Long-term antithrombotic protection by in vivo depletion of platelet glycoprotein VI in mice

Abstract

Coronary artery thrombosis is often initiated by abrupt disruption of the atherosclerotic plaque and activation of platelets on the subendothelial layers in the disrupted plaque. The extracellular matrix protein collagen is the most thrombogenic constituent of the subendothelial layer; therefore, a selective inhibition of the collagen activation pathway in platelets may provide strong antithrombotic protection while preserving other platelet functions. Here we demonstrate that treatment of mice with a monoclonal antibody against the activating platelet collagen receptor glycoprotein VI (GPVI; JAQ1) results in specific depletion of the receptor from circulating platelets and abolished responses of these cells to collagen and collagen-related peptides (CRPs). JAQ1-treated mice were completely protected for at least 2 wk against lethal thromboembolism induced by infusion of a mixture of collagen (0.8 mg/kg) and epinephrine (60 microg/ml). The tail bleeding times in JAQ1-treated mice were only moderately increased compared with control mice probably because the treatment did not affect platelet activation by other agonists such as adenosine diphosphate or phorbol myristate acetate. These results suggest that GPVI might become a target for long-term prophylaxis of ischemic cardiovascular diseases and provide the first evidence that it is possible to specifically deplete an activating glycoprotein receptor from circulating platelets in vivo.

Figure 1

JAQ1 induces transient thrombocytopenia. Mice received purified IgG (a) or Fab fragments (b) of the indicated mAb intraperitoneally in 200 μl sterile PBS. Platelet counts were determined at the indicated times using an improved Neubauer hemocytometer. Results are expressed as the mean platelet count ± SD for groups of each six mice.

Figure 1

JAQ1 induces transient thrombocytopenia. Mice received purified IgG (a) or Fab fragments (b) of the indicated mAb intraperitoneally in 200 μl sterile PBS. Platelet counts were determined at the indicated times using an improved Neubauer hemocytometer. Results are expressed as the mean platelet count ± SD for groups of each six mice.

Figure 2

Platelets from JAQ1-treated mice do not respond to CRP and collagen. (a) Two-color flow cytometric analysis of platelets from JAQ1-treated or control mice 3 d after Ab injection. Diluted whole blood was stimulated with 10 μM ADP or 10 μg/ml CRP for 2 min and subsequently incubated with antifibrinogen^FITC and anti–P-selectin^PE Abs for 10 min at RT and analyzed directly. Platelets were gated by FSC/SSC characteristics and Fl3 intensity (anti–mouse GPIb α^PE/Cy5). The data shown are representative of six mice per group. Similar results were obtained on days 7 and 14 after Ab injection. (b) Heparinized prp from the indicated mice was stimulated with collagen (50 μg/ml), ADP (10 μM), or PMA (50 ng/ml). Light transmission was recorded on a Fibrintimer 4 channel aggregometer. (c) Heparinized prp from control mice was incubated with stirring in the presence of irrelevant rat IgG2a (20 μg/ml; ○) or JAQ1 (20 μg/ml; ▿) for 5 min before the addition of the indicated concentrations of collagen. In parallel, prp from JAQ1-treated mice was tested (□). Results are expressed as the maximum (max.) platelet aggregation ± SD for groups of each six mice.

Figure 2

Platelets from JAQ1-treated mice do not respond to CRP and collagen. (a) Two-color flow cytometric analysis of platelets from JAQ1-treated or control mice 3 d after Ab injection. Diluted whole blood was stimulated with 10 μM ADP or 10 μg/ml CRP for 2 min and subsequently incubated with antifibrinogen^FITC and anti–P-selectin^PE Abs for 10 min at RT and analyzed directly. Platelets were gated by FSC/SSC characteristics and Fl3 intensity (anti–mouse GPIb α^PE/Cy5). The data shown are representative of six mice per group. Similar results were obtained on days 7 and 14 after Ab injection. (b) Heparinized prp from the indicated mice was stimulated with collagen (50 μg/ml), ADP (10 μM), or PMA (50 ng/ml). Light transmission was recorded on a Fibrintimer 4 channel aggregometer. (c) Heparinized prp from control mice was incubated with stirring in the presence of irrelevant rat IgG2a (20 μg/ml; ○) or JAQ1 (20 μg/ml; ▿) for 5 min before the addition of the indicated concentrations of collagen. In parallel, prp from JAQ1-treated mice was tested (□). Results are expressed as the maximum (max.) platelet aggregation ± SD for groups of each six mice.

Figure 2

Platelets from JAQ1-treated mice do not respond to CRP and collagen. (a) Two-color flow cytometric analysis of platelets from JAQ1-treated or control mice 3 d after Ab injection. Diluted whole blood was stimulated with 10 μM ADP or 10 μg/ml CRP for 2 min and subsequently incubated with antifibrinogen^FITC and anti–P-selectin^PE Abs for 10 min at RT and analyzed directly. Platelets were gated by FSC/SSC characteristics and Fl3 intensity (anti–mouse GPIb α^PE/Cy5). The data shown are representative of six mice per group. Similar results were obtained on days 7 and 14 after Ab injection. (b) Heparinized prp from the indicated mice was stimulated with collagen (50 μg/ml), ADP (10 μM), or PMA (50 ng/ml). Light transmission was recorded on a Fibrintimer 4 channel aggregometer. (c) Heparinized prp from control mice was incubated with stirring in the presence of irrelevant rat IgG2a (20 μg/ml; ○) or JAQ1 (20 μg/ml; ▿) for 5 min before the addition of the indicated concentrations of collagen. In parallel, prp from JAQ1-treated mice was tested (□). Results are expressed as the maximum (max.) platelet aggregation ± SD for groups of each six mice.

Figure 3

GPVI is not detectable in platelets from JAQ1-treated mice for at least 2 wk. (a) Whole platelet proteins were separated by SDS-PAGE under nonreducing conditions and immunoblotted with FITC-labeled JAQ1 (anti-GPVI) or EDL1 (anti-GPIIIa). Bound mAb was detected by HRP-labeled rabbit anti-FITC and ECL. (b) Washed platelets from control, FcR γ-chain–deficient (FcRγ^−/−) and JAQ1-treated (day 7) mice were stimulated with 10 μg/ml convulxin (Cvx). Control platelets were preincubated with irrelevant rat IgG2a or JAQ1 (20 μg/ml) for 5 min before the addition of Cvx. (c) Washed platelets from the indicated mice were incubated with FITC-labeled convulxin (5 μg/ml) for 15 min at RT and then analyzed on a FACScan™ (Becton Dickinson). The data shown are representative of six mice per group.

Figure 3

GPVI is not detectable in platelets from JAQ1-treated mice for at least 2 wk. (a) Whole platelet proteins were separated by SDS-PAGE under nonreducing conditions and immunoblotted with FITC-labeled JAQ1 (anti-GPVI) or EDL1 (anti-GPIIIa). Bound mAb was detected by HRP-labeled rabbit anti-FITC and ECL. (b) Washed platelets from control, FcR γ-chain–deficient (FcRγ^−/−) and JAQ1-treated (day 7) mice were stimulated with 10 μg/ml convulxin (Cvx). Control platelets were preincubated with irrelevant rat IgG2a or JAQ1 (20 μg/ml) for 5 min before the addition of Cvx. (c) Washed platelets from the indicated mice were incubated with FITC-labeled convulxin (5 μg/ml) for 15 min at RT and then analyzed on a FACScan™ (Becton Dickinson). The data shown are representative of six mice per group.

Figure 3

GPVI is not detectable in platelets from JAQ1-treated mice for at least 2 wk. (a) Whole platelet proteins were separated by SDS-PAGE under nonreducing conditions and immunoblotted with FITC-labeled JAQ1 (anti-GPVI) or EDL1 (anti-GPIIIa). Bound mAb was detected by HRP-labeled rabbit anti-FITC and ECL. (b) Washed platelets from control, FcR γ-chain–deficient (FcRγ^−/−) and JAQ1-treated (day 7) mice were stimulated with 10 μg/ml convulxin (Cvx). Control platelets were preincubated with irrelevant rat IgG2a or JAQ1 (20 μg/ml) for 5 min before the addition of Cvx. (c) Washed platelets from the indicated mice were incubated with FITC-labeled convulxin (5 μg/ml) for 15 min at RT and then analyzed on a FACScan™ (Becton Dickinson). The data shown are representative of six mice per group.

Figure 4

Absence of a ∼60-kD protein in platelets from JAQ1-treated mice. Whole cell lysates of platelets from control and JAQ1-treated (day 7) mice were subjected to two-dimensional gel electrophoresis and the gels were silver stained. First dimension: IEF (linear pH gradient 3–10). Second dimension: SDS-PAGE under reducing conditions. The molecular weight marker is shown on the right. The position of the ∼60-kD protein (pI: ∼5.6) that is absent in platelets from JAQ1-treated mice is indicated by arrows.

Figure 6

Reduced adhesion to collagen and abolished procoagulant response of GPVI-depleted platelets. (a) Platelets from JAQ1-treated mice (day 7) bind normal amounts of plasma vWF in the presence of botrocetin (2 μg/ml; solid line). Bound vWF was detected by FITC-labeled anti-vWF Abs (10 μg/ml). No binding was detected in the absence of botrocetin (shaded area). Normal activation of β1-integrins on platelets from JAQ1-treated mice in response to thrombin (0.1 U/ml). Resting (shaded area) or thrombin activated (solid line) platelets were incubated with FITC-labeled 9EG7 (5 μg/ml) for 15 min at RT and analyzed directly. (b) Washed platelets from control or JAQ1-treated mice (day 7) were incubated in collagen-coated microtiter plates in the presence or absence of MgCl₂ (1 mM)/CaCl₂ (1 mM) for the indicated times and adherent platelets were quantitated fluorimetrically. The data shown are from a single experiment, representative of five identical experiments and expressed as the mean of triplicate readings ± SD. (c) Flow cytometric analysis of annexin V-FITC binding to platelets from control and JAQ1-treated (day 7) mice activated with a combination of collagen (50 μg/ml) and thrombin (0.01 U/ml).

Figure 6

Reduced adhesion to collagen and abolished procoagulant response of GPVI-depleted platelets. (a) Platelets from JAQ1-treated mice (day 7) bind normal amounts of plasma vWF in the presence of botrocetin (2 μg/ml; solid line). Bound vWF was detected by FITC-labeled anti-vWF Abs (10 μg/ml). No binding was detected in the absence of botrocetin (shaded area). Normal activation of β1-integrins on platelets from JAQ1-treated mice in response to thrombin (0.1 U/ml). Resting (shaded area) or thrombin activated (solid line) platelets were incubated with FITC-labeled 9EG7 (5 μg/ml) for 15 min at RT and analyzed directly. (b) Washed platelets from control or JAQ1-treated mice (day 7) were incubated in collagen-coated microtiter plates in the presence or absence of MgCl₂ (1 mM)/CaCl₂ (1 mM) for the indicated times and adherent platelets were quantitated fluorimetrically. The data shown are from a single experiment, representative of five identical experiments and expressed as the mean of triplicate readings ± SD. (c) Flow cytometric analysis of annexin V-FITC binding to platelets from control and JAQ1-treated (day 7) mice activated with a combination of collagen (50 μg/ml) and thrombin (0.01 U/ml).

Figure 6

Reduced adhesion to collagen and abolished procoagulant response of GPVI-depleted platelets. (a) Platelets from JAQ1-treated mice (day 7) bind normal amounts of plasma vWF in the presence of botrocetin (2 μg/ml; solid line). Bound vWF was detected by FITC-labeled anti-vWF Abs (10 μg/ml). No binding was detected in the absence of botrocetin (shaded area). Normal activation of β1-integrins on platelets from JAQ1-treated mice in response to thrombin (0.1 U/ml). Resting (shaded area) or thrombin activated (solid line) platelets were incubated with FITC-labeled 9EG7 (5 μg/ml) for 15 min at RT and analyzed directly. (b) Washed platelets from control or JAQ1-treated mice (day 7) were incubated in collagen-coated microtiter plates in the presence or absence of MgCl₂ (1 mM)/CaCl₂ (1 mM) for the indicated times and adherent platelets were quantitated fluorimetrically. The data shown are from a single experiment, representative of five identical experiments and expressed as the mean of triplicate readings ± SD. (c) Flow cytometric analysis of annexin V-FITC binding to platelets from control and JAQ1-treated (day 7) mice activated with a combination of collagen (50 μg/ml) and thrombin (0.01 U/ml).

Figure 5

JAQ1 induces internalization of GPVI in vivo Fc independently. Mice were injected with 100 μg biotinylated JAQ1 or DOM1 (anti-GPV) and platelets were analyzed at the indicated time points. (a) Flow cytometric analysis of platelets from JAQ1-treated or control mice. Diluted control blood was incubated with biotinylated rat IgG2a (10 μg/ml; control [ctrl]) or JAQ1 (10 μg/ml; in vitro) for 6 h at 37°C. Subsequently, these samples and samples from JAQ1-treated mice (ex vivo) were incubated with FITC-labeled JAQ1 (5 μg/ml), streptavidin (Strep; 5 μg/ml), or convulxin (Cvx; 5 μg/ml) for 15 min at RT and analyzed directly. Platelets were gated by FSC/SSC characteristics and Fl2 intensity (anti–mouse GPIIb/IIIa^PE). (b) Detection of surface-bound biotinylated DOM1 (anti-GPV) ex vivo. The staining was performed as described for biotinylated JAQ1. Results in panels a and b are expressed as mean log Fl1 ± SD (n = 6). (c) Top: whole platelet proteins were separated by SDS-PAGE under reducing conditions and biotinylated JAQ1 was detected with HRP-labeled streptavidin/ECL. For detection of GPVI and GPIIIa, the proteins were separated under nonreducing conditions and immunoblotted with FITC-labeled JAQ1 or EDL1 followed by HRP-labeled rabbit anti-FITC/ECL. (d) Mice were injected with 100 μg Fab fragments of JAQ1 and platelets were analyzed in a Western blot for the presence of GPVI and GPIIIa after 48 h. These platelets did not aggregate in response to collagen (50 μg/ml), CRP (30 μg/ml), or Cvx (10 μg/ml), whereas ADP (10 μM) induced normal aggregation.

Figure 5

JAQ1 induces internalization of GPVI in vivo Fc independently. Mice were injected with 100 μg biotinylated JAQ1 or DOM1 (anti-GPV) and platelets were analyzed at the indicated time points. (a) Flow cytometric analysis of platelets from JAQ1-treated or control mice. Diluted control blood was incubated with biotinylated rat IgG2a (10 μg/ml; control [ctrl]) or JAQ1 (10 μg/ml; in vitro) for 6 h at 37°C. Subsequently, these samples and samples from JAQ1-treated mice (ex vivo) were incubated with FITC-labeled JAQ1 (5 μg/ml), streptavidin (Strep; 5 μg/ml), or convulxin (Cvx; 5 μg/ml) for 15 min at RT and analyzed directly. Platelets were gated by FSC/SSC characteristics and Fl2 intensity (anti–mouse GPIIb/IIIa^PE). (b) Detection of surface-bound biotinylated DOM1 (anti-GPV) ex vivo. The staining was performed as described for biotinylated JAQ1. Results in panels a and b are expressed as mean log Fl1 ± SD (n = 6). (c) Top: whole platelet proteins were separated by SDS-PAGE under reducing conditions and biotinylated JAQ1 was detected with HRP-labeled streptavidin/ECL. For detection of GPVI and GPIIIa, the proteins were separated under nonreducing conditions and immunoblotted with FITC-labeled JAQ1 or EDL1 followed by HRP-labeled rabbit anti-FITC/ECL. (d) Mice were injected with 100 μg Fab fragments of JAQ1 and platelets were analyzed in a Western blot for the presence of GPVI and GPIIIa after 48 h. These platelets did not aggregate in response to collagen (50 μg/ml), CRP (30 μg/ml), or Cvx (10 μg/ml), whereas ADP (10 μM) induced normal aggregation.

Figure 5

JAQ1 induces internalization of GPVI in vivo Fc independently. Mice were injected with 100 μg biotinylated JAQ1 or DOM1 (anti-GPV) and platelets were analyzed at the indicated time points. (a) Flow cytometric analysis of platelets from JAQ1-treated or control mice. Diluted control blood was incubated with biotinylated rat IgG2a (10 μg/ml; control [ctrl]) or JAQ1 (10 μg/ml; in vitro) for 6 h at 37°C. Subsequently, these samples and samples from JAQ1-treated mice (ex vivo) were incubated with FITC-labeled JAQ1 (5 μg/ml), streptavidin (Strep; 5 μg/ml), or convulxin (Cvx; 5 μg/ml) for 15 min at RT and analyzed directly. Platelets were gated by FSC/SSC characteristics and Fl2 intensity (anti–mouse GPIIb/IIIa^PE). (b) Detection of surface-bound biotinylated DOM1 (anti-GPV) ex vivo. The staining was performed as described for biotinylated JAQ1. Results in panels a and b are expressed as mean log Fl1 ± SD (n = 6). (c) Top: whole platelet proteins were separated by SDS-PAGE under reducing conditions and biotinylated JAQ1 was detected with HRP-labeled streptavidin/ECL. For detection of GPVI and GPIIIa, the proteins were separated under nonreducing conditions and immunoblotted with FITC-labeled JAQ1 or EDL1 followed by HRP-labeled rabbit anti-FITC/ECL. (d) Mice were injected with 100 μg Fab fragments of JAQ1 and platelets were analyzed in a Western blot for the presence of GPVI and GPIIIa after 48 h. These platelets did not aggregate in response to collagen (50 μg/ml), CRP (30 μg/ml), or Cvx (10 μg/ml), whereas ADP (10 μM) induced normal aggregation.

Figure 7

Bleeding time of JAQ1-treated mice. Bleeding times were determined in mice 7 d after injection of 100 μg nonimmune IgG2a or JAQ1 (n = 15 per group). As a control, mice received 100 μg F(ab′)₂ fragments of JON/A (anti-GPIIb/IIIa) 24 h before the experiment (n = 6). Where necessary, bleeding was manually stopped at the 10 min time point to prevent death. Each point represents one individual.

Figure 8

JAQ1 induces long-term protection from intravascular thrombosis. Thromboembolism in response to a bolus injection of a mixture of collagen (0.8 mg/kg body weight) and epinephrine (60 μg/kg body weight). (a) Mortality in control mice and mice treated with 100 μg JAQ1 at the indicated times before challenge. (b) Platelet counts in control and JAQ1-treated mice 3 min after infusion of collagen/epinephrine (n = 8 per group). (c) Top: representative histology of the lungs (original magnification: ×100); obstructed vessels are indicated by arrowheads. Bottom: immunohistochemical detection of platelets in the thrombi (original magnification: ×400). Acetone-fixed frozen sections were reacted with a platelet-specific Ab (anti–GPIb-IX) and counterstained with hematoxylin. The red HRP reaction product shows high density of platelets in the thrombus.

Figure 8

JAQ1 induces long-term protection from intravascular thrombosis. Thromboembolism in response to a bolus injection of a mixture of collagen (0.8 mg/kg body weight) and epinephrine (60 μg/kg body weight). (a) Mortality in control mice and mice treated with 100 μg JAQ1 at the indicated times before challenge. (b) Platelet counts in control and JAQ1-treated mice 3 min after infusion of collagen/epinephrine (n = 8 per group). (c) Top: representative histology of the lungs (original magnification: ×100); obstructed vessels are indicated by arrowheads. Bottom: immunohistochemical detection of platelets in the thrombi (original magnification: ×400). Acetone-fixed frozen sections were reacted with a platelet-specific Ab (anti–GPIb-IX) and counterstained with hematoxylin. The red HRP reaction product shows high density of platelets in the thrombus.

Figure 8

JAQ1 induces long-term protection from intravascular thrombosis. Thromboembolism in response to a bolus injection of a mixture of collagen (0.8 mg/kg body weight) and epinephrine (60 μg/kg body weight). (a) Mortality in control mice and mice treated with 100 μg JAQ1 at the indicated times before challenge. (b) Platelet counts in control and JAQ1-treated mice 3 min after infusion of collagen/epinephrine (n = 8 per group). (c) Top: representative histology of the lungs (original magnification: ×100); obstructed vessels are indicated by arrowheads. Bottom: immunohistochemical detection of platelets in the thrombi (original magnification: ×400). Acetone-fixed frozen sections were reacted with a platelet-specific Ab (anti–GPIb-IX) and counterstained with hematoxylin. The red HRP reaction product shows high density of platelets in the thrombus.