Platelet protein disulfide isomerase is required for thrombus formation but not for hemostasis in mice

Abstract

Protein disulfide isomerase (PDI) derived from intravascular cells is required for thrombus formation. However, it remains unclear whether platelet PDI contributes to the process. Using platelet-specific PDI-deficient mice, we demonstrate that PDI-null platelets have defects in aggregation and adenosine triphosphate secretion induced by thrombin, collagen, and adenosine diphosphate. Such defects were rescued by wild-type but not mutant PDI, indicating that the isomerase activity of platelet surface PDI is critical for the regulatory effect. PDI-deficient platelets expressed increased levels of intracellular ER protein 57 (ERp57) and ERp72. Platelet PDI regulated αIIbβ3 integrin activation but not P-selectin exposure, Ca(2+) mobilization, β3-talin1 interaction, or platelet spreading on immobilized fibrinogen. Inhibition of ERp57 further diminished αIIbβ3 integrin activation and aggregation of activated PDI-deficient platelets, suggesting distinct roles of PDI and ERp57 in platelet functions. We found that platelet PDI is important for thrombus formation on collagen-coated surfaces under shear. Intravital microscopy demonstrates that platelet PDI is important for platelet accumulation but not initial adhesion and fibrin generation following laser-induced arteriolar injury. Tail bleeding time in platelet-specific PDI-deficient mice were not significantly increased. Our results provide important evidence that platelet PDI is essential for thrombus formation but not for hemostasis in mice.

Figure 1

Characterization of platelet-specific PDI–deficient mice. (A) Polymerase chain reaction (PCR) analysis of WT and PDI CKO mice with primers for floxed PDI (422 bp) and PF4-Cre (450 bp). (B) Lysates of neutrophils, endothelial cells, and platelets isolated from WT and PDI CKO mice were immunoblotted with indicated antibodies. Band density of PDI, ERp57, ERp72, and GRP78 in WT platelets (gray bars) is shown as 100%. White bars, PDI-null platelets. Data represent mean ± SD (n = 3-6 mice per group). **P < .01; ***P < .001 vs WT platelets after Student t test. (C) Flow cytometric analysis shows the surface expression of ERp57 and ERp72 on resting (dotted line) and thrombin-activated (black line) WT and PDI-null platelets. The gray histogram represents the fluorescence intensity of control IgG on thrombin-activated platelets. The geometric mean fluorescence intensity of antibodies was normalized to that of control IgG, and data are shown as a fold increase (mean ± SD, n = 3).

Figure 2

Aggregation and ATP secretion of WT and PDI-null platelets and the rescue effect of recombinant PDI. Platelets isolated from WT (black line) and PDI CKO (gray line) mice were stimulated with 0.05 U/mL thrombin (A), 0.5 μg/mL collagen (B), and 2.5 μM ADP (C). (i) Platelet aggregation and quantitative graphs. (ii) ATP secretion. In some experiments, PDI-null platelets were pretreated with 50 μg/mL wtPDI (black dotted line) or dmPDI (gray dotted line) and activated with an agonist. Quantitative results of aggregation and ATP secretion are presented as mean ± SD (n = 4). **P < .01 vs WT platelets after ANOVA and Dunnett test; ##P < .01 vs PDI-null platelets treated with wtPDI after Student t test. (D,E) Mouse platelets were pretreated with His-tagged wtPDI or dmPDI and activated with or without thrombin in the presence of 1 mM EGTA under a stirring condition. Platelets were washed with HEPES-Tyrode buffer (D) or carbonate buffer (0.1 M Na₂CO₃, pH 9.0) (E). Binding of recombinant PDI was analyzed by flow cytometry using a Dylight 488–conjugated anti-polyHis antibody. PDI binding is shown as a fold increase by the ratio of the geometric mean intensity value of the anti-His antibody on PDI-treated vs untreated platelets (mean ± SD, n = 3). *P < .05; **P < .01 vs resting platelets; and #P < .05 vs activated WT platelets treated with wtPDI or dmPDI after Student t test.

Figure 3

Inhibitory effect of surface ERp57 on α_IIbβ₃ integrin activation, P-selectin exposure, aggregation, and ATP secretion of PDI-null platelets. Mouse platelets were pretreated with nonimmune sheep immunoglobulin G (IgG) or blocking anti-ERp57 antibodies (50 μg/mL) and stimulated with 0.05 U/mL thrombin (thr). (A,B) α_IIbβ₃ integrin activation and P-selectin exposure were analyzed by flow cytometry as described in “Materials and methods.” Binding of anti-activated α_IIbβ₃ (JON/A) and anti–P-selectin antibodies to platelets was calculated by the ratio of the geometric mean intensity value of antibodies to that of control IgG. Then, antibody binding to resting WT platelets treated with nonimmune sheep IgG was normalized as 1 (white bar in the left panel). Data represent mean ± SD (n = 3). **P < .01 vs control IgG; ##P < .01 vs WT platelets after Student t test. (C,D) Platelet aggregation and ATP secretion were measured as described in Figure 2 (mean ± SD, n = 3). **P < .01 vs control IgG; ##P < .01 vs WT platelets after Student t test.

Figure 4

Platelet PDI does not regulate the β₃–talin1 interaction, protein phosphorylation, or platelet spreading on immobilized FG. (A) WT and PDI-null platelets (3 × 10⁷ platelets in 0.3 mL) were treated with or without 0.05 U/mL thrombin (Thr) for 45 seconds in an aggregometer and lysed with ice-cold lysis buffer. Lysates were immunoprecipitated and immunoblotted as described in “Materials and methods.” Data represent mean ± SD (n = 4). (B) Protein phosphorylation levels were determined by immunoblotting with lysates of thrombin-activated WT and PDI-null (CKO) platelets using mouse IgG2b (mIgG2b) or an anti-phosphotyrosine antibody (4G10). The representative blot was obtained from 3 independent experiments. (C-E) Mouse platelets (8 × 10⁶ platelets in 0.4 mL) were incubated on FG-coated surfaces for 2 h at 37°C in the presence of 0.025-0.05 U/mL thrombin. Adherent and spread platelets were stained with rhodamine-conjugated phalloidin. (C) Representative images. Bar = 10 μm. (D) Number of adherent (but not spread, black bars) and fully spread (white and gray bars) platelets. **P < .01 vs WT platelets (total number of adherent and spread platelets) after Student t test. (E) Platelet spreading was analyzed by the surface area, which was measured by the number of pixels divided by the number of platelets in the field. Data represent mean ± SD (n = 3).

Figure 5

Platelet PDI is important for thrombus formation on collagen-coated surfaces under arteriolar shear. Blood drawn from WT and PDI CKO mice was perfused over collagen-coated surfaces at a wall shear rate of 1000 s⁻¹ for 1 minute. In some experiments, wtPDI or dmPDI was added to blood drawn from PDI CKO mice and then blood was perfused through the chamber. Adherent thrombi were stained with rhodamine-conjugated phalloidin and analyzed as described in “Materials and methods.” (A) Representative images. Bar = 10 μm. (B,C) Surface coverage and thrombus volume were measured and presented as mean ± SD (n = 4). ***P < .001 vs WT platelets after ANOVA and Dunnett test; ##P < .01; and ###P < .001 vs PDI-null platelets treated with wtPDI after Student t test.

Figure 6

Platelet PDI plays a critical role in platelet accumulation but not initial adhesion and fibrin generation at the site of laser-induced arteriolar wall injury. Intravital microscopy was performed as described in “Materials and methods.” (A) Representative fluorescence images associated with platelets (red) and fibrin (green) are shown over the course of 180 seconds after vascular injury. White arrows show the direction of blood flow. (B-E) The median integrated fluorescence signals of anti-CD42c (F platelet) and anti-fibrin antibodies (F fibrin) were obtained from 28-30 thrombi in 3 WT or 6 CKO mice and is presented as a function of time. Little fluorescence signal was observed by fluorescently labeled control IgG (data not shown). (C,E) The fluorescence signal is shown at 60, 120, and 180 seconds after vascular injury. *P < .05; **P < .01 vs WT mice after Kruskal-Wallis test. (F) After infusion of wtPDI or dmPDI (100 μg), Dylight 649–conjugated rat anti-mouse CD42c and Dylight 488–conjugated anti-polyHis antibodies or mouse IgG1 were infused into PDI CKO mice. The representative images were obtained from 18-20 thrombi in 3 PDI CKO mice per group and are shown as T_max for platelet thrombus formation. (G) The median integrated fluorescence signal of anti-polyHis antibodies (PDI binding) is presented as a function of time.

Figure 7

Tail bleeding time and blood loss in platelet-specific PDI–deficient mice. Tails of WT (circle) and PDI CKO (triangle) mice were amputated, and bleeding time (A) was monitored as described in “Materials and methods.” (B) Blood loss during the bleeding time assay was determined by measuring the absorbance at 575 nm of hemoglobin (Hb). Horizontal bars represent the median of bleeding times and Hb content for each group of animals (n = 6).