Zhongren Ding, Key Laboratory of Molecular Medicine, Ministry of Education, Shanghai, China and Department of Biochemistry and Molecular Biology, Fudan University Shanghai Medical College, Shanghai 200032, China. Tel.: +86 21 5423 7896; fax: +86 21 6403 3738. E-mail: email@example.com
Summary. Background: In our previous in vitro study, we reported a constitutively active chimeric P2Y12 (cP2Y12) and found that AR-C78511 is a potent inverse agonist at this receptor. The role of cP2Y12 in platelet activation and thrombosis is not clear.
Objectives: To investigate the physiologic implications of cP2Y12 for platelet activation and thrombus formation, and to evaluate the antiplatelet activity of AR-C78511 as an inverse agonist.
Methods and Results: We generated transgenic mice conditionally and platelet-specifically expressing cP2Y12. High-level expression of cP2Y12 in platelets increased platelet reactivity, as shown by increased platelet aggregation in response to multiple platelet agonists. Moreover, transgenic mice showed a shortened bleeding time, and more rapid and stable thrombus formation in mesenteric artery injured with FeCl3. The constitutive activity of cP2Y12 in platelets was confirmed by decreased platelet cAMP levels and constitutive Akt phosphorylation in the absence of agonists. AR-C78511 reversed the cAMP decrease in transgenic mouse platelets, and exhibited a superior antiplatelet effect to that of AR-C69931MX in transgenic mice.
Conclusions: These findings further emphasize the importance of P2Y12 in platelet activation, hemostasis, and thrombosis, as well as the prothrombotic role of the constitutive activity of P2Y12. Our data also validate the in vivo inverse agonist activity of AR-C78511, and confirm its superior antiplatelet activity over neutral antagonists.
ADP plays a key role in platelet function, hemostasis, and thrombosis. Although ADP itself is a weak platelet activator, it is released from dense granules during platelet activation, and amplifies the platelet responses induced by other platelet agonists, and stabilizes platelet aggregates. The ADP-induced signal is mediated by two G-protein-coupled receptors (GPCRs), Gq-coupled P2Y1 and Gi-coupled P2Y12, and signaling through both pathways is necessary for ADP-induced platelet activation .
P2Y12 plays a central role in platelet activation and thrombosis, and has been by far the most successful target for antiplatelet therapy [2–4]. The discovery of constitutive activity of GPCRs and inverse agonists has significantly changed our understanding of receptor activation, disease pathogenesis, and mechanisms of drug action. GPCRs can be active in the absence of agonists (i.e. have constitutive activity), owing to receptor overexpression  or receptor mutation, both of which have been reported to be the cause of human diseases, such as the familial syndrome of hypocalcemia with hypercalciuria , Bartter’s syndrome , melanoma , thyroid adenomas , and others . To treat such diseases, classical GPCR antagonists that block agonist binding to the receptors are ineffective, whereas the inverse agonists have therapeutic advantages . Using a recombinant system, we have generated a chimeric P2Y12 construct (henceforth referred to as cP2Y12) with constitutive activity, and have identified a potent inverse agonist (AR-C78511) of cP2Y12 . We hypothesize that the constitutive activity of cP2Y12 is prothrombotic, and thus inverse agonists of this receptor will be more effective platelet inhibitors.
In this study, we generated transgenic mice conditionally and platelet-specifically expressing cP2Y12. We confirmed the constitutive activity of cP2Y12 in platelets and its prothrombotic role. Moreover, the superior antiplatelet effect of AR-C78511 as an inverse agonist was also validated.
Materials and methods
Generation of mice expressing cP2Y12 in platelets (double transgenic mice expressing both αIIb–rtTA–VP16 and β-gal–CMV–cP2Y12)
Conditional transgenic mice platelet specifically expressing cP2Y12 were generated as described in Data S1.
To induce transgene cP2Y12 expression, a 0.2 mg mL−1 doxycycline (Sigma-Aldrich, St Louis, MO, USA) solution was administered through drinking water 2–3 weeks before analysis . The water with doxycycline was changed every 3–4 days, and the container was wrapped in aluminum foil to ensure the activity of doxycycline .
RT-PCR and real-time PCR analysis
Total RNA was extracted from mouse platelets and megakaryocytes with Trizol reagent (Invitrogen, Carlsbad, CA, USA). Reverse transcription was carried out with the reverse transcription reagent kit (Takara, Dalian, China), as described in Data S1.
Confocal microscopy of mouse bone marrow (BM) cells and platelets was performed as described in Data S1.
Western blot analysis
Washed mouse platelets (0.25 mL) were stimulated with or without agonists for the appropriate time, lysed, and then subjected to western blot analysis as described in Data S1.
Preparation of mouse platelets and platelet aggregation
Blood was collected from the abdominal aorta of pentobarbital sodium-anesthetized mice into syringes containing 3.8% citrate (9 : 1). Washed platelets were prepared and resuspended in Tyrode’s buffer as described previously . Platelet aggregation was analyzed with a lumi-aggregometer (Model 400VS; Chrono-Log, Havertown, PA, USA) [15,16].
Measurement of cAMP in mouse platelets
Intracellular cAMP in platelets was measured as described previously [12,17], and cAMP conversion from ATP was calculated as described by Berlot , with the following formula: cAMP conversion from ATP = [3H]cAMP/([3H]ATP + [3H]cAMP) × 103.
Mouse blood was collected from the tail, and was drawn into tubes containing 3.8% citrate (6 : 1). Platelet numbers were then determined on a Sysmex XS-800i automated cell counter (Sysmex Corporation, Kobe, Japan).
Bleeding time measurement
The bleeding time was measured by 0.2-cm tail-tip transection. The tail was immediately immersed in 0.9% isotonic saline at 37 °C, and the time required for arrest was defined as the bleeding time.
In vivo thrombosis
Intravital microscopy was performed as described previously [19,20]. Thrombus formation in mesenteric arterioles were observed in real time for ∼ 30 min after injury.
All data are expressed as mean ± standard error of the mean. Unless otherwise stated, differences between the groups were analyzed by one-way anova followed by a Newman–Keuls test with GraphPad Prism 5 (Graphpad Inc, San Diego, CA, USA). P < 0.05 was considered to be statistically significant.
Generation of transgenic mice platelets specifically expressing cP2Y12
Previously, we have shown that a chimeric P2Y12 with the C-terminus replaced by the corresponding P2Y1 is constitutively active when stably expressed in CHO-K1 cells . To evaluate the role of cP2Y12 in platelet activation, hemostasis, and thrombosis, we generated trangenic mice that express cP2Y12 in the megakaryocytic/platelet lineage using the Tet-on system to conditionally express cP2Y12 under the control of the platelet-specific human promoter αIIb (Fig. 1). We first generated two mouse lines: the first one expresses the reverse Tet-responsive transcriptional activator (rtTA) under the control of the human platelet-specific αIIb promoter (αIIb–rtTA–VP16); the second is the responder line with a bidirectional minimal cytomegalovirus (CMV)–rtTA-responsive promoter driving the β-galactosidase gene (as a reporter gene) and our gene of interest, the cP2Y12 gene (β-gal–CMV–cP2Y12) (Fig. 1).
We obtained 19 founders with αIIb–rtTA–VP16 integrated into mouse genomic DNA, as detected by PCR analysis of mouse tail genomic DNA. Four lines with higher αIIb–rtTA–VP16 integration were further confirmed by RT-PCR analysis, and chosen for subsequent crossbreeding and further experiments (Fig. 2A,B). Western blot and immunohistochemistry analysis also confirmed transgene expression in platelets and megakaryocytes of the transgenic mice (Fig. 2C–E).
Nine responders expressing β-gal–CMV–cP2Y12 were generated, as shown by PCR analysis of genomic DNA with primers specific for the transgenes (Fig. 3A). By crossbreeding the αIIb–rtTA–VP16 line with the β-gal–CMV–cP2Y12 line, we finally obtained three mouse lines (lines 24, 28, and 32) expressing both transgenes simultaneously (Fig. 3B). β-Gal–CMV–cP2Y12 expression in platelets was further validated by RT-PCR amplifying the β-gal gene (Fig. 3C) and by western blot detecting hemagglutinin (HA)-tagged cP2Y12 (Fig. 3D) in the three lines pretreated with doxycycline. Immunofluorescence histochemistry also confirmed doxycycline-induced cP2Y12 expression (Fig. 3E) in platelets of the double transgenic mice. Lines 24, 28 and 32 were bred together to generate homozygous transgenic mice, as confirmed by breeding analysis, and these were used for subsequent experiments. The general health and lifespan of transgenic mice were normal.
cP2Y12 expression is inducible and megakaryocytic/platelet lineage-specific
To determine the expression level of cP2Y12, we used real-time PCR to amplify cDNA reverse transcribed from mRNA. As shown in Fig. S1A, treatment with 0.05 mg mL−1 doxycycline induced high-level cP2Y12 expression in transgenic mouse platelets. Doxycycline-dependent cP2Y12 expression was further confirmed by western blot with an antibody against the HA-tag at the N-terminus of cP2Y12 (Fig. S1B). Doxycycline induction was time-dependent and dose-dependent; treatment with 0.05 mg mL−1 for 20 days induced maximal cP2Y12 expression under our conditions (Fig. S1C). RT-PCR detection of the β-gal gene in liver, spleen, kidney, lung, BM and platelets demonstrated that transgene expression was detectable only in platelets and BM from doxycycline-treated transgenic mice (Fig. S1D), confirming the megakaryocytic/platelet lineage-specific transgene expression.
Hyperaggregatory responses of transgenic mouse platelets
Functional studies showed that platelet aggregation in response to ADP (10 μm) or 2MeSADP (100 nm, data not shown) was strongly increased in doxycycline-treated transgenic mice as compared with transgenic mice without doxycycline treatment and wild-type mice (Fig. 4A). Platelet aggregation induced by other agonists, such as thrombin, the protease-activated receptor-4 (PAR4) agonist AYPGKF, and the thomboxane A2 analog U46619, was also drastically enhanced in transgenic mice (Fig. 4A), consistent with the central role of P2Y12 in platelet activation . Real-time PCR revealed that transgenic mouse platelets had similar P2Y1, P2Y12, PAR4 and thomboxane receptor (TP) expression levels as the wild-type platelets (Fig. 4B), further supporting the idea that the increased platelet reactivity in transgenic mouse platelets is the result of expression of cP2Y12 rather than changes in expression of other receptors.
We also evaluated basal platelet function by measuring the expression levels of P-selectin, using fluorescein isothiocyanate-labeled anti-P-selectin antibody, and the active form of αIIbβ3, using phycoerythrin-labeled JON/A antibody, under basal conditions in platelets from wild-type and induced transgenic mice by fluorescence-activated cell sorting analysis. We found that the expression of P-selectin and that of the active form of αIIbβ3 were similar in platelets from wild-type and transgenic mice (Fig. S2).
Decreased bleeding time in transgenic mice
The bleeding time reflects primary hemostasis in vivo. The doxycycline-treated transgenic mice showed a significantly shorter bleeding time than transgenic mice without doxycycline treatment and wild-type mice (Fig. 4C), consistent with the hyperaggregatory responses of platelets as a result of expression of cP2Y12 in transgenic mice. The enhanced hemostasis was not caused by platelet number change, because transgenic mice and wild-type mice had similar platelet counts (Fig. 4D).
Enhanced thrombosis in transgenic mice expressing cP2Y12
Given the finding that the transgenic mouse platelets had an increased aggregation response to stimuli, we investigated whether cP2Y12 expression also prompted increased thrombosis in vivo. We carried out intravital microscopy experiments with an FeCl3-injured mouse mesenteric arteriole thrombosis model [19,20]. As shown in Fig. 5, thrombus formed more rapidly and stably in mesenteric arterioles of transgenic mice than in those of wild-type mice (Fig. 5A; Video Clips S1 and S2). Both time to first thrombus and occlusion time (Fig. 5B) after injury in cP2Y12 transgenic mice (time to first thrombus, 2.7 ± 1.0 min; occlusion time, 7.0 ± 0.8 min) were significantly decreased as compared with wild-type mice (time to first thrombus, 4.5 ± 0.3 min; occlusion time, 12 ± 0.3 min) (Fig. 5B), indicating that cP2Y12  made the transgenic mice more sensitive to thrombogenic stimuli, and consequently more vulnerable to thrombosis.
cP2Y12 in platelets of transgenic mice is constitutively activated and can be inhibited by the P2Y12 inverse agonist AR-C78511
So far, we have shown that cP2Y12 expression in platelets increases platelet reactivity and thrombogenesis. Previously, we have found that this cP2Y12 is constitutively activated in CHO-K1 cells, and that the constitutive activity can be antagonized by a potent inverse agonist, AR-C78511 . To address whether the cP2Y12 in platelets of transgenic mice was also constitutively activated, we compared the cAMP levels, a widely accepted hallmark reflecting platelet P2Y12 activation status , in platelets from transgenic mice and wild-type mice. To facilitate cAMP assay, platelets were pretreated with forskolin. As shown in Fig. 6A, in the absence of the P2Y12 agonist ADP, the platelet cAMP level of transgenic mice was significantly decreased as compared with that of wild-type mice. The platelet cAMP level in transgenic mice in the absence of P2Y12 agonists was ∼ 73% of that in wild-type mice, indicating that cP2Y12 in transgenic mouse platelets is constitutively activated. The cAMP level in transgenic mouse platelets was further decreased upon ADP stimulation, indicating that cP2Y12 in trangenic platelets retains a normal response to P2Y12 agonists, as found in our previous studies in cell lines . In line with our previous cell line results, the decreased cAMP level in transgenic mouse platelets was reversed by AR-C78511  (Fig. 6A), a potent P2Y12 inverse agonist .
The serine-threonine kinase Akt has been established as an important downstream signal molecule of the P2Y12–Gi pathway in platelets . Therefore, we evaluated whether this downstream signaling molecule is activated in transgenic mouse platelets expressing cP2Y12 by measuring Akt phosphorylation. As expected, Akt was constitutively phosphorylated in the platelets of transgenic mice in the absence of P2Y12 agonists, and the phosphorylation was further enhanced by stimulation with 2-MeSADP, a more potent analog of ADP (Fig. 6B), consistent with our previous finding that cP2Y12 has a normal response to 2MeSADP .
Finally, we investigated the role of the constitutive activity of cP2Y12 in platelet activation. As shown in Fig. 6C, serotonin at 10 μm caused shape change alone, and did not induce aggregation of washed platelets from wild-type mice. In contrast, the same concentration of serotonin elicited robust aggregation of platelets from transgenic mice, because the cP2Y12 in transgenic mouse platelets is already constitutively activated. Very importantly, we found that the serotonin-induced platelet aggregation was abolished by AR-C78511; in contrast, AR-C69931MX, a neutral P2Y12 agonist , caused only slight inhibition at the concentration (100 nm) that abolished ADP-induced platelet aggregation in wild-type mice (Fig. 6C). Even at 1 μm, AR-C69931MX still only slightly inhibited serotonin-induced platelet aggregation in transgenic mice (Fig. S3), in contrast to the abolition by AR-C78511 at 100 nm (Fig. 6C). Together, these findings suggest that P2Y12 antagonists with potent inverse agonist activity, such as AR-C78511, may be superior to neutral P2Y12 antagonists for thrombotic diseases related to P2Y12 expression or gain-of-function mutation.
P2Y12 plays a central role in platelet activation, hemostasis, and thrombosis, and has therefore been intensively investigated. We previously reported a chimeric P2Y12 with constitutive activity and a normal response to P2Y12 agonists. To investigate the role of this cP2Y12 in platelet activation and thrombosis, in this study we generated transgenic mice conditionally and platelet-specifically expressing cP2Y12. The transgenic mice platelet-specifically expressed cP2Y12 upon doxycycline administration, showed increased platelet reactivity to multiple platelet agonists, and enhanced thrombogenicity. AR-C78511, a potent P2Y12 inverse agonist identified in our previous study , inhibited the constitutive activity of cP2Y12 in platelets of the transgenic mouse. More importantly, AR-C78511 completely inhibited serotonin-induced platelet aggregation in transgenic mice, whereas AR-C69931MX, a neutral P2Y12 antagonist , did not.
We generated the transgenic mice by using a Tet-on system under the control of a platelet-specific promoter, αIIb, and only platelets and megakaryocytes expressed transgenes upon administration of doxycycline; this circumvented possible embryonic lethality caused by cP2Y12 and the resultant platelet hyperreactivity. The enhanced platelet reactivity, increased thrombosis and shortened bleeding time in transgenic mice is consistent with the impaired platelet activation, thrombosis and prolonged bleeding in P2Y12 knockout mice [23,24]. Transgenic mice have similar platelet numbers as wild-type mice, indicating that cP2Y12 expression did not influence platelet number. This is in accordance with previous studies showing that P2Y12 knockout did not influence platelet production [23,24].
Platelet P2Y1, P2Y12, PAR4 and TP expression remained unchanged in transgenic mice as compared with wild-type mice, indicating that expression of cP2Y12 does not affect the expression of P2Y1, P2Y12, PAR4 and TP on platelets. The normal P2Y1, P2Y12, PAR4 and TP expression on transgenic mouse platelets also ruled out the possibility that the enhanced platelet response to ADP, thrombin and U46619 is caused by increased expression of platelet P2Y1, P2Y12, PAR4 and TP, further supporting the important role of cP2Y12, which is overexpressed at a 4.6-fold higher level than endogenous P2Y12, in enhancing platelet reactivity (Fig. S4).
Constitutive activity of GPCRs has been widely reported in recombinant systems, and has also been reported clinically. In our previous study, we reported a constitutively active chimeric P2Y12 in a recombinant system. The present study confirmed its constitutive activity in platelets of transgenic mice, as shown by decreased platelet cAMP levels and constitutive Akt phosphorylation in the absence of ADP receptor agonists. Strikingly, serotonin alone induced robust platelet aggregation in transgenic mice, in agreement with the idea that serotonin alone does not trigger platelet aggregation, because it activates only the Gq pathway, and the Gi or Gz pathway must be complemented for normal platelet aggregation . Therefore, the robust platelet aggregation in transgenic mice induced by serotonin alone found in this study establishes that the overexpressed cP2Y12 in platelets is constitutively activated and functions physiologically.
Theoretically, antagonists with inverse agonist activity have therapeutic advantages over neutral receptor antagonists for the treatment of diseases caused by constitutive receptor activation as a result of receptor overexpression or gain-of-function mutation. Actually, most of the clinically effective receptor antagonists are found to have inverse agonist activity, to different extents. More and more inverse agonists are being subjected to clinical trials . Whether the superior antiplatelet activity of AR-C78511 as a potent inverse agonist over AR-C69931MX revealed in this study can be translated into better antithrombotic role is not known, and is currently under investigation. In this context, the transgenic mice that we have generated also provide a useful animal model with which to study the inverse agonist activity of P2Y12 antagonists.
Although the constitutive activity of P2Y12 has not been reported clinically so far, numerous studies have shown that some P2Y12 polymorphisms are related to increased platelet activation and thrombotic risk [26–31], which may be a result of P2Y12 constitutive activation caused by receptor mutation or overexpression. In line with this hypothesis, increased P2Y12 expression has been implicated in the gain-of-function P2Y12-H2 haplotype , which makes the H2 carriers highly susceptible to thrombotic risks . Consistently, we recently found that type II diabetes mellitus patients, who are characterized by platelet hyperreactivity, have enhanced P2Y12 expression on platelets (Yan Zhang, Jianqin Ye, Liang Hu, Zhichao Fan, Si Zhang, Yan Yan, Xunbin Wei, Satya P. Kunapuli, Zhongren Ding, unpublished data). Hence, our transgenic mouse model will enable evaluation of the implications of disease states for thrombotic complications, and could serve as a good animal model for use in disease models to evaluate thrombotic complications.
In conclusion, we have generated a transgenic mouse line and demonstrated that P2Y12 constitutive activation increases platelet reactivity and thrombosis. We have also shown that a P2Y12 inverse agonist has a superior antiplatelet effect to that of a neutral antagonist. Further studies are necessary to explore the potential superior antithrombotic effects of P2Y12 inverse agonists.
We thank D. Wilcox, Medical College of Wisconsin, for providing the plasmid containing human αIIb promoter; K. Ravid, Boston University School of Medicine, for providing the pBI-G vector; and P. L. Gross, McMaster University, for critical reading of the manuscript. This work was supported by the National Natural Science Foundation of China (No. 30772564) and the Program for Professor of Special Appointment (Eastern Scholar) at Shanghai Institutions of Higher Learning to Z. Ding, and a grant from the National Institutes of Health National Heart, Lung and Blood Institute (grant HL60683) to S. P. Kunapuli.
Disclosure of Conflict of Interests
The authors state that they have no conflict of interest.