Identification of a Racially Dimorphic Variant in the Human Platelet PAR4 Thrombin Receptor Altering Platelet Function and Pharmacologic Inhibition

Compared to whites, blacks have a 2-fold higher incidence of coronary heart disease (CHD), and black race is an independent predictor of worse survival after CHD events after accounting for other confounding variables (e.g., socioeconomic, demographic, etc.). However, there has been a paucity of literature considering racial differences in platelet function. We recently reported results from the Platelet RNA And eXpression-1 (PRAX1) study of platelets in healthy black (n=70) and white (n=84) subjects (Edelstein et al, Nat. Med 2013). Human platelets express two thrombin receptors, protease activated receptor (PAR) 1 and PAR4. We demonstrated 1) a 3.7-fold increased PAR4-mediated aggregation kinetics and greater calcium mobilization in platelets from black subjects compared to whites, and 2) phosphatidylcholine transfer protein (PC-TP) was a mediator of this racial difference. These findings have potential clinical significance because greater platelet-mediated thrombosis could contribute to the worse outcomes in blacks than whites after coronary events. Additionally, in the presence of vorapaxar (an FDA-approved PAR1 inhibitor for patients with coronary and peripheral vascular disease) PAR4 is the primary means by which thrombin activates platelets, and the risks and benefits of vorapaxar and other PAR inhibitors by race are unknown.

Because PC-TP expression only accounted for 18% of the observed variance in PAR4 function, we have considered additional mechanisms to explain the racial difference in thrombin-induced PAR4-mediated platelet reactivity. Although platelets from blacks express 14% more PAR4 protein than whites, this difference does not explain the variance in platelet PAR4 function. Genome-wide quantitative trait locus analysis identified common 3 SNPs in the PAR4 gene (F2RL3) as associated with PAR4-induced platelet aggregation (P = 4.79 x 10^-11). Importantly, the allele frequency of these SNPs differs by race in both the 1000 Genomes dataset and in PRAX1 (P= 4.31 x 10^-16). One of these SNPs, rs773902, determines if residue 120 in PAR4 is an alanine (81% in whites vs. 37% in blacks) or a threonine (63% in blacks vs. 19% in whites). A second less frequent, non-synonymous F2RL3SNP, Phe296Val, was only observed in blacks and was associated with loss of PAR4 function. This variant appears to have a dominant negative effect since only one copy of PAR4-296Val abolished the enhanced PAR4-AP induced platelet aggregation associated with PAR4-Thr120.

A proximal step in PAR4-induced platelet signal transduction is Gq activation leading to hydrolysis of phosphatidylinositol 4,5-biphosphate to IP₃ and diacylglycerol, followed by increased cytoplasmic calcium. Platelets with the PAR4-Thr120 variant exhibited a 65% greater Ca²⁺ flux than PAR4-Ala120 homozygotes (P = 0.03). To eliminate other sources of inter-individual variation we cloned and expressed each variant in 293 cells, which lack endogenous PAR4 expression, and measured the generation of IP₃ in response to PAR4-AP treatment. Compared to PAR4-Ala120 expressing cells, PAR4-Thr120 expressing cells generated 41% higher IP₃ levels (P = 0.03).

Because PAR inhibitors are either currently approved (vorapaxar) or in development, we examined platelet function in the presence of PAR1 and PAR4 antagonists. PAR4 genotype had no effect on vorapaxar inhibition of PAR1 signaling or PAR4 signaling in the presence of vorapaxar. YD-3, a selective inhibitor of PAR4 and not PAR1, was also tested. In stark contrast to vorapaxar, PAR4 genotype had a significant effect of the efficacy of PAR4 inhibition by the compound YD-3. Platelets from subjects homozygous for the PAR4-Thr120 were not inhibited by any tested concentration of YD-3 while PAR4-Ala120 homozygotes were inhibited in a dose-dependent manner (P = 6x10^-8).

In summary, we have identified variation in the PAR4 gene that alters PAR4 function and accounts for 48% of the observed racial difference in platelet PAR4 signaling. These results, couple with other racial data (Tourdot et al., submitted), have two critical clinical implications for anti-platelet therapy: (1) Patients expressing the PAR4-Thr120 variant (mostly black) may not be adequately protected on current anti-platelet drugs, and (2) PAR4 antagonists currently in development may not be effective in PAR4-Thr120 expressing platelets. This suggests a critical need for new and effective therapies for these patients.