Key role of glycoprotein Ib/V/IX and von Willebrand factor in platelet activation-dependent fibrin formation at low shear flow

Judith M. E. M. Cosemans, Saskia E. M. Schols, Lucia Stefanini, Susanne de Witt, Marion A. H. Feijge, Karly Hamulyák, Hans Deckmyn, Wolfgang Bergmeier and Johan W. M. Heemskerk

Data supplements

Article Figures & Data


  • Figure 1

    Platelet-dependent fibrin formation under flow. Coverslips with or without adhered platelets were perfused with pooled plasma under recalcification at a low shear rate of 250 s−1. (A) Representative phase-contrast images taken at 580-640 seconds after start (ie, during the first minute of fibrin formation). Image recording with a Nikon Diaphot 200 microscope as described in “Methods.” Right panels show differential, subtracted images (bars, 10 μm). Note the parallel fibrin fibers in the direction of flow on perfusion with tissue factor (1pM). (B) High magnification images of star-like fibrin fibers on platelets. (C) Quantification of fibrin fibers from subtracted images. Increases in mean pixel intensity are shown. (D) Thrombin activity in plasma samples collected from the flow chamber outlet. Means ± SE (n = 3-4 experiments); ***P < .001 vs control.

  • Figure 2

    Morphologic activation score of fibrin-forming platelets. (A), Different morphologies of adhered platelets superfused with normal plasma at low shear rate of 250 s−1. Start of fibrin formation was at 580 seconds. Numbers in italic refer to platelets with different structures: 1, platelet with pseudopods and core; 2, platelet with lamellipods and core; 3, translucent platelet without core, forming ruffled edges; 4, bleb-forming platelet with rounded structure (bars, 10 μm). (B), Fluorescence images after postlabeling with OG488–annexin A5. (C), Time-dependent increase in OG488–annexin A5 fluorescence during plasma perfusion (arbitrary fluorescence units per cell, mean ± SE, n = 12). Arrow points to start of fibrin formation. (D), Representative [Ca2+]i rises of single adhered platelets during plasma perfusion, expressed as pseudo-ratio F/Fo values. Where indicated, PPACK (40μM) or 6B4 mAb (20 μg/mL) was present in plasma. Microscopic images and Ca2+ traces recorded with a Nikon Diaphot 200 microscope as described in “Methods.”

  • Figure 3

    Effects of PGE1 on platelet activation and fibrin formation. Platelets were superfused with plasma at 250 s−1 in the absence (control) or presence (0.5μM) of PGE1. Bright-field and contrast images (Nikon Diaphot 200 microscope, see “Methods”) were taken at 0-60 seconds after start of fibrin formation (bars, 10 μm). (A) Reduced fibrin on platelets after PGE1 treatment (average pixel intensities of subtracted images). (B) Lower [Ca2+]i rises with PGE1 in Fluo-4–loaded platelets. (C) Lower OG488–annexin A5 staining with PGE1. (D) Reduced OG488-fibrin(ogen) staining (integrated fluorescent intensities). Means ± SE; n ≥ 3; *P < .05, ***P < .001 vs control.

  • Figure 4

    Role of platelet GpIb-V-IX in fibrin formation under flow. Platelets were superfused with plasma at 250 s−1 in the presence (10 μg/mL) of isotype control mAb 2D4 or one of the following anti-GpIbα mAbs: 2D2 (block of thrombin binding site), 12G1 (block of shear-induced VWF binding) or 6B4 (block of shear- and ristocetin-induced VWF binding). (A) Representative microscopic images (Nikon Diaphot 200 microscope, see “Methods”) were taken at 0-60 seconds after start of fibrin formation: phase contrast and OG488-fibrin(ogen) (bars, 10 micron). (B) Lag time to start of fibrin formation. (C) Quantification of fibrin formation from subtracted images (60 seconds). (D) Staining with OG488-fibrin(ogen). (E) Fractions of platelets positive for OG488–annexin A5. Means ± SE; n = 3-4; ***P < .001 vs control mAb.

  • Figure 5

    Role of plasma VWF in platelet fibrin formation under flow. Platelets were superfused with pooled plasma from control subjects (control) or from 2 patients with VWD at shear rate of 250 s−1. Ristocetin (37.5 μg/mL) was added as indicated. Microscopic images were taken at start and after 60 seconds of fibrin formation. (A) Lag time to start of fibrin formation. (B) Quantification of OG488-fibrin(ogen) binding. Means ± SE; n = 3-4; *P < .05, ***P < .001 vs control plasma.

  • Figure 6

    Roles of murine GpIb-V-IX and VWF on platelet-dependent fibrin formation under flow. Adhered platelets from wild-type (WT) mice, transgenic mice lacking the extracellular domain of GpIbα (IL4Rα/GpIbα), or mice deficient in VWF were superfused with autologous plasma at a shear rate of 250 s−1. (A) Representative bright-field images (bars, 20 μm) after 0-60 seconds of fibrin formation, as well as images of OG488-fibrin(ogen). Images were recorded as described.31 (B) Lag time to start of fibrin formation. (C) Fractions of fibrin-forming platelets during 25 minutes perfusion. Means ± SE; n = 4-6 mice; ***P < .001 vs WT.

  • Figure 7

    Rapid fibrin formation on ionomycin-stimulated platelets. Adhered platelets were prestimulated with ionomycin (10μM) and CaCl2 (1mM), then perfused with control plasma (Ctrl) or plasma from a patient with VWD in the presence or absence of 6B4 mAb (10 μg/mL). (A) Representative phase contrast and OG–annexin A5 fluorescence images (Nikon Diaphot 200 microscope, see “Methods”) after ionomycin stimulation (bars, 10 μm). (B) Fractions of type 3 and 4 (procoagulant) platelets before plasma perfusion. (C) Platelet-dependent fibrin formation after 240 seconds. (D) Lag time to fibrin formation after plasma perfusion. Means ± SE; n = 3-4.

  • Figure 8

    Contribution of GpIb-V-IX and VWF to fibrin formation in whole blood. Platelets were superfused with recalcified whole blood at a shear rate of 250 s−1 for 5 minutes, after which the flow was stepwise reduced every 2.5 minutes to reach 125 s−1 at 10 minutes. Blood from healthy control subjects (Ctrl) or type 1 VWD patients was supplemented with labeled fibrinogen (150 μg/mL), AF647–annexin A5 (0.5 μg/mL) and/or 12G1 mAb (10 μg/mL). Bright-field and confocal fluorescence images were captured at 10 minutes (t = 0) with a Biorad/Zeiss laser-scanning microscope system (see “Methods”). (A) Representative images taken after 60 seconds, showing clusters of PS-exposing platelets and a growing fibrin network. Arrow indicates thrombus (bars, 10 μm). (B), Time plots showing fluorescence accumulation of AF546-fibrin(ogen) (blue) and AF647–annexin A5 (red). (C), High-resolution confocal images of OG488-fibrin(ogen) and AF647–annexin A5 fluorescence. Data are representative of 3 or more experiments.


  • Table 1

    Effect of various treatments on platelet activation and fibrin formation under flow

    TreatmentFraction of adhered platelets
    Types 3 and 4PS-exposingFibrin-forming
    Initial (0 min)
        Control0.12 ± 0.070.06 ± 0.030
    After perfusion (12 min)
        Control0.79 ± 0.040.89 ± 0.020.96 ± 0.04
        PPACK0.14 ± 0.04*0.10 ± 0.05*0
        Cathepsin G0.10 ± 0.03*0.16 ± 0.09*0.02 ± 0.01*
        Annexin A50.08 ± 0.01*N.d.0.01 ± 0.01*
    • Adhered platelets were superfused with recalcified plasma at 250 seconds−1, as described in Figure 1. Plasma was incubated with PPACK (40μM) or unlabeled annexin A5 (20 μg/mL), as indicated. Alternatively, platelets were pretreated with cathepsin G (400nM). After 0-12 minutes of perfusion, adhered platelets were classified as types 1-4 (pseudopods, lamellipods plus core, translucent without core, and bleb-forming, respectively) from phase-contrast images. Adhered platelets were counted as PS-exposing (staining with OG488-annexin A5) and forming fibrin fibers. Values reported are means ± SE (n = 3-6 experiments).

    • N.d. indicates not determined.

    • * P < .01 compared with control.