Blockade of the purinergic P2Y12 receptor greatly increases the platelet inhibitory actions of nitric oxide Nicholas S. Kirkbya,b, Martina H. Lundberga, Melissa V. Chana, Ivana Vojnovica, Antonia B. Solomonb, Michael Emersonb, Jane A. Mitchellb,1, and Timothy D. Warnera,1,2 aThe William Harvey Research Institute, Barts and the London School of Medicine and Dentistry, Queen Mary University of London, London EC1M 6BQ, United Kingdom; and bNational Heart and Lung Institute, Imperial College London, London SW3 6LY, United Kingdom Edited by Marek Witold Radomski, Trinity College Dublin, Dublin, Ireland, and accepted by the Editorial Board July 31, 2013 (received for review November 1, 2012) Circulating platelets are constantly exposed to nitric oxide (NO) guanylyl cyclase, causing an increase in intraplatelet cGMP (13). released from the vascular endothelium. This NO acts to reduce Although this cGMP pathway is distinct from the cAMP pathway platelet reactivity, and in so doing blunts platelet aggregation and that is modulated by PGI2 and P2Y12, earlier studies have shown thrombus formation. For successful hemostasis, platelet activation that activation of these two inhibitory pathways produces a syn- and aggregation must occur at sites of vascular injury despite the ergistic antiplatelet effect (18). Further, it has been shown that constant presence of NO. As platelets aggregate, they release a NO donor drug can boost the effectiveness of P2Y12 inhibitors secondary mediators that drive further aggregation. Particularly against thrombus formation ex vivo (19, 20). Here we report that significant among these secondary mediators is ADP, which, acting treatment of platelets with P2Y12 receptor blockers dramatically increases the antiplatelet effects of NO, an increase that is very through platelet P2Y12 receptors, strongly amplifies aggregation. much greater than that already established for PGI2. This reveals Platelet P2Y12 receptors are the targets of very widely used anti- that activation of platelet P2Y12 receptors by ADP could provide thrombotic drugs such as clopidogrel, prasugrel, and ticagrelor. Here a common switching mechanism by which platelet sensitivity to we show that blockade of platelet P2Y12 receptors dramatically the antithrombotic effects of NO and PGI2 is rapidly reduced to enhances the antiplatelet potency of NO, causing a 1,000- to allow platelet activation, as required for hemostasis. It also raises 100,000-fold increase in inhibitory activity against platelet aggre- the important possibility that for individual patients, the effec- gation and release reactions in response to activation of receptors tiveness of P2Y12 receptor blockers may vary on the basis of their for either thrombin or collagen. This powerful synergism is ex- vascular production of NO and PGI2. This provides a rationale plained by blockade of a P2Y12 receptor-dependent, NO/cGMP- for the commonly reported disconnect between the results of ex insensitive phosphatidylinositol 3-kinase pathway of platelet acti- vivo testing of platelet responses and thrombotic outcomes. vation. These studies demonstrate that activation of the platelet Results and Discussion ADP receptor, P2Y12, severely blunts the inhibitory effects of NO. The powerful antithrombotic effects of P2Y12 receptor blockers P2Y12 Receptor Blockade Greatly Increases the Inhibition of Platelet may, in part, be mediated by profound potentiation of the effects Aggregation by NO. To study interactions between P2Y12 receptor of endogenous NO. activation and the inhibitory effect of NO on platelet aggregation, we used thrombin as the stimulus for platelet activation. Thrombin anti-platelet therapy | atherothrombosis | prostacyclin is a physiologically relevant agonist that provides a strong signal for ADP release. Similar to previous studies of this type (8), we found 1 U/mL thrombin to provide a maximal aggregation and secretion ctivation of platelets in a damaged blood vessel leads to response that was relatively insensitive to P2Y12 receptor blockade Aa well-characterized sequence of events, including the im- (Fig. S1). By using 1 U/mL thrombin as an agonist, we could portant release of secondary mediators of aggregation, notably – therefore study the effects of P2Y12 receptor blockade on re- ADP and thromboxane A2 (1 3). These mediators are the tar- sponses to NO without the confounding effects of a differing gets of antithrombotic drugs taken by many millions of patients baseline. In control conditions, the NO donor diethlyammonium as prophylactic protection against heart attacks and strokes. In (Z)-1-(N,N-diethylamino)diazen-1-ium-1,2-diolate (DEA/NONOate), particular, inhibition of thromboxane A2 production by aspirin which spontaneously releases NO at physiological pH (half-life ∼2 explains the antithrombotic utility of this drug, whereas blockade min), produced only weak inhibition of thrombin (1 U/mL)-induced of the ADP receptor, P2Y12, by drugs such as clopidogrel, pra- washed platelet aggregation, even at supraphysiological concen- sugrel, and ticagrelor provides a second, and possibly even trations (Fig. 1A). However, in the presence of the P2Y12 receptor stronger, antithrombotic protection (4). Within the circulation, blocker prasugrel active metabolite (PAM), both the potency and platelets are constantly exposed to antithrombotic mediators, the maximal effect of DEA/NONOate were dramatically in- A − ± notably prostaglandin I2 (PGI2), which is derived from the vas- creased (Fig. 1 ; logIC50 DEA/NONOate, 3.0 0.4; DEA/ cular endothelium, and nitric oxide (NO), which can be pro- NONOate +PAM, −7.9 ± 0.2; P < 0.0001; n = 8). This clearly duced by the vascular endothelium (5) and platelets themselves suggests a powerful interaction between NO and P2Y12 blockade. (6, 7). This leads to the understanding that there is an important interplay between endothelial-derived antiaggregatory mediators and platelet-derived proaggregatory mediators. For example, it Author contributions: N.S.K., M.H.L., J.A.M., and T.D.W. designed research; N.S.K., M.H.L., M.V.C., and I.V. performed research; A.B.S. and M.E. contributed new reagents/analytic has been demonstrated that activation of platelet P2Y12 recep- tools; N.S.K., M.H.L., M.V.C., and T.D.W. analyzed data; and N.S.K. and T.D.W. wrote tors by ADP causes inhibition of platelet adenylyl cyclase, which the paper. reduces the ability of PGI2 acting through prostaglandin IP re- Conflict of interest statement: T.D.W. has received consultancy fees from AstraZeneca and ceptors to elevate cAMP, and thus reduces platelet aggregation. Daichii Sankyo/Eli Lilly relating to clinical development of P2Y12 inhibitors. Hence, blockade of platelet P2Y12 receptors increases the sen- This article is a PNAS Direct Submission. M.W.R. is a guest editor invited by the sitivity of platelets to the antiaggregatory and antisecretory effects Editorial Board. of PGI2 (8, 9). This interaction is readily understandable, as both Freely available online through the PNAS open access option. 1 P2Y12 receptors and PGI2 have actions on a common signaling J.A.M. and T.D.W. contributed equally to this work. pathway, cAMP. 2To whom correspondence should be addressed. E-mail: [email protected]. Compared with PGI2, NO is less potent as an inhibitor of This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10. platelet activation (10–17). NO acts by directly activating platelet 1073/pnas.1218880110/-/DCSupplemental. 15782–15787 | PNAS | September 24, 2013 | vol. 110 | no. 39 www.pnas.org/cgi/doi/10.1073/pnas.1218880110 Downloaded by guest on September 28, 2021 Vehicle DEA/NONOate PAM (3µM) DEA/NONOate Aggregation in washed A (1µM) +PAM platelets - -20 thrombinthrombin thrombinthrombin thrombinthrombin thrombinthrombin n i b 0 m o r 20 h t f o 40 n o Fig. 1. Blockade of the P2Y receptor sensitizes pla- i 60 12 t Light transmission i b i 80 Vehicle telets to the inhibitory effect of NO against thrombin- nh I Inhibition of thrombin- 100 PAM * induced aggregation, secretion, and Rap1 activation. induced aggregation (%) aggregation induced -12 -10 -8 -6 -4 -2 Thrombin (1 U/mL)-induced aggregation of washed 1 min platelets, determined by changes in light transmission, [DEA/NONOate] log (M) was only weakly inhibited by the NO donor, DEA/ A n = Aggregation in blood GPIIb/IIIa activation ADP+ATP secretion NONOate ( )( 8), but this effect was greatly in- B C D creased in platelets pretreated with the P2Y receptor -20 -20 -20 12 blocker PAM (3 μM). Traces show the effects of 1 μM 0 0 0 DEA/NONOate in the absence and presence of PAM. 20 20 20 40 40 40 Similarly, inhibition by DEA/NONOate of platelet ag- 60 60 60 gregation in hirudin-anticoagulated whole blood stim- (% inhibition)(% PAC-1 bindingPAC-1 μ + 80 Vehicle 80 Vehicle 80 ulated with SFLLRN-amide (10 M) AYPGKF-amide (30 Inhibition of PAR- Vehicle PAM μ B 100 PAM * 100 * Inhibition of thrombin- 100 PAM * M) and measured by impedance aggregometry ( ), induced aggregation (%) thrombin (1 U/mL)-induced GPIIb/IIIa activation by flow -12-10 -8 -6 -4 -2 -12 -10 -8 -6 -4 induced ATP+ADP release (%) -12 -10 -8 -6 -4 cytometry (C)andADP+ATP secretion, measured by [DEA/NONOate] log M [DEA/NONOate] log M [DEA/NONOate] log M amodified luminometric method (D), were also strongly potentiated by P2Y receptor blockade (n = 4 for each). Rap1 activation 12 E Thrombin (1 U/mL) also stimulated rapid activation of Rap1b-GTP 0.6 Vehicle Rap1, as determined by affinity purification of Rap1-GTP PAM * total Rap1b and Western blotting of lysates of platelets stimulated 0.4 under stirring conditions for 15 s (E). Rap1 activation was Thrombin (1U/ml) - + + + + + + + + - little altered by DEA/NONOate or P2Y12 receptor block- 0.2 (active / total) (active ade alone but was strongly inhibited by the combina- DEA/NONOate (uM) 0 0 0.01 1 100 0 0.01 1 100 - Rap1 activation tion. Platelet lysates incubated with GTPγS were included 0.0 Prasugrel-AM (3uM) -----++++- on blots as a positive control for affinity isolation of GTP S (pull-down control) - --------+ -10 -8 -6 -4 Rap1-GTP (n = 4 for each).