Cardiovascular and Hemostatic Disorders: SOCE and Ca²⁺ Handling in Platelet Dysfunction

Abstract

Among the Ca²⁺ entry mechanisms in platelets, store-operated Ca²⁺ entry (SOCE) plays a prominent role as it is necessary to achieve full activation of platelet functions and replenish intracellular Ca²⁺ stores. In platelets, as in other non-excitable cells, SOCE has been reported to involve the activation of plasma membrane channels by the ER Ca²⁺ sensor STIM1. Despite electrophysiological studies are not possible in human platelets, indirect analyses have revealed that the Ca²⁺-permeable channels involve Orai1 and, most likely, TRPC1 subunits. A relevant role for the latter has not been found in mouse platelets. There is a body of evidence revealing a number of abnormalities in SOCE or in its molecular regulators that result in qualitative platelet disorders and, as a consequence, altered platelet responsiveness upon stimulation with multiple physiological agonists. Platelet SOCE abnormalities include STIM1 and Orai1 mutations. This chapter summarizes the current knowledge in this field, as well as the disorders associated to platelet SOCE dysfunction.

1 Introduction

Platelets are anucleated cell fragments, of approximately 0.5–3.0 μm diameter, discoid, and irregularly shaped that play an essential role in hemostasis. Platelets, also called as thrombocytes, form clots, along with other plasma factors that prevent hemorrhage. Platelets were identified by Bizzonero in 1882 (Bizzonero 1882), and the same year thrombocytopenia, a disorder characterized by a relative low platelet concentration, was reported as a cause of impaired hemostasis (Hayem 1882).

Platelets derive from larger bone marrow cells called megakaryocytes. Intravital two-photon microscopy of the bone marrow sinusoids has confirmed the generation of protrusions of megakaryocytes, known as proplatelets, that extend into the bone marrow sinusoids where cytoplasmic fragments, called preplatelets, are release and further mature into platelets within the circulation (Junt et al. 2007). More recently, an alternative mechanism for platelet generation has been described by Nishimura and coworkers. This mechanism, which might be induced by interleukin-1α and occurs independently of the megakaryopoiesis modulator, thrombopoietin, may restore circulating platelets during situations of increased platelet consumption (Nishimura et al. 2015). Despite platelets lack nucleus, they contain most intracellular organelles and subcellular structures. Platelets possess an internal membrane structure, the open canalicular system, that is continuous with, and part of, the plasma membrane and might facilitate the interaction of the plasma membrane with internal organelles. In addition, platelets exhibit a dynamic cytoskeleton, consisting of tubulin microtubules and actin filaments, which support the discoid shape of resting platelets and provide the contractile mechanism necessary for platelet function: shape changes, pseudopod formation, internal contraction, and secretion of different regulatory factors (Rosado and Sage 2000a; Fox 2001). Platelets possess a membranous system known as the dense tubular system (DTS), the analogue of the endoplasmic reticulum (ER) in other cells, which represents the major intracellular Ca²⁺ store, a reduced number of mitochondria, lysosomes, and Golgi apparatus (Ebbeling et al. 1992; Lu et al. 2013). Ultrastructural studies have revealed that platelets possess two types of secretory granules: the δ-granules and the α-granules. δ-granules store serotonin, Ca²⁺, ATP, ADP, and pyrophosphate (Dell’Angelica et al. 2000). α-granules contain platelet factor IV, von Willebrand factor, thrombospondin, and fibrinogen. In addition, α-granules might store immunoglobulins taken from the extracellular medium by endocytosis (Harrison and Cramer 1993). Secretion of δ- and α-granules is a critical event in hemostasia. Platelet granule exocytosis has been found to be modulated by the cytosolic Ca²⁺ concentration in different ways: (1) modulation of Ca²⁺-dependent PKC isoforms (Yoshioka et al. 2001), (2) activation of scinderin and other Ca²⁺-dependent actin filament regulatory proteins (Rodriguez Del Castillo et al. 1992; Marcu et al. 1996), (3) induction of dense core granule secretion by activation of Munc13-4 (Shirakawa et al. 2004; Ren et al. 2010), and (4) activation of dense granule exocytosis by Rap1GAP2/synaptotagmin-like protein 1 and Rab8/synaptotagmin-like protein 4 (Neumuller et al. 2009; Hampson et al. 2013). We have recently reported that thrombin-stimulated granule secretion depends on Ca²⁺ mobilization from intracellular stores and only δ-granule secretion is partially dependent on Ca²⁺ influx from the extracellular medium through TRPC6 channels. Analyses of the kinetics of δ- and α-granule exocytosis have revealed that platelet stimulation with thrombin induces rapid release of α-granules which precedes δ-granules secretion (Lopez et al. 2015).

Cytosolic Ca²⁺ is a major signaling element in platelets. Changes in cytosolic Ca²⁺ concentration ([Ca²⁺]_i) modulate a variety of functions in most cellular models, including platelets, such as granule secretion and aggregation (Rosado and Sage 2000c). In platelets, agonists increase [Ca²⁺]_i by stimulation of Ca²⁺ release from the intracellular stores, facilitation of Ca²⁺ entry from the extracellular stores, or both. The major intracellular Ca²⁺ store in platelets is the DTS. Ca²⁺ has been reported to be stored in the DTS by the sarco-/endoplasmic reticulum Ca²⁺-ATPase (SERCA) isoform 2b, which is highly sensitive to thapsigargin but insensitive to 2,5-di-(tert-butyl)-1,4-benzohydroquinone (TBHQ) (Papp et al. 1992; Cavallini et al. 1995). In addition, a second agonist-releasable Ca²⁺ store has been described in human platelets on the base of its sensitivity to TBHQ and high concentrations of thapsigargin (Cavallini et al. 1995). This Ca²⁺ compartment, located in lysosomal-like acidic organelles (López et al. 2005), has been reported to be refilled by SERCA3 isoforms (Bobe et al. 1994). Agonists, such as ADP, release Ca²⁺ selectively from the DTS (Lopez et al. 2006a), while strong agonists like thrombin induce Ca²⁺ release from the DTS and the acidic stores through activation of protease-activated receptor (PAR)-1 and PAR-4. In addition, thrombin is able to induce selective Ca²⁺ release from the acidic stores through the activation of the glycoprotein GPIb-IX-V (Jardin et al. 2007a). The acidic store plays a relevant role in thrombin-induced platelet aggregation (Ben Amor et al. 2009).

Physiological agonists are able to activate both platelet aggregation and granule secretion by releasing the finite amount of Ca²⁺ accumulated in the intracellular stores; nevertheless, Ca²⁺ entry through plasma membrane channels is necessary to achieve full activation of platelet functions (Jardin et al. 2007b; Braun et al. 2009; Lopez et al. 2015). Different Ca²⁺ influx pathways have been described in platelets, including receptor-operated Ca²⁺ entry (ROCE) through P2_X1 receptors (MacKenzie et al. 1996; Vial et al. 2003), second messenger-operated Ca²⁺ entry (SMOCE) activated by PKC and Src tyrosine kinases (Rosado and Sage 2000d; Harper and Sage 2010), and store-operated Ca²⁺ entry (SOCE), a process controlled by the filling state of the intracellular Ca²⁺ stores (Putney 1986; Desai et al. 2015), that is a major mechanism for Ca²⁺ entry in these cells.

2 SOCE in Platelets

SOCE was identified three decades ago by James Putney (1986) as a mechanism for Ca²⁺ entry activated by discharge of the intracellular Ca²⁺ stores. The first demonstration of SOCE in platelets was reported by Sage and coworkers in 1989 (Sage et al. 1989). In this study, quenching of fura-2 fluorescence at the excitation wavelength of 360 nm (the isosbestic point) by Mn²⁺ was used as a marker for Ca²⁺ entry in human platelets, avoiding interference of the changes in [Ca²⁺]_i with the signal. Using this maneuver, the authors reported robust Mn²⁺ influx upon stimulation with thrombin after discharge of the internal Ca²⁺ stores. After this first description, the identification of the molecular components of SOCE in platelets has occurred in parallel with the investigations in other cell types, as SOCE in platelets share a number of features with other non-excitable cells.

SOCE is a mechanism by which discharge of receptor-releasable intracellular Ca²⁺ pools leads to the activation of a Ca²⁺ entry through plasma membrane channels. Ca²⁺ entering the cytosol then refills the intracellular Ca²⁺ compartments leading to the termination of the Ca²⁺ influx. Therefore, the simplest model to account for the activation of SOCE should include at least two components: a Ca²⁺ sensor in the lumen of the intracellular stores and a Ca²⁺-permeable channel in the plasma membrane. The identification of these elements focused extensive investigation on this field until the characterization in 2005 of STIM1 (STromal Interaction Molecule 1) as the ER Ca²⁺ sensor (Roos et al. 2005; Zhang et al. 2005) and the identification of Orai1 as the pore-forming subunit of the CRAC (Ca²⁺ release-activated Ca²⁺) channel in 2006 (Feske et al. 2006; Mercer et al. 2006; Peinelt et al. 2006; Prakriya et al. 2006; Soboloff et al. 2006).

As in other cells, in human platelets, the Ca²⁺ sensor of the intracellular stores was identified as the protein STIM1 (Lopez et al. 2006b). As mentioned in previous chapters, STIM1 is a Ca²⁺-binding protein with a putative single transmembrane domain that might be located both in intracellular membranes, including the ER (Luik et al. 2006; Wu et al. 2006) and acidic Ca²⁺ stores (Zbidi et al. 2011), and in the plasma membrane (Manji et al. 2000; Lopez et al. 2006b; Spassova et al. 2006; Mignen et al. 2007). STIM1 contains different functional domains (the reader is referred to previous chapters of the book for a detailed description of STIM1 structure). Briefly, the STIM N-terminal region is located in the luminal compartment of the Ca²⁺ stores (or the extracellular medium if located in the PM) and contains the canonical, Ca²⁺-binding, EF-hand motif (Zhang et al. 2005), as well as a “hidden,” EF-hand domain, which does not bind Ca²⁺, and the sterile-α-motif (SAM) domain involved in protein-protein interaction. The STIM1 cytosolic region comprises three conserved coiled-coil domains (CC), named CC1, CC2, and CC3; the CRAC modulatory domain (Derler et al. 2009), which includes the STIM1 homomerization domain (Yu et al. 2011); the C-terminal inhibitory domain (that regulates the interaction of STIM1 with the regulatory protein SARAF (Jha et al. 2013)); a serine-/proline-rich region; and a lysine-rich region at the end of the C-terminus, which binds to membrane phospholipids and interact with target channels (Huang et al. 2006; Jardin et al. 2013) and might be involved in enlargement of ER-plasma membrane clusters (Sauc et al. 2015). We have recently found that upon store depletion, SARAF transiently dissociates from STIM1 and interacts with the C-terminal region of Orai1, a mechanism that might be involved in the activation of SOCE (Albarran et al. 2016a). SARAF acts as a modulator of STIM1-dependent channel proteins, including the store-operated Orai1 and TRPC1 subunits (Albarran et al. 2016a, b) and the arachidonate-regulated channels (ARC; Albarran et al. 2016c), which are regulated by plasma membrane resident SARAF, whose expression in the plasma membrane is controlled by STIM1 (Albarran et al. 2016d). The CC2 and CC3 domains encompass the STIM1-Orai1-activating region (SOAR, amino acid residues 344–442) (Yuan et al. 2009), also known as OASF (Orai-activating small fragment; amino acids 233–450/474) (Muik et al. 2009), CAD (CRAC-activating domain; amino acids 342–448) (Park et al. 2009), and CCb9 (amino acids 339–444) (Kawasaki et al. 2009).

The characterization of STIM1 as the Ca²⁺ sensor in the intracellular stores allowed the identification of the Ca²⁺-permeable channels in the plasma membrane. Electrophysiological analysis of store-mediated currents revealed the presence of two different currents: a highly Ca²⁺-selective current (I _CRAC) and a heterogeneous set of nonselective currents called store-operated currents (I _SOC). I _CRAC is a non-voltage-operated current that shows a large current amplitude at negative potentials and approaches the zero current level at very positive potentials (Hoth and Penner 1992). The CRAC channels show a single channel conductance <1 pS and are highly selective for Ca²⁺ over monovalent cations, although the CRAC channels might lose their selectivity in divalent-free solutions, a situation that allows Na⁺ to permeate the channels (Bakowski and Parekh 2002; Prakriya and Lewis 2002). On the other hand, the I _SOC currents are most likely mediated by a family of cation-permeable channels (SOC channels) that exhibit a greater conductance than CRAC channels. I _SOC currents have been described in a number of cell types, including vascular endothelial cells (Vaca and Kunze 1994; Fasolato and Nilius 1998), human A431 carcinoma cells (Luckhoff and Clapham 1994; Kiselyov et al. 1999), smooth muscle cells (Golovina et al. 2001; Trepakova et al. 2001), or liver cells (To et al. 2010), among others. Store-operated currents have not been recorded in platelets due to the limitation of these cells to perform electrophysiological studies; however, I _CRAC has been recorded in its precursor, megakaryocytes (Tolhurst et al. 2008).

The protein Orai1 has been proposed as the pore-forming subunit of the CRAC channel mediating I _CRAC (Feske et al. 2006; Huang et al. 2006; Luik et al. 2006), and Orai1, together with the mammalian homolog of the Drosophila transient receptor potential (TRP) channels, TRPC1, has been reported to form SOC channels (Desai et al. 2015; Berna-Erro et al. 2016; Ambudkar et al. 2017).

As mentioned in previous chapters of this book, the role of Orai1 in I _CRAC was identified by gene mapping in patients with the severe combined immune deficiency (SCID) syndrome, attributed to loss of I _CRAC, and whole-genome screen of Drosophila S2 cells (Feske et al. 2006; Vig et al. 2006; Yeromin et al. 2006). The Orai family consists of three conserved homologs: Orai1, Orai2, and Orai3 (Feske et al. 2005; Mercer et al. 2006; Zhang et al. 2006; Gwack et al. 2007; Rothberg et al. 2013). Orai1 is a protein with four putative transmembrane domains, with the N- and C-terminal domains facing the cytosol and exposing two loops (1 and 3) to the extracellular medium and loop (2) located intracellularly. Both N- and C-termini are required for the interaction with and regulation by STIM1 (Park et al. 2009; Yuan et al. 2009; Derler et al. 2013; Palty and Isacoff 2015; Palty et al. 2015). Crystallization of Drosophila Orai1 revealed that the CRAC channel is a hexameric complex permeable to Ca²⁺ as well as to monovalent ions in the presence of divalent cations (Hou et al. 2012). The architecture of Orai1-/TRPC1-forming SOC channels still remains unresolved. In addition to the store-operated, STIM1-activated CRAC and SOC channels, Orai1 participates in the formation of heteropentameric (3 Orai1 and 2 Orai3) Ca²⁺ channels regulated by arachidonate, a store-independent channel also regulated by STIM1 located in the plasma membrane (Thompson and Shuttleworth 2013; Zhang et al. 2014). Orai1 and STIM1 expression is mediated by the nuclear factor kappa B (NF-κB), which, in turn, is upregulated in megakaryocytes by the serum- and glucocorticoid-inducible kinase 1 (SGK1), an important process for Orai1 expression and SOCE in platelets (Lang et al. 2013). NF-κB activity is downregulated by 1,25(OH)₂ vitamin D3, whose formation is inhibited by klotho; thus klotho-deficient mice exhibit a reduced STIM1/Orai1 expression at the transcript and protein level, as well as attenuated platelet Ca²⁺ signaling and activation (Borst et al. 2014).

The canonical TRP (TRPC) channels were earlier identified in mammalian cells as cation-permeable channels that mediate receptor-stimulated Ca²⁺ influx (Wes et al. 1995; Zhu et al. 1995). The discovery of TRPC1 led to the identification of other six members of the TRPC channels subfamily as well as the remaining members of the TRP superfamily (28 in total) (Venkatachalam and Montell 2007). TRP channels show six transmembrane domains (S1–S6), with the N- and C-termini located in the cytosol and a reentrant loop between S5 and S6. TRPC1 forms tetrameric, polymodal, nonselective cation-permeable channels with significant Ca²⁺ permeability. Crystallographic analysis of TRPC1 is not available, but recent studies have revealed the structure of TRPV1 and TRPA1 using electron cryo-microscopy (Cao et al. 2013; Liao et al. 2013; Paulsen et al. 2015) reporting several features that might be applicable to other TRP channels. These analyses have revealed that the TRP channels show a tetrameric architecture that exhibits a fourfold symmetry around a central ion pore formed by transmembrane domains S5 and S6 and the pore loop. TRP channels contain a 25 amino acid motif, called TRP domain, encompassing the TRP box (EWKFAR) just C-terminal to S6 (Montell et al. 2002) that interacts with both the S4–S5 linker and also with an N-terminus helix located prior to S1 (Liao et al. 2013).

The first description of Orai1 in platelets was reported in 2008, where Orai1 was found to be involved a Ca²⁺ signalplex activated by Ca²⁺ store depletion that includes TRPC1, TRPC6, the type II IP₃ receptor, and SERCA3 in human platelets (Redondo et al. 2008). The expression of Orai1 in human platelets and murine megakaryocytes was confirmed by quantitative RT-PCR analysis (Tolhurst et al. 2008). Orai1 expression has also been reported in mouse platelets, and Orai1-deficient mice were found to impair platelet SOCE and thrombus formation, thus revealing the functional role of Orai1 in SOCE and platelet aggregation in mouse platelets (Braun et al. 2009). The role of Orai1 in mouse platelet function was confirmed in mice expressing an inactive Orai1 form (Orai1-R93W), which displayed a markedly attenuated SOCE, reduced integrin activation, and impaired degranulation upon activation with low agonist concentrations (Bergmeier et al. 2009). On the other hand, the first description of TRPC1 at the protein level in platelets was reported in 2000 (Rosado and Sage 2000b) where we found that TRPC1 interacts with the type II IP₃ receptor upon depletion of the intracellularCa²⁺ stores. A previous study reported the presence of TRPC1 transcripts in the megakaryocytic cell lines MEG01, DAMI, and HEL, although the analysis was unable to detect TRPC1 mRNA in human platelets, probably as a result of rapid processing of these transcripts during platelet generation (Berg et al. 1997).

STIM1 has been found to interact with endogenously expressed TRPC1 (Lopez et al. 2006b) and Orai1 proteins (Jardin et al. 2008) in human platelets. Studies in different cell types have revealed that STIM1 undergoes a conformational change upon Ca²⁺ store depletion as a result of Ca²⁺ displacement from the STIM1 EF-hand domain. It has been hypothesized that at rest, STIM1 SOAR region is hidden from the plasma membrane channels (Fahrner et al. 2013). Discharge of the intracellular stores leads to Ca²⁺ dissociation from the EF-hand, which, in turn, changes the EF-hand-SAM conformation exposing hydrophobic domains and facilitating the formation of STIM1 dimers/oligomers (Stathopulos et al. 2006, 2009). The conformational change that occurred in the N-terminal region is transmitted to the C-terminus through the transmembrane domain, thus bringing the C-termini together (Ma et al. 2015). The CC1 region, which is clamping the remaining C-terminal portion in a tight state, releases it allowing the interaction of the SOAR region with Orai1 C- and N-termini (Derler et al. 2013; Stathopulos et al. 2013; Fahrner et al. 2014). The interaction between STIM1 and the Orai1 C-terminus involves four positively charged residues (K382, K284, K385, and K386), two aromatic amino acids (Y361 and Y362), and four hydrophobic residues (L347, L351, L373, and A376) located in the SOAR region and the amino acids L273, L276, R281, L286, and R289 located in Orai1 C-terminus, leading to what has been called the STIM1-Orai1 association pocket (SOAP) (Stathopulos et al. 2013). The interaction between STIM1 and Orai1 N-terminus remains unsolved. The activation of TRPC1 by STIM1 has been reported to require both the SOAR region, which is important for the STIM1-TRPC1 interaction (Lee et al. 2014), and also the electrostatic interaction between two negatively charged aspartates in TRPC1 (D639 and D640) with the positively charged amino acids K684 and K685 located in STIM1 polybasic domain (Zeng et al. 2008).

A recent study has reported that Orai1, TRPC1, and STIM1 might reproduce I _SOC, thus demonstrating that SOC channels involved the participation of these proteins (Desai et al. 2015). We have presented evidence for the existence of functional interactions between Orai1 and TRPCs under the influence of STIM1 and propose that SOC channels in human platelets are composed of heteromeric complexes that include TRPCs and Orai proteins (Jardin et al. 2008, 2009). These findings suggest that the participation of Orai1 in SOCE in human platelets might occur either by forming self-contained ion channels highly selective for Ca²⁺ and activated by STIM1, consistent with the CRAC channel hypothesis as suggested for murine megakaryocytes (Tolhurst et al. 2008), or a model in which SOC channels are formed by a combination of TRPCs and Orai proteins.

The role of the STIM1 homolog, STIM2, in SOCE in platelets is still unclear. Gilio et al. (2010) have reported that while STIM1- and Orai1-deficient mice exhibit impaired glycoprotein VI-dependent Ca²⁺ signals, PS exposure, and thrombus formation, STIM2-deficient mice show normal responses. In human platelets, STIM2 and STIM1 have been found to be expressed in acidic Ca²⁺ stores, including lysosome-related organelles and dense granules. Upon discharge of the acidic stores, using bafilomycin A1, STIM2 interacts with STIM1, SERCA, and Orai1 channels in the plasma membrane (Zbidi et al. 2011). Furthermore, treatment with thapsigargin leads to the formation of a signaling complex including STIM1, STIM2, Orai1, Orai2, and TRPC1 (Berna-Erro et al. 2012), thus suggesting a possible role for STIM2 in SOCE in human platelets.

3 Molecular Pathophysiology of SOCE Mechanisms in Platelets

Changes in cytosolic Ca²⁺ concentration play a key role in the regulation of platelet responsiveness, and several disorders have been found to be mediated by abnormalities in Ca²⁺ signaling, especially in SOCE. As described above, Ca²⁺ entry through plasma membrane channels is necessary to achieve full activation of platelet functions (Jardin et al. 2007b; Braun et al. 2009), and SOCE is a major mechanism for Ca²⁺ influx in these cells (Adam et al. 2016). Consequently, abnormalities in the key elements of SOCE or its molecular regulators are expected to produce platelet dysfunctions that might underlie hemostatic disorders. Therefore, identification of the molecular basis of altered cytosolic Ca²⁺ homeostasis will shed new light on the pathophysiology of cardiovascular disorders.

3.1 Functional Abnormalities Concerning STIM1, Orai1, and TRPC Proteins

Grosse et al. (2007) reported the first piece of evidence suggesting that STIM1 plays an important role in platelet function and mutations of human STIM1 and other proteins implicated in SOCE can be involved in inherited thrombocytopenias. Later, it was demonstrated that platelets isolated from mice lacking expression of Stim1 gene showed both an impaired agonist-induced Ca²⁺ release from intracellular stores and SOCE activation that results in an impaired platelet function and reduced thrombus formation (Varga-Szabo et al. 2008b; Ahmad et al. 2011). Similar results were observed in Orai1-deficient mice, where platelets show attenuated SOCE and ROCE, as well as impaired agonist-induced platelet activation. This defect in platelet activation is stronger when platelets are activated via collagen receptor than when platelets are activated via G protein-coupled receptors (Braun et al. 2009; Bergmeier et al. 2013). In both STIM1- and Orai1-deficient mice platelets, SOCE inhibition causes an altered integrin activation, dense granule release, and phosphatidylserine (PS) exposure (Varga-Szabo et al. 2008b; Braun et al. 2009). However, TRPC1-deficient mice platelets showed a normal SOCE and platelet function compared to wild-type mice platelets, suggesting that Orai1 is the major SOC channels in mice platelets (Varga-Szabo et al. 2008a; Braun et al. 2009).

The functional role of STIM1 and Orai1 in the control of agonist-stimulated platelet aggregation has also been demonstrated in human platelets using a different approach. Functional knockdown of STIM1 or Orai1 by electrotransjection of cells with STIM1 (25–139) antibody or with anti-Orai1 (288–301) antibody, which impaired STIM1-Orai1complexes formation upon platelet stimulation, decreased thrombin-induced platelet aggregation and prevented ADP-stimulated aggregation as compared to cells treated with IgG of the same origin of antibodies used and containing the same concentration of preservatives. Similar results were obtained by external application of the anti-STIM1 (25–139) antibody or anti-TRPC1 antibody that bocks STIM1-TRPC1 interaction without affecting STIM1-Orai1 complexes formation, suggesting a functional role of TRPC1 in human platelet aggregation (Galan et al. 2009), in contrast to published in TRPC1-deficient mice (Varga-Szabo et al. 2008a). These discrepancies might be attributed to differences in the activation of SOCE in distinct species.

In platelets from diabetes mellitus type 2 patients, the loss of the association of STIM1 with Orai1, TRPC1, and TRPC6 results in a reduced SOCE that might be involved in the pathogenesis of the altered platelet responsiveness observed in diabetic patients (Jardin et al. 2011). As mentioned above, TRPC6 has been shown to be involved in both SOCE and non-SOCE pathways by association with different Ca²⁺-handling proteins. Hence, depletion of intracellular Ca²⁺ stores promotes the association of TRPC6 with Orai1 and STIM1, while the non-SOCE activator OAG displaces TRPC6 from Orai1 and STIM1 and induces its association with TRPC3 (Jardin et al. 2009). Different studies suggested a functional role for TRPC6 in platelets modulating both Ca²⁺ entry and aggregation through its interaction with calmodulin and IP₃Rs (Dionisio et al. 2011) or with Orai1-STIM1-forming complexes (Jardin et al. 2011). However, the role of TRPC6 in platelet function is still unclear, and controversial studies were published using TRPC6-deficient mice (Paez Espinosa et al. 2012; Ramanathan et al. 2012; Harper et al. 2013; Albarran et al. 2014). Paez Espinosa et al. (2012) proposed that TRPC6 plays an essential role in platelet function since TRPC6-deficient mice showed an increased bleeding time and reduced thrombus formation. In contrast, two studies have showed a decrease in [Ca²⁺]_i in platelets lacking TRPC6 at resting conditions, without playing a role in the regulation of SOCE and platelet aggregation (Ramanathan et al. 2012; Albarran et al. 2014). Berna-Erro and coworkers proposed that TRPC6 regulates [Ca²⁺]_i controlling the passive Ca²⁺ leak rate from agonist-sensitive intracellular Ca²⁺ stores in resting platelets (Albarran et al. 2014).

Since the role of STIM1 and Orai1 in the mechanism of SOCE was identified, several loss-of-function mutations in both proteins have been related to different human diseases, collectively termed CRAC channelopathies, characterized by severe combined immunodeficiency (SCID)-like disease, autoimmunity, tubular aggregate myopathy, dental enamel maturation defect, and ectodermal dysplasia. Gain-of-function mutations of Orai1 are related with tubular aggregate myopathy and Stormorken-like syndrome, while gain-of-function mutations of STIM1 are associated with tubular aggregate myopathy, Stormorken syndrome, and York platelet syndrome. However, only Stormorken syndrome and York platelet syndrome are widely characterized by platelet disorders that result in bleeding diathesis and thrombocytopenia (Feske 2010; Lacruz and Feske 2015). The initial studies in genetically modified mice demonstrated that replacement of an acidic residue of aspartate by glycine at position 84 in the EF-hand motif of STIM1 induces a gain-of-function mutation of STIM1 that reduces its ability to bind Ca²⁺, mimicking store depletion and therefore inducing a constitutive activation of STIM1 and CRAC channels that promote an activated procoagulant state of platelet due to an increased [Ca²⁺]_i in resting conditions. In vivo and in vitro functional assays demonstrated that these mutant platelets are unresponsiveness to the immunoreceptor tyrosine-based activation motif (ITAM)-containing collagen receptor complex GPVI-FcR γ-chain that causes platelet dysfunction, thrombocytopenia, and fatal bleeding disorders in mice (Grosse et al. 2007). Recently, it has been described that STIM1 (D84G) mutation also induces a constitutive activation of STIM1 and CRAC channels in human myoblast, causing tubular aggregate myopathy, but platelet function was not investigated in these patients (Bohm et al. 2013). Similar altered bleeding phenotype described above in EF-hand mutant mice has also been observed in human patients with Stormorken syndrome. This rare autosomal-dominant genetic disease was first reported in 1985, which, beside thrombocytopenia and bleeding diathesis, is characterized by congenital miosis, asplenia, headache, ichthyosis, tubular aggregate myopathy, and proximal muscle weakness (Stormorken et al. 1985, 1995; Stormorken 2002). Misceo and coworkers demonstrated that platelets isolated from Stormorken syndrome patients show an increased [Ca²⁺]_i in resting conditions and a markedly reduced both TRAP- and thapsigargin-induced SOCE compared with platelets isolated from healthy donors. As consequence of these alterations in Ca²⁺ handling, platelets from Stormorken syndrome patients are much less responsive to stimulation than those from healthy donors, and they are in a constitutively activated procoagulant state, characterized by phosphatidylserine (PS) exposure and other platelet activation markers at the plasma membrane, that promotes a decrease in lifespan of circulating platelets and a low blood platelet count, causing the mild bleeding disorder (Misceo et al. 2014), as previously Stormorken and coworkers described in 1995 (Stormorken et al. 1995). Gain-of-function mutation STIM1 (R304W) has been shown to cause the clinical phenotypes related with Stormorken syndrome (Misceo et al. 2014; Morin et al. 2014; Nesin et al. 2014). Based on structural analyses, the residue R304 is localized close to STIM1-Orai1-activating region (SOAR) in C-terminal region of STIM1. The lateral chain of the residue R304 formed a hydrogen bond with the E318 and Q314 residues, located in the first coiled-coil domain, leading the occlusion of SOAR under resting conditions (Korzeniowski et al. 2010; Cui et al. 2013; Morin et al. 2014; Lacruz and Feske 2015). Dissociation of this inhibitory helix of coiled-coil domain/SOAR interaction is required for the association of SOAR with Orai1 and the subsequent channel activation (Korzeniowski et al. 2010; Feske and Prakriya 2013). Substitution of tryptophan residue to arginine residue in position 304 increased amphipathic properties of the coiled-coil domain, avoiding the hydrogen bond with E318 and Q314 residues and leading to the release of SOAR domain that results in the constitutive activation of STIM1 and of the Orai1 channel (Shen and Demaurex 2012; Yang et al. 2012; Morin et al. 2014). Another gain-of-function mutation of STIM1, STIM1 (I115P), is mainly involved in the named York platelet syndrome (YPS) that is also associated with bleeding tendency caused by platelet abnormalities and thrombocytopenia. As described above in Stormorken syndrome, platelets are in a constitutive activated state and also show structural and functional abnormalities, including giant electron-opaque organelles and massive, multilayered target bodies and deficiency of platelet Ca²⁺ storage in delta granules. The mutation in the residue I115, located in the hidden second EF-hand domain of STIM1, induces a constitutive activation of STIM1 that increases STIM1- and Orai1-forming complexes and promotes an elevated and sustained [Ca²⁺]_i. YPS patients share the STIM1 (R304W) mutation described in Stormorken syndrome patients, as well as certain symptoms such as muscle weakness with skeletal muscle atrophy (White and Gunay-Aygun 2011; White et al. 2013; Markello et al. 2015).

Although mutations in Orai1 have not been widely associated with thrombocytopenia and bleeding disorders, the loss-of-function mutation Orai1 (R93W) in mice platelets and the equivalent Orai1 (R91W) mutation in human patients, responsible for SCID, promote defects in agonist-induced Ca²⁺ influx as well as Ca²⁺-regulated platelet functions, such as integrin activation, dense granule release, and PS exposure. However, no spontaneous bleeding or clotting disorders were observed in both cases (Bergmeier et al. 2009, 2013). The replacement of a conserved residue of arginine by tryptophan at position 91, located in the conducting pore at the cytoplasmic end of transmembrane region 1 of Orai1, avoids CRAC channel activation and Ca²⁺ influx currents through this channel (Feske et al. 2006; Thompson et al. 2009). These studies demonstrated that mutations in STIM1 or Orai1 proteins are not responsible for the clinical phenotype of Scott’s syndrome, a disorder of platelet coagulant activity associated with a markedly impaired PS exposure similar to that described for Orai1 (R91W) and Orai1 (R93W) platelets (Bergmeier et al. 2009). In addition, gain-of-function mutation Orai1 (P245L) has been related with a Stormorken-like syndrome characterized by congenital miosis and tubular aggregate myopathy, but platelet dysfunction and bleeding disorders have not been described (Nesin et al. 2014). The residue P245 is located within the fourth transmembrane helix of Orai1 and plays an essential role in the stabilization of the closed state of the channel. The mutation Orai1 (P245L) stabilizes Orai1- and STIM1-forming complexes and suppresses the slow Ca²⁺-dependent inactivation of the channel (Misceo et al. 2014; Nesin et al. 2014; Palty et al. 2015).

3.2 Homer1

Homer1 is a cytoplasmic adaptor family protein that contains a class II EVH1 [Ena (Enabled)/VASP (vasodilator-stimulated phosphoprotein) homology 1] domain that binds to a proline-rich PXXF motif located in different Ca²⁺-handling protein involved in SOCE, such as TRPC1 (Yuan et al. 2003; Kim et al. 2006), STIM1 (Jardin et al. 2012), and IP₃R (Yuan et al. 2003; Kim et al. 2006). Homer1 also binds with Orai1, although this interaction is mediated by STIM1 (Jardin et al. 2012). In platelets, Homer1 has been shown to facilitate both STIM1-Orai1- and TRPC1-IP₃R-forming complexes upon cell stimulation. Impairment of Homer1 function by electrotransjection of the PPKKFR peptide, which emulates PPXXF Homer1-binding motif, into cellular cytoplasm abolished agonist-stimulated aggregation in platelets as consequence of reduction of the maintenance of STIM1- and Orai1-forming complexes, suggesting that Homer1 regulates SOCE and, in turns, platelet function (Jardin et al. 2012).

4 Conclusions and Perspectives

Summarizing, SOCE is a major mechanism for Ca²⁺ influx in platelets and plays a relevant role in platelet function. In human platelets, two platelet disorders due to SOCE dysfunction have been characterized, the Stormorken and York platelet syndromes, which course with bleeding diathesis and thrombocytopenia. The mechanism underlying both syndromes includes gain-of-function mutations in STIM1. Similarly, gain-of-function mutations in Orai1 result in less characterized Stormorken-like syndromes. The precise analysis of still uncharacterized platelet dysfunctions might provide evidence for the existence of further loss- or gain-of-function mutations. Further studies should also be focused on the involvement of regulatory SOCE proteins, such as SARAF, CRACR2A, or STIMATE, as well as TRPC channels, including TRPC1 and TRPC6, in the development of platelet disorders that might be responsible for still uncharacterized pathological phenotypes.