Introduction: Human platelets express two thrombin receptors, protease activated receptor (PAR) 1 and PAR4 with different affinities for thrombin and different signaling properties. PAR1 and PAR4 have non-identical tissue expression patterns and repertoire of protease activators. We have shown a significant difference in activation kinetics and calciummobilization between platelets from healthy white and black subjects stimulated with PAR4-AP (Nat Med 2013). Part of these racial differences appears to be mediated by PAR4 signaling through Gq to Rap1 and PKC (ATVB 2014). Given the importance of platelet thrombus formation in ischemic arterial disease, these findings may contribute to the known worse outcomes in blacks compared to whites after acute coronary syndromes. Genomic approaches demonstrated that ~50% of the racial variance in PAR4-AP-induced platelet aggregation and calcium flux is caused by an Ala120Thr substitution in PAR4 (encoded by rs773902) and that the Thr120 variant is the "hyperreactive" compared to Ala120 (Blood 2014). Importantly, the Thr120 allele frequency is 63% in blacks and 19% in whites. However, the effect of the physiologic activator, thrombin, on this platelet PAR4 variant (independent of race) has not been characterized. The goals of this work were to perform a detailed characterization of the PAR4 Ala120Thr effect on thrombin-induced platelet function, and to assess pharmacogenetic effects of the PAR4 Ala120Thr variant on standard anti-platelet agents and novel PAR4 inhibitors.

Results:We recruited 130 healthy, multi-racial donors and genotyped for rs773902. Subjects were excluded who had reduced arachidonic acid aggregation or who had the rare PAR4 Val296 variant. Preliminary studies showed heterozygotes had intermediate phenotypes to all assays, so most comparisons were between washed platelets homozygous for Thr120 or Ala120. The thrombin dose response curve for Thr120 homozygotes was left-shifted relative to Ala120 homozygotes (p=0.022 for genotype effect; n=10 per genotype). The concentration of thrombin required to produce 50% maximal aggregation (ED50) was 17% lower for Thr120 homozygotes than for Ala120 homozygotes (0.041 ± 0.002 U/mL vs. 0.049 ± 0.002 U/mL, respectively; p<0.001). The PAR4-AP dose response for Thr120 homozygotes was also left-shifted compared to Ala120 homozygotes (p<0.0001 for genotype effect; n=11 per genotype), as would be expected for functional variants in PAR4. The PAR4-AP ED50 was 2.5-fold lower for Thr120 homozygotes than for Ala120 homozygotes (41.0 uM ± 5.0 uM vs. 101.0 ± 4.5 uM respectively; p<0.001) Compared to Ala120 homozygotes, Thr120 homozygote platelets showed higher thrombin-induced ATP release and P-selectin exposure, indicating higher levels of dense and alpha granule release, and higher calcium mobilization, indicating Gq coupling (n=4-7 per genotype).

Next, washed platelets were incubated with varying concentrations of YD-3 (PAR4 blocker), ASA (COX inhibitor), 2-MeSAMP (P2Y12 antagonist) or the PAR1 antagonist, vorapaxar, and assessed for aggregation after stimulation with 0.1 U/mL thrombin. The concentration of vorapaxar required to inhibit 50% of maximal platelet aggregation (IC50) was 2.6 fold lower for Ala120 homozygotes relative to Thr120 homozygotes (45.5 ± 7.6 nM vs. 118.6 ± 40.9 nM, p<0.0001; n=3 per genotype). The IC50 of 2-MeSAMP was substantially lower for Ala120 homozygotes relative to Thr120 homozygotes (4.0 ± 5.0 uM vs. 10.9 ± 4.8 uM, p=0.056; n=4 per genotype). YD-3 and ASA were unable to inhibit thrombin-induced aggregation regardless of the PAR4 variant.

Conclusions: These data indicate that the common Ala120Thr variant alters platelet reactivity to physiologic agonists, that the Thr120 variant has decreased sensitivity to PAR1 and P2Y12 inhibition, and predict the PAR4 Thr120 variant would alter the risk-benefit of PAR1 inhibition. Since current antiplatelet therapy is based largely on platelet studies done with white subjects, novel PAR4 inhibitors may provide improved anti-thrombotic therapy for patients with the Thr120 risk allele.

Disclosures

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.

This icon denotes a clinically relevant abstract

Sign in via your Institution