GPVI and alpha2beta1 play independent critical roles during platelet adhesion and aggregate formation to collagen under flow

Abstract

The roles of the 2 major platelet-collagen receptors, glycoprotein VI (GPVI) and integrin alpha2beta1, have been intensely investigated using a variety of methods over the past decade. In the present study, we have used pharmacologic and genetic approaches to study human and mouse platelet adhesion to collagen under flow conditions. Our studies demonstrate that both GPVI and integrin alpha2beta1 play significant roles for platelet adhesion to collagen under flow and that the loss of both receptors completely ablates this response. Intracellular signaling mediated by the cytoplasmic adaptor Src homology 2 domain-containing leukocyte protein of 76 kDa (SLP-76) but not by the transmembrane adaptor linker for activation of T cells (LAT) is critical for platelet adhesion to collagen under flow. In addition, reduced GPVI receptor density results in severe defects in platelet adhesion to collagen under flow. Defective adhesion to collagen under flow is associated with prolonged tail-bleeding times in mice lacking one or both collagen receptors. These studies establish platelet-collagen responses under physiologic flow as the consequence of a close partnership between 2 structurally distinct receptors and suggest that both receptors play significant hemostatic roles in vivo.

Figure 1.

Pharmacologic inhibition of integrin α₂β₁ severely reduces deposition of mouse platelets on fibrillar collagen under flow. (A) Percent surface coverage area of platelets after whole mouse blood was treated with 30 μg/mL of the blocking anti-α2 antibody Ha1/29 (□), 5 mM EDTA (▪), or control hamster IgG (HamIgG; ). The results shown were determined by image analysis of the phase-contrast images and are the mean ± SEM (n = 5). Statistically significant inhibition was seen at each shear rate for Ha1/29 (P < .05) and EDTA (P < .05). (B) Representative phase-contrast images captured at 400, 700, 1000, and 1300 s^–1 after 4 minutes of perfusion. The lines seen in the upper left portions of some panels are visual artifact and do not arise due to abnormalities in the collagen-coated surface.

Figure 2.

Pharmacologic inhibition of integrin α₂β₁ and/or glycoprotein VI reduces deposition of human platelets on collagen under flow. (A) Whole blood was treated with 10 μg/mL of the blocking anti-α2 antibody 6F1 (▪), the blocking anti-GPVI antibody 11A12 (□), both 6F1 and 11A12 (), or 20 μg/mL mouse IgG (mIgG; ) for at least 30 minutes prior to being perfused over a collagen-coated surface (“undiluted”). To facilitate comparison with the mouse platelet studies shown in Figure 1, this experiment was also performed using human blood diluted with an equal volume of Tyrode buffer (“diluted”). The results shown are the mean ± SEM (n = 4-5). Statistically significant inhibition was seen with all treatments (P < .05) at all shear rates with undiluted blood, and at 1300 s^–1 with diluted blood. (B) Representative phase-contrast images after 4 minutes of perfusion of heparinized whole human blood over an immobilized collagen surface. The lines seen in the upper left portions of some panels are visual artifact and do not arise due to abnormalities in the collagen-coated surface.

Figure 3.

Platelets derived from mice genetically deficient in α₂β₁ integrin and/or GPVI-FcRγ demonstrate critical roles for both receptors during platelet deposition on fibrillar collagen under flow. Whole blood from the specified mice was perfused over an immobilized collagen surface for 4 minutes. (A) Surface coverage area of platelets from α₂β₁^–/– (), GPVI-FcRγ^–/– (□), GPVI-FcRγ^–/–; α₂β₁^–/– (▪), and wild-type () after flow. The results shown are the mean ± SEM (n = 6-11). Statistical significance (P < .01) was seen between the knockouts and the control at and between 400 and 1300 s^–1. (B) Representative phase-contrast images after perfusion. The lines seen in the upper left portions of some panels are visual artifact and do not arise due to abnormalities in the collagen-coated surface.

Figure 4.

Platelet deposition on collagen under flow requires signal transduction through the adaptor SLP-76. Whole blood from the indicated mice was perfused over collagen-coated glass slides at wall shear rates at and between 400 and 1300 s^–1 for 4 minutes. (A) Percent surface coverage of SLP-76–deficient (▪;n = 5) and the wild-type (□;n = 3) platelets after whole blood perfusion over a collagen-coated surface. The results shown are the mean ± SEM. Statistical significance was seen between the SLP-76–deficient platelets and the control (P < .01) at the specified shear rates. (B) Representative phase-contrast images after perfusion. The lines seen in the upper left portions of some panels are visual artifact and do not arise due to abnormalities in the collagen-coated surface.

Figure 5.

Platelet deposition on collagen under flow does not require signal transduction through the adaptor LAT. Whole blood from the indicated mice was perfused over collagen-coated glass slides at wall shear rates at and between 400 and 1300 s^–1 for 4 minutes. (A) Surface coverage of LAT-deficient (▪; n = 6) and wild-type (□; n = 3) platelets. The results shown are the mean ± SEM. There is a statistically significant difference between LAT-deficient and wild-type platelets at 1300 s^–1 (P < .05). (B) Representative phase-contrast images after perfusion. The lines seen in the upper left portions of some panels are visual artifact and do not arise due to abnormalities in the collagen-coated surface.

Figure 6.

Low levels of GPVI reduce deposition of platelets on collagen under shear stress. Whole blood from the indicated mice was perfused over collagen-coated glass slides at wall shear rates at and between 400 and 1300 s^–1 for 4 minutes. (A) Percent surface coverage of GPVI-FcRγ^–/– (▪), low-GPVI (□), and wild-type () platelets. The results shown are the mean ± SEM after analysis of the phase-contrast images (n = 5). Statistical significance was seen between GPVI-FcRγ^–/–, low-GPVI, and wild-type at and between 400 s^–1 and 1300 s^–1 (P < .01). (B) Representative phase-contrast images after flow. The lines seen in the upper left portions of some panels are visual artifact and do not arise due to abnormalities in the collagen-coated surface.

Figure 7.

Integrin α_IIbβ₃ is required for platelet aggregate formation but not for primary adhesion of platelets to collagen under flow ex vivo. (A) Percent surface coverage area of platelets from α₂β₁^–/–, GPVI-FcRγ^–/–, and wild-type mice after being treated with 30 μg/mL blocking anti-α_IIbβ₃ antibody Leo.H4 (□) or control rat IgG (RatIgG; ▪) at 1300 s^–1. The results shown were determined by image analysis of the phase-contrast images and are the mean ± SEM (n = 4). Statistical significance was seen between treated and untreated wild-type platelets (P < .01). (B) Representative phase-contrast images after flow.

Figure 8.

Mice lacking platelet-collagen receptors exhibit prolonged tail-bleeding times. Tail-bleeding times of wild-type (BalbC and C57Bl/6), α₂β₁^–/–, GPVI-FcRγ^–/–, and GPVI-FcRγ^–/–;α₂β₁^–/– were determined. Approximately 2 mm of tail was cut from unanesthetized mice, the tail immersed in 37°C saline, and the bleeding time recorded as the moment when bleeding ceased. Each point represents an individual animal (n = 10-17). Mouse strain is indicated in parentheses. WT indicates wild type. The horizontal bars indicate the average tail-bleeding times.