oxidase system (Quinn et al., 1989), and introduction of Rap1 into neutrophil‐like HL60 cells results in a stimulation of this oxidase (Gabig et al., 1995). Also, in blood platelets, Rap1 may play an important role, since Rap1, in particular Rap1B, is highly expressed (Torti and Lapetina, 1994).
Platelets are anuclear cell fragments that adhere to sites of injury in a blood vessel and aggregate to stop bleeding. Adhesion and aggregation are accompanied by a profound change in the morphology of platelets due to remodelling of the actin‐based cytoskeleton (Hartwig, 1992; Fox, 1993). Thrombin is the most potent stimulator of platelets, but a large number of other activators has been described (Kroll and Schafer, 1989; Siess, 1989). When platelets are activated, Rap1 is translocated to the platelet cytoskeleton (Fischer et al., 1990, 1994). This translocation is a relatively slow process that depends on aggregation of platelets. In addition, it has been reported that Rap1 associates with p120RasGAP and PLCγ1 after platelet stimulation (Torti and Lapetina, 1992). Also, prostaglandin I2 (PGI2), a strong negative regulator of platelet function, affects Rap1 by inducing the phosphorylation of a serine residue at the C‐terminal end of Rap1 (Kawata et al., 1989; Siess et al., 1990). This phosphorylation induces a translocation of Rap1 from the plasma membrane to the cytosol (Lapetina et al., 1989).
Despite all these results, it is still unclear what the function of Rap1 is and in which signal transduction pathway Rap1 is involved. This is partly due to the lack of information on signals that may activate endogenous Rap1, i.e. induce the conversion of the inactive GDP‐bound form of the protein to the active GTP‐bound form. Unfortunately, no antibodies are available that can immunoprecipitate Rap1 to measure the ratio of GTP:GDP bound to Rap1. Therefore, we have developed an alternative, indirect assay to measure Rap1 activation. This assay is based on the differential affinity of Rap1GTP versus Rap1GDP for the Rap binding domain of RalGDS (RBD) as shown by Wittinghofer and co‐workers (Herrmann et al., 1996): whereas the Rap binding domain of RalGDS binds to the GTP‐bound form of Rap1 with a very high affinity (KD =10 nM), the affinity for the GDP‐bound form is undetectably low. Here we report that in platelets α‐thrombin stimulation results in a dramatic increase of Rap1 which can associate with RBD in vitro, indicating an increase of Rap1 in the GTP‐bound state. We have investigated this activation further and found that most, if not all, activators of platelet function activate Rap1, whereas the negative regulator PGI2 inhibits Rap1 activation. Furthermore, we show that an increase in the intracellular Ca2+ concentration is both necessary and sufficient to activate Rap1. Finally, we show that Rap1 activation occurs independently of, and probably prior to, platelet aggregation and the association of Rap1 with the cytoskeleton. From these results, we conclude that Rap1 is activated rapidly and strongly by Ca2+‐mediated signalling after platelet stimulation, suggesting a critical role for Rap1 in platelet activation.
α‐Thrombin induces rapid activation of Rap1
Freshly isolated human platelets were stimulated with α–thrombin for various time periods and lysed. Rap1 was precipitated with the polyhistidine‐tagged Rap binding domain of RalGDS (RBD) bound to nickel beads and identified by Western blotting using a monoclonal antibody directed against Rap1. This antibody is specific for Rap1, and does not recognize Ras or R‐ras (not shown). A rapid and strong increase in the amount of Rap1 that bound to RBD in vitro was observed (Figure 1A). This increase was visible within 5 s after thrombin addition and reached its maximum level at ∼30 s. At this latter time point, >50% of total Rap1 could be precipitated with RBD (Figure 1B). Since RBD associates exclusively with the GTP‐bound form of Rap1 in vitro, with no detectable affinity for the GDP‐bound form of Rap1 (Herrmann et al., 1996), we conclude that α‐thrombin induces a rapid conversion of the majority of Rap1 to the GTP‐bound, active state.
Treatment of platelets with agonists other than α‐thrombin (Benton et al., 1982; Kroll and Schafer, 1989; Siess, 1989; Inazu et al., 1991) also caused activation of Rap1; the thrombin receptor‐activating peptide (TRAP), collagen, the thromboxane A2 (TxA2) analogue U46619, ADP and platelet‐activating factor (PAF) all induced rapid and strong Rap1 activation (Figure 2). Wheat germ agglutinin and lysophosphatidic acid (LPA) treatment also resulted in Rap1 activation (data not shown). From these results, we conclude that Rap1 is activated rapidly and strongly by signalling events common to most, if not all platelet agonists.
Rap1 activation occurs independently of secretion and platelet aggregation
During thrombin‐induced platelet activation, TxA2 and ADP are released by platelets. Both act in positive feedback loops to enhance further the ongoing platelet activation (Kroll and Schafer, 1989; Siess, 1989). As TxA2 and ADP induce Rap1 activation, we investigated whether thrombin‐induced Rap1 activation is a consequence of these processes. Treatment of platelets with indomethacin inhibits TxA2 formation (Shen and Winter, 1977); addition of phosphoenolpyruvate and pyruvate kinase (PEP:PK) scavenges ADP, thus inhibiting ADP signalling (Ammit and O'Neill, 1991; Van Willigen et al., 1996). Addition of both inhibitors together, at concentrations causing complete inhibition of both pathways, resulted in only a partial inhibition of Rap1 activation (Figure 3). From these results, we conclude that although both TxA2 and ADP do contribute to thrombin‐induced activation of Rap1, they are not essential for thrombin to induce Rap1.
Previously, it was shown that Rap1 translocates to the platelet cytoskeleton in a mainly aggregation‐dependent manner (Fischer et al., 1994). We therefore investigated the effects of disruption of the cytoskeleton and inhibition of platelet aggregation on thrombin‐induced Rap1 activation. Treatment of platelets with cytochalasin D prevents activation‐dependent actin polymerization and remodelling of the cytoskeleton (Dash et al., 1995). This treatment totally prevented the translocation of Rap1 to the cytoskeleton (Figure 4A). Cytochalasin D, however, did not affect thrombin‐induced activation of Rap1 (Figure 4B), showing that Rap1 activation occurs independently of translocation to the cytoskeleton. Furthermore, in platelets that were not allowed to aggregate by the addition of the RGDS peptide, which blocks integrin function (Hynes, 1992; Calvete, 1994), Rap1 activation was not impaired (Figure 5). Also, inhibition of aggregation by not stirring the platelet suspension did not affect α‐thrombin‐induced activation of Rap1 (data not shown). From these results, we conclude that Rap1 activation occurs independently of Rap1 translocation to the cytoskeleton and of platelet aggregation.
Increase in intracellular Ca2+ is necessary and sufficient for Rap1 activation
One of the signalling events shared by platelet agonists is the activation of phospholipase C (PLC), which releases diacylglycerol (DAG) to activate protein kinase C (PKC), and inositol‐1,4,5‐triphosphate (InsP3) to mobilize Ca2+ from intracellular stores (Kroll and Schafer, 1989; Siess, 1989). Figure 6A demonstrates that inhibition of PLC by U73122 (Okamoto et al., 1995) strongly reduced thrombin‐induced activation of Rap1, suggesting that either PKC or Ca2+ is an important factor in Rap1 regulation. We therefore tested whether an increase in the intracellular Ca2+ concentration is important for the activation of Rap1. Inhibition of the cytosolic Ca2+ increase by chelation of intracellular Ca2+ with BAPTA‐AM (Watson et al., 1995) completely blocked activation of Rap1 by α‐thrombin. On the other hand, increasing the concentration of intracellular Ca2+ with ionomycin (Cavallini and Alexandre, 1994; Doni et al., 1994) or thapsigargin (Authi et al., 1993; Cavallini et al., 1995) strongly activated Rap1 (Figure 6A). This indicates that an increased level of Ca2+ is both necessary and sufficient to activate Rap1. The Ca2+ effect is not due simply to Ca2+‐induced production and release of TxA2 or secretion of ADP, since inhibitors of both signalling events, indomethacin and PEP:PK, caused only a slight reduction of Ca2+‐induced Rap1 activation (Figure 6A). Moreover, this reduction is probably due to the slightly lower Ca2+ level induced by thapsigargin and ionomycin in the presence of indomethacin and PEP:PK (data not shown).
Inhibition of PKC by several inhibitors, i.e. staurosporin, bisindolylmaleimid and calphostin C (Tamaoki, 1991; Toullec et al., 1991), did not at all or only marginally affect the activation of Rap1 by α‐thrombin (Figure 6B). Consistent with this, activation of PKC by the phorbol ester phorbol 12‐myristate 13‐acetate (PMA) (Kroll and Schafer, 1989; Siess, 1989) activated Rap1 only slightly and after a prolonged period of time (Figure 6B). From these results, we conclude that Rap1 activation is mediated by an increase in intracellular Ca2+, which is both necessary and sufficient, whereas activation of PKC does not play an essential role.
Prostaglandin I2 (PG I2) inhibits thrombin‐ and Ca2+ ‐induced Rap1 activation
PGI2, a potent inducer of cAMP production, is a strong antagonist of platelet activation (Kroll and Schafer, 1989; Siess, 1989). We therefore tested the effect of PGI2 on thrombin‐induced activation of Rap1. Figure 7A shows that addition of PGI2 inhibited thrombin‐induced activation of Rap1 completely. Furthermore, when platelets were first treated with thrombin and subsequently with PGI2, the initial activation of Rap1 was followed by a rapid down‐regulation (Figure 7B). Interestingly, the activity of Rap1 (again) correlated with the level of intracellular Ca2+ (Figure 7C). However, Ca2+‐induced activation of Rap1 by thapsigargin and ionomycin was also fully blocked by PGI2 (Figure 7D), even though the intracellular Ca2+ levels induced by thapsigargin or ionomycin were still high in the presence of PGI2 (data not shown; Siess and Lapetina, 1989; Nakamura et al., 1995). These results show that Rap1 activity is under tight negative control of PGI2, which acts both upstream and downstream of Ca2+.
PGI2 activates adenylate cyclase resulting in an increase in the cAMP level. It is intriguing that Rap1 is an in vivo substrate for cAMP‐dependent protein kinase A (PKA) (Kawata et al., 1989; Siess et al., 1990). This phosphorylation, which results in a lower electrophoretic mobility of Rap1 (Siess et al., 1990; Siess and Grünberg, 1993), occurs only slowly and was still very limited 2 min after PGI2 treatment of the platelets (Figure 8A; Grünberg et al., 1995). At this time point, however, PGI2 had already completely inhibited Rap1 activity (see Figure 7A). Furthermore, α‐thrombin was able to activate phosphorylated Rap1 apparently normally. This was shown by pre‐treating platelets with PGI2 for various times, followed by thrombin stimulation: whereas platelets became fully responsive to thrombin stimulation within 60 min after PGI2 addition (due to the instability of PGI2), Rap1 phosphorylation remained high (Figure 8A; Siess and Grünberg, 1993; Grünberg et al., 1995). Under these conditions, thrombin activated the phosphorylated, shifted form of Rap1 (Figure 8B). We conclude, therefore, that the PGI2‐induced inhibition of Rap1 activation is not caused by phosphorylation of the GTPase.
We have developed a novel assay to identify the active GTP‐bound state of Rap1. This assay is based on the observation that the GTP‐bound form of Rap1 associates with the RBD of RalGDS with high affinity in vitro, whereas no interaction can be detected with the GDP‐bound form of Rap1. When we incubated cell lysates of resting and α‐thrombin‐stimulated human blood platelets with polyhistidine‐tagged RBD bound to Ni2+‐NTA–agarose beads, we observed a large amount of Rap1 associated with RBD only in stimulated platelets. Based on the in vitro binding affinity, we conclude that this increase is due to an induction of the GTP‐bound form of Rap1 by α‐thrombin. Alternatively, Rap1 might be constitutively GTP‐bound but complexed to a factor in resting platelets that prevents it from associating with RBD. Stimulation with thrombin would then lead to dissociation of this factor, rendering Rap1 available for RBD binding. To discriminate between the two possibilities, the ratio of GDP and GTP bound to Rap1 in platelets needs to be measured. Unfortunately, suitable antisera for Rap1 immunoprecipitation, which are essential for such an experiment, are still not available. Irrespective of this, in functional terms there may be no difference between the two mechanisms, as both result in free GTP‐bound, active Rap1.
The assay was used to monitor the activation of Rap1 in platelets. We observed that all agonists tested that activate platelets also activate Rap1. This could imply that Rap1 is either involved in a signalling pathway common to all of these agonists or that Rap1 activation is a secondary event, induced by positive feedback loops (TxA2 release and ADP secretion) or by platelet aggregation, for instance. However, inhibition of both TxA2 production and ADP signalling as well as aggregation did not affect α‐thrombin‐induced activation of Rap1 significantly. This strongly suggests that Rap1 activation is one of the early events in platelet activation. The observed translocation of Rap1 to the cytoskeleton is not involved in the activation of Rap1, since it occurs much later than Rap1 activation. Furthermore, inhibition of actin polymerization by cytochalasin D, which totally blocked Rap1 association with the cytoskeleton, did not inhibit Rap1 activation. Therefore, we conclude that Rap1 is involved in a signalling pathway common to all agonists tested. A common denominator for all these agonists is the increase in intracellular Ca2+ by mobilization of Ca2+ from internal stores and by influx of extracellular Ca2+. Indeed, we found that increasing the concentration of intracellular Ca2+ with either ionomycin or thapsigargin induced Rap1 activation. In addition, BAPTA‐AM, which chelates intracellular Ca2+, inhibited thrombin‐induced Rap1 activation. Apparently, Rap1 activation is mediated by a rise in intracellular Ca2+. The mechanism by which the Ca2+ increase results in Rap1 activation, however, is unclear and currently under investigation.
PGI2, which activates adenylate cyclase to increase the levels of cAMP in platelets, inhibits thrombin‐induced activation of Rap1. This is in agreement with the observation that PGI2 inhibited thrombin‐induced increases in intracellular Ca2+. However, thapsigargin‐ or ionomycin‐induced activation of Rap1 was also inhibited by PGI2, indicating that the antagonist also affects a factor downstream of Ca2+. Although Rap1 is phosphorylated after PGI2 treatment of platelets, it is unlikely that this phosphorylation causes the inhibition. We conclude this from the observation that (i) PGI2‐induced phosphorylation of Rap1 was a much later event than inhibition of Rap1, and (ii) thrombin could activate the phosphorylated form of Rap1. It should be noted here that the detection of the phosphorylated form of Rap1 by a mobility shift is an accurate method, as it has been shown by others that PGI2‐induced phosphorylation invariantly is accompanied by the shift in mobility (Siess et al., 1990). Another candidate for mediating the PGI2 inhibition is Rap1GAP. Phosphorylation of this protein is induced by elevated levels of cAMP, at least in insect cells and the SK‐MEL–3 cell line (Polakis et al., 1992; Rubinfeld et al., 1992). Whether this occurs in platelets and whether it affects GAP activity is still unknown. It should be noted that phosphorylation of Rap1 correlates with the PGI2‐induced translocation of Rap1 from the plasma membrane to the cytosol (Lapetina et al., 1989). This translocation may be another form of inactivation by PGI2, which clearly interferes at a different level of Rap1‐mediated signalling.
From the rapid activation of a major fraction of Rap1 and the stringent control of Rap1 activity by PGI2, we anticipate that Rap1 plays a critical role in agonist‐induced, calcium‐mediated events in platelet activation. Interestingly, Ca2+‐induced signalling is involved in the activation of integrin αIIbβ3, resulting in the exposure of binding sites for fibrinogen and, subsequently, platelet aggregation (Kroll and Schafer, 1989; Siess, 1989). It is tempting, therefore, to speculate that Rap1 is involved in this process. Support for this possibility comes from the recent finding that R‐ras, a close relative of Rap1, is involved in the activation of integrins in various cell lines (Zhang et al., 1996). A function of Rap1 in the activation of integrins is not incompatible with the flat revertant phenotype of Krev‐1/Rap1 in transformed fibroblasts.
Materials and methods
Production of his‐tagged RalGDS‐RBD
The cDNA encoding the 97 amino acids spanning RBD was isolated from pGEX‐RGF97 (Herrmann et al., 1996) as a BamHI–XhoI fragment. The BamHI site was blunted with Klenow DNA polymerase. Subsequently, the fragment was inserted into the pET‐15b vector (Novagen) digested with NdeI (and blunted with Klenow DNA polymerase) and XhoI. The construct was transformed into Escherichia coli (strain BL21). Protein production was initiated by addition of isopropyl β‐d‐thiogalactopyranoside (IPTG) to the culture. The fusion protein was affinity purified on a Ni2+‐NTA–agarose column (Qiagen) from the supernatant of bacteria lysed by sonication and freeze–thaw cycles in a sucrose‐containing buffer.
Isolation and stimulation of platelets
Freshly drawn venous blood from healthy volunteers (with informed consent) was collected into trisodium citrate (0.1 vol. of 130 mM trisodium citrate). The donors claimed not to have taken any medication during the previous 10 days. The blood was centrifuged at 200 g for 15 min at room temperature to yield platelet‐rich plasma (PRP). Then, 0.1 vol. of ACD (2.5% trisodium citrate, 1.5% citric acid, 2% d–glucose) was added to the PRP to lower the pH of the plasma to 6.5 and thus prevent platelet activation during further isolation. Platelets were purified from PRP by centrifugation at 700 g for 15 min at room temperature. The platelet pellet was resuspended in HEPES/Tyrode buffer (10 mM HEPES, 137 mM NaCl, 2.68 mM KCl, 0.42 mM NaH2PO4, 1.7 mM MgCl2, 11.9 mM NaHCO3, pH 7.4) containing 5 mM d‐glucose at 2 × 108 platelets/ml. Platelets were left at room temperature for at least 30 min to ensure a resting state. Samples of 0.5 ml were used for the experiments. Purified platelets were incubated in a lumiaggregometer (CHRONO‐LOG corporation) at 37°C. In the standard assay, incubation with agonists was without stirring. Without stirring, platelets only change their shape but do not aggregate, whereas in a stirred suspension (900 r.p.m.) platelet shape change and aggregation occur. Platelet agonists used in the study are: α‐thrombin which was added to the platelets for 1 min at a final concentration of 0.1, 0.2 or 0.25 U/ml (as indicated), TRAP (6mer: SFLLRN) (10 μM), collagen (5 μg/ml), U46619 (TxA2 analogue) (1 μM), ADP (10 μM) and PAF (200 nM). PMA was added to the platelets at a final concentration of 10 nM, thapsigargin, an inhibitor of intracellular Ca2+ ATPases, was used at 100 nM concentration as was ionomycin, a Ca2+ ionophore. In the case of ionomycin, 1 mM CaCl2 was added to the platelet suspension just prior to stimulation. PGI2 was used at a concentration of 10 or 20 ng/ml and incubated with the platelets for 2 min (unless indicated otherwise). The PLC inhibitor U73122 (1 μM) was present during 3 min pre‐incubation. The inactive component U73343 used as control had no effect on Rap1 activity (data not shown). BAPTA‐AM (30 μM), an intracellular Ca2+ chelator, was pre‐incubated with the platelet suspension for 30 min. Indomethacin (30 μM) was added to the platelets for 10 min to inhibit TxA2 formation. The ADP scavenger PEP:PK was freshly prepared prior to every experiment and added to the platelets at a final concentration of 0.28 mM PEP and 3 U/ml PK 1 min before stimulation. PKC inhibitors were used as follows: staurosporin (1 μM), 5 min incubation, bisindolylmaleimid (5 μM), 1 min incubation, calphostin C (5 μM), 5 min incubation. Cytochalasin D was added to the platelet suspension at a concentration of 5 μg/ml 5 min prior to platelet stimulation. The peptide RGDS was used at a concentration of 100 μM and was added to the platelets 1 min prior to stimulation. RGDS binds to the ligand binding site of the platelet integrin αIIbβ3. It inhibits binding of fibrinogen to the integrin and therefore blocks platelet aggregation. RGDS does not activate outside‐in signalling of integrin αIIbβ3 (Hynes, 1992; Calvete, 1994).
Rap1 activation assay using RalGDS‐RBD
Platelets were lysed by addition of 1 vol. of cold 2× RIPA lysis buffer [150 mM NaCl, 100 mM Tris–HCl pH 7.4, 2% NP‐40, 1% deoxycholic acid (DOC), 0.2% SDS, 2 mM sodium orthovanadate, 2 mM phenylmethylsulfonyl fluoride (PMSF), 2 μM leupeptin, 2 μM aprotinin] to the platelet suspension. Lysis was performed at 4°C for 10–30 min. Lysates were clarified by centrifugation at maximal speed in an Eppendorf centrifuge for 10 min at 4°C. Five μg of RalGDS‐RBD coupled to Ni2+‐NTA–agarose beads (Qiagen) were added to the supernatant and incubated at 4°C for 30–90 min with slight agitation. Beads were washed four times in 1× RIPA. After the final wash, Laemmli sample buffer was added to the samples. Next, proteins were fractionated by SDS–PAGE and transferred to polyvinylidene difluoride membranes (Immobilon‐P, Millipore). The antibody used specifically to detect Rap1 was a monoclonal antibody directed against Rap1 (Transduction Laboratories). Immune complexes were detected by enhanced chemiluminescence (Amersham). All experiments shown here were performed at least three times with the same result to exclude donor‐specific effects.
PRP was prepared as described above. ACD was added and platelets were incubated with 3 μM Fura‐2‐AM for 45 min at 37°C. Surplus Fura‐2‐AM and plasma were removed by gel filtration over a Sepharose 2B column equilibrated in HEPES/Tyrode buffer. Measurement of the cytosolic Ca2+ concentration was performed using a Hitachi F4500 fluorescence spectrophotometer by a dual wavelength program (excitation was measured at 340 nm and 380 nm, emission at 510 nm).
Samples of 5×108 platelets/ml were lysed in cold 2× CSK buffer (100 mM Tris–HCl, 20 mM EGTA, 2% Triton X‐100, 2 mM sodium orthovanadate, 2 mM PMSF, 2 μg/ml aprotinin, 2 μg/ml leupeptin) after stimulation. Lysis was for at least 10 min at 4°C. Lysates were centrifuged at maximal speed in an Eppendorf centrifuge for 10 min at 4°C. The pellet containing the actin cytoskeleton of the lysed platelets was washed once with 1× CSK buffer and centrifuged as before. Laemmli sample buffer was added to the pellet. Amounts of protein representing equal numbers of platelets were used for SDS–PAGE.
We thank Fred Wittinghofer for communicating the high affinity of RBD for Rap1 prior to publication, Marcel Spaargaren for the RBD construct and our colleagues for support, discussions and critically reading the manuscript. This work was supported by a grant provided by the Netherlands Heart Foundation (grant 94.136).
- Copyright © 1997 European Molecular Biology Organization