Advertisement

Article Free access | 10.1172/JCI6982

The PlA2 polymorphism of integrin β3 enhances outside-in signaling and adhesive functions

K. Vinod Vijayan,1 Pascal J. Goldschmidt-Clermont,2 Christine Roos,2 and Paul F. Bray1

1Department of Medicine and Pathology, Johns Hopkins University School of Medicine, Baltimore, Maryland 21205, USA2Heart and Lung Institute, and Division of Cardiology, Department of Medicine, The Ohio State University, Columbus, Ohio 43210, USA

Address correspondence to: Paul F. Bray, Baylor College of Medicine, Section of Thrombosis, One Baylor Plaza, BCM 286, Room N1319, Houston, Texas 77030, USA. Phone: (713) 798-3470; Fax: (713) 798-3415; E-mail: pbray@bcm.tmc.edu.

Find articles by Vijayan, K. in: JCI | PubMed | Google Scholar

1Department of Medicine and Pathology, Johns Hopkins University School of Medicine, Baltimore, Maryland 21205, USA2Heart and Lung Institute, and Division of Cardiology, Department of Medicine, The Ohio State University, Columbus, Ohio 43210, USA

Address correspondence to: Paul F. Bray, Baylor College of Medicine, Section of Thrombosis, One Baylor Plaza, BCM 286, Room N1319, Houston, Texas 77030, USA. Phone: (713) 798-3470; Fax: (713) 798-3415; E-mail: pbray@bcm.tmc.edu.

Find articles by Goldschmidt-Clermont, P. in: JCI | PubMed | Google Scholar

1Department of Medicine and Pathology, Johns Hopkins University School of Medicine, Baltimore, Maryland 21205, USA2Heart and Lung Institute, and Division of Cardiology, Department of Medicine, The Ohio State University, Columbus, Ohio 43210, USA

Address correspondence to: Paul F. Bray, Baylor College of Medicine, Section of Thrombosis, One Baylor Plaza, BCM 286, Room N1319, Houston, Texas 77030, USA. Phone: (713) 798-3470; Fax: (713) 798-3415; E-mail: pbray@bcm.tmc.edu.

Find articles by Roos, C. in: JCI | PubMed | Google Scholar

1Department of Medicine and Pathology, Johns Hopkins University School of Medicine, Baltimore, Maryland 21205, USA2Heart and Lung Institute, and Division of Cardiology, Department of Medicine, The Ohio State University, Columbus, Ohio 43210, USA

Address correspondence to: Paul F. Bray, Baylor College of Medicine, Section of Thrombosis, One Baylor Plaza, BCM 286, Room N1319, Houston, Texas 77030, USA. Phone: (713) 798-3470; Fax: (713) 798-3415; E-mail: pbray@bcm.tmc.edu.

Find articles by Bray, P. in: JCI | PubMed | Google Scholar

Published March 15, 2000 - More info

Published in Volume 105, Issue 6 on March 15, 2000
J Clin Invest. 2000;105(6):793–802. https://doi.org/10.1172/JCI6982.
© 2000 The American Society for Clinical Investigation
Published March 15, 2000 - Version history
Received: April 2, 1999; Accepted: January 25, 2000
View PDF
Abstract

Genetic factors are believed to influence the development of arterial thromboses. Because integrin αIIbβ3 plays a crucial role in thrombus formation, we analyzed receptor adhesive properties using Chinese hamster ovary and human kidney embryonal 293 cells overexpressing the PlA1 or PlA2 polymorphic forms of αIIbβ3. Soluble fibrinogen binding was no different between PlA1 and PlA2 cells, either in a resting state or when αIIbβ3 was activated with anti-LIBS6. PlA1 and PlA2 cells bound equivalently to immobilized fibronectin. In contrast, significantly more PlA2 cells bound to immobilized fibrinogen in an αIIbβ3-dependent manner than did PlA1 cells. Disruption of the actin cytoskeleton by cytochalasin D abolished the increased binding of PlA2 cells. Compared with PlA1 cells, PlA2 cells exhibited a greater extent of polymerized actin and cell spreading, enhanced tyrosine phosphorylation of pp125FAK, and greater fibrin clot retraction. These adhesion differences appear to depend on a signaling mechanism sensitive to receptor occupancy. Thus, the PlA2 polymorphism altered integrin-mediated functions of adhesion, spreading, actin cytoskeleton rearrangement, and clot retraction.

Introduction

With few exceptions, defects in single genes do not cause ischemic coronary disease, and it is likely that contributions of multiple inherited and environmental influences culminate in the event recognized as a myocardial infarction. Platelet aggregation mediated by integrin αIIbβ3 (glycoprotein IIb-IIIa) plays a key role in unstable ischemic coronary syndromes (1), and the possibility that the PlA2 polymorphism of integrin β3 might contribute to the genetic component of ischemic vascular disease has been given credence by some (2–5) but not other (6, 7) clinical epidemiology studies (for review, see ref. 8). Platelet studies on donors of known PlA status have suggested a possible effect on platelet function. For example, most anti-PlA1 antibodies completely inhibit the aggregation of PlA1/A1 platelets and only retard or partially inhibit aggregation of PlA1/A2 platelets (9, 10). Feng et al. found that the PlA2 polymorphism was associated with a lower threshold of platelet aggregation (11), and we have observed that compared with PlA2-negative individuals, PlA2-positive platelets have a lower threshold for αIIbβ3 activation and α granule release (12).

Using platelets in formal studies of the consequences of the PlA2 polymorphism on αIIbβ3 function presents several challenges. First, the marked heterogeneity in in vitro functional assays among ostensibly normal donors presents difficulties in the detection of small differences. An example of this heterogeneity comes from estimates of the Kd for fibrinogen binding to αIIbβ3, which have varied by 2 orders of magnitude (13–15). Secondly, age, medications, gender, cigarette smoking, co-morbid conditions, or other unknown variables affect platelet function and limit both the interpretation of the data obtained and the pool of potential donors. Lastly, PlA2-homozygous individuals represent less than 2% of the population (16), making these platelets difficult to obtain. Stable cell lines overexpressing the αIIbβ3 have been used to elucidate many of the kinetic and signaling properties of αIIbβ3 receptor function (17, 18). The aim of the current study was to determine whether the PlA2 polymorphism could influence functions relevant to the process of thrombosis by using Chinese hamster ovary (CHO) and 293 cell lines overexpressing the 2 PlA forms of αIibβ3. We found that PlA2-expressing cells demonstrated increased adhesive functions compared with their PlA1-expressing counterparts, and these differences appear to be the result of a differential dependency on outside-in signaling pathways. These functional differences provide physiologic support for the epidemiologic association between the genetic and platelet components of coronary artery disease.

Methods

Materials. The CHO cell line CHO–K1 (CCL-61) and human embryonal kidney cell line 293 (CRL 1573) were obtained from the American Type Culture Collection (Rockville, Maryland, USA). All reagents that do not specify the name of the manufacturers were from Sigma Chemical Co. (St. Louis, Missouri, USA).

Generation of cell lines. The cDNA for β3 (19) corresponding to the PlA1 polymorphism was mutated to code for the PlA2 polymorphism using the Altered Sites II in vitro Mutagenesis Kit (Promega Corp., Madison, Wisconsin, USA) as described previously (20) and sequenced to confirm authenticity. The cDNAs for αIIb (21) and for β3 (either PlA1 or PlA2) were engineered into the LK444 vector (22) and used to transfect CHO cells. In addition, the cDNA for αIIb was engineered into the pZeoSV2 vector (Invitrogen Corp., Carlsbad, California, USA) and the cDNA for β3 (either PlA1 or PlA2) engineered into the pcDNA3.1 vector (Invitrogen Corp.) and used to transfect 293 cells. Both CHO and 293 cells were transfected using lipofectin (Life Technologies Inc., Gaithersburg, Maryland, USA) (23) with the respective plasmids for both αIIb and β3 subunits. Control CHO cells (designated LK444) were transfected with the LK444 parental vector whereas the control 293 cells (designated Pc/Z) were transfected with empty pcDNA3.1 and pZeoSV2 vectors. Stable CHO cell lines expressing the neomycin resistance gene were selected in the presence of 250 μg/mL G418 (Life Technologies) and cultured in αMEM (Life Technologies) containing 10% FBS (Gemini Bio-Products, Calabasas, California, USA.), and antibiotics (penicillin, streptomycin, amphotericin; Life Technologies). Stable 293 cell lines expressing the neomycin and zeocin resistance genes were selected in the presence of 300 μg/mL G418 and 100 μg/mL zeocin (Invitrogen Corp.) and cultured in DMEM (Life Technologies) containing 10% FBS containing the same antibiotics as above. αIIbβ3-expressing cells were sorted by flow cytometry using the αIIbβ3 complex-specific mAb, P2 (Immunotech, Marseilles, France) to select for high and equivalent expression. These cells were pooled and used for the study.

Flow cytometry. CHO and 293 cells were grown to 70–80% confluency and detached using 0.05% trypsin (Life Technologies). After neutralization with complete media, the cells were suspended in PBS (pH = 7.4; 0.137 M NaCl, 4.3 mM Na2HPO4, 1.4 mM KH2PO4, 2.7 mM KCl) with 2% BSA and incubated with either 3 μg/mL P2, 2 μg/mL anti-β3 mAb (clone SZ21; Immunotech), or 3 μg/mL control mouse IgG (Pierce Chemical Co., Rockford, Illinois, USA) for 1 hour on ice. The cells were washed, incubated with a secondary goat anti-mouse FITC-labeled antibody, and analyzed on a FACScalibur flow cytometer (Becton Dickinson, San Jose, California, USA), with fluorescence data acquired in the logarithmic mode and light scattering data acquired in the linear mode (24). The mean channel number, which corresponded to cell-fluorescence intensity, was used as the measure of surface-expressed αIIbβ3 for the whole population.

Western blot analysis. Cells were lysed in 15 mM HEPES, pH 7.0, 145 mM NaCl, 0.1 mM MgCl2, 10 mM EGTA, 1% Triton X-100, 1 mM NaVO4, 250 μg/mL 4-2-aminoethyl-benzene sulfonylfluoride, 15 μg/mL of protease inhibitors chymostatin, antipain, and pepstatin, and 55 μg/mL of the protease inhibitor leupeptin. Fifteen micrograms of cell lysate or 2 μg platelet lysate were separated in 7% SDS polyacrylamide gels under nonreducing conditions, transferred to nitrocellulose, and then subjected to Western blotting using a 1:500 dilution of monoclonal anti-αIIb (132.1) and monoclonal anti-β3 (AP-3) (gifts from Peter Newman, Blood Research Institute, Milwaukee, Wisconsin, USA) or 1:1000 dilution of SZ21 antibody. After the addition of horseradish peroxidase–labeled goat anti-mouse (1:3000 dilution), blots were developed using enhanced chemiluminescence (ECL; Amersham Life Science, Piscataway, New Jersey, USA) (25).

Binding to soluble fibrinogen. CHO cells were trypsinized and rinsed as above, and 5 × 105 cells were resuspended in 50 μL Tyrode’s buffer (138 mM NaCl, 2.9 mM KCl, 12 mM NaHCO3, 0.36 mM Na2HPO4, 5.5 mM glucose, pH 7.4) containing 1 mg/mL BSA, 1 mM MgCl2, 0.1 mM CaCl2, and 20 mM HEPES, pH 7.4. The cells were incubated with varying concentrations of FITC-conjugated fibrinogen (0–100 μg/mL) in the presence or absence of 0.5 μM αIIbβ3-activating mAb LIBS-6 (gift of Mark Ginsberg, The Scripps Research Institute, La Jolla, California, USA) for 30 minutes at 22°C. Cells were analyzed for FITC fluorescence by flow cytometry. Specific fibrinogen binding was defined as the difference in mean fluorescence between LIBS-6–activated and nonactivated cells.

Adhesion to immobilized ligands. Cells were trypsinized and rinsed as above, suspended in 0.01% BSA in Tyrode’s buffer for adhesion studies to fibrinogen (Enzyme Research Laboratories Inc., South Bend, Indiana, USA) or in 0.01% BSA in HBSS (136 mM NaCl, 5.3 mM KCl, 0.33 mM Na2HPO4, 0.44 mM KH2PO4, 5.5 mM glucose, pH 7.4) for adhesion studies to fibronectin (Life Technologies). Cells were further labeled with 51chromium (51Cr; Amersham) by incubating in 25 μCi/200 μL at 37°C for 1 hour with gentle shaking. Cells were washed 3 times with Tyrode’s buffer or HBSS and resuspended at a concentration of 3.5 × 105 cells/mL in Tyrode’s buffer or HBSS with 1.8 mM CaCl2 and 0.49 mM MgCl2. Adhesion to immobilized fibrinogen and fibronectin were performed as described previously (26). Twenty-four–well tissue culture plates (Falcon, Lincoln Park, New Jersey, USA) were coated with varying concentrations (2.5, 5, 12.5, and 20 μg/mL) of fibrinogen or fibronectin in Tyrode’s buffer or HBSS, respectively, for 2 hours at 22°C. Wells were washed twice with Tyrode’s buffer or HBSS and incubated with 500 μL of 5 mg/mL of heat-inactivated BSA for 2 hours at 22°C, with constant shaking on a platform shaker to block nonspecific binding to the plates. Control wells were coated only with 500 μL of 5 mg/mL heat-inactivated BSA. Before the adhesion assay the wells were washed twice with Tyrode’s buffer or HBSS. For adhesion studies 0.7 × 10551Cr-labeled cells in 0.2 mL were added to each well and incubated at 37°C for 5 minutes. Unbound cells were removed by inverting the plates. Each well was washed 3 times with 200 μL of Tyrode’s buffer or HBSS by briskly tilting the plate 10 times at an angle of 90° along an axis parallel to that of the length of the plate, followed by 10 times along an axis perpendicular to that of the length of the plate. The buffer was removed by inverting the plate, and the excess liquid was dabbed on a filter paper. Note that each of the cell lines under comparison was present on the same plate and hence exposed to identical adhesion and washing conditions. The bound cells were lysed with 0.1% SDS, and the lysate was counted on a gamma counter. The specific percent of cells bound to fibrinogen in each well was calculated as: (cpm from fibrinogen/fibronectin-coated wells) – (cpm from BSA-coated wells) × 100/total cpm added to each well.

Specificity of adhesion to fibrinogen was investigated by incubating the 51Cr-labeled cells for 10 minutes with 1 of the following inhibitors: 2.1 μg/mL 10E5, a blocking mAb against αIIbβ3 (gift from Barry Coller, Mt. Sinai Hospital, New York, New York, USA); 5 μM integrelin, a blocking cyclic peptide specific for αIIbβ3 (gift from COR Therapeutics Inc. South San Francisco, California, USA); 200 μM RGD peptide; 10 μg/mL anti-αvβ3 mAb LM609 (Chemicon International Inc., Temecula, California, USA); 200 μM RGE peptide; or 10 μg/mL mouse IgG. In certain experiments 51Cr-labeled cells were incubated for 30 minutes with either 10 μM cytochalasin D, 12 μM bisindolylmaleimide, or control DMSO. A 5-minute adhesion assay was then performed as described above. In certain experiments 10- or 15-minute adhesion assays were also performed.

Confocal microscopy. Glass coverslips were coated with 12.5 μg/mL fibrinogen in Tyrode’s buffer for 2 hours at 22°C. Fibrinogen-coated coverslips were incubated with 105 CHO cells in 100 μL for 5 minutes at 37°C. Unbound cells were removed and cells fixed with 3.5% paraformaldehyde for 8 minutes, permeabilized with 0.1% NP-40 for 15 minutes, and transferred to PBS containing 10% FCS for 20 minutes at 22°C. Cells were stained for actin with 0.6 μM rhodamine-conjugated phalloidin (Molecular Probes, Eugene, Oregon, USA) for 30 minutes at 22°C. The coverslips were washed twice and mounted in permafluor (Lipshaw Immunon, Pittsburgh, Pennsylvania, USA). Fluorescent images were obtained with Zeiss LSM 410 confocal imaging system attached to a Zeiss C-apo ×40 water immersion lens microscope (Carl Zeiss Inc., Thornwood, New York, USA). The degree of fluorescence and the area of cell spreading were quantitated by scoring for individual cells using the Metamorph image analysis system (Universal Imaging Corp., Westchester, Pennsylvania, USA).

Clot retraction assay. CHO cells grown to confluence were harvested as described above and resuspended in αMEM containing 28 mM CaCl2 and 25 mM HEPES, pH 7.4. A total of 4 × 106 cells in 300 μL were mixed with 200 μL of fibronectin-depleted plasma, 250 μg of fibrinogen, 5 μg of aprotinin in a 75 × 100-mm glass tube (Fisher Scientific, Pittsburgh, Pennsylvania, USA). Clot formation was initiated by the addition of 2.5 U thrombin (Sigma). After incubation at 37°C for varying time periods, clot retraction was quantitated by measuring the volume of liquid not incorporated into the clot, as described (27). The percentage of clot retraction was calculated as: (amount of liquid collected)/(the total amount of liquid before the clot) × 100. Percentage of clot retraction was normalized for CHO cell αIIbβ3-receptor density by dividing by the mean fluorescence intensity (clot retraction units). In certain experiments, cells were pretreated for 15 minutes with 10 μg/mL of LM609 or 5 μM of integrelin.

Tyrosine phosphorylation of pp125FAK. Twenty-four–well tissue culture plates were coated with either 12.5 μg/mL of fibrinogen or 2.5 mg/mL heat-treated BSA. Two hundred microliters of 3.5 × 105 CHO cells/mL were added to each well and incubated for 5 minutes at 37°C in 5% CO2. The adherent cells were lysed with ice-cold lysis buffer (described above), and the lysate was incubated for 30 minutes on ice, clarified by centrifugation at 14,000 g for 20 minutes, and the protein content determined using then Bio-Rad protein assay kit (Bio-Rad Laboratories, Inc., Hercules, California, USA). One hundred fifty microliters of proteins from each lysate were precleared with protein A-Sepharose (Pharmacia Biotech, Piscataway, New Jersey, USA) and incubated with 2 μg of anti-pp125FAK antibody (Upstate Biotechnology Inc., Lake Placid, New York, USA) for 16 hours at 4°C and precipitated using protein A-Sepharose beads. After washing the beads 3 times with ice-cold lysis buffer, proteins were eluted in boiling Laemmli sample buffer containing 1 mM orthovanadate. Proteins were separated on a 7.5% SDS-polyacrylamide gel, transferred to nitrocellulose, probed with antiphosphotyrosine 4G10 antibody (Upstate Biotechnology) or anti-pp125FAK antibody, and developed as described above. The signals were scanned using PhotoShop 5 software and the densitometric quantitation performed using NIH Image software [developed at the Research Services Branch of the National Institute of Mental Health in Bethesda, Maryland, USA (http://rsb.info.aov/nih-image/default.html)].

Results

Overexpression of the PlA1 and PlA2 isoforms of αIIbβ3 in CHO cells. Wild-type CHO cells were stably transfected with expression plasmids containing the cDNAs for αIIb and β3 (either the PlA1 or Pl A2 forms) or the empty parental LK444 vector. Flow cytometric analysis with a mAb specific for the αIIbβ3 complex revealed equivalent levels of receptor expression on the cell lines generated with the PlA1 and PlA2 isoforms, designated as lines A1 and A2, respectively (Figure 1a). The neomycin-resistant control CHO cells, designated LK444, did not express detectable αIIbβ3. The mAb SZ21 was used to distinguish the 2 forms of β3 because it binds the PlA2 isoform of β3 with a much lower affinity than the PlA1 isoform (25). In addition, Western immunoblot analysis demonstrated the presence of αIIb and β3 in cell lysates of A1 and A2 cells, but not LK444 cells, (Figure 1b). As with the flow cytometry, the PlA1 form, but not PlA2 form, of β3 could be immunodetected using SZ21 antibody (not shown).

Characterization of stable αIIbβ3–expressing CHO cell lines. (a) Flow cytomFigure 1

Characterization of stable αIIbβ3–expressing CHO cell lines. (a) Flow cytometric analysis. LK444, A1, and A2 cell lines were labeled with 3 μg/mL isotype-matched control IgG (upper panel), 3 μg/mL αIIbβ3 complex–specific mAb P2 (middle panel), or 2 μg/mL SZ21 mAb that distinguishes PlA1 from PlA2 (lower panel). LK444, solid line; A1, heavy dotted line; A2, fine dotted line. (b) Western immunoblotting. Fifteen micrograms of CHO cell lysates or 2 μg of platelet lysates were separated by nonreduced SDS-PAGE, transferred to nitrocellulose, and probed with either 132.1 (αIIb-specific) or AP3 (β3-specific) mAbs.

Binding to soluble fibrinogen. We assessed initially whether the PlA2 isoform was in a conformation favoring soluble ligand binding. Similar to previous studies where αIIbβ3-expressing CHO cells did not bind soluble fibrinogen in response to physiologic agonists (18), we observed no binding of soluble fibrinogen or the mAb PAC-1 (specific for the activated conformation of αIIbβ3) to the “resting” states of either the A1 and A2 cell lines (data not shown). In addition, neither the thrombin receptor-activating peptide (100 μM) nor epinephrine (20 μM) caused fibrinogen binding to either cell line. However, the αIIbβ3-activating mAb LIBS-6 induced equivalent fibrinogen binding to both the PlA1 and PlA2 CHO cell lines, both at subsaturating (Figure 2) and saturating (20–100 μg/mL; data not shown) fibrinogen concentrations. The concentration of fibrinogen producing half-maximal binding appeared to be no different for A1 cells and A2 cells. Thus, although both isoforms were functional in CHO cells, neither the resting nor activated conformation of PlA2 resulted in greater soluble fibrinogen binding compared with PlA1, and we conclude that there is no significant difference in the binding kinetics of soluble fibrinogen to the 2 isoforms of β3.

Binding to soluble fibrinogen. LK444 (open squares), A1 (open diamonds), anFigure 2

Binding to soluble fibrinogen. LK444 (open squares), A1 (open diamonds), and A2 (open circles) cells were incubated with 5 μM LIBS-6 to activate αIIbβ3 and varying concentrations of FITC-labeled fibrinogen, then analyzed by flow cytometry. The graph is expressed as SEM of 4 experiments. Best-fit curves for LK444, A1, and A2 cells were generated using the logarithmic mode in CA Cricket Graph III version 1.5.2. The inset shows an expanded view of the fibrinogen concentrations between 0 and 1 μg/mL.

Adhesion to immobilized ligands. Because in vivo thrombus growth requires platelet adhesion to fibrinogen, we next examined the ability of αIIbβ3-expressing CHO cells to bind immobilized ligands. Both A1- and A2-expressing cells bound in significantly greater numbers than did the LK444 to immobilized fibrinogen, suggesting αIIbβ3 mediated adhesion (Figure 3a). Whereas there was a slight trend toward greater binding to soluble fibrinogen in A2 cells compared with A1 (Figure 2), significantly greater A2 cell adhesion was observed over a range of immobilized fibrinogen concentrations (P ≤ 0.001). In contrast, there was equivalent adhesion of A1 and A2 cells to fibronectin (a ligand for endogeneous CHO cell α5β1) (Figure 3b), suggesting that the differential adhesion due to the PlA polymorphism was ligand specific. Similarly, no difference in adhesion to an immobilized anti-β3 antibody (AP3) was observed between A1 and A2 cells (data not shown). Surface expression of αIIbβ3 was not detectably different between the A1 and A2 cell lines and could not account for the observed difference in adhesion (Figure 3c). Specificity of binding to fibrinogen by both the A1 and A2 cell lines was further demonstrated by more than 90% inhibition with αIIbβ3-specific blocking antibody 10E5, integrelin, and RGD peptides (Figure 4). RGE peptides showed no inhibition, and the LM609 mAb, which recognizes the chimeric hamster-human αvβ3 receptor, inhibited little of the A1 and A2 cell adhesion to fibrinogen (Figure 4), indicating adhesion was mediated primarily through αIIbβ3. In addition, these assays were performed in the presence of 1.8 mM calcium and no manganese, conditions not permissive for the αvβ3–mediated adhesion to fibrinogen (28).

Adhesion of CHO cell lines to immobilized ligands. The 51Cr-labeled LK444 (Figure 3

Adhesion of CHO cell lines to immobilized ligands. The 51Cr-labeled LK444 (open squares), A1 (open diamonds), and A2 (open circles) cells were allowed to adhere to immobilized fibrinogen (a) or fibronectin (b) for 5 minutes. The results are expressed as ± SEM for 5–21 observations. Compared with A1 cells, A2 cells demonstrated 80% greater adhesion at 2.5 μg/mL (P < 0.001), 60% greater adhesion at 5 μg/mL (P < 0.001), 31% greater adhesion at 12.5 μg/mL (P < 0.001), and 61% greater adhesion at 20 μg/mL (P = 0.001). There was no difference in adhesion between A1 and A2 cells at any concentration of fibronectin (P ≥ 0.22). (c) Mean fluorescence intensity of P2 binding (αIIb-specific) to the 3 CHO cell lines determined within 24 hours of the adhesion assay.

Adhesion of A1 and A2 CHO cells is αIIbβ3 specific. Adhesion was as in FiguFigure 4

Adhesion of A1 and A2 CHO cells is αIIbβ3 specific. Adhesion was as in Figure 3, except that A1 and A2 cells were first incubated for 10 minutes with buffer or 2.1 μg/mL 10E5, 10 μg/mL LM609, 5 μM integrelin, 200 μM RGD peptide, or 200 μM RGE peptide. Cells were allowed to bind for 5 minutes before washing. Binding without inhibitors is displayed as 100%. The experiment was performed 3 times; error bars were too narrow to be seen with the 10E5, integrelin, and RGD inhibition. Binding of A1 and A2 cells after inhibition with 10E5, integrelin, and RGD was equivalent to basal binding of the LK444 control line.

Studies in 293 cells. Although CHO cell lines were generated by cell sorting and should not have been subject to clonal variation, we generated a second set of cell lines to confirm the functional difference associated with the PlA2 polymorphism. Wild-type 293 cells were stably transfected as described in Methods. Equivalent levels of receptor expression on the cell lines generated with the PlA1 (called A1.2) and PlA2 (called A2.2) isoforms were seen by flow cytometry (Figure 5a). Both A1.2 and A2.2 cells, but not the vector-only control (called Pc/Z) cells, showed αIIb and β3 with Western blot analysis. SZ21 antibody was used to distinguish the 2 forms of β3 (Figure 5b).

Characterization of stable αIIbβ3-expressing 293 cell lines. (a) Flow cytomFigure 5

Characterization of stable αIIbβ3-expressing 293 cell lines. (a) Flow cytometric analysis. Pc/Z, A1.2, and A2.2 cell lines were labeled with 3 μg/mL isotype-matched control IgG (upper panel), 3 μg/mL αIIbβ3 complex–specific mAb P2 (lower panel). Pc/Z, solid line; A1.2, heavy dotted line; A2.2, fine dotted line. (b) Western immunoblotting of the 293 cell lines. Fifteen micrograms of 293 cell lysates or 2 μg of platelet lysates were separated by nonreduced SDS-PAGE, transferred to nitrocellulose, and probed with 132.1 (αIIb-specific), AP3 (β3-specific), or SZ21 (PlA1-specific) mAbs.

Both the A1.2 and A2.2 cells bound in significantly greater numbers than the Pc/Z to immobilized fibrinogen (Figure 6). A2.2 cells exhibited greater binding over A1.2 cells after 5 minutes over a range of immobilized fibrinogen concentrations. (Figure 6a). The observed difference in adhesion between A1.2 and A2.2 cells was not because of difference in their surface expression of αIIbβ3 (Figure 6b). Furthermore, this adhesion was abolished by the 10E5 antibody but not by LM609 antibody or mouse IgG (Figure 6c). Thus, 2 independent sets of cell lines from 2 different parental strains yielded similar data regarding greater adhesion of PlA2-expressing cells compared with PlA1-expressing cells. This makes clonal variation a highly unlikely explanation for these data and greatly strengthens the conclusion that the different adhesive properties are intrinsic to the PlA polymorphism.

Adhesion of 293 cell lines to immobilized fibrinogen. (a) Pc/Z (open squareFigure 6

Adhesion of 293 cell lines to immobilized fibrinogen. (a) Pc/Z (open squares), A1.2 (open diamonds), and A2.2 (open circles) cell adhesion was performed as in Figure 3. The results are expressed as ±SEM of 9 observations for all concentrations. Compared to A1 cells, A2 cells demonstrated 26% greater adhesion for 2.5 μg/mL, 25% greater adhesion for 5 μg/mL, 23% greater adhesion for 12.5 μg/mL, and 37% greater adhesion for 20 μg/mL. The increased adhesion of A2 over A1 was significant at P = 0.005 for 2.5 μg/mL, P = 0.05 for 5 μg /mL, P = 0.006 for 12.5 μg/mL. (b) Mean fluorescence intensity of P2 binding to the three 293 cell lines, determined within 24 hours of the adhesion assay. (c) Inhibition of A1.2 and A2.2 cell lines. A1.2 and A2.2 cells were first incubated for 10 minutes with buffer or 2.1 μg/mL 10E5, 10 μg/mL LM609, 10 μg/mL mouse IgG. Cells were allowed to bind for 5 minutes before washing. Binding without inhibitors is displayed as 100%. Results are expressed as ± SEM of 3 observations. Error bars were too narrow to be seen with the 10E5 inhibition.

Role of actin cytoskeleton in the differential adhesion to fibrinogen. During the course of these studies, it appeared that the bound A2 cells exhibited a different morphology than the bound A1 cells. The actin cytoskeleton plays an important role in a large number of cellular functions, including cell adhesion, shape, and motility (29) and is rapidly remodeled in response to external stimuli. We used confocal microscopy of rhodamine-phalloidin–stained actin and observed more F actin in the periphery of A2 cells compared with A1 cells (Figure 7a, upper row). With additional time for adhesion, substantially greater cell spreading was observed in the A2 cells compared with the A1 cells (Figure 7a, lower row). Quantitative image analysis revealed a greater actin staining (P < 0.001; Figure 7b) and greater surface area (P = 0.01; Figure 7b) for A2 cells adhered to fibrinogen compared with A1 cells. Figure 8a shows how cytochalasin D abolishes the difference in adhesion between the A1 and A2 cell lines over a range of fibrinogen concentrations at the 5-minute time point. Cytochalasin D inhibited both A1 and A2 cell adhesion, but the overall percent of inhibition was greater in the A2 cells. The inhibitory effects of cytochalasin D on both cell lines was time dependent and largely lost after the 5-minute time point (Figure 8b), perhaps because of the recruitment of other noncytoskeletal mechanisms that stabilize cell adhesion. However, the ability of cytochalasin D to eliminate the adhesion difference between A1 and A2 cells persisted at these later time points. Surface expression of αIIbβ3 was not detectably different between the A1 and A2 cell lines in these experiments (Figure 8c). These findings indicate that an intact cytoskeletal architecture is required for the rapid stable adhesion of both A1 and A2 cells to fibrinogen, but the PlA2 isoform of β3 demonstrates a more robust and brisk actin reorganization.

(a) Confocal microscopy of the actin cytoskeleton in CHO cells adhered to fFigure 7

(a) Confocal microscopy of the actin cytoskeleton in CHO cells adhered to fibrinogen. Cells adherent to coverslips coated with fibrinogen were stained with rhodamine phalloidin to detect F actin. A greater actin polymerization was seen for A2 cells bound to fibrinogen for 5 minutes compared with A1 cells (upper row). After 15 minutes of incubation, greater spreading was seen for A2 compared with A1 cells (lower row). (b) The morphometric analysis of actin staining and cell spreading is shown to the right in the upper and the lower graphs, respectively. At 5 minutes F-actin staining of A2 compared with A1 cells was significantly greater (P < 0.001). Actin staining is shown as arbitrary units of fluorescence defined by Metamorph software. Results are expressed as ± SEM of values obtained from 50 cells. At 15 minutes the greater surface area for A2 cells compared with A1 cells adhered to fibrinogen was significant at P = 0.01. Surface area is shown as arbitrary units of cell area. Results are expressed as ± SEM of values obtained from 35 cells.

Effect of cytochalasin D on CHO cell adhesion. Cells were treated with eithFigure 8

Effect of cytochalasin D on CHO cell adhesion. Cells were treated with either DMSO or 10 μg/mL of cytochalasin D for 30 minutes, washed twice, and adhesion assays performed as above. (a) Adhesion for 5 minutes to different concentrations of fibrinogen. *P ≤ 0.008 for A1 vs. A2; †P > 0.46 for A1 vs. A2; #P < 0.03 for A1 (DMSO vs. cyto-D), and P < 0.001 for A2 (DMSO vs. cyto-D). The results are expressed as ± SEM of 6 observations. (b) Adhesion to 12.5 μg/mL fibrinogen for different time points. *P ≤ 0.007 for A1 vs. A2; †P ≤ 0.08 for A1 vs. A2; #P ≤ 0.0008 for A1 and A2 (DMSO vs. cyto-D at 5 minutes), and P > 0.19 for A1 (DMSO vs. cyto-D at 10 and 15 minutes), and P = 0.06 for A2 (DMSO vs. cyto-D at 10 and 15 minutes). (c) Mean fluorescence intensity of P2 binding to the 3 CHO cell lines. Error bars are too narrow to be seen.

Analysis of signaling pathways that influence adhesion. We investigated postreceptor occupancy signaling pathways that might affect reorganization of the actin cytoskeleton and hence, cell adhesion. Focal adhesion kinase, pp125FAK, a cytoplasmic tyrosine kinase, is an important regulator of signaling in focal contacts. Cell spreading and focal contact is dictated partly by the tyrosine phosphorylation status of this protein, which in turn is regulated by protein kinase C (PKC) (30). Incubation of the A1 and A2 cells with the PKC inhibitor, bisindolylmaleimide, caused a greater inhibition in adhesion of A2 cells (83.3%; P = 0.02) than A1 cells (32.7%; P = 0.173) (data not shown), so we assessed pp125FAK phosphorylation in the CHO cell lines (Figure 9a). A small but consistent increase in pp125FAK phosphorylation was observed in the A2 cells compared with the A1 cells that had adhered to fibrinogen (Figure 9a, lane 7 versus lane 8). This difference depended upon adhesion to fibrinogen because it was not observed in suspension cells (Figure 9a, lanes 1–3) or in cells incubated on albumin (Figure 9a, lanes 4–6). In most experiments LK444 cells showed little pp125FAK phosphorylation. Figure 9b shows the ratio of phosphorylated pp125FAK to immunoprecipitated pp125FAK in 4 experiments. Although the difference in signals between A1 and A2 cells was modest, this data raised the possibility that downstream αIIbβ3 signaling might be different in A2 cells compared with A1 cells.

pp125FAK tyrosine phosphorylation of CHO cells. (a) Immunoblot showing tyroFigure 9

pp125FAK tyrosine phosphorylation of CHO cells. (a) Immunoblot showing tyrosine phosphorylation of pp125FAK (P-FAK). LK444, A1, and A2 cells either in suspension (lanes 1–3), or adhered to BSA (lanes 4–6) or fibrinogen (lanes 7–9) were solubilized. P-FAK was immunoprecipitated, separated by SDS-PAGE, and blotted with an antiphosphotyrosine antibody (upper panel). The lower panel shows the same filter stripped and reprobed with an anti–P-FAK antibody. (b) Densitometric quantitation of the P-FAK phosphorylation (ratio of tyrosine phosphorylation P-FAK to total P-FAK immunoprecipitated in arbitrary units) in 4 different experiments. FAK phosphorylation in panel a is experiment 1 displayed in b.

Fibrin clot retraction represents another outside-in signaling (downstream) cellular function involving αIIbβ3 that also depends on actin cytoskeletal rearrangement. We found that both A1 and A2 cells, but not LK444 cells, could retract a fibrin clot (Figure 10a). Compared with A1 cells, A2 cells exhibited a small but significant increase in fibrin clot retraction at all time points. This clot retraction could be inhibited by an αIIbβ3-specific peptide, but not by a blocking anti-αvβ3 antibody (Figure 10, b and c). These clot retraction studies provided additional evidence for functional differences in A1 and A2 cells that was downstream of receptor occupancy.

Fibrin clot retraction. (a) Clot retraction of LK444 (open squares), A1 (opFigure 10

Fibrin clot retraction. (a) Clot retraction of LK444 (open squares), A1 (open diamonds), or A2 (open circles) cells was performed as described in Methods. The increased fibrin clot retraction of A2 over A1 was significant at P = 0.002, P = 0.003, P = 0.004, P = 0.03 for 30, 60, 90, and 120 minutes, respectively. The results are ± SEM of 6 experiments performed. Error bars were too small to be displayed. (b) A1 cells and (c) A2 cells demonstrate that clot retraction is mediated through αIIbβ3. Cells were incubated with buffer (open squares), 10 μg/mL anti αvβ3 LM609 (open circles), or 5 μM integrelin (open diamonds), and clot retraction was performed as above. The results are ± SEM of 3 experiments performed. Error bars are too small to be displayed.

Discussion

In this work we have studied the adhesive and signaling properties of cell lines expressing the PlA1 or PlA2 polymorphism of integrin β3. Compared with PlA1-expressing cells, the PlA2-expressing cells exhibited increased adhesion to immobilized fibrinogen with greater cell spreading and F-actin content and increased clot retraction, all of which suggest differences in outside-in signaling events. Should similar properties hold true in platelets, this genetic alteration of integrin β3 might account for some component of the reported associations between PlA2 and acute coronary thrombotic syndromes.

We used a CHO and 293 cell system to address functional differences between the PlA1 and PlA2 forms of αIIbβ3 to overcome some of the limitations of platelets, and several issues related to our experimental conditions should be mentioned. First, because cell lines derived from single-cell clones can acquire properties that might affect function beyond that induced by the exogenously expressed cDNAs, we generated stable populations of overexpressing CHO and 293 cells by cell sorting. These “pools” of clones make genetic drift a much less likely explanation for the differences we observed. Through cell sorting, we maintained the equivalent expression of αIIbβ3 receptors and did not perform studies if there was a greater than 10% difference in surface density of αIIbβ3 between the A1 and A2 cell lines. Importantly, we always observed greater A2 cell binding even when the surface density of αIIbβ3 was greater on A1 compared with A2 cells (data not shown), further substantiating a qualitative rather than quantitative difference in the 2 β3 isoforms. Second, the low levels of chimeric hamster-human αvβ3, expressed in our cell lines had little contribution to cell adhesion, as demonstrated by the minor degree of inhibition by LM609. The use of a calcium-containing buffer, which does not support αvβ3-fibrinogen binding (28), may have contributed to the αIIbβ3 specificity of our studies. Lastly, although these studies were performed under static conditions, shear was applied during the washing procedure. Indeed, in complementary studies (31), shear was found to augment the difference in A2 cell binding to fibrinogen relative to A1 cells.

The Leu→Pro substitution at amino acid 33 of integrin β3 results in a conformation change in the extracellular domain (32). With respect to soluble fibrinogen at subsaturating concentrations, we found this change in conformation had no effect on binding in either the resting or activated state of the receptor (Figure 2). This argues against any difference in the apparent Kd between the 2 forms of β3 for fibrinogen. This was consistent with findings by Bennett et al., who found no difference in soluble fibrinogen-binding kinetics to platelets from donors of PlA1/A1 and PlA1/A2 genotype (33). Because thrombus formation involves platelet binding to both soluble and immobilized fibrinogen, we studied this latter process and found significantly greater αIIbβ3-mediated binding of A2 cells compared with A1 cells, regardless of the concentration of immobilized fibrinogen used or the time of incubation. Adhesion data (Figure 6a) from a second set of cell lines generated in a different parental strain further confirmed the functional differences between the 2 forms of β3. In contrast, adhesion of CHO cells to fibronectin (a ligand for endogeneous α5β1) showed no difference by PlA genotype. Immobilization of fibrinogen at the densities used in these experiments is known to result in a conformation that differs significantly from that of soluble fibrinogen (34), and this might explain the difference we observed in binding between the soluble and immobilized forms (Figures 2 and 3).

Our findings indicated that the increased adhesion of A2 cells was due to a greater reorganization of the actin cytoskeleton. This conclusion is based on: (a) greater F-actin content seen in the periphery of A2 cells adhered to fibrinogen for 5 minutes compared with A1 cells (Figure 7); (b) greater spreading of A2 cells compared with A1 cells after 15 minutes (Figure 7); and (c) the difference in adhesion between A1 and A2 cells was abolished in the presence of cytochalasin D (Figure 8). Rapid remodeling of cytoskeletal machinery in A2 cells compared with A1 could also explain the greater clot retraction seen in A2 cells. Certain lines of evidence suggested the greater reorganization of the actin cytoskeleton seen in the A2 cell line was because of differences in postreceptor occupancy signaling. A small increase in pp125FAK phosphorylation was consistently observed in A2 cells adhered to fibrinogen compared with A1 cells (Figure 9, a and b). This difference occurred only after ligand binding and required integrin αIIbβ3. Tyrosine phosphorylation of pp125FAK represents one of the early-recognized events during outside-in signaling, and our findings are consistent with evidence that tyrosine phosphorylation of pp125FAK correlates with cell spreading, focal adhesion plaque formation, and stress fiber assembly (35). In addition, A2 cells demonstrated greater reduction in adhesion after PKC inhibition with bisindolylmaleimide. Because PKC regulates tyrosine phosphorylation of pp125FAK and spreading in platelets (30), a theoretical mechanism to explain these data would be as follows: compared with A1 cells, adhesion of A2 cells to fibrinogen induces greater activation of PKC with subsequent activation of MAPK, which influences rapid cytoskeletal changes that favor stronger and sustained adhesion of A2 cells compared with A1 cells. Further studies are underway to address this hypothesis.

In summary, cells overexpressing the PlA2 form of the αIIbβ3 exhibited several functional differences compared with its PlA1 counterpart: greater adhesion to immobilized fibrinogen, greater spreading and actin reorganization, and greater clot retraction. It remains unclear how these pathways are linked to the Leu → Pro substitution at amino acid 33 of β3. Perhaps the conformational change in the β3 extracellular domain favors the interaction with an associated activation molecule, such as one containing an immunoreceptor tyrosine–based activation motif. Our findings support a prothrombotic feature of the PlA2 isoform of β3 that provides biologic plausibility to the clinical associations that have been described. Given the prevalence of PlA2 in the population, it should not be too surprising that the functional differences between the PlA1 and PlA2 forms of β3 are modest. More substantial changes may have produced a phenotype that would not have survived negative evolutionary pressure. Considering both the central role of αIIbβ3 in thrombus formation and its abundance on platelets, even a modest functional alteration could have profound effects over an individual’s lifetime.

Acknowledgments

Supported in part by grant H57488 from the National Institutes of Health.

References
  1. Lefkovits, J, Plow, EF, Topol, EJ. Platelet glycoprotein IIb/IIIa receptors in cardiovascular medicine. N Engl J Med 1995. 332:1553-1559.
    View this article via: PubMed CrossRef Google Scholar
  2. Weiss, EJ, et al. A polymorphism of a platelet glycoprotein receptor as an inherited risk factor for coronary thrombosis. N Engl J Med 1996. 334:1090-1094.
    View this article via: PubMed CrossRef Google Scholar
  3. Carter, AM, Ossei-Gerning, N, Wilson, IJ, Grant, PJ. Association of the platelet Pl (A) polymorphism of glycoprotein IIb/IIIa and the fibrinogen Bβ 448 polymorphism with myocardial infarction and extent of coronary disease. Circulation 1997. 96:1424-1431.
    View this article via: PubMed Google Scholar
  4. Walter, DH, Schachinger, V, Elsner, M, Dimmeler, S, Zeiher, AM. Platelet glycoprotein IIIa polymorphisms and risk of coronary stent thrombosis. Lancet 1997. 350:1217-1219.
    View this article via: PubMed CrossRef Google Scholar
  5. Ardissino, D, et al. Prothrombotic genetic risk factors in young survivors of myocardial infarction. Blood 1999. 94:46-51.
    View this article via: PubMed Google Scholar
  6. Ridker, PM, Hennekens, CH, Schmitz, C, Stampfer, MJ, Lindpaintner, K. PlA1/A2 polymorphism of platelet glycoprotein IIIa and risks of myocardial infarction, stroke, and venous thrombosis. Lancet 1997. 349:385-388.
    View this article via: PubMed CrossRef Google Scholar
  7. Herrmann, SM, et al. The Leu33/Pro polymorphism (PlA1/PlA2) of the glycoprotein IIIa (GPIIIa) receptor is not related to myocardial infarction in the ECTIM study. Thromb Haemost 1997. 77:1179-1181.
    View this article via: PubMed Google Scholar
  8. Bray, PF. Integrin polymorphisms as risk factors for thrombosis. Thromb Haemost 1999. 70:337-344.
  9. Van Leeuwen, EF, Leeksma, OC, Van Mourik, JA, Engelfriet, CP, Von dem Borne, AE. Effect of the binding of anti-Zwa antibodies on platelet function. Vox Sang 1984. 47:280-289.
    View this article via: PubMed Google Scholar
  10. Kunicki, T.J. 1989. Biochemistry of platelet-associated isoantigens and alloantigens. In Platelet immunology: molecular and clinical aspects. T.J. Kunicki and J.N. George, editors. Lippincott-Raven Publishers. Philadelphia, PA. 99–120.
    View this article via: PubMed Google Scholar
  11. Feng, D, et al. Increased platelet aggregability associated with platelet GPIIIa PlA2 polymorphism: the Framingham offspring study. Arterioscler Thromb Vasc Biol 1997. 19:1142-1147.
    View this article via: PubMed Google Scholar
  12. Michelson, A.D., et al. 1999. Platelet GPIIIa PlA polymorphisms display different sensitivity to agonists. Circulation. In press.
    View this article via: PubMed Google Scholar
  13. Kornecki, E, Niewiarowski, S, Morinelli, TA, Kloczewiak, M. Effects of chymotrypsin and adenosine diphosphate on the exposure of fibrinogen receptors on normal human and Glanzmann’s thrombasthenic platelets. J Biol Chem 1981. 256:5696-5701.
    View this article via: PubMed Google Scholar
  14. Marguerie, GA, Edgington, TS, Plow, EF. Interaction of fibrinogen with its platelet receptor as part of a multistep reaction in ADP-induced platelet aggregation. J Biol Chem 1980. 255:154-161.
    View this article via: PubMed Google Scholar
  15. Bennett, JS, Vilaire, G. Exposure of platelet fibrinogen receptors by ADP and epinephrine. J Clin Invest 1979. 64:1393-1401.
    View this article via: JCI PubMed CrossRef Google Scholar
  16. Kim, HO, et al. Gene frequency of the five major human platelet antigens in African, American, white and Korean populations. Transfusion 1995. 35:863-867.
    View this article via: PubMed CrossRef Google Scholar
  17. O’Toole, TE, et al. Integrin cytoplasmic domains mediate inside-out signal transduction. J Cell Biol 1994. 124:1047-1059.
    View this article via: PubMed CrossRef Google Scholar
  18. O’Toole, TE, et al. Affinity modulation of the αIIbβ3 integrin (platelet GPIIb-IIIa) is an intrinsic property of the receptor. Cell Regul 1990. 1:883-893.
    View this article via: PubMed Google Scholar
  19. Rosa, JP, et al. Cloning of glycoprotein IIIa from human erythroleukemia cells and localization of the gene to chromosome 17. Blood 1988. 72:593-600.
    View this article via: PubMed Google Scholar
  20. Sohn, RH, Chen, J, Koblan, KS, Bray, PF, Goldschmidt-Clermont, PJ. Localization of a binding site for phosphatidylinositol 4,5-bisphosphate on human profilin. J Biol Chem 1995. 270:21114-21120.
    View this article via: PubMed CrossRef Google Scholar
  21. Bray, PF, et al. Platelet glycoprotein IIb: chromosomal localization and tissue expression. J Clin Invest 1987. 80:1812-1817.
    View this article via: JCI PubMed CrossRef Google Scholar
  22. Gunning, P, Leavitt, J, Muscat, G, Ng, SY, Kedes, L. A human beta-actin expression vector system directs high-level accumulation of antisense transcripts. Proc Natl Acad Sci USA 1987. 84:4831-4835.
    View this article via: PubMed CrossRef Google Scholar
  23. Felgner, PL, et al. Lipofectin: a highly efficient, lipid-mediated DNA-transfection procedure. Proc Natl Acad Sci USA 1987. 84:7413-7417.
    View this article via: PubMed CrossRef Google Scholar
  24. Faraday, N, Goldschmidt-Clermont, P, Dise, K, Bray, PF. Quantitation of soluble fibrinogen binding to platelets by fluorescence-activated flow cytometry. J Lab Clin Med 1994. 123:728-740.
    View this article via: PubMed Google Scholar
  25. Weiss, EJ, et al. A monoclonal antibody specific for platelet GPIIIa (SZ21) distinguishes PlA1 from PlA2. Tissue Antigens 1995. 46:374-381.
    View this article via: PubMed Google Scholar
  26. Vijayan, VK, Zingde, SM. Heat-treated bovine serum albumin is a better blocking reagent for adhesion studies. Biotechniques 1995. 18:430-432.
    View this article via: PubMed Google Scholar
  27. Leng, L, Kashiwagi, H, Ren, X-D, Shattil, SJ. Rho A and the function of platelet integrin αIIbβ3. Blood 1998. 91:4206-4215.
    View this article via: PubMed Google Scholar
  28. Suehiro, K, Smith, JW, Plow, EF. The ligand recognition specificity of β3 integrins. J Biol Chem 1996. 271:10365-10371.
    View this article via: PubMed CrossRef Google Scholar
  29. Hall, A. Rho GTPasethe actin cytoskeleton. Science 1998. 279:509-514.
    View this article via: PubMed CrossRef Google Scholar
  30. Haimovich, B, Kaneshiki, N, Ji, P. Protein kinase C regulates tyrosine phosphorylation of pp125 FAK on platelets adhered to fibrinogen. Blood 1996. 87:152-161.
    View this article via: PubMed Google Scholar
  31. Roos, CM, et al. Adhesiveness of CHO cells expressing the PlA2 polymorphism of GPIIb-IIIa. Circulation 1998. 98:I-526. (Abstr.)
  32. Valentin, N, Newman, P. Human platelet alloantigens. Curr Opin Hematol 1994. 1:381-387.
    View this article via: PubMed Google Scholar
  33. Bennett, JS, Vilaire, G, Catella-Lawson, F, Rut, AR, Fitzgerald, G. The PlA2 alloantigen does not alter the affinity of GPIIb-IIIa for fibrinogen or RGD-containing peptides. Blood 1997. 90:154a. (Abstr.)
  34. Moskowitz, KA, Kudryk, B, Coller, BS. Fibrinogen coating density affects the conformation of immobilized fibrinogen: implications for platelet adhesion and spreading. Thromb Haemost 1998. 79:824-831.
    View this article via: PubMed Google Scholar
  35. Defilippi, P, et al. pp125 FAK tyrosine phosphorylation and the focal adhesion assembly: studies with phosphotyrosine phosphatase inhibitors. Exp Cell Res 1995. 221:141-152.
    View this article via: PubMed CrossRef Google Scholar
Version history
  • Version 1 (March 15, 2000): No description
Advertisement
Advertisement