Negative Feedback Regulation of Human Platelets via Autocrine ...

THEJOURNAL OF Blo~oc~ciu. CHEMISTRY 0 1994 by The American Society for Biochemistry and Molecular Biology, Inc.

Vol. 269, No. 19, Issue of May 13, pp. 13874-13879, 1994 Printed in U.S.A.

Negative Feedback Regulationof Human Platelets via Autocrine Activation of the Platelet-derived Growth Factor a-Receptor* (Received for publication, January 3, 1994, and in revised form, March 7, 1994)

Flemming S . VassbotnlOn, Ole Kristian HavnenPII, Carl-Henrik Heldinl,and Holm HolmsenO From the $Ludwig Institute for Cancer Research, Biomedical Center, S-751 23 Uppsala, Sweden a n d the §Department of Biochemistry and Molecular Biology, University of Bergen, N-5009Bergen, Norway

Human platelets contain platelet-derived growth factor (PDGF) in theira-granules which is released during platelet exocytosis. We show by immunoprecipitation and 12'I-PDGF binding experiments that human platelets have functionally activePDGF a-receptors, but not P-receptors. ThePDGF a-receptor (PDGFR-a) wasidentified as a 1'70-kDaglycosylated protein-tyrosine kinase as found in other cell types. Stimulation of platelets with 0.1 unit/ml thrombin resulted in a significant increase (2-&fold) of the tyrosine phosphorylation of the PDGFR-a, as determined by immunoprecipitation with phosphotyrosine antiserum as well as with PDGFR-a antiserum. The observed thrombin-induced autophosphorylation of the PDGFR-a wasinhibited by the addition of a neutralizing monoclonal PDGF antibody. Thus, our results suggest that theplatelet PDGFR-a is stimulated in an autocrine manner by PDGF secreted during platelet activation. Preincubation of platelets with PDGF inhibited thrombin-induced platelet aggregation and secretion of ATP + ADP and j.3-hexosaminidase.Thrombin-induced platelet aggregationwasalsoreversedwhen PDGF was added 30 s after thrombin stimulation. Inhibition of the autocrine PDGF pathway during platelet activation by the PDGF antibody led to a potentiation of thrombin-induced P-hexosaminidasesecretion.Thus, the PDGFR-a takes part in a negative feedbackregulation during platelet activation. Our demonstration of PDGF a-receptors on human platelets and its inhibitory function during platelet activation identifies a new possible role of PDGF in the regulation of thrombosis.

gesting normal physiological roles of PDGF during wound healing, placental growth, early embryogenesis, as well as during glial cell differentiation ( 5 ) . Structurally, PDGF i s a disulfide-linked dimer composed of two homologous polypeptide chains denoted A and B (1). The PDGF isoforms, PDGF-AA, - A B , a n d -BB, mediate their effects by binding to two protein-tyrosine kinase receptors (PDGFR), denoted the a- and P-receptors (6-8). Ligand binding induces receptor dimerization and autophosphorylation, which correlates with activation of the receptor kinases (9-11). The a-receptor binds both PDGF A- and B-chains with high affinity, whereas the P-receptor binds only the PDGF B-chain. Both receptor types mediate stimulation of DNA synthesis and mitosis (1).Interestingly, the PDGF P-receptor and a-receptor have agonistic and antagonistic effects, respectively, on the chemotactic response of human fibroblasts (12, 13). Human platelets contain all three PDGF isoforms, mainly as PDGF-AB heterodimers (14, 15). PDGF is an a-granule component (16) a n d is secreted during the platelet exocytosis induced by thrombin, collagen, or ADP (17). Human platelets have been shown previously to bind lZ5I-PDGF,and the addition of exogenous PDGF was found to modulate collagen-induced platelet responses(18). We have therefore examinedthe possibility that human platelets have functional PDGF receptors that can be activated by PDGF secreted during platelet activation. Our results demonstrate the presence of PDGF a-recepin human platelets. tors (PDGFR-a) but not PDGF p-receptors Activation of the platelets by thrombin resulted in activation andtyrosinephosphorylation of PDGFR-a in an autocrine manner. The observed autophosphorylation of the PDGFR-a was independent of exogenous PDGF and the autocrine recepthe addition of neutralizing tor activation was inhibited by Platelet-derived growth factor (PDGF)' is a potent mitogen PDGF antibodies. Moreover, the addition of exogenous PDGF and chemotactic factor for mesenchymally derived cells (reor PDGF antibodies inhibitedor potentiated thrombin-induced viewed in Ref. 1). Receptors for PDGF have recently also been platelet responses, respectively. found on other cell types, e.g. capillary endothelial cells (2, 3) and neuronal cells (4). Several observations suggest that PDGF MATERIALS AND METHODS may be involved in disorders with excessive cell proliferation, Cell Culture and Platelet Isolation"AG1518 human foreskin fibroconsuch as cancer, atherosclerosis, and chronic inflammatory ditions (reviewed in Ref. 5 ) . Evidence has been presented sug- blasts, passage 15-30 (Human Genetic Mutant Cell Repository, Camden, NJ), were cultured in Eagle's minimum essential medium supplemented with 10%fetal calf serum, 100 units/ml penicillin, and 50 pdml streptomycin in humidified 5% CO,, 95% air at 37 "C. * This work was supported in part by the Blix' Foundation. The costs Platelets from human blood, anticoagulated with acid citrate dexof publication of this article were defrayed in part by the payment of trose, were transferred by gel-filtration (Sepharose CL-2B; Pharmacia page charges. This article must therefore be hereby marked "aduertise- LKB Biotechnology Inc.) into a calcium-free Tyrode's buffer (pH 7.3) rnent" in accordance with 18 U.S.C. Section 1734 solelyto indicate this containing 0.2% bovine serum albumin (Miles Laboratories) and 5 mM fact. glucose as described previously (19). The platelet concentration in gel9 Supported by a grant from the Norwegian CancerSociety. To whom correspondenceshould be addressed Dept. of Biochemistry and Molecu- filtered platelets (GFP)was then adjusted to 3.5 x 10' cellslml as measured by a Coulter counter. Porcine PDGF was prepared as described lar Biology, University of Bergen, N-5009 Bergen, Norway. I( Supported by a grant from the Norwegian Research Council for (20). Antibodies-ThemonoclonalPDGFantibodymAb 6Dll (subclone Science and Humanities. The abbreviations used are: PDGF, platelet-derived growth factor; 6Dll El, kindly provided by Dr. Inger Hagen, Center for Industrial GFP, gel-filtered platelets; PBS, phosphate-buffered saline; PDGFR, Research, Oslo, Norway), neutralizes the biological activity of platelet-derived growth factor receptor(s1;mAb 6Dl1, monoclonal anti- PDGF-BB and -AB (21), and also cross-reacts with PDGF-AA (22). The rabbit antisera PDGFR-3and PDGFR-7 wereraised against peptides in body against PDGF.

13874

13875

Antithrombotic Activity of the PDGF a-Receptor stimulation:

FIG.1. H u m a n platelets have functionally activePDGF a-receptors.Human platelets (left panel) or fibroblasts (right panel)were incubated withporcine PDGF or recombinant human PDGF-BB (100ng/ml) or vehicle. Cell lysates were immunoprecipitatedwithcontrolantiserum (Ctr.)or antiserum specific for the PDGF a- or P-receptors (a and P, respectively), and the beads were subjected to an in vitrokinaseassay by incubation with [y-"PIATP; samples were analyzed by SDS-gel electrophoresis followed by exposure using PhosphorImager. a The PDGF a-receptor (170 kDa) and 6-receptor (180 kDa) bands are indicated by arrows, to the left and to the right, respectively.

n

E a

800

antiserum:

Ctr.

a

r porclne PDGF 1 r PDGF-BE 1 P Ctr. a P Ctr. a P

1 1 -200

-

c

- 97 human platelets

human fibroblasts

€-

V

v

v

5

600

0

n m m I

400

LL (3

n a

vl'-

thrombin (0.1 U/ml)

200

N

0

competition with:

-

AA

BB

FIG.2. Cross-competition for binding of lZ5I-PDGF-BBto hum a n platelets. The bindingof '"I-PDGF-BB to human platelets in the absence (-1 or presence of 200 ng/ml unlabeled PDGF-AA (AA) or PDGF-BB (BB)a t for 2 h a t 0 "C was determined. The data represent total cell-associated ''I radioactivity -c 1 S.D., n = 4. the C termini of the murine PDGFR-P (23) and the human PDGFR-a (8), respectively. Therabbitphosphotyrosineantiserum(anti-PY) FIG.3. Immunostaining of PDGF on resting and thrombin-ac(24) was affinity-purified using a phosphotyrosine-Sepharose CL-4B tivated platelets. Human platelets were incubated with vehicle or 0.1 unitlml thrombin, stained with the monoclonal anti-PDGF antibody column. '2,51-PDGF-BBBinding Experiments-Human recombinant PDGF- 6Dl1, and visualized with fluorescein isothiocyanate-conjugated rabbit BB expressed in yeast was labeled with lZ5I, according to the Bolton- anti-mouse antibodies. Resting platelets were also stained with 6Dll Hunter procedure, to specific activities of 12,000-23,000 c p d n g (25). after preincubation of the antibodies with 400 ng/ml PDGF-BB (control ). Aliquots of 0.5 ml of GFP were incubated with 1 ng/ml '"I-PDGF-BB with or without 200 ng/ml unlabeled PDGF-AA or PDGF-BB for 2 h on CL-4B (Pharmacia LKB Biotechice. The platelets were then centrifuged for 10 min at 2,800 x g and incubation time). Protein A-Sepharose washed four times withice-cold PBS (136.7 mM NaCl, 2.7 mM KCI, 13.1 nology Inc.) was used to collect the immune complexes. The immunolysis and once with 20 mM mM Na,HPO,, 1.5 mM KH,PO,) containing 1mg/ml bovine serum albu- precipitates were washed four times in buffer Tris-HCI (pH 7.5), 150 mM NaC1. Kinase assays wereperformed in 35 pl min, 0.9 mM CaCI,, and 0.5 mM MgCI,. The platelet pellets were from cut a of 20 mM HEPES (pH7.5). 10 mhl MnCI,, 1mhl dithiothreitol, containing the Eppendorf,tube and the ''.'I radioactivitywasmeasuredin 10 pCi of [y-:"PIATP (Amersham; specific activity, 3000 Ci/mmol) for 30 y-counter. Immunohistochemistry-GFP (0.5 ml) was incubated with or without min a t room temperature (26). Immune complexes were eluted by heat0.1 unitlml thrombin for 60 s a t 37 "C and fixed by addition of 50 mM ing the beadsa t 95 "C for 4 min in 35 pl of 2 x sample buffer (4%SDS, formaldehyde containing 4mM EDTA. The platelets were transferred to 2 8 mercaptoethanol, 0.2 M Tris-HCI (pH 8.8), 0.5 M sucrose, 5 mht EDTA, slides coated with0.1% polylysine. The PDGF antibodymAb 6Dll was 0.01% bromphenol blue). Samples were subjected to SDS-gel electroadded to parallel sections overnighton ice. To assure the specificity of phoresis in 7.5% acrylamide gels; gels were then fixed, dried, and ana6Dl1, the antibody solutions were preincubated overnight on ice with or lyzed by a PhosphorImager (Molecular Dynamics)for measurement of the relative amount of radioactivity in the receptor bands. without 400ng/ml PDGF-BB. The slides were then washed in PBS and incubated with secondary fluorescein-conjugated rabbit anti-mouse im- Platelet Aggregation and Secretion-GFP were pretreatedwith PDGF, PDGF antibodies, or vehicle for 60 s a t 37 "C before incubation munoglobulin (dilution 1:3000 in PBS) for 2 h. In Vitro Immune Complex Kinase Assay-Human platelets and sub- with thrombin(0.04-0.3 unitlml final concentration)or vehicle for 60 s. confluent human fibroblasts weresolubilized in lysis buffer (0.5M NaCI, After 60-s incubation with thrombin, sampleswere withdrawn for sub0.02 M Tris-HC1 (pH 7.4), 0.5% Triton, 1%Trasylol, 1 mM phenylmeth- sequent analysis. Platelet aggregation was measured in a Payton dual ylsulfonyl fluoride, 1 mM dithiothreitol) for 20 min a t 4 "C. The lysates channelaggregation module connected toaLineartwo-channelrewere then centrifuged a t 10,000 x g for 15 min, and the supernatants corder and quantified by the initial increase in light transmission. were subjected to immunoprecipitations with PDGF receptor specific Dense granule secretion was determined as the extracellular appearance of ADP plus ATP, measured by a luciferin-luciferasemethod (27). antiserum, affinity-purified anti-PY, or control preimmune serum (2 h

13876

of the PDGF a-Receptor

Antithrombotic Activity

A 3500 -

T

3000 2500 -

2000

-

T

1500 -

FIG.4. Autocrine activation of PDGF a-receptors during platelet activation. Human platelets were preincubated with recombinant PDGF-BB (PDGF) or 6 D l l antibodies (monoclonal anti-PDGF) for 1 min before incubation with thrombin for 5 min a t 37 "C. Cell lysateswereimmunoprecipitatedwith anti-phosphotyrosine ( A ) or PDGFR-a antiserum ( B ) and subjected to in vitro kinase assayby incubation with [y32PlATP, before analysis by SDS-gel electrophoresis and autoradiography. A, the relative radioactivity of the PDGF a-receptor bands as measured by PhosphorImager analysesafter SDS-gel electrophoresis. Bars illustrate ? 1 S.D. B , immunoprecipitation of treated platelets (left panel) and control fibroblasts (right panel)with PDGFR-a antiserum. The relative radioactivity of the a-receptor bands(arrow)is indicated on the base line. The data are representative of three different experiments.

1000 -

500

-

preincubation: ctr. ctr. thrombin U/ml: ctr.

B preincubation: thromMn Ulml:

Ctr.

0.05

0.1

monoclonal

r anti-PDGF 1

PDGF

Ctr.

anti-PDGF ctr. ctr. 0.05 0.1

PDGF

0.05

0.1

0.05

0.1

PDGF Ctr.

Mr

(X163

a

I"-+ 1

- 97ralativr radioactlvlty (x of Ch):

_r

100

235

143

204

116103

human platelets

100

2109

human fibroblasts

Secretion of acid hydrolases from lysosomes was measured fluorometri- probably due to dimerization of the receptors by the receptor cally by the extracellular concentrationof P-hexosaminidase (28). antiserum (26). Notably, the 32Pcontent of the PDGFa-receptor

from platelet lysates increasedapproximately 200% on incubation with porcine or recombinant human PDGF-BB (Fig. 1). HumanPlatelets Have PDGF a- but Not P-ReceptorsThus, resting human platelets have PDGF a-receptors which Humanplateletshave beenshownpreviously to bind "'Ican be activated by exogenous PDGF. PDGF (18).To determine whether platelets have functional To investigate further the presence of PDGF receptors on PDGF receptors, human platelets, or human fibroblasts as a human platelets, cross-competition experimentswith 1251positive control, were incubated with porcine PDGF, recombiPDGF-BB and unlabeled PDGF-AA or -BB (200 ng/ml) were nant PDGF-BB (100 ng/ml), or vehicle for 30 min on ice. Cell lysates were immunoprecipitated with PDGFreceptor-specific performed. Approximately 40% of the total platelet-associated "'1 radioactivity was specific binding of '"I-PDGF-BB. antisera or preimmune serum, washed, and incubated with [y-32P]ATPbefore SDS-gel electrophoresis. From both human PDGF-AA was found to completely block the specific binding of platelets and human fibroblasts thePDGFR-a antiserum pre- '"I-PDGF-BB (Fig. 2). Thus, in agreement with the in vitro cipitated a 170-kDa 32P-phosphorylated component correspond- kinase assay (Fig. 11, the binding experiments showed an exing to the mature glycosylated PDGF a-receptor(Fig. 1,arrows clusive presence of the PDGF a-receptor on human platelets. Relocalization of PDGF to thePlatelet Membrane afterPlateto the left 1. The PDGF P-receptor was identified in the human fibroblasts as a 180-kDa protein (Fig. 1,arrow to the right) but let Activation-PDGF is secreted during plateletactivation (17) was not found in the platelets. In agreement with previous and may thus bind to the membrane PDGF a-receptors after observations, the PDGF receptor bands contained a relative externalization. We therefore performed immunohistochemical high amount of radioactivity even in the absence of ligand, stainings of resting orthrombin-activated platelets with a RESULTS

Antithrombotic Activity

of

the PDGF a-Receptor

13877

monoclonal antibody (mAb 6Dll) against PDGF (21). The mAb 6 D l l gave a granular staining of the resting platelets(Fig. 31, consistent with theknown localization of PDGF in the a-granules. In contrast, the thrombin-activated platelets appeared in clusters, and the PDGF-positive staining was localized extracellularly to theplateletmembrane (Fig. 3).Thegranular staining of resting platelets was blocked by preincubation of the antibody with400 ng/ml of PDGF-BB (Fig. 3). Autocrine Activationof the PDGF a-Receptors during Platelet Activation-The association of PDGF to the membraneof activatedplateletsmayindicate that the endogenous platelet PDGF binds and stimulates the PDGF a-receptor in an auto.--w crine manner after release from the a-granule.To examine this possibility, platelets were preincubated with PDGF-BB (100 0 ng/ml), mAb 6 D l l anti-PDGF (5 pg/ml), or vehicle for 1 min a, m -U and then thrombin(0.05 and 0.1 unit/ml) or vehicle was added n v for a further 5 min at 37"C. For comparison, human fibroblasts I-2 were incubated in parallel with or without PDGF. The cells a + rno were lysed, immunoprecipitatedwith phosphotyrosine anti3 serum (anti-PY, Fig.4A)or withPDGFR-a antiserum (Fig. 4 B ) a x and incubated with[-y-32P]ATPbefore SDS-gel electrophoresis. a The relative amountof radioactivity ineach receptor band was n Q determined by a PhosphorImager. As shown in Fig. 4A, PDGF increased the tyrosine phosphorylation of a componentin platelets of the samesize as the PDGF a-receptor by approximately c .--w0 250%. Interestingly, thrombin also increased thephosphorylaa," tion of the putative PDGF a-receptor a indose-dependent manL O 0 + ner; at 0.1 uniffml thrombin increased thephosphorylation by a,o m + 524% of control. The notion that thephosphorylated component .LC immunoprecipitated with anti-PY antiserum was identical to x 0 the PDGF a-receptor was supported by the finding that similar results were obtainedby immunoprecipitation with PDGFR-a I Qa antiserum (Fig. 4B1. Moreover, the thrombin-induced phospho1 I rylation of the PDGFR-a, as immunoprecipitated with anti-PY 0.00 0.05 0.10 0.15 0.20 0.25 0.30 as well as PDGFR-a antiserum, was completely inhibited by thrombin (U/ml) preincubation of the platelets with neutralizing PDGF antibodFIG.5. Inhibition of thrombin-induced platelet activation by ies (Fig. 4). This strongly suggests that the activation of the PDGFR-a during thrombin-stimulation was due to autocrine PDGF. Human platelets were preincubated withvehicle (open circles) or porcine PDGF (closed circles) for 60 s before incubation with different activation by endogenous PDGFsecreted from theplatelet concentrations of thrombin. Aggregation was measured in a Payton a-granules. Thrombin also increased the phosphorylation of the aggregation module ( A ) ,and the samples were analyzed for ATP plus ADP ( B ) and P-hexosaminidase (C) secretion after 60 s a t 37 "C. The PDGFR-ainplateletspreincubatedwithPDGF(datanot data are representative of six separate experiments made from different shown) donors;bursillustrate S.E. Thesecretionispresented as percent PDGF Znhibits Thrombin-induced Platelet Activation-The secretion of total platelet content. effects of activation of thePDGFa-receptor by exogenous PDGF on thrombin-induced platelet aggregation andsecretion were examined. Plateletswerepreincubatedwith porcine a negative feedback mechanism by inhibiting platelet activaPDGF or vehicle for 60 s and further with differentconcentra- tion. Platelets were preincubated with PDGFantibodies for 60 tions of thrombin (0-0.3 uniuml), samples were thenanalyzed s before thrombin incubation (0.01-0.2 uniffml), and secretion for secretion from dense granules or lysosomes after 60 s. As of P-hexosaminidase was measured. Thethrombin-induced lyshown in Fig. 5 , PDGF inhibitedboth thrombin-induced aggre- sosomal secretion was significantly potentiated by the PDGF gation and platelet secretion ofATPplus ADP fromdense gran- antibodies; at 0.3-0.4 uniffml thrombin,the mAb 6 D l l inules and @-hexosaminidase from lysosomes at thrombin con- creased the secretion by 600% (Fig. 7A). Platelets were also centrations less than 0.1 uniffml. Similar effects of PDGF on preincubated with different concentrations of PDGF or PDGF platelet secretion was also observed after 30- and 60-s incuba- antibodies before incubation with 0.05 unit/ml thrombin. As tion with thrombin (data not shown). Since PDGF is released shown inFig. 7B, exogenous PDGF-BB a t 100 ng/ml inhibited from activated platelets, the effect of PDGF on thrombin-in- secretion of @-hexosaminidaseby 50%. In contrast, the PDGF duced aggregation wasalso studied by addition of PDGF before antibodies (2 pg/ml) increased the thrombin-induced secretion or after thrombin (0.1 unit/ml) incubation. PDGF was found to by more than 200%. Thus, the autocrine PDGF pathway during reverse the platelet aggregation when added s30 after incuba- platelet activation most likely function in a negative feedback regulation of platelet activation. tion with thrombin (Fig. 6). Neutralizing PDGFAntibodies Potentiate Thrombin-induced DISCUSSION @-HexosaminidaseSecretion-The above results showed that exogenous PDGF inhibited platelet activation and that the Platelet agonists like thrombin, collagen, and epinephrine PDGF a-receptor was activated an in autocrine manner during stimulate tyrosinephosphorylation of several proteins in platethrombin stimulation. The ability of the PDGF neutralizing lets, suggesting that such phosphorylation is involved in plateantibodiestoinhibittheautocrinePDGFpathway (Fig. 4) let activation (29-31). However, cloning of the membrane reprompted us to investigate whether this pathway functions as ceptors of these agonists have shown that neither of them

E

: b e

I

Antithrombotic Activity of the PDGF &-Receptor

13878

A

-

30 sec

M

30 sec

ThromblnPDGFINaCI

PDGFINsCI Thrombln

PDGF

PDGF

FIG.6. Reversion of thrombin-induced aggregation by PDGF. Platelets were incubated with 100 ng/ml porcine PDGF or vehicle (NaC1) 30 s prior (A) or after ( E ) stimulation with 0.1 unit/ml thrombin. The aggregation was measured in a Payton dual channel aggregation module.

contain a cytoplasmic tyrosine domain (321, and the tyrosine phosphorylation in platelets has so far been recognized as secondary to activation ofpp60"" (33) and Src-related proteintyrosine kinases like pp60fynand ~ ~ 6 (34, 2 ~ 35).' The ~ present identification of PDGF a-receptors in human platelets is thus the first demonstration of the presence of a tyrosine kinase receptor in these cells. The receptor was identified by immunoprecipitation as a 170-kDa componentcorresponding in size t o the mature glycosylated protein as found in other cell types (8). Only PDGF a-receptors and not P-receptors were found, as verified by receptor binding studies with cross-competition of lZ5I-PDGF-BB by high doses of unlabeled PDGF-AA. The PDGFR-a tyrosine kinase of resting platelets was stimulated by the addition of PDGF. Interestingly, thrombin significantly increased the autophosphorylation of the PDGFR-a independent of exogenous PDGF, as determined by immunoprecipitation with anti-PY as well as PDGFR-a antiserum (Fig. 4). This effect of thrombin was completely inhibited by preincubation of the platelets with PDGF neutralizing antibodies. Thus, the observed thrombin-induced phosphorylation of the PDGF a-receptor was most likely due to the release of endogenous PDGF froma-granules, rather thansecondary phosphorylation of the PDGF a-receptor by an intracellular signal transduction pathway activated by thrombin. In fact, immunohistochemical staining showed that the PDGF-positive component was transferred afterthrombin activation from a granular intracellular pattern to a circular staining in and around the platelet membrane. Moreover, thrombin has previously been found to induce tyrosine phosphorylation of protein components of 170-180 kDa, corresponding in size to the PDGF a-receptor (30, 31). The concept of autocrine-positive feedback during platelet activation is well established; platelet secretion of ADP and serotonin from dense granules, secretion of a-granule components like PF 4, factor VIIUvon Willebrand, thrombospondin, and fibrinogen, as well as prostanoid formation from liberated arachidonate, have all been shown t o amplify the platelet response (36). However, platelet granule components that function in a negative feedback pathway have not been reported. Previous work has shown that addition of PDGF inhibits certain responses in resting (37) and collagen-activated platelets (18). We found that exogenous PDGF (50-100 ng/ml) inhibited thrombin-induced platelet aggregation and secretion at physiological dosesof less that 0.1 uniffml of thrombin. In fact PDGF was able to reverse thrombin-induced aggregation when added 30 s after thrombin stimulation, suggesting that a rapid and efficient inhibitory pathway is initiated by the PDGFR-a. The

0.00

0.02

0.04

0.06

0.08

0.22

0.10

thrombin (U/ml)

0 0

1

2 3 4 pg/rnl Mob 6D11

1

I

0

50

100

5

6

150

ng/ml PDGF

FIG.7. Effects of PDGFandPDGF antibodies on thrombininduced P-hexosaminidase secretion. A, platelets were preincubated with PDGF antibodies or vehicle for 60 s a t 37 "C before incubation with different concentrations of thrombin for 2 min at 37 "C. The secretion of P-hexosaminidase in the absence of PDGF antibodies was set at 100%. E , platelets were preincubated with different concentrations of PDGF (closed circles)or PDGF antibodies (open circles) before incubation with 0.05 uniffml thrombin. The data are representative of three differentexperiments performed in triplicate, the bars illustrate 1 S.D.

Antithrombotic Activity of the PDGF a-Receptor ability of thePDGF-neutralizing antibodies to inhibitthe thrombin-induced activation of the PDGFR-a allowed us to study the functional significance of this autocrine loop during platelet activation. We found that thePDGF antibodies significantly potentiated the thrombin-induced lysosomal secretion of P-hexosaminidase. Dose-response studies showed thatthe thrombin-inducedsecretion was inhibited or potentiated by preincubation with exogenous PDGF or PDGF antibodies, respectively. It is therefore likely that our observed autocrine activation of the PDGFR-a elicitsa negative feedback pathway during platelet activation. The inhibitory signal pathway of the PDGFR-a in activated platelets remains to be identified. Several signal transduction molecules which interact with the PDGF receptors have been identified, including phospholipase C-y (381, the GTPase activation protein of Ras (39, 401, the regulatory subunit ( ~ 8 5of ) phosphatidylinositol 3”kinase (41, 421, and members of the c-Src family kinase (43), as well as the adaptor molecules Nck (44) and Grb2 (45). Interestingly, phospholipase C - y , GTPase activation protein, phosphatidylinositol 3’-kinase, and the Src proteins are present in human platelets(32) and are reported to be tyrosine phosphorylated during platelet activation; it is thus possible that certainof these molecules are involved in the inhibitoryPDGF pathway. The PDGFa-receptor has been shown toinhibitthe chemotacticresponseinduced by the PDGF P-receptor in human fibroblasts (12, 13); the molecular mechanism behind this effect is not known. Since platelet activation and chemotaxis both involve cell shape changes, it is possible that similar components in platelets and fibroblasts, respectively, are involved in these inhibitory signal transduction pathways. Secretion of PDGF from platelets, endothelial cells, smooth muscle cells, and activated macrophages represent a source of PDGF generally recognized as important for chemotaxis and cell growth in atherosclerotic lesions (reviewed in Ref. 46). Our observations suggest that thrombosis and atherosclerosis are regulated by PDGF in a complex manner. Elevated levels of PDGFmay prevent platelet aggregation, as well as inhibit chemotaxis by activation of the PDGF a-receptor (12, 13) of smooth muscle cells (47). Production of nitric oxide (48) and prostaglandin I, (49) on endothelial surfaces inhibit platelet activation. In addition,secretion of PDGF from endothelial antithrombogenic mechanisms cells (50) could take part in the of the vessel wall. Our demonstration of PDGF a-receptors on human platelets and itsinhibitory function during platelet activation identifiesa new possible role of PDGF in theregulation of thrombosis.

7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41.

Acknowledgments-We technical assistance.

thank G. M. Aarbakke and C. A. Cook for REFERENCES

1. Heldin, C.-H., and Westermark, B. (1990) Cell Regul. 1, 555-566 2. Bar, R. S., Boes, M., Booth, B. A,, Dake, B. L., Henley, S., and Hart, M. N. (1989) Endocrinology 124, 1841-1848 3. Smits, A., Hermansson, M., Nister, M., Karnushina, I., Heldin, C.-H., Westermark, B., and Funa, K. (1989) Growth Factors 2, 1-8 4. Smits, A,, Kato, M., Westermark, B., Nister, M., Heldin, C.-H., and Funa, K. (1991) Proc. Natl. Acad. Sei. U. S. A. 88, 8159-8163 5. Raines, E. W., and Ross, R. (1993) in Biology ofplatelet-derived GrowthFactor (Westermark, B., and Sorg, C., eds) pp. 74-115, S. Karger, Basel, Switzerland 6. Yarden, Y., Escobedo, J. A,, Kuang, W. J., Yang-Feng, T. L., Daniel, T. O.,

42. 43. 44. 45. 46. 47. 48. 49. 50.

13879

Tremble, P. M., Chen, E. Y., Ando, M. E., Harkins, R. N., Francke, U., Friend, V.A,, Ullrich, A,, and Williams, L. T. (1986) Nature 323, 226-232 Matsui, T., Heidaran, M., Miki, T.,Popescu, N., La-Rochelle, W., Kraus, M., Pierce, J., and Aaronson, S. (1989) Science 243, 800-804 Claesson-Welsh, L., Eriksson, A., Westermark, B., and Heldin, C.-H (1989) Proc. Natl. Acad. Sei. U. S. A. 88,49174921 Heldin, C.-H., Ernlund, A,, Rorsman, C., and Ronnstrand, L. (1989) J. Bid. Chem. 264,8905-8912 Bishayee, S., Majumdar, S., Khire, J., and Das, M. (1989) J. Biol. Chem. 264, 11699-11705 Seifert, R. A,, Hart, C. E., Phillips, P. E., Forstrom, J. W., Ross, R., Murray, M. J., and Bowen-Pope, D. F. (1989) J. Biol. Chem. 264,8771-8778 Siegbahn, A,, Hammacher, A., Westermark, B., and Heldin, C.-H. (1990) J . Clin. Invest. 85, 916-920 Vassbotn, E S., Ostman, A., Siegbahn, A,, Holmsen, H., and Heldin, C.-H. (1992) J. Biol. Chem. 287, 15635-15641 Hammacher, A,, Hellman, U., Johnsson, A,, Ostman, A,, Gnnnarsson, K., Westermark, B., Wasteson, and Heldin, C.-H. (1988) J. B i d . Chem. 283, 16493-16498 Hart, C. E., Bailey, M., Curtis, D. A,, Osborn, S., Raines, E., Ross, R., and Forstrom, J. W. (1990) Biochemistry 29, 166-172 Kaplan, D. R., Chao, F. C., Stiles, C. D., Antoniades, H. N., and Scher, C. D. (1979) Blood 53,1043-1052 Witte, L. D., Kaplan, K.L., Nossel, H. L., Lages, B. A,, Weiss, H. J., and Goodman, D. S. (1978) Circ. Res. 42, 402409 Bryckaert, M. C., Rendn, F., Tobelem, G., and Wasteson, A. (1989) J. Biol. Chem. 264,4336-4341 Tysnes, 0.B., Aarbakke, G. M., Verhoeven, A. J. M., and Holmsen, H. (1985) Thromb. Res. 40,329-338 Holmsen, H., Male, R., Rongved, S., Langeland, N., and Lillehaug, J. (1989) Biochem. J. 260,589-592 Vassbotn, F. S., Langeland, N., Hagen, I., and Holmsen, H. (1990) Biochim. Biophys. Acta 1054,246-249 Vassbotn, F. S., Ostman, A,, Langeland, N., Holmsen, H., Westemark, B., Heldin, C.-H., and Nister, M. (1994) J. Cell. Physiol. 158, 381-389 Claesson-Welsh, L., Hammacher, A,, Westermark, B., Heldin, C.-H., and Nister, M. (1989) J. Biol. Chem. 264, 1742-1747 Ek, B., and Heldin, C.-H. (1984) J . Biol. Chem. 259, 11145-11152 Ostman, A., Backstrom, G., Fong, N., Betsholtz, C., Wernstedt, C., Hellman, U., Westermark, B., Valenznela, P., and Heldin, C.-H. (1989) Growth Factors 1,271-281 Eriksson, A., Siegbahn, A,, Westermark, B., Heldin,C.-H.,andClaessonWelsh, L. (1992) EMBO J. 11, 543-550 Steen, V. M., and Holmsen, H. (1985) Thromb. Haemostasis 54, 680-683 Dangelmaier, C. A., and Holmsen, H. (1980)AnaL Biochem. 104, 182-191 Nakurama, S., and Yamamura, H. (1989) J. Biol. Chem. 264, 7089-7091 Ferrell, J. E., Jr., and Martin, G. S. (1988) Mol. Cell. Biol. 8, 3603-3610 Golden, A,, and Brugge, J. S. (1989)Proc. Natl. Acad. Sci. U. S. A . 86,901-905 Shattil, S., and Brugge, J. S. (1991) Curr. Opin. Cell Biol. 3, 869-879 Golden, A,, Nemeth, S. P., and Brugge, J. S. (1986) Proc. Natl. Acad. Sci. U. S. A . 83,852-856 Horak, I. D., Corcoran, M. L., Thompson, P. A,, Wahl, L. M., and Bolden, J. B. (1990) Oncogene 5, 597402 Zhao, Y. H., Kreuger, J. G., and Sndol, M. (1990) Oncogene 5, 1629-1635 Holmsen, H. (1990) in Heamatology (Williams, W. J., Beutler, E., Erslev, A. J., and Lichtman, M. A., eds) pp. 1200-1233, McGraw Hill Inc., New York Gerrard, J. M., Israels, S. J., and Friesen, L. L. (1985) N o m . Rev. FI:Hematol. 27,267-273 Wahl, M. I., Olashaw, N. E., Nishibe, S., Rhee, S. G., Pledger, W. J., and Carpenter, G. (1989) Mol. Cell. Biol. 9, 2934-2943 Molloy, C. J.,Bottaro, D. P., Fleming, T. €?, Marshall, M. S., Gibbs, J. B., and Aaronson, S. A. (1989) Nature 342, 711-714 Kazlanskas, A., Ellis, C., Pawson, T., and Cooper, J. A. (1990) Science 247, 1578-1581 Auger, K. R., Serunian, L. A,, Soltoff, S. P., Libby, P., and Cantley, L. C. (1989) Cell 57, 167-175 Coughlin, S. R., Escobedo, J. A,, and Williams, L. T.(1989) Science 243, 11911194 Kypta, R. M., Goldberg, Y., Ulug, E. T.,and Courtneidge, S. A. (1990) Cell 82, 481-492 Meisenhelder, J., and Pawson, T.(1992) Mol. Cell. Biol. 12, 5834-5842 Lowenstein, E. J., Daly, R. J., Batzer, A. G., Li, W., Margolis, B., Lammers, R., Ullrich, A., Skolnik, E. Y., Bar-Sagi, D., and Schlessinger, J. (1992) Cell 70, 431-442 Ross. R. (1993)Nature 362. 801-809 Koyama, N., Morisaki, N.,Saito, Y., and Yoshida, S. (1992) J. Biol. Chem. 267, 22806-22812 Furchgott, R. F. (1983) Circ. Res. 53, 557-573 Botting, R., and Vane, J. R. (1989)Arch. Mal. Coeur. 82, 11-14 DiCorleto, P. E., and Bowen-Pope, D.F. (1983) Proc. Natl. Acad. Sci. U. S. A. 81, 1919-1923

A,,