Mice with reduced protease-activated receptor 4 reactivity show decreased venous thrombosis and platelet procoagulant activity

Abstract

Background: Hypercoagulation and thrombin generation are major risk factors for venous thrombosis. Sustained thrombin signaling through protease-activated receptor (PAR) 4 promotes platelet activation, phosphatidylserine exposure, and subsequent thrombin generation. A single nucleotide polymorphism in PAR4 (rs2227376) changes proline to leucine extracellular loop 3, which decreases PAR4 reactivity and is associated with a lower risk for venous thromboembolism (VTE) in a genome wide association studies meta-analysis.

Objectives: The goal of this study was to determine the mechanism for the association of rs2227376 with a reduced risk of VTE using mice with a homologous mutation (PAR4-P322L).

Methods: Venous thrombosis was examined using our recently generated PAR4-P322L mice in the inferior vena cava stasis and stenosis models. Coagulation and clot stability were measured using rotational thromboelastometry. Thrombin-generating potential was measured in platelet-rich plasma. Phosphatidylserine surface expression and platelet-neutrophil aggregates were analyzed using flow cytometry.

Results: Mice heterozygous (PAR4^P/L) or homozygous (PAR4^L/L) at position 310 had reduced sizes of venous clots at 48 hours. PAR4^P/L and PAR4^L/L platelets had progressively decreased phosphatidylserine in response to thrombin and convulxin, in addition to decreased thrombin generation and decreased PAR4-mediated platelet-neutrophil aggregation.

Conclusion: The leucine allele in extracellular loop 3, PAR4-322L, leads to fewer procoagulant platelets, decreased endogenous thrombin potential, and reduced platelet-neutrophil aggregation. This decreased ability to generate thrombin and bind to neutrophils offers a mechanism for PAR4's role in VTE, highlighting a key role for PAR4 signaling.

Keywords: animal model; blood platelets; protease-activated receptor 4; single nucleotide polymorphisms; thrombin receptor.

FIGURE 1

Protease-activated receptor 4 - P322L (PAR4-P322L) reduces venous thrombosis after complete inferior vena cava (IVC) stasis. (A) Venous clots within the IVC were harvested from wild-type (PAR4–322^P/P [P/P]), heterozygous PAR4–322^P/L (P/L), homozygous PAR4^L/L (L/L), and global PAR4^−/− knockout (−/−) mice 48 hours after complete IVC ligation. (B) Clot incidence and (C) thrombus weight were also recorded 24 hours after IVC ligation. Dots represent individual mice. Statistical significance was determined by (A, C) unpaired t-tests and (B) Fisher’s exact test. PAR4^P/L, mice heterozygous; PAR4^L/L, mice homozygous. **P < .01. PAR4, protease-activated receptor 4.

FIGURE 2

Protease-activated recptor 4-P322L (PAR4-P322L) reduces venous thrombosis after partial inferior vena cava (IVC) stenosis. (A) Thrombi were excised from the surrounding tissue and weighed 48 hours after IVC stenosis. Dots represent individual mice. Statistical significance was determined by unpaired t-tests. PAR4^P/L, mice heterozygous; PAR4^L/L, mice homozygous. **P < .05.

FIGURE 3

Protease-activated receptor 4-P322L (PAR4-P322L) mice have smaller thrombi after inferior vena cava stenosis. Venous clots were excised from the surrounding tissue, embedded in paraffin, and sectioned for (A) hematoxylin and eosin or (B) Carstairs staining. Images were taken at 4× magnification and stitched together when necessary to display the entire clot (A and C). The stitching is indicated with the vertical dotted line. Higher-resolution images (200×) are presented in (B) and (D).

FIGURE 4

Blood from mice without protease-activated receptor 4 (PAR4^−/−) show impaired clot formation. Rotational thromboelastometry was performed to measure coagulation parameters by generating a thromboelastogram. (A) representative traces for each genotype are shown. (B) Clotting time, (C) clot formation time, (D) maximum clot firmness, (E) alpha angle, and the amplitudes at (F) 10 (A10) and (G) 20 (A20) minutes were extracted from the rotational thromboelastometry traces. Dots represent individual mice. Statistical significance was determined by unpaired t-tests. *P < .05, ****P < .0001. L/L, homozygous PAR4^L/L; P/L, heterozygous PAR4–322^P/L; P/P, wild-type PAR4–322^P/P.

FIGURE 5

Protease-activated receptor 4 (PAR4-P322L) decreases platelet-mediated thrombin generation. (A–D) Platelet-rich plasma was obtained from wild-type (WT), heterozygous PAR4–322^P/L (P/L), homozygous PAR4–322^L/L (L/L), and PAR4^−/− knockout (−/−) mice and assessed for thrombin-generating potential. (A) Thrombin generation curves were created using a fluorogenic thrombin substrate over 1 hour. (B) The area under the curve for each genotype was compared with the provided control human plasma (control high and control low) and interpreted as the total thrombin generated over that hour. (C) The peak concentration of thrombin generated and (D) the time it took to reach that peak were also recorded. Statistical significance was determined by unpaired t-tests. *P < .05. PAR4, protease-activated receptor 4.

FIGURE 6

Protease-activated receptor 4-P322L (PAR4-P322L) reduces platelet phosphatidylserine surface expression. Phosphatidylserine exposure was measured with flow cytometry using FITC-tagged lactadherin. Washed mouse platelets were stimulated with convulxin (1 nM) in combination with 200 μM of the protease-activated receptor (PAR)4-agonist peptide (PAR4-AP; AYPGKF), 1 nM of thrombin, and 5 nM of thrombin. Lactadherin-positive events were recorded as the percentage of CD41-positive cells that were also positive for FITC. Dots represent individual mice. Statistical significance was determined by unpaired t-tests. *P < .05, **P < .01, ***P < .001, ****P < .0001. −/−, PAR4^−/− knockout; L/L, homozygous PAR4^L/L; P/L, heterozygous PAR4–322^P/L; P/P, Unstim, unstimulated; wild-type PAR4–322^P/P.

FIGURE 7

Protease-activated receptor 4-P322L (PAR4-P322L) decreases platelet-neutrophil aggregation. Platelet-neutrophil complexes were measured using flow cytometry to record the percentage of LyG+ neutrophils that also expressed CD41. Lysed whole blood from wild-type (PAR4–322^P/P [P/P]), heterozygous PAR4–322^P/L (P/L), homozygous PAR4–322^L/L (L/L), and PAR4^−/− knockout (−/−) mice were incubated with (A) no agonist, or (B) 50 μM, (C) 200 μM, or (D) 400 μM of the protease-activated receptor (PAR)4-agonist peptide (AYPGKF). Statistical significance was determined by unpaired t-tests. *P < .05, **P < .01, ***P < .001, ****P < .0001.