Wolfgang Bergmeier, Cardeza Foundation and Department of Medicine, Thomas Jefferson University, Philadelphia, PA, USA. Tel.: +215 955 2093; fax: +215 923 3836. E-mail: email@example.com
Summary. A rise in the intracellular calcium (Ca2+) concentration is a major component of the signaling mechanisms regulating platelet function in thrombosis and hemostasis. Previous studies, however, failed to identify many key molecules regulating Ca2+ signaling in platelets. Here, we review recent findings, which identified CalDAG-GEFI as a critical Ca2+ sensor that links increases in intracellular Ca2+ to integrin activation, TxA2 formation, and granule release in stimulated platelets. Furthermore, we summarize work that lead to the discovery of STIM1 and Orai1 as key regulators of store-operated calcium entry (SOCE) in platelets. A short discussion on the usefulness of each molecule as a potential new target for antiplatelet therapy is included.
Platelet activation is an integral component of hemostasis and thrombosis. Activation of platelets by most stimulatory agents leads to an increase in the concentration of cytosolic Ca2+ [Ca2+]i. Platelet responses that are directly dependent on an increase in [Ca2+]i include integrin activation, release of the second wave mediators ADP and thromboxane A2 (TxA2), spreading, and platelet procoagulant activity [1–4]. The source of Ca2+ for cell activation can be either intracellular or extracellular (Fig. 1). Intracellular Ca2+ is released from the sarcoplasmatic reticulum (SR) by inositol 1,4,5-triphosphate (IP3), produced by activated phospholipase C (PLC). The major mechanism for entry of extracellular Ca2+ is store-operated Ca2+ entry (SOCE), a process controlled by the Ca2+ concentration in the SR. Depletion of SR Ca2+ stores triggers the activation of Ca2+ release activated calcium (CRAC) channels in the plasma membrane . In this review, we will summarize recent studies that identified molecules critical for SOCE and Ca2+ signaling in platelets.
CalDAG-GEFI links increases in intracellular Ca2+ to the signaling pathways regulating integrin activation, TxA2 formation, and granule release
CalDAG-GEFI and platelet function
CalDAG-GEFI (CD-GEFI, RasGRP2) is a member of the CalDAG-GEF/RasGRP family of intracellular signaling molecules involved in the activation of small G proteins of the Ras family [6,7]. We and others have shown that CD-GEFI is predominantly expressed in only few cell types, namely certain neurons, platelets/megakaryocytes, and neutrophils [7,8]. CD-GEFI contains binding sites for Ca2+ and diacylglycerol (DAG) and a guanine nucleotide exchange factor (GEF) domain catalyzing the activation of Rap1 and Rap2, key regulators of integrin activation in platelets and other cells .
Studies by Shattil and colleagues provided the first evidence that CD-GEFI may be a critical regulator of platelet integrin activation, as αIIbβ3 activation in megakaryocytes correlated with the expression level of CD-GEFI . In our own studies, we used CD-GEFI-deficient mice to demonstrate its importance for the activation of Rap1 and β3 integrins in platelets [8,11]. CD-GEFI-deficient platelets showed marked defects in αIIbβ3-mediated aggregation in response to all tested physiological agonists. The activation of β1 integrins was also significantly impaired. Mice lacking CD-GEFI were characterized by a strongly increased bleeding time and protection in models of systemic and arterial thrombosis. Furthermore, CD-GEFI−/− mice showed an impaired inflammatory response because of a defect in the activation of β1 and β2 integrins in neutrophils .
Several of our results suggest that CD-GEFI is a highly efficient sensor for Ca2+, while it may not be regulated by changes in the intracellular DAG concentration. First, CD-GEFI−/− platelets show normal aggregation in response to phorbol ester (DAG mimetic) but fail to aggregate when stimulated with Ca2+ ionophore . Second, Rap1 activation in CD-GEFI−/− platelets is markedly delayed , suggesting that CD-GEFI mediates the rapid but reversible activation of Rap1, which was previously identified as a Ca2+-dependent mechanism . Third, CD-GEFI-mediated integrin activation is independent of signaling by PKC and P2Y12/Gαi . In previous studies, signaling by Ca2+ and PKC/P2Y12 were identified as independent yet synergistic pathways regulating the activation of αIIbβ3 [14,15] and Rap1 [16,17].
In addition to the defect in integrin activation, we observed impaired granule release [8,12] and TxA2 generation (unpublished observation) in CD-GEFI−/− platelets. A central enzymatic step in the formation of TxA2 is the phosphorylation of cytosolic phospholipase A2 (cPLA2) by (ERK) MAP kinases . In platelets, both Ca2+ and PKC signaling have been implicated in ERK1/2 activation . Our data demonstrate a strong correlation between reduced TxA2 generation and impaired activation of ERK1/2 and Rap1 in CD-GEFI−/− platelets. Inhibition of P2Y12 signaling blocked the residual ERK activation and TxA2 production in the mutant platelets. These data suggest that activated Rap1, regulated by Ca2+/CD-GEFI and ADP/Gαi signaling, controls both integrin activation and ERK-mediated TxA2 generation. Our studies further indicate that CD-GEFI supports granule release through its role in TxA2 generation, as we were able to restore PKC activation and dense granule release in knockout platelets by addition of thromboxane A2 analog, U46619.
CD-GEFI as a target for antiplatelet therapy
Inhibition of P2Y12 by inhibitors like Clopidogrel (PLAVIX) is currently one of the most powerful antithrombotic strategies used in the clinic . Based on our studies, we propose that intervention with CD-GEFI signaling could prove to be an equally powerful antithrombotic approach. Targeting CD-GEFI should be safe, as its expression is limited to few cell types [7,8]. Furthermore, CD-GEFI inhibition should markedly reduce thrombotic complications because of its combined effect on several aspects of platelet function. Signaling by PKC/P2Y12/Gαi, however, should allow for some hemostatic function in platelets treated with CD-GEFI inhibitors, thus reducing the risk of bleeding. The absence of spontaneous bleeding in CD-GEFI−/− mice may confirm this hypothesis. It is important to note, however, that bleeding complications have been reported for dogs with loss-of-function mutations in CD-GEFI . In humans, an intronic mutation in CD-GEFI has been suggested as a molecular defect underlying an impaired hemostatic and inflammatory response observed in patients suffering from leukocyte adhesion deficiency (LAD) type-III (also called LAD-1/variant) . However, very recent studies in LAD-III patients indicate that mutations in Kindlin-3 (FERMT3), not CD-GEFI, account for the observed phenotype .
STIM1 and Orai1 are key regulators of store-operated Ca2+ entry in platelets
A series of recent studies using novel transgenic mice markedly advanced our understanding of the molecular machinery regulating SOCE in platelets. Grosse et al.  were the first to demonstrate that stromal interaction molecule 1 (STIM1) plays an important role in platelet calcium signaling. STIM1 is a type I transmembrane protein, which contains a Ca2+ binding EF hand motif in its ER region. In resting cells, this EF hand domain is occupied by Ca2+. Upon cellular activation, Ca2+ is released from the ER into the cytoplasm and STIM1 is no longer occupied by Ca2+. As a result, STIM1 translocates within the cell to communicate with calcium channels expressed in the plasma membrane . Mice expressing an activating EF hand mutant of STIM1 were characterized by macrothrombocytopenia, increased bleeding, and preactivation of circulating platelets caused by increased basal Ca2+ levels . In contrast, genetic deletion of STIM1 in mice resulted in decreased Ca2+ flux in platelets and markedly impaired platelet activation and adhesion under static and flow conditions in vitro. STIM1−/− mice were protected from arterial thrombosis and ischemic brain infarction, while they showed only moderately increased tail bleeding times . The four-transmembrane spanning CRAC channel moiety Orai1 (CRACM1) was identified as the SOC channel mediating STIM1-induced Ca2+ entry in activated platelets. Platelets from mice lacking Orai1  or from mice expressing a loss-of-function mutation of Orai1 (Orai1R93W)  showed strongly impaired SOCE, while agonist-induced Ca2+ release from internal stores was normal. In both studies, impaired SOCE resulted in defective integrin activation and granule release, with the defects being most prominent in platelets activated via the collagen receptor, GPVI. Consistent with the defect in GPVI-mediated SOCE, Orai1−/− mice were protected from collagen-driven arterial and systemic thrombosis. In our own studies with Orai1R93W mice, we observed normal aggregation and adhesion to collagen under conditions of venous shear stress, suggesting that Orai1-mediated SOCE is particularly important for platelet adhesion at arterial shear rates. At static and low-shear conditions, impaired phosphatidylserine (PS) exposure was the most prominent defect observed in Orai1R93W platelets. It has long been recognized that sustained increases in [Ca2+]i are critical for the ability of platelets to switch from a pro-adhesive to a pro-coagulant, PS-positive state [29,30]. In the absence of Orai1, platelets fail to maintain elevated levels of [Ca2+]i necessary for PS exposure.
These studies demonstrate the critical role of STIM1 and Orai1 in platelet SOCE. Semi-quantitative detection of RNA and protein levels identified low-level expression for STIM2 and Orai2/3, paralogs of Orai1 and STIM1, in platelets. At present, it remains unclear if and how Orai2, Orai3, and STIM2 contribute to platelet SOCE. Members of the canonical transient receptor potential (TRPC) family, especially TRPC1, have also been implicated in the regulation of platelet SOCE. However, while studies in human platelets identified a critical role of TRPC1 in SOCE [31,32], no detectable defect in Ca2+ entry was observed in mouse platelets lacking TRCP1 .
STIM1/Orai1 as targets for antiplatelet therapy
The studies outlined above suggest both STIM1 and Orai1 as potential new antithrombotic targets. While deficiency in STIM1 or Orai1 led to marked protection in mouse models of systemic, arterial, and ischemia-induced thrombosis, only a minor increase in tail bleeding times was observed in the mutant mice. It is to be expected that inhibition of Orai1 (and probably STIM1) will not have a major effect on hemostasis in humans, as no obvious bleeding or clotting disorder was observed in patients expressing Orai1R91W, a naturally occurring mutation equivalent to the R93W mutation introduced in Orai1R93W mice. However, these patients suffered from severe combined immunodeficiency (SCID) , a genetic disorder in which both ‘arms’ (B cells and T cells) of the adaptive immune system are severely malfunctional. Mouse studies confirmed the central role of STIM1 and Orai1 in lymphocyte [35,36] as well as in mast-cell [37,38] function. Furthermore, STIM1/Orai1-mediated SOCE is important for endothelial-cell proliferation . Thus, targeting STIM1 or Orai1 in antiplatelet intervention may be complicated because of unwanted side effects.
We would like to thank A. Graybiel, J. Crittenden, and S. Feske for helpful discussions. Many of the described studies were originally performed in the laboratory of D. Wagner; thanks to Denisa for ongoing support. We would like to apologize to all of our colleagues, whose important work we were not able to cite because of the limited word count of this review.
Disclosure of Conflict of Interests
The authors state that they have no conflict of interest.