Platelet-Derived Growth Factor Inhibits Platelet Activation in Heparinized Whole Blood

Thrombosis Research 95 (1999) 185–196

REGULAR ARTICLE

Platelet-Derived Growth Factor Inhibits Platelet Activation in Heparinized Whole Blood Frode Selheim1, Holm Holmsen1 and Flemming S. Vassbotn1,2 Department of Biochemistry and Molecular Biology, University of Bergen, Bergen, Norway; 2 Haukeland University Hospital, Department of Otolaryngology/Head and Neck Surgery, Bergen, Norway.

1

(Received 6 November 1998 by B. Østerud; revised/accepted 9 February 1999)

Abstract We previously have demonstrated that human platelets have functionally active platelet-derived growth factor a-receptors. Studies with gel-filtered platelets showed that an autocrine inhibition pathway is transduced through this tyrosine kinase receptor during platelet activation. The physiological significance of this inhibitory effect of platelet-derived growth factor on gel-filtered platelets activation is, however, not known. In the present study, we investigated whether platelet-derived growth factor inhibits platelet activation under more physiological conditions in heparinized whole blood, which represents a more physiological condition than gel-filtered platelets. Using flow cytometric assays, we demonstrate here that platelet-derived growth factor inhibits thrombin-, thrombin receptor agonist peptide SFLLRN-, and collagen-induced platelet aggregation and shedding of platelet-derived microparticles from the platelet plasma membrane during platelet aggregation in stirred heparinized whole blood. The inhibitory effect of platelet-derived growth factor was dose dependent. However, under nonaggregating conditions (no stirring), we could not demonstrate any significant effect of plateletderived growth factor on thrombin- and thrombin Abbreviations: GFP, gel-filtered platelets; GPRP, glycyl-l-prolyll-arginyl-l-proline; FITC, fluorescein isothiocyanate; PI 3-K, phosphoinositide 3-kinase; PMP, platelet-derived microparticles; PE, phycoerythrin; TRAP, thrombin receptor agonist peptide. Corresponding author: F. Selheim, Department of Biochemistry, ˚ rstadveien 19, N-5009 Bergen, Norway. University of Bergen, A Fax: 147 (55) 58 64 00; E-mail: ,[email protected]..

receptor agonist peptide-induced platelet surface expression of P-selectin. Our results demonstrate that platelet-derived growth factor appears to be a true antithrombotic agent only under aggregating conditions in heparinized whole blood.  1999 Elsevier Science Ltd. All rights reserved. Key Words: Whole blood; Platelet-derived growth factor; Flow cytometry; Inhibition of platelet activation

P

latelet-derived growth factor (PDGF) is a potent mitogen and chemotactic factor for cells of mesenchymal origin (reviewed in references [1] and [2]). Receptors for PDGF also are expressed in other cell types, for example, capillary endothelial cells [3] and neuronal cells [4], as well as human platelets [5]. Evidence has been presented suggesting that PDGF is involved in disorders with excessive cell proliferation, such as cancer, atherosclerosis, and chronic inflammatory conditions (reviewed in [6]). We have previously reported a new possible function of PDGF in regulation of thrombosis [5]. PDGF exists as homo- and heterodimers of disulfide-linked A- and B-polypeptide chains; the two chains exhibit 60% amino acid identity in their mature parts of slightly more than 100 amino acids each [7–9]. All three PDGF isoforms have been identified and purified from natural sources; the AB heterodimer is the major form found in human platelets [10,11]. The PDGF isoforms, PDGF-AA, PDGF-AB, and PDGF-BB, mediate their effects by binding to two different tyrosine kinase recep-

0049-3848/99 $–see front matter  1999 Elsevier Science Ltd. All rights reserved. PII S0049-3848(99)00038-9

186

F. Selheim et al./Thrombosis Research 95 (1999) 185–196

tors (PDGFR), denoted the a- and b-receptors [7–9]. Both receptor types bind SH2-containing intracellular substrates and stimulate mitosis of mesenchymal cells. Bryckaert et al. [12,13] showed that exogenous PDGF inhibited thrombin- and collagen-induced platelet responses. We previously have reported that human platelets have functionally active PDGF a-receptors. Studies with gel-filtered platelets (GFP) showed that activation of the PDGF a-receptor inhibits thrombin-induced platelet aggregation and secretion. Incubation of platelets with neutralizing PDGF antibodies demonstrated that the PDGF secreted from platelets takes part in a negative feedback regulation of platelet activation [5]. The physiological significance of this inhibitory effect of PDGF on gel-filtered platelet activation, however, is not known. Studies of platelet activation in heparinized whole blood better reflects the in vivo situation [14] since measurements of the platelet function is done in their usual surroundings (erythrocytes, leukocytes, plasma proteins) with a normal physiological concentration of the divalent cations calcium and magnesium. The present study therefore was designed to determine whether PDGF inhibits thrombin-, thrombin receptor agonist peptide (TRAP) SFLLRN-, and collageninduced platelet aggregation and shedding of platelet-derived microparticles (PMP) in stirred heparinized whole blood. We also investigated the effect of PDGF on surface expression of P-selectin [15] and CD63 [16] under nonaggregating conditions (no stirring). Flow cytometric assays of whole blood were used to allow measurement of platelet activation and the presence of PMP in their natural milieu with minimal manipulation of the samples [17–19]. Our results demonstrate that PDGF appears to be a true antithrombotic agent as it inhibits thrombin-, TRAP- and collagen-induced platelet aggregation and shedding of PMP in stirred whole blood. Possible targets for the antithrombotic effect of the PDGF a-receptor are discussed.

1. Materials and Methods 1.1. Materials Human recombinant PDGF-BB was generously provided by Prof. Carl-Henrik Heldin (Ludwig In-

stitute for Cancer Research, Uppsala, Sweden) or obtained from Upstate Biotechnology (Lake Placid, NY, USA). Porcine PDGF prepared as previously described [20,21] also was used. The synthetic TRAP SFLLRN was kindly provided by Prof. Nils Olav Solum (Rikshospitalet, Oslo, Norway). Collagen fibrils purchased from Nycomed Arzneimittel GmbH. Co. (Mu¨nchen, Germany) were stored at 48C, while stock solutions of bovine thrombin (Hoffman-La Roche) and TRAP were stored at 2208C. Collagen, thrombin, and TRAP were diluted with 0.15 M NaCl to desired concentrations just before the experiments. Lysophilized PDGF-BB (10 mg) rehydrated in 10 mM acetic acid and stored at 2208C, was diluted with phosphate-buffered saline (PBS; 136.7 mM NaCl, 2.7 mM KCl, 13.1 mM Na2HPO4, and 1.5 mM KH2PO4) before use. Vehicle for PDGF-BB was 10 mM acetic acid diluted with PBS. (N-[2-hydroxyethyl] piperazine-N9[2ethanesulfonic acid]) (Hepes) was purchased from Sigma Chemical Co.(St. Louis, MO, USA). The peptide glycyl-l-prolyl-l-arginyl-l-proline (GPRP) was supplied by Calbiochem (San Diego, CA, USA). Unlabeled CaliBRITE beads, fluorescein isothiocyanate (FITC)-labelled CaliBRITE beads, and phycoerythrin (PE)-labelled CaliBRITE beads were obtained from Becton Dickinson Immunocytometry (BDIS; San Jose, CA, USA). Heparin was obtained from Novo Nordisk A/S (Bagsv’rd, Denmark), and paraformaldehyde was from Fluka Chemika. All other chemicals were of analytical grade or better.

1.1.1. Antibodies FITC-conjugated anti CD42b, an mouse IgG1 antibody against the human membrane glycoprotein gpIb (170KD), was purchased from PharMingen (San Diego, CA, USA). R-phycoerythrin (R-PE)conjugated anti-human CD62, a monoclonal antibody directed against P-selectin (formerly known as CD62P [22], GMP-140 [23], or PADGEM [24]), was purchased from BDIS. R-PE-conjugated antihuman CD63, an monoclonal antibody against a lysosomal granule-membrane glycoprotein expressed on the surface of activated platelets and with resting platelets in the lysosomes[16] was purchased from Immunotech Coulter (Marseilles, France). CD63 and P-selectin recently have been reported to be present in minor amounts also in dense granules [25–27]. FITC-conjugated anti-

F. Selheim et al./Thrombosis Research 95 (1999) 185–196

187

Fig. 1. Illustration of the flow cytometric light scatter profiles for platelets and microparticles in whole blood. The figure shows flow cytometric data obtained from fixed whole blood labelled with FITC-conjugated antihuman platelet chicken antibodies. P and M denote the gates for platelets and the submicron sizes microparticles, respectively. Panel A (dot plot) and panel B (threedimensional plot) shows stirred (500 rpm) unstimulated whole blood at 378C for 90 seconds. Panel C (dot plot) and panel D (three-dimensional plot) shows whole blood preincubated with vehicle at 378C for 60 seconds before stirring with 4 mM of TRAP for 30 seconds.

human platelet chicken antibodies was from Biopool (Umea˚, Sweden).

1.2. Methods 1.2.1. Blood sampling Venous blood was obtained from healthy donors who had not taken any medications for at least 2 weeks before blood donation. To avoid platelet activation during blood drawing, a 19-gauge needle was used, and the first milliliter of blood was discarded. Blood was collected by antecubital venipuncture directly into polypropylene tubes containing 1.5 U/mL heparin (final concentration) as anticoagulant. Measurements of platelet aggregation and secretion in whole blood were started within 15 minutes of blood collection. 1.2.2. Flow cytometric measurement of P-selectin and CD63 surface expression Whole blood flow cytometric analysis was based on the method described by Shattil et al. [17]. Briefly, 5 ml aliquots of whole blood were added to polystyrene tubes containing 30 ml of HEPES-buffered sa-

line, the FITC-conjugated platelet identifier antiCD42b (10 mL), and 10 mL of R-PE-conjugated activation-dependent monoclonal antibody (antiCD62 or anti CD63). TRAP or thrombin and recombinant human PDGF-BB or their vehicles were included in these samples at various concentrations. For assays with thrombin, 1.25 mM GPRP was included in the samples to prevent fibrin polymerisation [28]. The samples were incubated at 378C for 20 minutes without stirring and then fixed with 0.2% paraformaldehyde in PBS (1:19). Blood samples prepared as described above were analysed in a FACSort flow cytometer (BDIS). BDIS calibration beads were used for calibration and colour compensation. The instrument was formatted for two-colour analysis with a 530/30-band pass filter in the fluorescein channel, and 585/42-band filter in the PE channel. The light scatter and fluorescence data were collected using logarithmic amplification, and a fluorescence threshold was set to include only those cells that had bound FITCconjugated anti-CD42b. In this way, only cells expressing the human glycoprotein GPIb were detected. Only platelets identified by both fluorescein

188

F. Selheim et al./Thrombosis Research 95 (1999) 185–196

positively and light scatter gates were analyzed for PE fluorescence. Ten thousand GPIb positive cells were analysed per sample, and the activationdependent binding of anti-CD62-PE or anti-CD63PE was expressed as percentage of platelets positive for these antibodies. The samples without agonist were used as negative control.

1.2.3. Flow cytometric measurements of platelet aggregation and PMP formation in whole blood Aggregations of platelets were performed in polypropylene tubes by using a Payton (Toronto, Ontario, Canada) dual channel aggregation module. Anticoagulated whole blood was preincubated with recombinant PDGF-BB or vehicle for 1 minute under stirring condition (500 rpm) at 378C, before stirring with various concentration of activating agents (thrombin, TRAP, or collagen). Aliquots of whole blood were fixed with 19 parts of 0.2% paraformaldehyde and then incubated with a saturating concentration of FITC-conjugated antihuman platelets chicken antibodies for 1 hour in the dark. Flow cytometric detection of platelet aggregation (single platelet disappearance) and PMP formation was performed essentially as described by Lindahl et al. [19] and Sims et al. [29], respectively. In brief, platelets and PMP were distinguished from electronic machine noise, erythrocytes, and white blood cells by gating on both FITC positively (FITC-conjugated anti-human platelet surface chicken antibodies) and light scatter profiles (Figure 1). Five thousand FITC-positive events were analysed per sample. PMP were discriminated from platelets on the basis of size (light scattering profile between 11 and 140 on Forward Side Scatter (FSC) scale). The degree of aggregation was calculated according to the formula: Aggregation (%)5(12NA/NC)3100

Where NA 5 Stirred agonist-stimulated platelet population and NC5Stirred unstimulated control platelet population. The statistical significance was determined by paired sample Student’s t-test.

2. Results 2.1. The Effect of PDGF on Platelet Aggregation and PMP Formation in Heparinized Whole Blood Heparinized whole blood was preincubated in the absence or the presence of human recombinant

PDGF-BB for 60 seconds under stirring conditions (500 rpm), before activation with increasing concentrations of thrombin (0–0.8 U/mL) or collagen (0–2.5 mg/mL) for 30–60 seconds. PDGF (100 ng/ mL) inhibited the thrombin-induced aggregation at 0.2 and 0.4 U/mL of thrombin by 31.968.7 (SEM) percent and 46.2613.2%, respectively (Figure 2A). Similarly, PDGF-BB inhibited the thrombininduced PMP formation at 0.2 and 0.4 U/mL of thrombin by 39.366.4% and 47.2615.1%, respectively (Figure 2B). An inhibitory effect of PDGF-BB on collageninduced platelet aggregation and PMP formation also was seen (Figures 3A and 3B). Preincubation with recombinant PDGF-BB inhibited collageninduced (0.625 mg/mL) aggregation by 62.6613.4%. Moreover, an inhibition of collagen-induced aggregation also was seen when PDGF-BB was added at the same time as collagen (45.766.0%, data not shown). A dose-response experiment with PDGF-BB showed that PDGF inhibited collagen-induced aggregation in a dose-dependent manner with maximal inhibition at 100 ng/mL PDGF (Figure 4). In agreement with the results obtained for thrombin above, 1.5 U of porcine PDGF-inhibited TRAP (SFLLRN)-induced platelet aggregation and PMP formation in stirred whole blood (Figure 5).

2.2. Flow Cytometric Measurement of P-Selectin and CD63 Surface Expression under Nonaggregating Conditions The above experiments (Figures 2–5) for heparinized whole blood and previously reported results with GFP [5] were obtained by stirring the samples, which allows both aggregation and secretion to occur. To examine whether platelet aggregation was required for the observed inhibitory effect of PDGF on platelet activation, we activated platelets in whole blood by TRAP (SFLLRN) in the absence of stirring. Under these conditions, platelets are activated in the physiological milieu of whole blood, without resulting in platelet-to-platelet adhesion. The surface expression of P-selectin and CD63 were low on unstimulated platelets (no added TRAP), being found on 2.010.1 (SEM) percent and 2.710.5% of platelets, respectively. A similar

F. Selheim et al./Thrombosis Research 95 (1999) 185–196

Fig. 2. Inhibition of thrombin-induced platelet aggregation and formation of PMP by PDGF in whole blood. Heparinized whole blood was preincubated with human recombinant PDGF-BB (100 ng/mL; filled squares) or vehicle (open squares) for 60 seconds at 378C under stirring conditions (500 rpm) before stirring with the indicated concentrations of thrombin for 60 seconds. Aggregation and PMP formation were detected in an FACSort flow cytometer. The percent platelet aggregation (A) was expressed as described in the Methods section. The amount of PMP formation (B) in the absence of thrombin was set at 100%. The data are average of six separate experiments performed in triplicate, six determinations at 0.4 U/mL of thrombin, and four determinations at 0.2 U/mL, 0.6 U/mL, and 0.8 U/mL of thrombin, respectively. Error bars represent 6SEM. p Values (vehicle1thrombin vs. PDGF 1 thrombin) were determined by paired samples Student’s t-test. *p,0.05 is considered to be statistically significant.

189

190

F. Selheim et al./Thrombosis Research 95 (1999) 185–196

Fig. 3. Inhibition of collagen-induced platelet aggregation and formation of PMP by PDGF in whole blood. Heparinized whole blood was preincubated with human recombinant PDGF-BB (100 ng/mL; filled circles) or vehicle (open circles) for 60 seconds at 378C under stirring conditions before stirring with different concentrations of collagen for 30 seconds. The amount of platelet aggregation (A) and PMP formation (B) were expressed as described in Figure 1. The data are average of four separate experiments performed in duplicate, four determinations at 0.625 mg/mL of collagen and three determinations at 1.25 mg/mL and 2.5 mg/mL of collagen, respectively. Error bars represent 6SEM. p values (vehicle1collagen vs. PDGF1collagen) were determined by paired samples Student’s t-test. *p,0.06 is considered to be statistically significant.

background expression of these markers was found with PDGF-BB alone, 2.360.2% (P-selectin) and 2.960.4% (CD63). TRAP-induced activation of platelets in heparinized whole blood increased the surface expression of P-selectin and CD63 in a dose-dependent

manner (Figure 6). However, under these nonaggregating conditions, PDGF-BB had no significant effect on platelet surface expression of P-selectin and CD63. In the experiment shown in Figure 6, the fixation was conducted 20 minutes after addition of PDGF-

F. Selheim et al./Thrombosis Research 95 (1999) 185–196

191

Fig. 4. Dose-dependent inhibition of collageninduced platelet aggregation by PDGF in whole blood. Heparinized whole blood was preincubated with various concentrations of recombinant PDGF-BB (25–200 ng/mL) for 60 seconds under stirring conditions before stirring with collagen (1.25 mg/mL) for 30 seconds. The data are representative of two separate experiments.

BB and TRAP. To examine whether a desensitisation of the inhibitory signal transduced through the PDGF a-receptor had occurred during this period, the blood samples were fixed (2% paraformaldehyde) 2 minutes after addition of PDGF-BB and

Fig. 5. Inhibition of TRAP (SFLLRN)-induced platelet aggregation and formation of PMP by PDGF in whole blood. Heparinized whole blood was preincubated with 1.5 U of porcine PDGF-BB, 1 U of PDGF being the amount of PDGF giving half-maximal DNA synthesis [21] or vehicle for 60 seconds at 378C under stirring conditions before stirring with 4 mM of TRAP for 30 seconds. The degree of platelet aggregation and PMP formation were expressed as described in Figure 1. The data are average of six separate experiments performed in triplicate. *p,0.05 for the differences between samples incubated with vehicle1TRAP vs. PDGF1 TRAP (paired samples Student’s t-test). Bars6SEM.

TRAP. The samples then were incubated with a saturating concentration of the platelet identifier anti-CD42b and the activation-dependent monoclonal antibody CD62-PE for 15 minutes before the flow cytometric analysis. Fixation 2 minutes

192

F. Selheim et al./Thrombosis Research 95 (1999) 185–196

Fig. 6. Effect of PDGF on TRAP (SFLLRN)induced surface expression of P-selectin and CD63. Heparinized whole blood was incubated with recombinant PDGF-BB (100 ng/ mL; filled circles) or vehicle (open circles), anti-CD42b-FITC, anti-CD62-PE or antiCD63-PE, and with the indicated concentrations of TRAP. The samples were incubated at 378C for 20 minutes without stirring and then fixed with 0.2% paraformaldehyde. The activation-dependent binding of antiCD62-PE or anti-CD63 was detected by flow cytometry as described in Materials and Methods. The measurements of surface expression are representative of 11 separate experiments performed in triplicate for P-selectin and of five separate experiments for the lysosomal marker CD63. Error bars represent 6SEM.

after addition of agonists resulted in a slightly lower P-selectin expression, but there were still no differences in the surface expression with or without the presence of PDGF-BB in the samples (data not shown). Since the effects of thrombin and TRAP on platelets are not always comparable [30], we performed studies by using thrombin and GPRP in the flow cytometric assay of P-selectin. In agreement with the present results for TRAP under nonaggregating conditions (Figure 6), 100 ng/mL PDGF-BB had no significant effect on thrombininduced platelet surface expression of P-selectin (Figure 7).

3. Discussion We previously have demonstrated the presence of PDGF a-receptors (PDGFR-a) but not PDGF b-receptors in human platelets, and that addition of exogenous PDGF to GFPs inhibits thrombininduced platelet aggregation and secretion under aggregating conditions [5]. In the present study,

we show that exogenous PDGF inhibits thrombin-, TRAP-, and collagen-induced platelet aggregation in the more physiologic milieu of heparinized whole blood. In addition, we found that PDGF inhibits shedding of PMP from the platelet plasma membrane during the aggregation of whole blood. Interestingly, the extent of the PDGF inhibition (in percentages) on platelet aggregation corresponded to the extent of PDGF inhibition on PMP formation. For example, PDGF inhibits the aggregation at 0.625 mg/mL of collagen by 62.6% and the corresponding extent of PMP formation with 67.2% at the same collagen concentration (Figures 3A and 3B). Similar agreement in the extent of the PDGF inhibition between aggregation and PMP formation is seen for thrombin (Figures 2A and 2B). However, under nonaggregating conditions (no stirring), we could not demonstrate any significant effect of PDGF-BB on TRAP-induced platelet surface expression of P-selectin and CD63. Similarly, we could not demonstrate any effect of PDGF on thrombin-induced surface expression of P-selectin. Thus, the dimeric PDGF protein only seems to inhibit platelet activation in association with plate-

F. Selheim et al./Thrombosis Research 95 (1999) 185–196

193

Fig. 7. Effect of PDGF on thrombin-induced surface expression of P-selectin. Heparinized whole blood was incubated with recombinant PDGF-BB (100 ng/mL; filled circles) or vehicle (open circles), anti-CD42b-FITC and antiCD62-PE, and with the indicated concentrations of thrombin and 1.5 mM GPRP. The samples were incubated at 378C for 20 minutes without stirring and then fixed with 0.2% paraformaldehyde. The activation-dependent binding of anti-CD62-PE was detected by flow cytometry as described in Materials and Methods. The measurements of surface expression are representative of four separate experiments performed in triplicate, four determinations at 0.2 U/mL, 0.4 U/mL 0.6 U/mL and 0.8 U/mL, respectively, and two determinations at 1.0 U/mL. Error bars represent 6SEM.

let aggregation but has no effect on surface expression of granule-located markers when aggregation is avoided in the absence of stirring. Similar effects on aggregation and secretion previously have been reported with the pharmacological phosphoinositide 3-kinase (PI 3-K) inhibitors wortmannin and LY294002 [31]. This investigation showed that these PI 3-K inhibitors inhibited TRAP-induced platelet aggregation and activation of the integrin glycoprotein IIb-IIIa. However, in parallel to our results with PDGF, these PI 3-K inhibitors did not affect TRAP-induced P-selectin surface expression (no stirring). Integrin receptors play a crucial role in regulation of cell migration by acting as a structural link between the extracellular matrix and the actin cytoskeleton [32]. Interestingly, the PDGF a-receptor has been shown to inhibit the chemotactic response induced by the PDGF b-receptor in human fibroblasts [33,34] and vascular smooth muscle cells [35]. Intracellular signal transduction for PDGF b-receptor induced chemotactic response has been shown to be dependent on PI 3-K activity [36], and it recently has been suggested that the PDGF a-receptor inhibits chemotaxis downstream of the PI 3-K pathway [37]. Since platelet activation and chemotaxis both involve cell shape changes, it is possible that these inhibitory signal transduction

pathways are related. Interestingly, PI 3-K inhibitors are reported to inhibit platelet activation [31]. PDGF-induced inhibition of platelet activation appears to be restricted to conditions where platelets are allowed to aggregate. Platelet-to-platelet adhesion (close cell contact) is known to activate platelets and thus can be regarded as part of the autocrine activation machinery of platelets. Our results suggest that PDGF may inhibit platelet activation by inhibition of this autocrine pathway. Local concentrations of PDGF are likely to be high close to the platelet membrane after activation and a-granula release. However, estimates of this concentration have to be speculative, only. We thus believe that inhibitory effect of exogenous PDGF at 25–100 ng/mL is achieved in vivo due to the local effect and the fact that total PDGF release during clotting of blood gives a concentration approximately 50 ng/mL in serum [12]. We previously have shown that PDGF antibodies potentiate platelet activation [5] demonstrating that endogenous and local PDGF effects do occur. PDGF generally is recognised as important for cell growth in atherosclerotic lesions [6], and PDGF-neutralising agents are found to inhibit restenosis after balloon catether deendothlialization of the carotid artery [38]. Moreover, in vascular smooth muscle cells and in monocytes [39,40] PDGF-BB has been shown to

194

F. Selheim et al./Thrombosis Research 95 (1999) 185–196

induce expression of tissue factor, a transmembrane glycoprotein that forms complex with factor VIIa and thereby initiates the coagulation cascade. These processes involve long-term effects dependent on chemotaxis and cell proliferation and are different from our studies on the direct effect of platelet activation. Thus, our observations suggest that PDGF may influence thrombosis and atherosclerosis in a complex manner. Concerning the above discussion, it is interesting that the involvement of PDGF in proliferative disease states, for example, during the restenosis process, is likely to be exerted primarily via b-receptors on smooth muscle cells [41]. Since an autocrine inhibition pathway is transduced through the human PDGF a-receptors, use of PDGF-neutralising agents that inhibits PDGF-BB and PDGF-AB but not PDGFAA, as described by Green et al. [41] might have the desired properties of a PDGF antagonist to prevent restenosis. We conclude that PDGF has antithrombotic activity under the more physiological conditions in whole blood than in GFP.

6.

7.

8.

9.

This study was supported by a grant from the Norwegian National Health Association. We thank Harry Mikkelsen for stimulating discussions. We also thank Ingrid Strand and Anita Rynningen for introducing us to flow cytometry.

10.

References 1. Heldin CH, Westermark B. Platelet-derived growth factor: Mechanism of action and possible in vivo function. Cell Regul 1990;1:555–66. 2. Bar RS, Boes M, Booth BA, Dake BL, Henley S, Hart MN. The effects of platelet-derived growth factor in cultured microvessel endothelial cells. Endocrinology 1989;124:1841–8. 3. Smits A, Hermansson M, Nister M, Karnushina I, Heldin CH, Westermark B, Funa K. Rat brain capillary endothelial cells express functional PDGF B-type receptors. Growth Factors 1989;2:1–8. 4. Smits A, Kato M, Westermark B, Nister M, Heldin CH, Funa K. Neurotrophic activity of platelet-derived growth factor (PDGF): Rat neuronal cells possess functional PDGF betatype receptors and respond to PDGF. Proc Natl Acad Sci USA 1991;88:8159–63. 5. Vassbotn FS, Havnen OK, Heldin CH, Holm-

11.

12.

13.

14.

15.

sen H. Negative feedback regulation of human platelets via autocrine activation of the platelet-derived growth factor alpha-receptor. J Biol Chem 1994;269:13874–9. Raines EW, Ross R. Platelet-derived growth factor in vivo. In: Westermark B, Sorg C, eds. Biology of platelet-derived growth factor. Basel, Switzerland: S. Karger; 1993:74–115. Yarden Y, Escobedo JA, Kuang WJ, Yang Feng TL, Daniel TO, Tremble PM, Chen EY, Ando ME, Harkins RN, Francke U, Friend VA, Ullrich A, Williams LT. Structure of the receptor for platelet-derived growth factor helps define a family of closely related growth factor receptors. Nature 1986;323:226–32. Matsui T, Heidaran M, Miki T, Popescu N, LaRochelle W, Kraus M, Pierce J, Aaronson S. Isolation of a novel receptor cDNA establishes the existence of two PDGF receptor genes. Science 1989;243:800–4. Claesson Welsh L, Eriksson A, Westermark B, Heldin CH. cDNA cloning and expression of the human A-type platelet-derived growth factor (PDGF) receptor establishes structural similarity to the B-type PDGF receptor. Proc Natl Acad Sci USA 1989;86:4917–21. Kaplan DR, Chao FC, Stiles CD, Antoniades HN, Scher CD. Platelet alpha granules contain a growth factor for fibroblasts. Blood 1979; 53:1043–52. Witte LD, Kaplan KL, Nossel HL, Lages BA, Weiss HJ, Goodman DS. Studies of the release from human platelets of the growth factor for cultured human smooth muscle cells. Circ Res 1978;42:402–9. Bryckaert MC, Rendu F, Tobelem G, Wasteson A. Collagen-induced binding to human platelets of platelet-derived growth factor leading to inhibition of P43 and P20 phosphorylation. J Biol Chem 1989;264:4336–41. Bryckaert MC, Rendu F, Tobelem G, Caen J. PDGF modifies phosphoinositide metabolism and inhibits aggregation and release in human platelets. Biochem Biophys Res Commun 1986; 135:52–7. Holmsen H. Significance of testing platelet functions in vitro. Eur J Clin Invest 1994;24 (Suppl 1):3–8. McEver RP. Properties of GMP-140, an induc-

F. Selheim et al./Thrombosis Research 95 (1999) 185–196

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

ible granule membrane protein of platelets and endothelium. Blood Cells 1990;16:73–80. Nieuwenhuis HK, van Oosterhout JJ, Rozemuller E, van Iwaarden F, Sixma JJ. Studies with a monoclonal antibody against activated platelets: Evidence that a secreted 53,000molecular weight lysosome-like granule protein is exposed on the surface of activated platelets in the circulation. Blood 1987;70:838–45. Shattil SJ, Cunningham M, Hoxie JA. Detection of activated platelets in whole blood using activation-dependent monoclonal antibodies and flow cytometry. Blood 1987;70:307–15. Abrams CS, Ellison N, Budzynski AZ, Shattil SJ. Direct detection of activated platelets and platelet-derived microparticles in humans. Blood 1990;75:128–38. Lindahl T, Kra˚gedal B, Larsson A. A new method for measurement of platelet concentration utilising flow cytometry [abstract]. In: XXIV Nordic Congress in Clinical Chemistry. Stockholm, Sweden. Haematology and Coagulation 1994;54:60. Vassbotn FS, Skar R, Holmsen H, Lillehaug JR. Absence of PDGF-induced, PKC-independent c-fos expression in a chemically transformed C3H/10T1/2 cell clone. Exp Cell Res 1992;202:98–104. Holmsen H, Male R, Rongved S, Langeland N, Lillehaug JR. Platelet-derived-growth-factor stimulated heterogeneous polyphosphoinositide metabolism and phosphate uptake in C3H fibroblasts. Biochem J 1989;260:589–92. Schlossman SF, Boumsell L, Giks W, Harlan JM, Kishimoto T, Morimoto C, Ritz J, Shaw S, Silverstein RL, Springer TA, Tedder TF, Todd RF. CD antigens 1993. Blood 1994;83: 879–80. Stenberg PE, McEver RP, Shuman MA, Jacques YV, Bainton DF. A platelet alphagranule membrane protein (GMP-140) is expressed on the plasma membrane after activation. J Cell Biol 1985;101:880–6. Hsu-Lin S-C, Berman CL, Furie BC, August D, Furie B. A platelet membrane protein expressed during platelet activation and secretion. Studies using monoclonal antibody specific for thrombin-activated platelets. J Biol Chem 1984;259:9121–6. Israels SJ, Gerrard JM, Jacques YV, McNicol

26.

27.

28.

29.

30.

31.

32.

33.

34.

195

A, Nishibori M, Baintaon DF. Platelet dense granule membranes contain both granulphysin and P-selectin (GMP-140). Blood 1992;80: 143–52. Nishibori M, Chem B, McNicol A, Shalev A, Jain N, Gerrard JM. The protein CD63 is in platelet dense granules, is deficient in a patient with Hermansky-Pudlak syndrome, and appears identical to granulphysin. J Clin Invest 1993;91:1775–82. Mirlashari MR, Ryningen A, Mikkelsen HM, Fukami MH. Differential secretion of blood platelet storage granules. Platelets 1996;7: 313–20. Michelson AD, Ellis PA, Barnard MR, Matic GB, Viles AF, Kestin AS. Downregulation of the platelet surface glycoprotein Ib-IX complex in whole blood stimulated by thrombin, adenosine diphosphate, or an in vivo wound. Blood 1991;77:770–9. Sims PJ, Faioni EM, Wiedmer T, Shattil SJ. Complement proteins C5b-9 cause release of membrane vesicles from the platelet surface that are enriched in the membrane receptor for coagulation factor Va and express prothrombinase activity. J Biol Chem 1988;263: 18205–12. Brass LF. Issues in the development of thrombin receptor antagonists. Thromb Haemost 1995;74:499–505. Kovacsovics TJ, Bachelot C, Toker A, Vlahos CJ, Duckworth B, Cantley LC, Hartwig JH. Phosphoinositide 3-kinase inhibition spares actin assembly in activating platelets but reverses platelet aggregation. J Biol Chem 1995;270:11358–66. Hynes RO. Integrins: Versatility, modulation, and signaling in cell adhesion. Cell 1992;69: 11–25. Vassbotn FS, Ostman A, Siegbahn A, Holmsen H, Heldin CH. Neomycin is a platelet-derived growth factor (PDGF) antagonist that allows discrimination of PDGF alpha- and beta-receptor signals in cells expressing both receptor types. J Biol Chem 1992;267:15635–41. Siegbahn A, Hammacher A, Westermark B, Heldin CH. Differential effects of the various isoforms of platelet-derived growth factor on chemotaxis of fibroblasts, monocytes, and granulocytes. J Clin Invest 1990;85:916–20.

196

F. Selheim et al./Thrombosis Research 95 (1999) 185–196

35. Koyama N, Morisaki N, Saito Y, Yoshida S. Regulatory effects of platelet-derived growth factor-AA homodimer on migration of vascular smooth muscle cells. J Biol Chem 1992; 267:22806–12. 36. Wennstrom S, Siegbahn A, Yokote K, Arvidsson AK, Heldin CH, Mori S, Claesson Welsh L. Membrane ruffling and chemotaxis transduced by the PDGF beta-receptor require the binding site for phosphatidylinositol 39 kinase. Oncogene 1994;9:651–60. 37. Yokote K, Mori S, Siegbahn A, Ronnstrand L, Wernstedt C, Heldin CH, Claesson Welsh L. Structural determinants in the platelet-derived growth factor alpha-receptor implicated in modulation of chemotaxis. J Biol Chem 1996; 271:5101–11. 38. Ferns GA, Raines EW, Sprugel KH, Motani

AS, Reidy MA, Ross R. Inhibition of neointimal smooth muscle accumulation after angioplasty by an antibody to PDGF. Science 1991;253:1129–32. 39. Taubam MB, Marmur JD, Rosenfield CL, Guha A, Nichtberger S, Nemerson Y. Agonistmediated tissue factor expression in cultured vascular smooth muscle cells. J Clin Invest 1993;91:547–52. 40. Ernofsson E, Siegbahn A. Platelet-derived growth factor-BB and monocyte chemotactic protein-1 induce human peripheral blood monocytes to express tissue factore. Thromb Res 1996;83:307–20. 41. Green LS, Jellinek D, Jenison R, Ostman A, Heldin CH, Janjic N. Inhibitory DNA ligands to platelet-derived growth factor B-chain. Biochemistry 1996;35:14413–24.