Targeting the anionic region of human protease-activated receptor 4 inhibits platelet aggregation and thrombosis without interfering with hemostasis

Abstract

Background: Human platelet activation and aggregation is a complex process. To date, many therapies have been developed targeting proteins that mediate this process to prevent unwanted activation. However, the current standard of care for acute coronary syndromes still has limitations, including bleeding risk.

Objective: To evaluate the protease-activated receptor 4 (PAR4) anionic cluster as a viable antiplatelet target by using a polyclonal antibody (CAN12).

Methods: We used western blotting, aggregation and secretion ex vivo to evaluate the ability of CAN12 to interact with PAR4 and inhibit platelet activation. The effects of CAN12 in vivo were evaluated with the Rose Bengal arterial thrombosis model and two models of hemostasis.

Results: CAN12 was able to interact with human PAR4 and delay PAR4 cleavage. In addition, CAN12 inhibited thrombin-induced human platelet aggregation and secretion in a dose-dependent manner. The specificity of CAN12 was agonist-dependent. In vivo, we determined that CAN12 was able to inhibit arterial thrombosis, and, using two independent methods, we found that CAN12 did not influence hemostasis.

Conclusion: Targeting the extracellular anionic cluster on PAR4 is a viable novel strategy as an antiplatelet therapy.

Keywords: G-protein-coupled receptors; antibodies; hemostasis; platelet aggregation inhibitors; protease-activated receptor-4 protein, human; thrombosis.

Figure 1. CAN12 reduces the rate of PAR4 cleavage

(A) The N-terminus of PAR4 with C⁵⁴ANDSDTLELPD (CAN12) and the thrombin cleavage site (*) indicated. (B) HEK293 cells (293) or cells transfected with mouse PAR4 (1.0 μg) or human PAR4 (1.0 μg) blotted with CAN12 (1:100) and actinin (1:1000). (C) Full gel for CAN12 interacting with hPAR4 expressed in HEK293 cells. (D) HEK293 cells (293) or cells transfected with hPAR4 (4.0 μg) or hPAR4-AAAA (D57A, D59A, E62A, D65A) (4.0 μg) blotted with CAN12, PAR4 (C-10), and actinin. (E) HEK293 cells transfected with HA-hPAR1 (4.0 μg) and V5-hPAR4 (0.1 μg) pretreated with buffer (PAR1PAR4), IgG (2 μg/ml), or CAN12 (2 μg/ml) for 10 min at room temperature. The cells were then activated with thrombin (100nM) for 0, 2, or 30 min at 37°C. Cleavage of PAR4 was measured by loss of the N-terminal V5 epitope. (F) Quantitation of percent of uncleaved PAR4 compared to 0 min as 100%. n=4 **p<0.01 vs. time zero

Figure 2. CAN12 inhibits thrombin induced aggregation and dense granule secretion

(A) Gel filtered platelets were isolated from healthy donors. Following treatment with goat IgG (2.0 μg/ml) or CAN12 (0.2 ng/ml – 2.0 μg/ml) for 10 min at 37°C, percent aggregation was determined for thrombin (0.5 nM). Representative curves shown. (B) Dose response curve for CAN12 inhibition of thrombin induced aggregation (0.5 nM). IC₅₀=10 ng/ml (2–49 ng/ml, 95% CI) (C) Dense granule secretion as determined by ATP (nM). *p<0.05 vs. IgG, NS=not significant

Figure 3. CAN12 specificity on aggregation

PRP was isolated from healthy donors. Following treatment with goat IgG (2.0 μg/ml) or CAN12 (dose response) for 10 min at 37°C, percent aggregation was determined for (A) AYPGKF (500 μM), (B) SFLLRN (25 μM), (C) ADP (5 μM), (D) collagen (1 μg/ml), (E) convulxin (5 nM). Representative curves and average percent aggregation shown. *p<0.05 vs. IgG, **p<0.01 vs. IgG, NS=not significant

Figure 4. CAN12 specificity on dense granule secretion

PRP was isolated from healthy donors. Following treatment with goat IgG (2.0 μg/ml) or CAN12 (dose response) for 10 min at 37°C, secretion was determined for (A) AYPGKF (500 μM), (B) SFLLRN (25 μM), (C) ADP (5 μM), (D) collagen (1 μg/ml), (E) convulxin (5 nM). *p<0.05 vs. IgG, **p<0.01 vs. IgG, NS=not significant

Figure 5. CAN12 inhibits aggregation through PAR4

Washed platelets were isolated from wild-type and PAR4^−/− mice. Following treatment with goat IgG (2.0 μg/ml) or CAN12(2.0 μg/ml) for 10 min at 37°C percent aggregation was evaluated for (A) wild-type and (B) PAR4^−/− platelets activated with AYPGFK (500 μM), ADP (5 μM) supplemented with 1 mg/ml fibrinogen, or collagen (1 μg/ml). **p<0.01 vs. IgG

Figure 6. CAN12 inhibits arterial thrombosis

(A) C57BL/6 mice were pretreated with saline, goat IgG (2 mg/kg), or CAN12 (1, 0.5, 0.25, 0.125 mg/kg) for 10 min and then subjected to the Rose Bengal carotid artery thrombosis model. Time to complete occlusion is indicated or the experiment was terminated at 90 min. (B) The concentration of platelets in the blood at termination of the experiment was determined. (C) 15 minutes after the initiation of carotid artery thrombosis, C57BL/6 mice were injected with goat IgG (2 mg/kg) or CAN12 (0.5 mg/kg) and the time to complete arterial occlusion was determined. The experiment was terminated at 90 min. (D) The concentration of platelets in the blood at termination of the experiment was determined. **p<0.01

Figure 7. CAN12 does not affect bleeding time

(A) C57BL/6 mice or PAR4^−/− mice were anesthetized and 3 mm of the tail was cut. The time to cessation of bleeding was determined or the experiment was terminated at 10 min. (B) The total amount of blood loss was determined by reading the absorbance of hemoglobin from lysed red blood cells and was compared to a standard curve. (C) C57BL/6 mice or PAR4^−/− mice were anesthetized and the saphenous vein was exposed and pierced. Once, bleeding ceased, the clot was disrupted. The procedure was repeated for 20 min. The average time of bleeding and (D) the number of clot formations were determined. **p<0.01, NS=not significant