Laminin stimulates spreading of platelets through integrin alpha6beta1-dependent activation of GPVI

Abstract

The extracellular matrix protein, laminin, supports platelet adhesion through binding to integrin alpha6beta1 In the present study, we demonstrate that human laminin, purified from placenta, also stimulates formation of filopodia and lamellipodia in human and mouse platelets through a pathway that is dependent on alpha6beta1 and the collagen receptor GPVI. The integrin alpha6beta1 is essential for adhesion to laminin, as demonstrated using an alpha6-blocking antibody, whereas GPVI is dispensable for this response, as shown using "knockout" mouse platelets. On the other hand, lamellipodia formation on laminin is completely inhibited in the absence of GPVI, although filopodia formation remains and is presumably mediated via alpha6beta1 Lamellipodia and filopodia formation are inhibited in Syk-deficient platelets, demonstrating a key role for the kinase in signaling downstream of GPVI and integrin alpha6beta1 GPVI was confirmed as a receptor for laminin using surface plasmon resonance spectroscopy and by demonstration of lamellipodia formation on laminin in the presence of collagenase. These results identify GPVI as a novel receptor for laminin and support a model in which integrin alpha6beta1 brings laminin to GPVI, which in turn mediates lamellipodia formation. We speculate that laminin contributes to platelet spreading in vivo through a direct interaction with GPVI.

Figure 1.

Adhesion and spreading of human and mouse platelets on laminin. (A) Washed murine platelets (3 × 10⁷/mL; left column) or washed human platelets (2 × 10⁷/mL; right column) were pretreated with 1 mM GRGDS peptide in the presence (bottom row) or absence (top row) of 20 μg/mL anti–integrin α₆β₁ antibody. Coverslips were coated with 50 μg/mL laminin from human placenta overnight at 4°C and then blocked with 1% fatty acid–free BSA for 2 hours at 4°C. After rinsing with Tyrode buffer, platelets were seeded on the coverslips and incubated at room temperature. After 30 minutes of incubation, unbound platelets were removed and adherent platelets were fixed by 3% paraformaldehyde, permeabilized with 0.3% Triton X-100, and F-actin stained using TRITC-labeled phalloidin. After F-actin staining, platelets were visualized using confocal fluorescent microscopy. (B) Murine platelets, washed as described in “Materials and methods” (3 × 10⁷/mL), were pretreated with or without 20 μg/mL anti–integrin α₂-blocking antibody (middle column) or anti–integrin α₆ antibody (right column) for 10 minutes and then seeded on the BSA-(top row), 50 μg/mL laminin-(middle row), or 50 μg/mL collagen-coated (bottom row) surfaces. Adherent platelets were fixed and visualized as described in “Materials and methods.” (C) Changes in morphology of adherent washed murine platelets (3 × 10⁷/mL; left column) or washed human platelets (2 × 10⁷/mL; right column) were visualized by time-lapse real-time imaging using videomicroscopy with DIC optics. The results are representative of 3 to 12 experiments. LM indicates laminin; col, collagen.

Figure 2.

Spreading but not adhesion of mouse platelets on laminin is dependent on Syk and PLCγ2. (A) Washed murine platelets (3 × 10⁷/mL) from control, Syk-deficient (Syk^-/-), or PLCγ2-deficient (PLCγ2^-/-) mice were pretreated with the α_IIbβ₃ blocker, lotrafiban (10 μM), before seeding on laminin-coated coverslips as described in Figure 1. Real-time imaging was performed using videomicroscopy with DIC optics. (B) (i) The number of adherent platelets was counted. The graph illustrates the mean number of percent wild-type adhesion ± SEM per image from at least 4 different images from 2 separate experiments. (ii) Total surface coverage of adherent platelets on laminin was measured using NIH image software. The graph illustrates the percent wild-type surface coverage ± SEM per image from at least 4 different images from 2 separate experiments. (iii) Number of filopodia-like projections in adherent platelets was counted. The graph illustrates the mean number of filopodia-like projection ± SEM per adherent platelet of at least 240 platelets from 2 separate experiments.

Figure 3.

Spreading but not adhesion of mouse platelets on laminin is dependent on FcR γ chain. (A) Washed murine platelets (3 × 10⁷/mL) from control or FcR γ chain–deficient mice (FcR γ chain^-/-) were pretreated with or without 1 mM GRGDS peptide. Coverslips were coated with 50 μg/mL laminin and blocked with fatty acid–free BSA as described in Figure 1. BSA-coated coverslips were used for negative control. GRGDS-treated platelets were seeded on BSA- or laminin-coated coverslips for 30 minutes at room temperature. After unbound platelets were removed, platelets were fixed, stained, and photographed as described in Figure 1. (B) (i) The number of adherent platelets was counted. The graph illustrates the mean number of adherent platelets ± SEM per 0.006 mm² from at least 8 different images from 3 separate experiments. (ii) Total surface coverage of adherent platelets on laminin-coated surfaces was measured. The graph illustrates the percent wild-type surface coverage ± SEM per 0.006 mm² from at least 8 different images from 3 separate experiments. ^**P < .01.

Figure 4.

Spreading but not adhesion of mouse platelets on laminin is dependent on GPVI. (A) Washed murine platelets (3 × 10⁷/mL) from control or GPVI-deficient mice (GPVI^-/-) were pretreated with or without 1 mM GRGDS peptide in the presence or absence of 20 μg/mL anti–integrin α₆ antibody. Laminin-coated coverslips were prepared as described in Figure 1. Platelets were seeded on laminin-coated coverslips for 30 minutes at room temperature. Adherent platelets were fixed, permeabilized, stained, and photographed using fluorescent microscopy as described in Figure 1. (B) The number of adherent platelets (i) and total surface coverage of adherent platelets on laminin (ii) were from at least 10 different images from 2 separate experiments.

Figure 5.

Laminin-induced protein phosphorylation is dependent on GPVI. Washed murine platelets (4 × 10⁸/mL) from control or GPVI-deficient mice (GPVI^-/-) were pretreated with 1 mM GRGDS peptide prior to adhesion to laminin as described in Figure 1. BSA-coated dishes were used for negative control; 300 μL of washed platelets were seeded on dishes for the indicated times at room temperature prior to stopping the reaction by addition of 300 μL of 2 × lysis buffer. (A) Proteins were separated by 6% to 20% gradient SDS-PAGE and protein-tyrosine phosphorylation visualized by Western blotting with 4G10. Protein loading was measured by Western blotting with anti-PLCγ2 pAb. (B) Fc receptor γ chain (i), SLP-76 (ii), LAT (iii), Btk (iv), Syk (v), or PLCγ2 (vi) were isolated by immunoprecipitation using specific antibodies before blotting with 4G10. Gels were reprobed with the antibody that was used in the immunoprecipitation studies. Tyrosine phosphorylation of Syk and PLCγ2 were quantified using Quantity One software for Macintosh. Optical density measurements were standardized by the recruitments of these proteins. The optical densities are shown above the corresponding lanes. The results are representative of 2 experiments.

Figure 6.

Adhesion to collagen but not laminin is inhibited by prior treatment with collagenase. (A) Coverslips were coated with 50 μg/mL collagen (i) or 50 μg/mL laminin (ii) as described in Figure 1. After rinsing with PBS, coverslips were incubated with 0.1 mg/mL collagenase in TES buffer containing 0.36 mM CaCl₂ for 2 hours at 35°C. After washing with PBS, coverslips were blocked with 2% fatty acid–free BSA for 2 hours. Washed murine platelets (3 × 10⁷/mL) were incubated with 2 mM GRGDS peptide and seeded on coverslips for 30 minutes at room temperature. Adherent platelets were fixed, permeabilized, stained, and photographed using fluorescent microscopy as described in Figure 1. (B) The number of adherent platelets was counted as described in Figure 3.

Figure 7.

Spreading of human platelets on laminin is mediated through GPVI. (A) Coverslips were coated with 50 μg/mL laminin or 50 μg/mL collagen and blocked with 2% BSA. Washed human platelets (2 × 10⁷/mL) were pretreated with PBS (i), 100 μg/mL GPVIex (ii), or 100 μg/mL GPVI-Fc₂ (iii) in the presence of 1 mM GRGDS. Platelets were seeded on BSA-, laminin- or collagen-coated coverslips for 30 minutes at room temperature. Unbound platelets were removed, and adhered platelets were fixed, stained, and visualized as described in Figure 1. (B) (i) The number of adherent platelets was counted. The graph illustrates the mean number of adherent platelets ± SEM per 0.006 mm² from at least 8 different images in 1 experiment, which is representative of 2 separate experiments. (ii) Total surface coverage of adherent platelets on laminin- or collagen-coated surfaces was measured. The graph illustrates the percent control surface coverage ± SEM per 0.006 mm² from at least 8 different images in 1 experiment, which is representative of 2 experiments. (C) Different concentrations of GPVI-Fc₂ were flowed over an immobilized laminin (i), collagen (ii), or a control surface. The arrows indicate the beginning and end of perfusion of GPVI-Fc_2. The results are shown from 1 experiment that is representative of 4 others. RU indicates resonance units.