Collagen - platelet interaction: Inhibition by a monoclonal antibody which binds a 90,000 dalton platelet glycoprotein

THROMBOSIS RESEARCH 53; 129-143, 1989 0049-3848/89 $3.00 t .OO Printed in the USA. Copyright (c) 1989 Pergamon Press plc. All rights reserved.

INHIBITION BY A MONOCLONAL - PLATELET INTERACTION: BINDS A 90,000 DALTON PLATELET GLYCOPROTEIN

COLLAGEN

Thomas M. Chian$,

ANTIBODY WHICH

Ai Jint, Karen A. Hasty*, and Andrew H. Kangil

+*

Memphis,

Memphis,

TN

+*

of Medicine , of Tennessee, 38104; Capital Hospital, Chinese Academy of Medical Sciencet, Beijing, China.

the Departments 'stration Medical Center ~~c-m~sf$yl~, and Anatomy and Neurobiology*, University

(Received 22.2.1988; accepted in original form 31.10.1988 by Editor J.H. Griffin)

ABSTRACT Polyclonal antiserum prepared to purified type I collagen receptor, a 65!000 molecular weight protein, isolated from human platelets reacted with two other proteins with molecular weights of 90,000 and 58,000 in immunoblots of solubilized platelet membranes. The immunoreactive proteins were purified to homogeneity with molecular Periodic seive chromatography and preparative gel electrophoreses. acid Schiff stain showed that both the 90,000 and 65,000 proteins were with glycoproteins. These purified proteins reacted immunoglobulin G (IgG) fractions isolated from antiserum raised antibody against the 65,000 protein and polyand monoclonal specific for the glycoprotein IIb-IIIa suggesting that these three To further proteins are immunocross reactive with GPIIb-IIIa. examine the immunocross-reactivity of these proteins, a monoclonal rrntihnrtv l.7Za.Z rlrifnrl rroainct the 4n nnn mlvr.nnrn+ein This bLJ-.,r-‘-----'. """Y"..~ WV" ..U..IIV Y6c.I”” I b.._ ““V monoclonal antibody also reacted with all three proteins in enzymesuggesting linked immunosorbent assays and transblot experiments The that these three possess a common antigenic determinant. the 90,000 glycoprotein also monoclonal antibody prepared to inhibited platelet aggregation induced by the addition of collagen and ADP but not a-thrombin and epinephrine-induced aggregation. This suggests that the monoclonal antibody binds a protein on the platelet surface which plays a role in platelet aggregation induced L.. _22-'*:,.-"L _c CL___ ^___i^C^ uy CL_ LLLC. auulLl"LL Cllebe dg"LLI>Lb. /”

Key words:

platelets,

collagen, monoclonal

129

,

antibody.

COLLAGEN-PLATELET INTERACTION

Vol. 53, No. 2

INTRODUCTION Platelets play an important role in the process of hemostasis and thrombosis. Following injury to blood vessels and in certain pathologic conditions, platelets adhere to the exposed subendothelial connective tissue, collagen in particular, and aggregate releasing several biologically active substances (l-5). The mechanism of collagen-induced platelet aggregation has been studied intensively in recent years by many investigators (6-13). Results from these laboratories suggest that the interaction between collagen and platelets is mediated through a specific receptor(s). In previous reports (14,15), we presented data on the isolation and purification of a type I collagen receptor following solubilization of the membrane proteins with 0.5% Triton X-100. The purified receptor is a glycoprotein with a molecular weight of 65,000 as estimated by its mobility during NaDodS04 polyacrylamide gel electrophoresis. The isolated receptor inhibits the binding of radiolabeled al(I) chains of collagen to platelets as well as the aggregation of platelets induced by type I collagen. We have also shown that antibody prepared against the 65,000 receptor competitively inhibits the binding of al(I) chains to platelets as well as the aggregation of platelets mediated by both al(I) chains from chick skin and native type I collagen from human skin (15). These observations indicate that the al(I) chain and type I collagen bind to platelets at the same site. In the present paper we show that, in addition to the 65,000 receptor, two other proteins isolated from platelet membranes with molecular weights of 90,000 and 58,000 cross-react immunologically with the anti-65,000 antiserum. In addition, a monoclonal antibody raised against the purified 90,000 protein also reacts with the 65,000 and 58,000 proteins. Collagen and ADP induced platelet aeereeation were both inhibited in the presence of -cIcI--cIm ~the monoclonal antibody. These results suggest that these three proteins are immunologically and functionally interrelated.

MATERIALS AND METHODS Prenaration of nlatelet membranes: Human platelet membranes were isolated according to the method developed by Barber and Jamieson (16). Ten units of platelet-rich plasma, purchased from a local blood bank, were sedimented by centrifugation and washed with albuffer containing 0.05M TrisHCl-0.15M NaCl-O.OOlM EDTA, pH 7.4 (Tris-EDTA) . The washed platelets were _ __~. iysed by the addition of a buffer containing U.UlM Tris-HCi and O.25M sucrose, pH 7.5, and sonicated for 10 seconds with a Branson sonifier. The membrane fraction was obtained by centrifuging the platelet lysate on a solution of 27% sucrose at 65,000 x g for 4 hours. The interface containing the platelet membranes was collected and the membranes were washed with Tris-EDTA buffer. The washed platelet membranes were either used directly or stored in test tubes at -20°C.

1. Abbreviations used: Tris-EDTA, 0.05M Tris-HCl-0.15M NaCl-O.OOlM EDTA; pH 7.4; ELISA, enzyme linked immunosorbent assay; NaDodS04-PAGE, sodium dodecyl sulfate polyacrylamide electrophoresis; PBS, 2omM gel phosphate/l30mM NaCl, pH 7.4; PAS, periodic acid Schiff stain; PRP, platelet-rich plasma; TBS, 0.02M Tris-HC1/0.5M NaC1/0.05% Tween 20, pH 7.5.

Vol. 53, No. 2

COLLAGEN-PLATELET INTERACTION

131

Pur fication 0 C: The isolated platelet membranes were solubilized with 0.2% sodium dodecylsulfate (NaDodSO4) for 30 minutes, The solubilized supernatant was loaded onto an Agarose A5m (BioRad) column (1.6 x 90cm) equilibrated with 0.02M Tris-HCl0.05M NaCl-0.2% NaDodSO4, pH 7.4, and the column was eluted with the same buffer. The optical density of the effluent was monitored using a Gilford spectrophotometer at 280 nm. Peak fractions were pooled, dialyzed, and lyophilized. The immunoreactive proteins were detected by a transblot procedure using a polyclonal antibody raised against the purified collagen receptor (Mr65,OOO). The lmmunoreactive fractions from the gel filtration column were dissolved in NaDodS04 sample buffer for NaDodS04-polyacrylamide gel (10%) electrophoresis (17) in the presence of 1OmM EDTA. These gels were electrophoresed overnight at a constant voltage of 35 volts. At the end of electrophoresis, gels were cut into strips, using an immunoblot as the guide. The protein was eluted from these gel strips in a protein elutor (Reliable Scientific, Memphis, TN) with 0.3M Tris buffer, pH 8.3. The effluent was filtered through Whatman No. 1 paper and was dialyzed with cold water containing protease inhibitors and lyophilized. The same procedure was repeated once more by preparative NaDodS04-PAGE using 7.5% gel. Prenarations of nolvclonal and monoclonal antibodies: Polyclonal antisera against the purified receptor (anti-65,000) were raised in rabbits. The details of the procedure were reported previously (15). Briefly, two New Zealand white rabbits (1.5 kg) were immunized intradermally with 100 pg of the purified 65,000 receptor emulsified with Freund's complete adjuvant. Each animal received a booster injection of 50 pg of the receptor in Freund's incomplete adjuvant at one, two and ten weeks. Serum samples were obtained prior to immunization and thereafter at two week intervals after boosting. All sera were heat-inactivated at 56°C for 30 minutes and stored at -20°C. Anti-receptor antibodies were detected by enzyme-linked immunosorbent assay (ELISA). The IgG fraction was isolated by an affinity column prepared by linking the 65,000 protein to Sepharose 2B. For the preparations of monoclonal antibodies to the 90,000 protein, BALB/c mice were immunized with 10 pg of the purified 90,000 protein emulsified in Freund's complete adjuvant injected into the foot pads. Four weeks later, the booster was given with the same immunogen using the same dosage and route but emulsified with Freund's incomplete adjuvant (IFA). After an additional period of 4 weeks, a final booster of 50 pg of the same immunogen in saline was given intravenously. The mouse wa killed 3 or 4 days later and the immune spleen cells in suspension (2 x 10B/ml) were fused with 8-Azoguanine selected myeloma cells (Sp2/0) at a ratio of 3:l (spleen:myeloma) using polyethylene glycol (PEG 1500; Aldrich Chemical Co., Milwaukee, WI) at 35% (w/v), pH 7.6-7.8, in serum-free DMEM at room temperature for 8 minutes. The hybrid cells were grown to confluency in master microtiter plates in selective medium. Hybrid cells were screened for desired antibody production as determined by ELISA and were cloned by limiting dilution. Seven to 14 days after subcloning, four strongly positive subclones (6.8, 7.7, 22.4, 72.21) were obtained. We selected 72.21 for further subclonin and expanded and injected into pristane-primed BALB/c mice at about 1 x 10P hybridoma cells/mouse (18). The monoclonal antibody obtained was purified from mouse ascitic fluid by chromatography on DEAE Affi-Gel Blue (BioRad). Fab' fragments were obtained by Agarose A1.5m column (2 x 95cm) chromatography of papain-digests of the purified IgG.

132

COLLAGEN-PLATELET INTERACTION

Vol. 53, No. 2

Polyclonal antibody raised against the GPIIb-IIIa complex was a gift from Dr. D. Phillips, Gladstone Foundation, San Francisco, CA. Monoclonal antibodies raised against GPIIb-IIIa complex and GPIIIa alone were also gifts from Dr. T. Kunicki, Blood Bank of Southwest of Wisconsin, Milwaukee, WI. Enzvme-linked immunosorbent assay: Inhibition and cross reactivity studies were performed by ELISA (19,20). Briefly, microtiter wells were coated with 0.20 ml of 0.05M Na2C03, pH 9.6, containing 10 pg/ml of the proteins. The plate was incubated at 37°C for 3 hours and washed 7 times with 0.15M NaGl containing 0.05% Tween-20 (Saline-Tween). Various dilutions of antisera in 0.02M phosphate/O.l3M NaCl, pH 7.4, (PBS) containing 0.05% Tween-20 (PBS-Tween) were added to each well and incubated at 37°C for 3 hours. Antibody and inhibition reagents were preincubated at 37°C for 30 The plate was again washed with min before putting into washed wells. Saline-Tween, and 0.2 ml of peroxidase-conjugated goat anti-rabbit or IgG and IgM mixture or individual subclasses of immunoglobulins (l/2000, Cappel Laboratories, West Chester, PA) was added to each well. The plate was then incubated for 2 hours at 37"C, washed with Saline-Tween, and enzyme The absorbance was substrate (5-aminosalicylic acid and H202) was added. read at 450 nm. NaDodSO,-Polvacrvlamide electrouhoresis: Samples were dissolved in NaDodSO4 sample buffer (containing 2 mM ,9-mercaptoethanol),boiled for 2 minutes, and loaded on 7.5% or 10% polyacrylamide slab gels. The slab gels were electrophoresed using Tris-glycine buffer, pH 8.3, for an overnight run at a constant voltage of 35 volts (17). Immunoblot assay: Samples were analyzed by NaDodS04-PAGE (7.5% or 10%). The gels were then transblotted onto nitrocellulose paper in buffer containing 0.025M Tris-HCl/O.l92M glycine/20% methanol, pH 8.3, for 5 hours with a constant current of 200mA. The nitrocellulose paper was transferred to a solution of 0.02M Tris/O.SM NaC1/0.05% Tween 20, pH 7.5, (TBS-Tween) containing 3% horse serum (V/V) for 3 hours. After 5 washes with TBS-Tween, the nitrocellulose paper was then transferred to various dilutions of first antibody solution (l/200 or l/500) with 1% horse serum. After an overnight incubation at 4"C, the nitrocellulose paper was washed 5 times with TBSTween and transferred to a second antibody solution (goat anti-rabbit IgG or mixture of anti-IgG and IgM conjugated to horseradish peroxidase, l/2000 dilution in TBS) and incubated for 2 hours at room temperature. The nitrocellulose paper was washed 5 time with TBS-Tween and then transferred to the same buffer (120 ml) which contained HRP color development solution (60 mg horseradish peroxidase color development reagent in 30 ml of cold methanol and 60 ~1 of ice cold 30% H20 ) (21-22). Preoaration of an affinity co1 umn and ourification of monoclonal antibody: Purified 90,000 glycoprotein was covalently linked to Sepharose 2B using a 1-ethyl-3-(3-dimethyl aminopropyl) carbodiimide according to the method described by other investigators (20,21). Briefly, the washed Sepharose 2B (5ml) was suspended in lOm1 of coupling buffer (3mM phosphate buffer, pH 6.3) and 300 pg of purified 90,000 glycoprotein in the presence of 50 mg of 1-ethyl-3-(3-dimethyl aminopropyl) carbodiimide and maintained at pH 6.3 for one hour. After the mixture was stirred overnight at 4"C, the gel was washed with 100 ml of phosphate-buffered saline and poured into a column. Ascitic fluid (lml) was loaded into the column, allowing the column to absorb for 30 min. and washed with phosphate buffered saline and eluted with 0.2M glycine pH 3.8. Both fractions were collected, dialyzed and lyophilized. The lyophilized material was dissolved in phosphate buffered saline and the protein concentration was determined.

Vol. 53, No. 2

COLLAGEN-PLATELET INTERACTION

133

Prenaration of nlatelet-rich plasma: Human blood collected from normal volunteers (not ingesting any drugs and after an overnight fast) was placed into polypropylene tubes containing 0.1 volumne of 3.8% sodium citrate. A two syringe technique was employed with blood collected in the second syringe being used for the preparation of platelet-rich plasma (PRP). PRP was prepared by centrifuging the titrated blood at room temperature for 10 minutes at 226 x g (5). Whole blood and PRP were exposed only to plastic surfaces or siliconized vessels. The platelet counts of the PRP ranged from 200,000 to 310,000 per cubic millimeter. Prenaration of type I collagen: Neutral salt-soluble collagen was extracted from the skin of three week-old White Leghorn chicks which had been rendered lathyritic by administration of /I-amino-propionitrile (Aldrich Chemical Co., Inc., Milwaukee, WI) for two weeks. Extracted collagen was purified by repeated differential precipitation with NaCl as previously described (23). Platelet azzrezometry: Platelet aggregation was studied by the turbidimetric method of Born (24). The transmittance of PRP was measured in a Payton Aggregometer, Model 300A. An aliquot of 0.45 ml of PRP or washed platelets in Tyrode's solution was pipetted into a 8 X 45 MI, siliconized cuvette and stirred at a constant speed of 1,100 rpm at 37°C. The test agent (antibody) was incubated with PRP for 10 minutes. Aggregating agents were added in volumes of 50 ~1. The changes in percent transmission were recorded continously using a Goerz (Gelman) recorder. Protein concentration determination: Protein concentrations were determined by the method of Lowry et al (25). The concentration of peptides was determined by amino acid analysis after acid hydrolysis as described (23).

RESULTS Isolated Immunoreactive proteins of platelet membrane preparations: in 0.2% NaDodS04 and subjected to platelet membranes were solubilized The proteins were then electrophoretically. transferred to NaDodS04-PAGE. When analyzed for immunoreactive proteins using the nitrocellulose paper. one in the region anti-65,000 antiserum, three protein bands were detected: of M, 58,000; another in the region of 65,000; and a third broad band in the region of 90,000 (Figure 1). slab gel of the broad band by preparative Further purification electrophoresis yielded a discrete band of protein with an apparent M, of No difference in the ratio of the three proteins were noted with 90,000. the inclusion of protease inhibitors (1 mM EDTA, 1 PM phenylmethylsulfonyl The stable ratio of the fluoride, 1 mM N-ethylmaleimide and 1 mM EGTA). three proteins in the absence of protease inhibitors suggest that the two low M, proteins exist in platelets and that they are not degraded products of high M, 90,000. The purity of the 90,000, The nuritv of 90.000 and 58.000 oroteins: Each of 65,000, and 58,000 proteins was examined with NaDodS04-PAGE (10%). these three proteins, 59,000 (Figure 2, lane l), 65,000 (Figure 2, lane 2), and 90,000 (Figure 2, lane 3) appeared as a single band, and were not contaminated with each other.

134

COLLAGEN-PLATELET INTERACTION

Vol. 53, No. 2

.

Figure 1. Transblot of solubilized olatelet membrane nroteins. Solubilized platelet membrane preparations (150 pg) prepared in the presence (lanes 1 & 2) and absence (lane 3) of protease inhibitors were subjected to NaDodSO4PAGE (10% gel) and electrophorectically transferred to nitrocellulose paper. Immunoreactive bands were localized using polyclonal anti-65,000 antiserum (lanes 2 & 3). Preimmune serum incubated transblot experiment is shown in lane 1. Lane 4 represents the molecular weight markers (Myosin, Bgalactosidase, phosphorylase, b, bovine serum albumin and ovalbumin, BioRad, Inc.) stained with Amido Black.

The nature of 90.000. 65.000. and 58.000 nroteins: Some of the characteristics of the 65,000 protein have been reported previously (15). Similar to the 65,000 protein, the 90,000 and 58,000 proteins can be destroyed by trypsin and pronase treatement (data not shown). The 90,000 and 65,000 proteins are glycoproteins based on the results of periodic acid Schiff Stain (PAS) (Figure 3). The 58,000 protein did not stain with PAS, suggesting it is not a glycoprotein. Lanes 1, 2, and 3 were purified protein samples of 58,000 65,000 and 90,000 respectively. Lane 4 contained the BioRad marker proteins. The migration of these three proteins were not changed under reduced or non-reduced conditions (data not shown).

Vol. 53, No. 2

COLLAGEN-PLATELET INTERACTION

12

135

3

Figure 2. NaDodSO,.-slab gel electronhoresis on 10% nolvacrvlamide of purified nroteins. Samples were separated on a 12% gel, stained with Coomassie brilliant blue and destained with 7% acetic acid-lo% methanol. Lane 1, 58,000 10 pg; lane 2, 65,000 10 pg; and lane 3, 90,000 10 pg.

Immunoreaction amens three oroteins: The effect of the 90,000, 65,000, and 58,000 proteins on the binding of the 65,000 protein to the anti-65,000 anti-serum was studied by ELISA technique. The 90,000, 65,000, and 58,000 proteins each inhibited the binding in a dose-dependent manner (Table I), although the degree of inhibition was dependent upon the individual protein. The inhibition was greatest with the 65,000 protein, as expected, followed by the 58,000 and, finally, the 90,000 protein. To elucidate the relationship between these three proteins and GPIIbIIIa, we have used polyclonal anti 65,000 and anti-GPIIb-IIIa antibodies to immunoblot the freshly isolated platelet membrane in the presence of protease inhibitors (the same inhibitors used in Figure 1). The result of the study showed that both anti-65,000 and anti-GPIIb-IIIa reacted with three identical proteins (Figure 4, lanes 3 and 4) compared to control (lanes 1 and 2). These results suggest that these three proteins are immuno cross-reactive with GPIIb-IIIa

136

COLLAGEN-PLATELET INTERACTION

Vol. 53, No. 2

Purified Fizure 3. Periodic acid Schiff stain of the ourified oroteins. analyzed with NaDodSO458,000, 65,000, and 90,000 proteins were The gel was then stained polyacrylamide slab gel electrophoresis (10%). with periodic acid then stained with periodic acid Schiff stain (lanes 1, 2, and 3 for 58,000, 65,000, and 90,000 respectively). The lane with BioRad marker proteins (lane 4) was stained with Coomassie brilliamt blue and The migration position of destained with 10% methanol-7% acetic acid. 58,000 protein is shown in arrow. TABLE I THE INHIBITORY EFFECT OF THREE PLATELET MEMBRANE PROTEINS ON THE BINDING OF THE 65,000 PROTEIN TO ITS POLYCLONAL ANTISERUM Absorbance, 450 nm Amount added 0 2.5 5 10

(pg)

None

90,000

65,000

58,000

1.05

0.82 0.56 0.42

0.43 0.37 0.12

0.67 0.45 0.41

The microtiter wells were coated with 0.2 ml of purified 65,000 protein (10 pg/ml) in 50 mM Na2C03, pH 9.6 at 37°C for 3 hours. l/300 dilutions of anti 65,000 serum were incubated with and without 90,000, 65,000, and 58,000 proteins at 37°C for 30 minutes. These mixtures were added to wells washed with Saline-Tween and incubated for a dilution of l/2000 of peroxidaseconjugated goat anti-rabbit IgG was added to the wells before incubation for 2 more hours. A substrate solution was added to each well after washing with Saline-Tween. The absorbance was read at 450 nm.

Vol. 53, No. 2

COLLAGEN-PLATELET INTERACTION

-

137

90,o 0 0

6 5,O 0 0 -- 5 8,0 0 0

-

1 2 3 4

Fizure 4. c late ets. Human platelets were isolated from a walking donor within 30 minutes in the presence of protease inhibitors (PMSF, NEM, EDTA and EGTA). These platelets were solubilized in NaDodS04 sample buffer and boiled for 3 minutes then analyzed with 10% NaDodS04-PAGE and transblotted onto nitrocellulose membrane. The polyclonal anti-65,000 serum (l/100 dilution) was used. Lane 1, preimmune serum (l/100 dilution as control); lane 2, control serum, lane 3, anti-65,000 antibodies,and lane 4, isolated IgG fraction from polyclonal anti-GPIIb-IIIa antiserum.

We have used an immunoprecipitation to precipitate GPIIb-IIIa from Triton solubilized platelet membranes and have analyzed the precipitate with 10% NaDodSO4-PAGE. The result of the study showed that the GPIIb-IIIa existed in the precipitated fraction (data not shown). Monoclonal antibodv raised against the 90.000 protein: We have raised a monoclonal antibody using the purified 90,000 glycoprotein. Among four clones examined, one had the highest reactivity to the antigen. This clone (72.21) was again subcloned by limiting dilution. The subtype of this clone The subclone (72-21-18) was used to obtain ascites fluid and was IgM. purified IgM study immuno cross-reactivity among the three proteins. The specificity of the monoclonal antibody was examined by using an enzyme-linked immunosorbent assay. The monoclonal antibody reacted equally to the 90,000 and 58,000 proteins but was less reactive to the 65,000 This result was confirmed in transblot experiments. protein (Table II). When solubilized platelet membrane proteins were analyzed by NaDodS04-PAGE and immunoblotted with monoclonal anti-90,000 antibody, the monoclonal antibody bound to the 90,000 protein as well as 65,000 and 58,000 proteins

138

COLLAGEN-PLATELET INTERACTION

Vol. 53, No. 2

-9 -65,000 -5 6,000

Transblot of olatelet membrane nroteins (lanes 1. 2. & 3) and Fizure 5. endozlvcosidase H treated membrane (lane 4) with monoclonal anti-90.000 antibody. Solubilized platelet membrane preparation (100 pg) was analyzed by NaDodS04-PAGE on 7.5% polyacrylamide gels transblotted to nitrocellulose paper and immunoblotted by using: lane 1, control ascites fluid; lanes 2, 3,and 4, the affinity purified monoclonal anti-90,000 antibody (20 pg/50ml, in the presence (lane 2) and absence (lane 3) of protease inhibitors and endoglycosidase H treated membrane (lane 4); lane 5 peroxidase conjugated anti-mouse antibody alone.

.

Fizure 6. The effect of the anti-90.000 monoclonal IzM isolated from ascites fluid on nlatelet azzreeation induced by various azerepatinp agents. An aliquot of 0.45 ml of platelet-rich plasma was incubated with the purified monoclonal IgM (b,30 pg; c, 60 pg) for 15 minutes before adding collagen (10 pg, A) a-thrombin (0.5 U, B) and ADP (4 Jo M; c). Control experiment (a) represents platelet-rich plasma preincubated with control IgG isolated from Sp2/0 ascites fluid. Control experiment - PRP was incubated with anti-90,000 and challenged with phosphate-buffered saline (panel A, line d).

Vol. 53, No. 2

COLLAGEN-PLATELET INTERACTION

139

(Figure 5, lanes 2 & 3). The endoglycosidase H treated platelet membrane still reacted with monoclonal anti-90,000 antibody (Figure 5, lane 4) and the migration of these bands was different from untreated samples (Figure 5, lanes 2 & 3) suggesting the reactive group is not carbohydrate moiety. These results suggest that these three proteins have a common antigenic determinant.

TABLE II CROSS REACTIVE PROPERTIES OF THE 90,000 AND OTHER PROTEINS

Coating Protein

Dilution (end point)*

106 lo5 106

90,000

65,000 58,000

Microtiter wells were coated separately with the purified 90,000, 65,000, and 58,000 proteins (0.15 ml of 10 pg/ml Na2C03 buffer) and incubated at 37°C for 3 hours. Different dilutions (in increments of 10 fold dilutions) of ascites fluid were added to each well after washing with Saline-Tween 20. Primary antibody was incubated at 37°C for 3 hours and the plates were then washed with Saline-Tween 20. Peroxidase conjugated anti-mouse (l/2,000) were added to wells and incubated for another 3 hours. At the end of the incubation, the wells were washed with Saline-Tween 20 and peroxidase substrate was added. The absorbance (450 run)was recorded with Mini-Reader II. *The end point (within 10 minutes) was obtained from the net value of the absorbance at 450 nm of sample well minus the absorbance of 450 nm of the control well

The effect of the anti-90.000 monoclonal immunozlobulin on olatelet aearezation: The effect of the monoclonal anti-90,000 immunoglobulin isolated from ascites fluid on platelet aggregation was examined in vitro (Figure 6). The addition of various amounts of the immunoglobulin to the PRP resulted in the inhibition of the collagen (6A), and ADP-(6C), but not shown) induced platelet a-thrombin (6B) and epinephrine-(data not aggregation. These results suggest that the 90,000 protein is important in platelet aggregation induced by some of the platelet agonists. DISCUSSION The identity of the platelet membrane component involved in the interaction with collagen has been the subject of investigation in several laboratories. We have previously isolated and purified a 65,000 protein from solubilized platelet membranes which specifically binds collagen and its Q chains and which inhibits collagen-induced platelet aggregation (15). By using an immunoblot technique, we have found that there are two other membrane proteins with molecular weights of 90,000 and 58,000 which also react with the polyclonal antibody raised against the 65,000 collagen receptor. The 90,000 and 58,000 proteins inhibited the interaction of the

140

Vol. 53, No. 2

COLLAGEN-PLATELET INTERACTION

polyclonal antibody raised against the 65,000 by competitive inhibition Antisera prepared to GPIIb/IIIa also recognized the ELISA (Table I). 90,000, 65,000, and 58,000 proteins in ELISA (Table III).

TABLE III REACTIVITY OF PURIFIED 90,000, 65,000, AND 58,000 PROTEINS TO RECEPTOR ANTIBODY AND POLYCLONAL AND MONOCLONAL ANTIBODIES TO GPIIb-IIIa

Coated Wells with 10 pg of Antisera of IgG Fraction

Anti-Receptor (65,000) Anti-GPIIb-IIIa (polyclonal) Anti-GPIIb-IIIa (monoclonal) Anti-GPIIIa (monoclonal) Blank

90,000

65,000

58,000

0.174a 0.215

0.184 0.268

0.194 1.198

0.577

0.631

0.621

0.060

0.062

0.070

0.060

0.060

0.068

Microtiter wells were coated individually with purified 90,000, 65,000, and 58,000 proteins in 0.05M Na2C03, pH 9.6 at 37°C for 3 hours and washed 7 times with Saline-Tween. Primary antibody was added to well. The dilution of antireceptor serum was l/200. All the other antibodies were IgG fractions isolated from serum or ascitic fluids and used at a concentration of 5 pg in each test well. These wells were incubated at 37°C for 3 hours. After washing 7 times with Saline-Tween, peroxidase-conjugated goat antirabbit or mouse IgG (l/2000) was added and incubated for another two hours, washed with Saline-Tween and the enzyme substrate was added. a: The absorbance was read at 450 run.

The relationship between the 90,000, 65,000, and 58,000 protein and polyclonal or monoclonal antibodies of GPIIb-IIIa was characterized by Polyclonal and monoclonal antibodies raised against GPIIb-IIIa ELISA. In contrast, a complex reacted with all three proteins (Table III). These results suggest monoclonal raised against GPIIIa failed to react. that these three proteins immunologically cross-react with GPIIb-IIIa complex but not at the epitope recognized by the anti-GPIIIa monoclonal antibody. The same anti-GPIIIa monoclonal antibody reacted with the purified GPIIIa (absorbance at 450 runwas 0.65) proving that the anti-GPIIIa antibody was active. The ratio of the three immunoblotted proteins (Figure 1) did not change in the presence and absence of protease inhibitors. These results suggest that the 65,000 and 58,000 proteins are not degraded products of the 90,000 protein.

Vol. 53, No. 2

COLLAGEN-PLATELET INTERACTION

141

To more precisely define the relationship between these three proteins, a monoclonal antibody was raised against the 90,000 protein. The monoclonal antibody against the 90,000 protein reacted also with the 65,000 and 58,000 proteins by the ELISA assay (Table II) as well as in immunoblots (Figure 5). These results further confirm that these three proteins share a common antigenic determinant. In support of this, the affinity purified collagen receptor also cross reacted immunologically with two monoclonal antibodies, PMI-1 which reacted with glycoprotein IIb and 22C4 which reacts with glycoptrotein IIIa (26). Furthermore, the isolate 65,000 receptor reacted with a specific antibody that is prepared to the glycoprotein IIbIIIa complex. These results indicate that these three proteins are immunologically cross reactive and functionally related to IIb/IIIa. The present finding of three immunoreactive bands differs from our earlier report that the anti-65,000 antiserum recognized only the 65,000 protein A possible explanation for this is that the 90,000 and 58,000 (15). proteins were not solubilized with 0.5% Triton in the earlier study, whereas they were solubilized by 0.2% NaDodS04 in the present study. ACKNOWLEDGEMENTS We wish to thank Ms. Virginia Woo and Mr. David Mensi for their expert technical assistance, Ms. Bwendolyn Scott and Ms. Kathy Pearson for excellent secretarial assistance in preparing the manuscript and Dr. L. Jennings for critical reading of the manuscript and purified GPIIb-IIIa and GPIIIa. This work was supported by grants from Veterans Administration Medical Research, National Institutes of Health (AM 16506) and American Heart Association (grant-in-aid, 84-799).

REFERENCES 1.

HUGHES, J. Accolement des plaquettes au collagen. C. B. Seances a. m. (Paris) 154:866-868, 1960.

2.

ZUCKER, M.B. and BORELLI, J. Platelet clumping procedured by connective tissue suspensions and by collagen. Proc.Soc.&.U. Med. m:779-787, 1962.

3.

HOVIG, T. Aggregation of rabbit blood platelets produced in vitro by saline "extract" of tendons. -. Throm -* Diath Haemorrh. 9:248-253, 1963.

4.

SPAET, T.H., CLINTRON, J., and SPIVACK, M. Some properties of the platelet-connective tissue mixed aggregation reaction. -.-.&IQ. Biol @. m:292-295, 1962. -a

5.

CHIANG, T.M., BEACHEY, E.H., and KANG, A.H. Interaction of a chick skin collagen fragment (al-CBS) with human platelets: biochemical Chem studies during the aggregation and release reaction. J. u. -* m:6916-6922, 1975.

6.

CHESNEY, C.M., HARPER, E., and COLMAN, R.W. Critical role of the carbohydrate side chains of collagen in platelet aggregation. J. Clin. Invest. =:2693-2701, 1972.

142

COLLAGEN-PLATELET INTERACTION

Vol. 53, No. 2

7.

PUETT, D., WASSERMAN, B.K., FORD, J.D., and GUNNIGNHAM, L.W. Collagenmediated platelet aggregation effects of collagen modification involving the protein and carbohydrate moieties. J. Clin. Invest. x:2495-2506, 1973.

8.

WILNER, G.D., NOSSEL, H.L., and PROCUPEZ, T.L. Aggregation of platelets by collagen: polar active sites of insoluble human collagen. &n. J. m. m:1074-1079, 1971.

9.

JAMIESON, G.A., URBAN, U.L., and BARBER, A.J. Enzymic basis for platelet-collagen adhesion as the primary step in haematosis. Nature (London) New Biol. 234:5-7, 1971.

10.

JAFFE, R., and DEYKIN, D. Evidence for a structural requirement for Invest. =:875-883, the aggregation of platelets by collagen. 2. m. 1974.

11.

MIJGGLI,H.L., and BAUMGARTNER, H.B. Collagen induced platelet aggregation: requirement for tropocollagen multimers. Thromb. m. 2~715-728, 1974.

12.

BRASS, L.F., and BENSUSAN, H.B. The role of collagen quaternary structure in the platelet-collagen interaction. J. m. Invest. j&:1480-1487, 1974.

13.

CHIANG, T.M., BEACHEY, E.H. and KANG, A.H. Binding of collagen al chains to human platelet. J. Biol. Chem. =:6347-6351, 1977.

14.

CHIANG, T.M., and KANG, A.H. Isolation and purification of a collagen al(I) receptor from human platelet membrane. J. Biol. Chem. =:75817586, 1982.

15.

CHIANG, T.M., KANG, A.H., DALE, J.B. and BEACHEY, E.H. Immunochemical studies of the purified human platelet receptor for the al(I) chain of chick skin collagen. z. Immunol. =:872-876, 1984.

16.

BARKER, A.J., and JAMIERSON, G.A. Characterization of plasma membranes from human platelets. J. Biol. Chem. =:6357-6365, 1970.

17.

LAMMELI, U.K. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. --. Nature 227.680-685, 1970.

18.

GEFTER, M.L., MARGUILES, T.H., and SHARFF, M.D. A simple method for polyethylene glycol promoted hybridization of mouse myeloma cells. Somatic u Genetic 3:231-236, 1977.

19.

VOLLER, A.D., BIRDWELL, E., and BARTLETT, A. The application of microplate enzyme-linked immunosorbent assays to show infectious disease. In Immunoenzvmatic Techniaue. C. Feldman, ed. Amsterdam, North-Holland, p. 167-173, 1976.

20.

AVRAMEAS, S., TANDOU, B., and TERNYNCK, T. Specificity of antibodies synthesized by immunocytes as detected by immunoenzyme techniques. Int. Arch Allerzy &&. Immunol. ,0:161-170, 1971. -0

Vol. 53, No. 2

COLLAGEN-PLATELET INTERACTION

143

21.

TOWBIN, H., STAEHLIN, I., and GORDON, J. Electrophorectic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedures and some applications. Proc. Natl. Acad. u. USA 76:4350-4354, 1979.

22.

GLASS, II, W.F., BRIGGS, R.C., and HAILICA, L.S. Identification of tissue specific nuclear antigens transferred to nitrocellulose from polyacrylamide gels. --. Science 211.70-73, 1981.

23.

KANG, A.H., PIEZ, K., and GROSS, J. Characterization of the a-chain of chick skin collagen and the nature of the NH2-terminal cross-link region. Biochemistrv 8:3648-3655, 1969.

24.

BORN, G.V.R. Aggregation of blood platelets by adenosine diphosphate and its reversal. --. Nature 194.927-929, 1962.

25.

LOWRY, O.H., ROSEBROUGH, N.J., FARR, A.L., and RANDALL, R.J. Protein measurement with Folin phenol reagent. J. Biol. Chem. m:265-275, 1951.

26.

SHADLE, P.J., GINSBERG, M.H., PLOW, E.F., and BARONEDES, S.H. Platelets-collagen adhesion: inhibition by a monoclonal antibody that binds glycoprotein IIb. J. Cell. Biol. =:2056-2060, 1984.