Analysis of the Interaction of Platelet Collagen Receptor Glycoprotein VI (GPVI) with Collagen

• DOI: 10.1074/jbc.m204029200
• Publication Date: 2002-11-23T05:06:07Z
• Authors: Yoshiki Miura; Tsuyoshi Takahashi; Stephanie M. Jung; Masaaki Moroi
• Keywords: GPVI; Fibrillogenesis

Glycoprotein VI (GPVI) is a platelet-specific glycoprotein that has been indicated to react with collagen and activate platelets. Its structure was recently identified by cDNA cloning (Clemetson, J. M., Polgar, J., Magnenat, E., Wells, T. N., Clemetson, K. (1999) Biol. Chem. 274, 29019–29024). However, the mechanism of interaction between GPVI not analyzed in detail because both are insoluble molecules. In this study, we expressed extracellular domain as soluble forms follows: monomeric form (GPVIex) dimeric fused human immunoglobulin Fc (GPVI-Fc2). Purified GPVIex strongly inhibited convulxin (Cvx)-induced platelet aggregation but only weakly induced collagen-related peptide. GPVI-Fc2, GPVIex, collagen-induced aggregation. The exhibits high affinity for collagen, concluded from measurements binding immobilized enzyme-linked immunosorbent assay surface plasmon resonance methods. GPVI-Fc2 bound dissociation constant (K D) 5.76 × 10−7m, too weak allow estimation parameter. Cvx did inhibit indicating site different Cvx. Taken together, our data indicate composed two chains GPVI. Furthermore, they suggest sites collagen-binding do need be formed Because shows fibrous results would present on platelet. Moreover, there no detectable GPVI, supporting previous observation reacts collagen. Collagen one major components vessel wall. When wall becomes damaged, platelets adhere activated exposed surface, leading thrombus formation. Many proteins were reported putative receptors, among them, glycoproteins have properties consistent them being relevant receptors under normal physiological conditions: (GP) 1The abbreviations used for: GP, glycoprotein; PBS, phosphate-buffered saline; CRP, peptide; ELISA, assay; SPR, resonance; Cvx, convulxin; HEK, embryonic kidney; PRP, platelet-rich plasma. other integrin α2β1 (GPIa/IIa). Platelets deficient either (1Handa M. Watanabe Kawai Y. Kamata Koyama Nagai H. Ikeda Thromb. Haemostasis. 1995; 73: 521-528Crossref PubMed Scopus (89) Google Scholar, 2Nieuwenhuis H.K. Akkerman J.W. Houdijk W.P. Sixma J.J. Nature. 1985; 318: 470-472Crossref (399) Scholar) or (3Sugiyama Okuma Ushikubi F. Sensaki S. Kanaji Uchino Blood. 1987; 69: 1712-1720Crossref 4Moroi Jung S.M. Shinmyozu Clin. Invest. 1989; 84: 1440-1445Crossref (371) show loss reactivity toward antibodies against α2β1, such 6F1 (5Coller B.S. Beer J.H. Scudder L.E. Steinberg M.H. 74: 182-192Crossref P1E6 (6Keely P.J. Parise L.V. 1996; 271: 26668-26676Abstract Full Text PDF (145) Scholar), Fab fragment an anti-GPVI antibody (7Sugiyama Ishibashi Int. Hematol. 1993; 58: 99-104PubMed Snake venom (Cvx) (8Francischetti I.M. Saliou B. Leduc Carlini C.R. Hatmi Randon Faili A. Bon C. Toxicon. 1997; 35: 1217-1228Crossref (96) peptide (CRP), which mimic triple helix (9Morton L.F. Hargreaves P.G. Farndale R.W. Young R.D. Barnes M.J. Biochem. 306: 337-344Crossref (285) can each specifically so useful tools analyzing function A number studies showed activation through tyrosine phosphorylation receptor γ-chain, Syk, phospholipase C, many (10Watson Berlanga O. Best D. Frampton Platelets. 2000; 11: 252-258Crossref (70) thus key activation. revealed belongs superfamily contains C2 immunoglobulin-like domains Arg residue transmembrane region makes salt bridge Asp γ-chain (11Clemetson J.M. Polgar Magnenat E. Wells T.N. Clemetson K.J. 1999; 274: 29019-29024Abstract (366) 12Miura Ohnuma Moroi Res. 98: 301-309Abstract (46) 13Jandrot-Perrus Busfield Lagrue A.H. Xiong X. Debili N. Chickering Le Couedic J.P. Goodearl Dussault Fraser Vainchenker W. Villeval J.L. 96: 1798-1807Crossref Scholar). Although its identified, reaction remains unclear. Our obtained suggested reactive (4Moroi 18Moroi Tomiyama Ordinas Diaz-Ricart 88: 2081-2092Crossref facilitate analysis these complex reactions, simplified model system assessed ability We protein dimer consisting molecules fusion GPVI-Fc (from immunoglobulin) form, These specific conformation constructed connected γ-chain. containing Ig without signal sequence (642 bp, residues 21–234 (12Miura Scholar)) PCR using template oligonucleotide TTAAGCTTCAGAGTGGACCGCTCCCCAAGC (with theHindIII underlined) AATCTAGAGGAATGAGACGGTCAGTTCAGCG XbaI forward reverse primer, respectively.Pfu turbo DNA polymerase (Stratagene, La Jolla, CA) used. product purified QIAquick gel Extraction Kit (Qiagen K.K., Tokyo, Japan), digested HindIII andXbaI, again, ligated pSecTag vector (Invitrogen). c-Myc His tag sequences at ligation site, epitopes COOH terminus. mixture transformed into Escherichia coli XL-10 (Stratagene). construct verified restriction enzyme digestion sequencing. linker κ deleted QuickChange Site-directed Mutagenesis (Stratagene) according manufacturer's instructions. plasmid, pGPVI mono, checked dideoxy sequencing entire coding portion. To prepare amplified pBOS-Fc (14Suda Takahashi Golstein P. Nagata Cell. 75: 1169-1178Abstract (2450) AATCTAGAGAGCCCAAATCTTGTGA AGGGCCCCGGCCGTCGCACTCAT primers inserted XbaI-ApaI pGPVI-mono. This recombinant designated pGPVI-Fc. Human kidney (HEK) 293 cells transfected mono pGPVI-Fc calcium phosphate precipitation method (15Chen Okayama Mol. 7: 2745-2752Crossref (4824) Stable cell lines expressing (GPVI-Fc2) selected medium 100 μg/ml Zeocin. For purification protein, culture centrifuged, supernatant applied column A-Sepharose (AmershamBiosciences). After extensive washing saline (PBS), eluted ImmunoPure Elution Buffer (Pierce). fractions concentrated ULTRAFREE®-15 (Millipore, Marlborough, MA) centrifugal filters Superdex 200 (Amersham Biosciences), equilibrated PBS (10 mm sodium phosphate, 154 NaCl, pH 7.4). HiTrap Ni2+-chelating Biosciences) column. NH2-terminal confirmed DQSGPLPKP amino acid sequencer (model 377A, Applied Biosystems). determine inhibitory effect aggregation, samples dialyzed HEPES-Tyrode buffer (136 2.7 KCl, 0.42 NaH2PO4, 12 mmNaHCO3, 5.5 glucose, 5 mmHEPES, Microtiter wells (ImmunoMax; Nalge Nunc. International Japan) reacted 50 μl bovine acid-soluble type I (20 0.9% NaCl) overnight 4 °C. CRP React-Bind maleic anhydride-activated polystyrene plate (Pierce) similar conditions, reacting well 20 washed once incubated blocking (PBS 0.2% serum albumin 0.05% Tween 20) 1 h room temperature. solutions added 2 h. extensively 20, 5000-fold diluted horseradish peroxidase-conjugated anti-human (American Qualex Antibodies, San Clemente, temperature times then color developed o-phenylenediamine ELISA kit (Nakalai Tesque, Kyoto, 20–30 min Experiments performed triplicate also repeated preparations. Binding albumin-coated measured background value, values subtracted level. dimerization (final concentration μm) preincubated various concentrations anti-Myc 9E10 30 cross-linked assessing amount biotinylated collagen-bound GPVIex. incubating reactants (5 μg/ml) h, detected VECTASTAIN® Elite (Vector Laboratories, Burlingame, CA). above. assay, estimated monoclonal above-described following modifications: amounts ng/ml) rabbit anti-mouse IgG (H+L)-horseradish peroxidase conjugate (Bio-Rad) developing described effects method. Data percentages control value presented mean ± S.E. three independent experiments. Analyses carried out 25 °C BIAcore 2000 (BIAcore AB, Uppsala, Sweden) HES HEPES, 150 EDTA, Bovine 10 acetate buffer, 5.0, covalently coupled CM5 chip (Biacore) Amine Coupling 5.5, sensor same procedure. Regeneration achieved running 15 HCl flow μl/min times. amine coupling reagent absence ligand blocked ethanolamine. several perfused over rate μl/min, changes recorded. sensorgram immobilized-collagen nonlinear curve fitting Langmuir isotherm BIAevaluation software (Biacore). injecting III (50 GPVI-Fc2at μl/min. preparation previously (16Jung 1998; 273: 14827-14837Abstract (108) Whole blood drawn cubital vein healthy volunteers 0.1 volume 3.8% citrate. plasma (PRP) obtained, prostaglandin I2 (Funakoshi, final μg/ml, centrifuged 900 ×g isolate 6.85 citrate, 130 mmglucose, 6.5. Then finally suspended 2–4 108platelets/ml. Platelet monitored whole aggregometer (Chrono-Log Corp., Haverton, PA) stirring 37 PRP buffer. (0.2 μg/ml), equine (1–2 Chrono-Log), (23.8 ng/ml); change transmission synthesized Mortonet al. (18Moroi Cross-linked activating activity 0.2 less. Crotalus durissus terrificusvenom (Miami Serpentarium Punta Gorda, FL) sulfo-NHS-biotin before collagens Koken Co., Ltd. (Tokyo, Japan). purchased Santa Cruz Biotechnology (Santa Cruz, Immunotech (Marseilles, France). (residues 21–234) (Myc-His6) protein. GPVIs secreted eukaryotic line, HEK cell, prevented misfolding any non-glycosylation proteins, occasionally happened when coli. About mg/liter mediums molecular mass 41 42 kDa non-reduced reduced conditions SDS-PAGE, respectively (Fig.1, lanes 3 4). 68 (Fig. 1,lanes 2). ∼180 filtration (data shown), confirming dimer. Both 204-11 2M. Moroi, Mizuguchi, Kawashima, Nagamatsu, Miura, Nakagaki, Ito, Jung, submitted publication. recognize native condition, suggesting conformations shown). Next, tested activities As shown Fig.2, even (100 it CRP-induced although very (Fig.3). contrast, Cvx-induced markedly 0.5 0.56 μm, shown).Figure 3Inhibitory aggregation.Inhibitory (A) GPVI-Fc2(B) legend Fig. 2, except (CRP-XL) agonist.View Large Image Figure ViewerDownload (PPT) abrogated preincubation B). affinities itself collagen-mimetic CRP. ligands (Fig.4). dose-dependent saturable immobilized-CRP A). EDTA affect impossible compare quantitatively extent bindings detect could strong Even sensitivity increased longer color-developing time, small increase development observed 400 contrast dose-dependent, becoming saturated C). preferably binds GPVI-Fc2. above may necessary induce test hypothesis, 9E10, enabled us make chain bind molecule Myc COOH-terminal. illustrates exhibited enhanced enhancement depending molar ratio At 1:2 (9E10:GPVIex), maximum, 3.4-fold higher than alone. compared measuring method, corresponded about 160 nmGPVI-Fc2, 10–20% curves excess rather decreases explained decrease dimerized where chance molecule, forming support hypothesis responsible negate possibility due having determined their compete manner (Fig.6 required 50% inhibition (IC50) calculated 4.3 0.4 μg/ml. instead apparent low possible explanation enhancing proposed basis multiple subunit (17Polgar Kehrel B.E. Wiedemann E.M. 272: 13576-13583Abstract (316) ones. result, able thereby help accumulate more surface. providing further evidence Cvx-binding site. almost completely shared Kinetic SPR chips acidic interactions flowing conditions. sensorgrams normalized subtracting signals 7 A) B) responses. kinetic experiments summarized TableI. table, constants calculation methods, fromk k off equilibrium equation units equilibrium. Curve fit better two-state one-state one, conformational after initial association (TableII). model, K D substantially those transition small, contribution (conformational change) binding. TheK least 10-fold reliable parameters experimental 8 measurable, 8.5 10−5m B).Table IKinetic modelk aaL+R⇄kdkaL·R;L,ligand;R,receptor.k dK DK equilibriumm−1s−1s−1mmCRP7.87 3.21 1033.22 1.57 10−25.27 3.40 10−65.26 10−6Collagen8.62 0.66 1035.04 0.67 10−35.78 0.38 10−75.76 0.37 10−7a L+R⇄kdkaL·R;L,ligand;R,receptor. Open table new tab Table IIKinetic a1aL+R⇄kd1ka1L·R⇄kd2ka2(LR);L,ligand;R,receptor.k d1k a2k d2K DbKD=kd1·kd2ka1·ka2m−1s−1s −1s −1mCRP1.65 0.54 1046.38 1.09 10−21.04 0.25 10−31.91 0.31 10−32.71 1.90 10−5Collagen1.69 0.40 1043.12 1.59 10−22.19 10−31.36 0.20 10−38.91 0.33 L+R⇄kd1ka1L·R⇄kd2ka2(LR);L,ligand;R,receptor.b KD=kd1·kd2ka1·ka2 8Sensorgrams solution (31.25 nm 35 (B). processed legend.RU, units.View experiments, analyzed. contain some experiment first analyze prepared "Experimental Procedures," Fig.9, indicates significant obtained. result clearly does solely attributable c