Store-operated calcium entry in vascular smooth muscle

Abstract

In non-excitable cells, activation of G-protein-coupled phospholipase C (PLC)-linked receptors causes the release of Ca²⁺ from intracellular stores, which is followed by transmembrane Ca²⁺ entry. This Ca²⁺ entry underlies a small and sustained phase of the cellular [Ca²⁺]_i increases and is important for several cellular functions including gene expression, secretion and cell proliferation. This form of transmembrane Ca²⁺ entry is supported by agonist-activated Ca²⁺-permeable ion channels that are activated by store depletion and is referred to as store-operated Ca²⁺ entry (SOCE) and represents a major pathway for agonist-induced Ca²⁺ entry. In excitable cells such as smooth muscle cells, Ca²⁺ entry mechanisms responsible for sustained cellular activation are normally considered to be mediated via either voltage-operated or receptor-operated Ca²⁺ channels. Although SOCE occurs following agonist activation of smooth muscle, this was thought to be more important in replenishing Ca²⁺ stores rather than acting as a source of activator Ca²⁺ for the contractile process. This review summarizes our current knowledge of SOCE as a regulator of vascular smooth muscle tone and discusses its possible role in the cardiovascular function and disease. We propose a possible hypothesis for its activation and suggest that SOCE may represent a novel target for pharmacological therapeutic intervention.

Introduction

Virtually, every cellular response is regulated by changes in intracellular free calcium levels ([Ca²⁺]_i), making this ion a universal intracellular mediator. Thus, understanding the mechanisms that control Ca²⁺ entry into cells becomes critically important. Increases in cytoplasmic Ca²⁺ signals can be generated either by release of Ca²⁺ from intracellular stores and/or by influx of Ca²⁺ from the extracellular fluid. The release of intracellular Ca²⁺ occurs from the endoplasmic reticulum or its specialized counterpart in muscle cells, the sarcoplasmic reticulum (SR), and is generally signaled by the formation of second-messengers, such as inositol 1,4,5-trisphosphate (IP₃) (Streb et al., 1983). However, it soon became apparent that the release of Ca²⁺ from intracellular stores is often followed by a sustained phase of Ca²⁺ entry from the extracellular space (Putney and McKay, 1999a). This led to the proposal by Putney that depleted Ca²⁺ stores (primarily in the endoplasmic reticulum) are able to gate the entry of extracellular Ca²⁺ where intracellular Ca²⁺ stores act as a capacitor, thus, leading to the term ‘capacitative calcium entry' that has been superceded more recently by the ‘store-operated calcium entry' (SOCE) (Putney, 1999b, Parekh and Putney, 2005). This concept was supported by the identification of a well-characterized store-operated current, the so-called Ca²⁺ release-activated Ca²⁺ current (Hoth and Penner, 1992), although the consensus is that the current mechanisms underlying capacitative calcium entry may in fact be a special case of SOCE, and not represent a generalized phenomenon.

In spite of many investigations of SOCE, the molecular details of the activation mechanisms of store-operated Ca²⁺ channels (SOCCs) remain fragmentary. In fact, the existence of different genes encoding SOCCs may account for the diverse activation mechanisms of this channel. There appears to be no general agreement regarding the nature of the Ca²⁺ store from which the signal emits, the identity of the Ca²⁺ sensor that monitors the filling state of the stores, the retrograde signal transduction mechanism that activates SOCCs or the molecular identity of SOCCs. This is probably due to the complexities of mechanisms involved in SOCE, as well as the peculiarity of various experimental methods employed. The effect of SOCE on vascular function may include changes in both the endothelium (Oike et al., 1994; Fasolato and Nilius, 1998; Freichel et al., 2001; Cioffi et al., 2005) and vascular smooth muscle cells (VSMCs). In this review, we will focus our discussion on SOCE in VSMCs. Moreover, we will highlight some of the challenges encountered in creating a unified hypothesis.

SR Ca²⁺-ATPase inhibitors and SOCE

Exploration of the functional significance of SOCE has been greatly aided by the use of agents, such as cyclopiazonic acid (CPA) and thapsigargin, which act as selective inhibitors of the sarcoplasmic/endoplasmic reticulum Ca²⁺-ATPase (SERCA) (Laporte et al., 2004). These drugs cause depletion of SR Ca²⁺ stores by inhibiting sequestration of Ca²⁺ ions without activation of G proteins, and are used to provide an important distinction between Ca²⁺ entering through SOCCs as opposed to receptor-operated Ca²⁺ channels (ROCCs). Undoubtedly, sustained Ca²⁺ influx, or cellular responses, activated by SR Ca²⁺-ATPase inhibitors can be considered markers for the involvement of SOCE in cell signalling.

In cultured VSMCs, depletion of SR Ca²⁺ stores with thapsigargin activates Ca²⁺ influx that is independent of the generation of inositol phosphate and resistant to the L-type voltage-operated Ca²⁺ channel (VOCC) blocker, nicardipine (Xuan et al., 1992). Numerous studies show that the SERCA inhibitors increase not only Ca²⁺ influx but also vascular tone of different blood vessels (Table 1). In all such cases, the contractions are sustained and dependent on the presence of extracellular Ca²⁺. However, the contractile response to SERCA inhibitors shows variable sensitivity to Ca²⁺ channel-blocking drugs. For instance, in the rat aorta, the majority of the contractions is nifedipine-sensitive (Kwan et al., 1994; Low et al., 1994; Tepel et al., 1994; Xuan and Glass, 1996; Noguera et al., 1997; Tosun et al., 1998), whereas in the rat pulmonary, renal and retinal arteries, the contraction is nifedipine-resistant (Gonzalez De La Fuente et al., 1995; Curtis and Scholfield, 2001; Snetkov et al., 2003). In rat femoral and carotid arteries, mouse anococcygeus, guinea-pig and cat fundus, Ca²⁺ entry stimulated by store depletion is partially nifedipine-sensitive (Gibson et al., 1994; Sekiguchi et al., 1996; Petkov and Boev, 1996a, 1996b; Nomura et al., 1997). These results suggest that smooth muscle contraction in response to SERCA inhibitors may be caused by Ca²⁺ entry through both VOCCs and SOCCs, with the relative importance of these entry pathways depending on the smooth muscle type, with SOCCs appearing to be of greater importance in tonic smooth muscles, for example guinea pig and cat gastric fundus, mouse anococcygeus and rat pulmonary artery. The SERCA inhibitors do not contract all smooth muscles, and in some cases, a poor correlation exists between increased intracellular Ca²⁺ and contraction (Snetkov et al., 2003). Huang et al. (2006) reported that Ca²⁺ entry through SOCC is not directly coupled to VSMC contraction in renal arteries (Huang et al., 2006). A role for SOCCs other than smooth muscle contraction was first suggested by Flemming et al. (2003) in rabbit cerebral arteries where the application of CPA induces a sustained increase in [Ca²⁺]_i in the presence of a VOCC blocker (D600), which is not associated with contractions. However, membrane depolarization with a K⁺-rich solution (in the absence of D600) also produces a sustained rise in [Ca²⁺]_i that is associated with smooth muscle contraction (Flemming et al., 2003). Since both CPA and a 35 mM KCl-containing solution raise [Ca²⁺]_i to similar levels, it is possible that CPA activation of SOCCs causes an increase in [Ca²⁺]_i in a cellular compartment that is spatially separated from contractile proteins, indicating that this spatially separate cellular compartment may include internal Ca²⁺ stores that are able to regulate local Ca²⁺ levels. Moreover, CPA activates a sustained, non-selective cation conductance in single myocytes isolated from the mouse anococcygeus. The current–voltage relationship for the CPA-induced current is linear with a reversal potential close to +30 mV in near physiological cation gradients. The reversal potential shifts to a more negative value upon removal of extracellular Ca²⁺, indicating that a large proportion of the current is carried by Ca²⁺ (Wayman et al., 1996a). This notion was supported by simultaneous recordings of current and intracellular Ca²⁺ levels, which showed that activation or inhibition of the current is accompanied by rises and falls in [Ca²⁺]_i correspondingly (Wayman et al., 1996b). However, the cellular mechanisms linking store depletion to the opening of SOCCs are not well understood (Castells and Droogmans, 1981; Putney, 2001; Flemming et al., 2002; Wilson et al., 2002).

Table 1.

(Vascular) smooth muscles in which SERCA pump inhibitors raise [Ca²⁺]_i and/or elicit contraction

Species	Smooth muscle	Remarks	Responses	References
Rat	Aorta	Abolished by nifedipine/nicardipine	Contraction and increased [Ca2+]i	Tepel et al., 1994; Kwan et al., 1994; Low et al., 1994; Noguera et al., 1997; Tosun et al., 1998; Xuan and Glass, 1996
	Carotid artery	Partially reduced by verapamil	Contraction and increased [Ca2+]i	Sekiguchi et al., 1996
	Coronary artery		No contraction and increased [Ca2+]i	Snetkov et al., 2003
	Femoral artery	Partially reduced by verpamil	Contraction and increased [Ca2+]i	Nomura et al., 1997
	Femoral artery		A small transient contraction and increased [Ca2+]i	Snetkov et al., 2003
	Mesenteric artery		No contraction and increased [Ca2+]i	Snetkov et al., 2003
	Pulmonary artery	Unaffected by nifedipine, verapamil reduced by tyrosine kinase inhibitors. pCa2+/tension curve unaffected		Gonzalez De La Fuente et al., 1995
	Pulmonary distal arterial smooth muscle	Abolished by nifedipine	Increased [Ca2+]i	Wang et al., 2004
	Intrapulmonary artery	Unaffected by diltiazem or the reverse mode Na+/Ca2+ antiport inhibitor KB-R7943	Contraction and increased [Ca2+]i	Snetkov et al., 2003
	Renal cortical interlobar arteries		Increased [Ca2+]i	Facemire et al., 2004
	Renal artery		No contraction	Snetkov et al., 2003
	Preglomerular vascular smooth muscle		Increased [Ca2+]i	Fellner and Arendshorst, 1999
	Basilar arteries	Unaffected by verapamil	Contraction	Bergdahl et al., 2005
	Ileum	Unaffected by nifedipine, methoxyverapmail	Increased [Ca2+]i	Ohta et al., 1995
	Spleen	Unaffected by nifedipine reduced by tyrosine kinase inhibitors		Burt et al., 1995
	Urinary bladder	Very weak contractile		Munro and Wendt, 1994
	A7r5 rat smooth muscle cell line	Not applicable	Increased [Ca2+]i	Byron and Taylor, 1995; Iwamuro et al., 1999; Iwasawa et al., 1997
Mouse	Anococcygeus	Partially reduced by nifedipine	Contraction and increased [Ca2+]i	Gibson et al., 1994; Wallace et al., 1999
Guinea-pig	Gastric fundus	Partially reduced by nifedipine		Petkov and Boev, 1996a
Cat	Gastric fundus	Partially reduced by nifedipine		Petkov and Boev, 1996b
Canine	Pulmonary arterial smooth muscle cells	Unaffected by nifedipine reduced by tyrosine kinase inhibitors	Contraction and increased [Ca2+]i	Doi et al., 2000; Wilson et al., 2002
	Renal arterial smooth muscle cells	Unaffected by nisoldipine reduced by tyrosine kinase inhibitors	Contraction and increased [Ca2+]i	Wilson et al., 2002
Rabbit	Pulmonary arterial smooth muscle cells	Unaffected by nifedipine	Increased [Ca2+]i	Kang et al., 2003
	Carotid artery smooth muscle		Increased [Ca2+]i	Kawanabe et al., 2002
	Pial artery		Contraction and increased [Ca2+]i	Flemming et al., 2002, 2003
	Choroidal artery	Abolished by nifedipine	Increased [Ca2+ ]i	Curtis and Scholfield, 2001
Human	Lower oesophageal sphinctor smooth muscle		Increased [Ca2+ ]i	Wang et al., 2003
	Bronchus			Cortijo et al., 1997
Porcine	Airway smooth muscle		Increased [Ca2+ ]i	Ay, 2004
Swine	Renal artery		Increased [Ca2+ ]i	Utz et al., 1999
Prostate cell line			Increased [Ca2+ ]i	Thebault et al., 2005

Abbreviation: SERCA, sarcoplasmic/endoplasmic reticulum Ca²⁺ ATPases.

A model for SOCE

Several mechanisms for activation of SOCE have been proposed (Berridge, 1995; Parekh and Penner, 1997; Gibson et al., 1998; Putney, 2001; Bolotina, 2004). Smani et al. (2004) presented a simple model that may explain how Ca²⁺ influx factor (CIF, produced upon depletion of Ca²⁺ stores) activates SOCCs using mouse aortic smooth muscle cells. After CIF activation of SOCCs, CIF induces displacement of inhibitory calmodulin (CaM) from Ca²⁺-independent phospholipase (iPLA₂), a key event leading to activation of iPLA₂ and generation of lysophospholipids; the latter, in turn, activate SOCCs in a plasma membrane—delimited manner. Upon refilling of the stores and termination of CIF production, CaM rebinds to iPLA₂ to resume its inhibition and so terminating the activity of SOCCs and Ca²⁺ entry (Smani et al., 2004). Studies by Trepakova et al. (2001) identified native CIF in support of this hypothesis. A novel 3-pS Ca²⁺-conducting channel that is activated by 1,2-bis(o-aminophenoxy)ethane-N,N,N′,N′-tetraacetic acid (BAPTA) and thapsigargin causes passive depletion of intracellular Ca²⁺ stores, and this channel is likely to be a native store-operated channel in VSMCs. The 3-pS channels are activated in inside-out membrane patches from smooth muscle cells immediately upon application of CIF extracted from mutant yeast cell lines (Trepakova et al., 2001). The existence of CIF that is produced by depleted stores and the idea that it may trigger activation of SOCCs was proposed more than a decade ago. However, the molecular structure of CIF has yet to be defined even though it is reportedly stable and can be partly purified (Kim et al., 1995). Moreover, functional studies demonstrate that the plasma membrane Na⁺/Ca²⁺ exchanger (NCX) is also involved in the regulation of vascular Ca²⁺ homeostasis by contributing to SOCE (Arnon et al., 2000). Evidence for an important role of NCX in SOCE came from a study showing that NCX is functionally expressed in cultured VSMCs from the human pulmonary artery and that Ca²⁺ entry via the reverse mode of NCX participates in store depletion-mediated elevation in [Ca²⁺]_cyt. Thus, blockade of NCX in its reverse mode may serve as a potential therapeutic approach for the management of pulmonary hypertension (Zhang et al., 2005a). The functional evidence for a positive role of NCX in Ca²⁺ filling is also supported by immunoblotting and immunfluorescence studies where NCX is expressed in cultured arterial myocytes (Juhaszova et al., 1994). Besides, expression of NCX1 (mainly NCX1.3) is detected in coronary artery smooth muscle cells (Slodzinski et al., 1995). Furthermore, nitric oxide (NO), which induces vascular relaxation by accelerating SERCA-dependent refilling of Ca²⁺ stores, would be expected to blunt CIF production and so terminate the activity of SOCCs and Ca²⁺ influx (Cohen et al., 1999).

Members of the canonical transient receptor potential family (TRPC), particularly TRPC1, are involved in SOCE in VSMCs (Golovina et al., 2001; Xu and Beech, 2001; Sweeney et al., 2002a; Bergdahl et al., 2005). Xu et al. (2006) suggested that TRPC5 is another component of SOCC. Studies on non-vascular cells have implicated that additional TRPC family members exist in association with SOCE, including TRPC3 (Liu et al., 2000; Zagranichnaya et al., 2005), TRPC4 (Hofmann et al., 2002; Strubing et al., 2003) and TRPC7 (Zagranichnaya et al., 2005). In addition, TRPC1 may also be linked to TRPP2 (polycystin-2) Ca²⁺ permeable channels (Tsiokas et al., 1999; Giamarchi et al., 2006). More recently, Roos et al. (2005) have demonstrated that a single membrane-spanning protein termed STIM1 (stromal-interacting molecule 1) plays an essential role in the activation of SOCCs. The STIM1 protein serves as a sensor of Ca²⁺ within the stores (Roos et al., 2005). Other studies provide strong evidence showing that Orai1 (the Greek mythological characters the Orai, which are the keepers of the gates of heaven) is a pore subunit of the store-operated Ca²⁺ release-activated Ca²⁺ channels (Feske et al., 2006; Prakriya et al., 2006). Soboloff et al. (2006) revealed a powerful gain in the SOCC function that is dependent on the presence of both STIM1 and Orai1 (Soboloff et al., 2005). STIM1 may interact with TRPC1 and is involved in SOCE (Lopez et al., 2006). Jackson (2006) suggested that given the similarities between Ca²⁺ handling in platelets and VSMCs, it is likely that STIM1 and possibly Orai1 as well take part in SOCE in VSMCs and human airway myocytes (Jackson, 2006; Peel et al., 2006).

Based on these reports, we propose a model for SOCE in smooth muscle excitation/contraction coupling, which takes into account that, in many cases, the responses to SERCA inhibitors and to IP₃-generating receptor agonists, both have VOCC-dependent and VOCC-independent components (Figure 1). In addition, the contractile response to the SERCA inhibitors is more variable than that to receptor agonists. The release of Ca²⁺ from intracellular stores in response to Ins(1,4,5)P₃ has two effects. Firstly, there is a rise in [Ca²⁺]_i—possibly amplified by Ca²⁺-induced Ca²⁺ release from ryanodine-sensitive stores—which activates Ca²⁺-dependent Cl⁻ channels, thus producing membrane depolarization that promotes Ca²⁺ entry via VOCCs. Wayman et al. (1996a) described that Ca²⁺ store depletion activates a biphasic inward current in mouse anoccygeous smooth muscle cells. An initial transient current upon the release of Ca²⁺ from the SR is due to activation of Ca²⁺-dependent Cl⁻ channels (Wayman et al., 1996a). Secondly, depleted Ca²⁺ stores generate CIF, which diffuses to the plasma membrane. A cascade of plasma-membrane-delimited reactions in which CIF displaces inhibitory CaM from the membrane-bound Ca²⁺-independent iPLA₂ leading to iPLA₂ activation and the generation of lysophospholipids that in turn stimulate SOCCs (Smani et al., 2004). Activation of SOCE via SOCCs is responsible for sustaining the contraction and refilling the stores upon removal of the agonist. On the other hand, activation of protein kinase C (PKC) by diacylglycerol following receptor stimulation may play an additional role in the sensitization of the contractile apparatus to Ca²⁺, further amplifying the response to receptor agonists. Besides, receptor agonists can also induce Ca²⁺ sensitization via Rho-associated kinase pathways (Ghisdal et al., 2003). The SERCA inhibitors, which deplete the stores and cause SOCE, fail to activate these sensitizing processes, thus explaining the observed low magnitude of contractions produced by CPA or thapsigargin, as compared with receptor agonists. Moreover, in pulmonary arteries and spleen, the cellular response activated by capacitative calcium entry in smooth muscle is reduced by tyrosine kinase inhibitors, suggesting a phosphorylation step via tyrosine kinase in the SOCE pathway. Finally, NO was found to modulate the SOCC activity via a guanosine-3′,5′-cyclicmonophosphate (cGMP)-dependent mechanism (Clementi and Meldolesi, 1997; Bolotina, 1999).

Figure 1.

A model for excitation/contraction coupling in a tonic smooth muscle cell in which sustained contraction involves Ca²⁺ entry through both VOCCs and SOCCs. (1) Physiologically, SOCE is initiated either by stimulation of receptors that couple through heterotrimeric GTP-binding protein (G proteins) to activate phospholipase C_β (PLC_β) or (2) by stimulation of receptors that couple through tyrosine phosphorylation to activate PLCγ (Parekh and Penner, 1997; Patterson et al., 2002). This results in breakdown of phosphoinositide and production of IP₃. (3) This second messenger activates IP₃ receptors, which are ligand-gated Ca²⁺ channels located in the SR. The resulting release of Ca²⁺ into the cytoplasm causes a transient increase in [Ca²⁺]_i, whereas (4) emptying of Ca²⁺ stores generates a retrograde signal that activates SOCCs in the plasma membrane, which are responsible for the sustained increase in [Ca²⁺]_i after the initial Ca²⁺ transient. (5) Depleted Ca²⁺ stores generate a key messenger molecule called Ca²⁺ influx factor (CIF), which diffuses to the plasma membrane. (6) A cascade of plasma-membrane-delimited reactions in which (7) CIF displaces inhibitory CaM from the membrane-bound iPLA₂, leading to iPLA₂ activation and the generation of lysophospholipids (8) that in turn activate SOCCs. Ca²⁺ release from the SR causes the sensor (i.e., Ca²⁺-sensing STIM1 protein) to aggregate in areas close to the plasma membrane and to interact with SOCC (i.e., Orai1), which is believed to be the store-operated channel (9). On the other hand, (10) SOCC may also provide direct depolarization, independently of Ca²⁺ -activated Cl⁻ channel VOCCs are opened by membrane depolarization due to initial Ca²⁺ -release from SR, which stimulates a Ca²⁺ -activated Cl⁻ channel (11). (12) The plasma membrane NCX is involved in the regulation of Ca²⁺ homeostasis in blood vessels by contributing to Ca²⁺ entry. Finally, (13) NO/cGMP inhibits SOCCs, possibly by enhanced re-filling of the Ca²⁺ -stores. (14) Ca²⁺ sensitization might occur via agonist-induced activation of either the small G protein RhoA/Rho-associated kinase (Rho K) pathway or protein kinase C (PKC). A, agonist; CICR, Ca²⁺ -induced Ca²⁺ release; CIF, calcium influx factor; CLCA, Ca²⁺-activated Cl⁻ channel; iPLA₂, Ca²⁺-independent phospholipase A2; CaM, calmodulin; CPA, cyclopiazonic acid; DAG, diacylglycerol; G, GTP binding proteins; cGMP, guanosine-3′,5′-cyclicmonophosphate; IP₃, inositol-1,4,5-trisphosphate; NCX, Na⁺/Ca²⁺ exchanger; NO, nitric oxide; PKC, protein kinase C; PLC, phospholipase C; R, receptor; S, putative Ca²⁺ sensor; SOCC, store-operated Ca²⁺ channel; STIM1, stromal-interacting molecule 1; TG, thapsigorgin; TK, tyrosine kinase; V_m, membrane potential; VOCC; voltage-operated Ca²⁺ channel.

Pharmacological inhibition of SOCE

A recent review by Putney (2001) discussed a number of drugs that possess inhibitory activity against SOCCs, but in most cases with less than optimal specificity. The SOCE inhibitors include cations (lanthanides, Gd³⁺ and divalent cations), P450 inhibitors (econazole, miconazole, clotrimazole and ketoconazole), cyclooxygenae inhibitors (niflumic acid, flufenamic acid and tenidap), lipoxygenase inhibitors (nordihydroguaiaretic acid and eicosatetraynoic acid), putative channel blockers ((SK&F 96365, SC38249, LU52396, L-651, 582, tetrandrine, 2-Aminoethyl diphenylborinate) and mechanism-based inhibitors (U73122 (phospholipase C inhibitor) and wortmannin (phosphatidylinositol kinase inhibitor)) (Putney, 2001). The simplest and most dependable SOCE inhibitors are Ca²⁺ mimics, for example, divalent cations, and the potent trivalent lanthanides.

There has been considerable progress in our understanding of the pharmacological profile of SOCCs in VSMCs. Flemming et al. (2003) characterized the pharmacological properties of store-operated channels in VSMCs of rabbit pial arterioles with the use of various SOCC inhibitors (Flemming et al., 2003) and demonstrated that SOCE is inhibited by Gd³⁺ in a concentration dependent manner (IC₅₀=101 nM). The inhibitory effect of other inhibitors in the same study includes: 10 μM La³⁺ (70% inhibition), 75 μM 2-Aminoethyl diphenylborinate (66% inhibition), 100 μM Ni²⁺ (57% inhibition), 10 μM wortmannin (76% inhibition) and 100 μM capsaicin (12% inhibition). Drugs that are ineffective include: 1 μM nifedipine, 10 μM SK&F96365, 10 μM LOE908, 10–100 μM ruthenium red, 100 μM sulindac, 0.5 mM streptomycin and a 1:10 000 dilution of Grammostolla spatula venom (Flemming et al., 2003). On the contrary, Wayman et al. (1996a) reported that SOCCs in anococcygeus smooth muscle cells are insensitive to Gd³⁺ or La³⁺ at concentrations of up to 400 μM (Wayman et al., 1996a). SOCCs in VSMCs of canine renal arteries show some degree of sensitivity to 100 μM Gd³⁺, while those in pulmonary arterial smooth muscle are resistant to Gd³⁺ (Wilson et al., 2002). SOCCs of rat intrapulmonary arteries are sensitive to 1 μM La³⁺ and these channels in main pulmonary arteries are blocked by La³⁺ only at much higher concentrations (>100 μM) (Robertson et al., 2000; Ng and Gurney, 2001). The SOCCs of the anococcygeus and ileal smooth muscles can be inhibited by 10 μM SK&F96365 (Wayman et al., 1996a; Zholos et al., 2000; Ng and Gurney, 2001).

A recent study showed that diethylstilbestrol (DES), a synthetic estrogenic agonist, elicits a rapid and reversible block of SOCCs in rat basophilic leukaemia cells, aortic smooth muscle and human platelets. DES also inhibits whole-cell Ca²⁺ release-activated Ca²⁺ currents and thapsigargin-induced capacitative calcium entry (Zakharov et al., 2004). In contrast, trans-stilbene, a close structural analog of DES that lacks hydroxyl and ethyl groups, had no effect on the Ca²⁺ release-activated Ca²⁺ current and on SOCE. Thus, DES is proposed to be an effective inhibitor of SOCCs in a diversity of cell types (Zakharov et al., 2004). Brueggemann et al. (2006) also reported the pharmacological characteristics of a store-operated current, including its sensitivity to DES, 2-Aminoethyl diphenylborinate or micromolar Gd³⁺, and compared the effects of these inhibitors on thapsigargin- or [Arg8]-vasopressin-activated SOCE in rat mesenteric artery VSMCs using fura-2 (Brueggemann et al., 2006).

Other important roles played by SOCE in response to hormones and neurotransmitters

SOCCs and VSMC proliferation

Considerable existing evidence supports an important role for Ca²⁺ in cell proliferation, where activation of SOCCs is thought to participate in the process. VSMC proliferation normally occurs during the development and progression of hypertension. Pulmonary vascular medial hypertrophy due to VSMC proliferation contributes to the increased pulmonary vascular resistance in patients with pulmonary hypertension. A rise in [Ca²⁺]_cyt promotes the growth of pulmonary artery VSMCs. Resting [Ca²⁺]_cyt, intracellular stored [Ca²⁺], SOCE and store-operated Ca²⁺ currents are greater in proliferating human pulmonary artery VSMCs than in growth-arrested cells (Sweeney et al., 2002a). In cells treated with an antisense oligonucleotide specifically designed to cleave TRPC1 mRNA (resulting in reduced mRNA and protein expression of TRPC1), the amplitudes of the store depletion-activated currents (I_SOC) and SOCE elicited by passive depletion of Ca²⁺ stores are reduced. Importantly, there is a 50% reduction in the growth rate of these cells, indicating that TRPC1 may encode a SOCC that plays a critical role in VSMC proliferation of the pulmonary artery by regulating SOCE-associated changes in [Ca²⁺]_cyt (Sweeney et al., 2002b).

Spontaneous SOCC activity

In addition to SOCC stimulation by store depletion, spontaneous channel activity has been also recorded in unstimulated smooth muscle cells. In freshly isolated rabbit portal vein myocytes, approximately 45% of outside-out patches contain spontaneous single-channel currents with a unitary conductance 23 pS, which have similar properties as to those of channel currents evoked by noradrenaline and the diacylglycerol analogue 1-oleoyl-2-acetyo-sn-glycerol (Albert and Large, 2001). The molecular identity of the channel is unknown. However, it is becoming increasingly evident that there are several similarities between these channels and the TRP and TRPL (transient receptor potential-like) channels previously described in Drosophilia photoreceptors (Harteneck et al., 2000). Firstly, in rabbit portal vein smooth muscle, the non-selective cation channels are activated by diacylglycerol in a PKC-independent manner (Helliwell and Large, 1997). The mammalian homologue hTRPC6, hTRPC3 and mouse TRPC7 are non-selective cation channels that are activated by diacylglycerol independently of PKC (Hofmann et al., 1999; Okada et al., 1999). Secondly, the relative permeability of some of these channels to divalent cations is similar. Thirdly, in the absence of activators, channels open and close spontaneously in native venous myocytes and in cells expressing TRP and TRPL channels (Hofmann et al., 1999). Lastly, the probability of channel opening is greatly increased at positive potential (Chyb et al., 1999; Hofmann et al., 1999).

Using whole-cell, perforated-patch recording method, Bae et al. (1999) described a basal non-selective cation current in freshly dispersed rabbit pulmonary artery myocytes and concluded that the non-selective cation conductance is a component of the resting membrane potential (Bae et al., 1999). Spontaneous SOCC activity could be one reason for the resting membrane potentials (−60 and −45 mV) of VSMCs being significantly less negative than the K⁺ equilibrium potential (E_K, about −85 mV, Nelson and Quayle, 1995; Kuriyama et al., 1998). Albert et al. (2003) suggested that in rabbit ear artery myocytes, there exists a constitutively active Ca²⁺-permeable cation channel that is regulated by external Ca²⁺ ions and suppressed by the tonic PKC activity. Such a constitutively active Ca²⁺-permeable cation current may contribute to the resting membrane conductance and basal Ca²⁺ influx in the arteries (Albert et al., 2003).

Molecular identity of SOCCs in VSMC

Emerging evidence links SOCE to TRP channels. It has long been known that smooth muscle contraction can occur independently of changes in the membrane potential; what is less clear is the nature of the Ca²⁺ permeation pathway that is stimulated without voltage activation. In this regard, ROCCs activated by ligand–receptor interaction and SOCCs are thought to be principal modes of voltage-independent Ca²⁺ entry. It is possible that ROCCs and SOCCs may be closely related members of the TRP channel family (McFadzean and Gibson, 2002). The TRP channel proteins were first identified in the Drosophila melanogaster fruit fly where a mutation led to visual defects due to defects in the Ca²⁺ influx pathway. There have since been a large number of TRP channel proteins identified and these can be classified into three categories. They all have six transmembrane domains and are non-selective ion channels.

1. TRPC, where C stands for classical or canonical due to the highest homology with the molecular identity of TRP channels in Drosophila. There are at least seven members (TRPC 1–7) of this subfamily, with TRPC1 being the most abundant in vascular tissue. TRPC2 is a pseudogene in man. However, the expression of the various isoforms of TRPC channels in the vasculature is likely to be highly species and vascular bed dependent. There is evidence suggesting that the ROC may in fact be a TRPC6 and possibly TRPC1 channel (Xu and Beech, 2001; Inoue, 2005). It is likely that ROC and SOC channels are made up of heteromeric combinations of various TRPC proteins.

2. TRPV, where V stands for vanilloid, as these channels (TRPV 1–6) are closely related to the vanilloid receptor.

3. TRPM, where M stands for melastatin (a tumor suppressor).

Although TRP channels are widely studied for their roles in ion conductance, it is also clear that they have several other principal functions as manifested by their sensory activities in perception of temperature, taste, pH, chemical stimuli (for example capsaicin) and osmolarity (Inoue, 2005).

Multiple homologues of TRPC proteins are expressed in VSMCs (Beech, 2005). When expressed heterologously, TRP channels generally form functional entities that exhibit electrophysiological properties characteristic of non-selective cation channels. There is growing support for the involvement of the TRPC1–7 family in the formation of Ca²⁺-permeable non-selective cation channels in VSMCs (Table 2). Several studies demonstrate that TRPC mRNA and TRPC proteins are expressed in several smooth muscle preparations. In rat pulmonary artery VSMCs, reverse transcription-PCR analysis revealed the expression of TRPC1, TRPC3, TRPC4, TRPC5 and TRPC6, while mRNA immunostaining identified proteins for TRPC1, TRPC3, TRPC4 and TRPC6 (Ng and Gurney, 2001). Moreover, reverse transcription-PCR and western blotting performed on RNA and protein isolated from distal intrapulmonary arteries and main pulmonary artery VSMCs revealed both mRNA and protein expression for TRPC1, TRPC4 and TRPC6, but not for TRPC2, TRPC3, TRPC5 or TRPC7 (Wang et al., 2003). In a number of murine and canine smooth muscle cell preparations, mRNA for TRPC4, TRPC6 and TRPC7 is detected but with no detection of mRNA for TRPC1, TRPC2 and TRPC5. In rat renal resistance arteries and aorta, mRNA and protein are probed for TRPC1, TRPC3, TRPC4, TRPC5 and TRPC6, while mRNA for TRPC2 and TRPC7 is undetectable (Facemire et al., 2004). Furthermore, in situ hybridization yielded strong labeling of TRPC1, TRPC3, TRPC4, TRPC5 and TRPC6 in endothelial and VSMCs of human coronary and cerebral arteries. TRPC7 is only expressed in endothelial cells but not in the underlying VSMCs. Results from immunohistochemical staining are in consistence with those from in situ hybridization (Yip et al., 2004).

Table 2.

Detection of TRPC in smooth muscle of various tissues

Tissue	TRPC1a	TRPC2	TRPC3	TRPC4a	TRPC5a	TRPC6	TRPC7	References
Aorta	+	+	+	+	+	+		Facemire et al., 2004
A7r5	+			+		+		Brueggemann et al., 2006
Rat mesentery artery	+			+		+		Brueggemann et al., 2006
Rat renal resistance artery	+		+	+	+	+		Facemire et al., 2004
Rat intralobar pulmonary arteries	+		+			+		Lin et al., 2004
Rat cerebral arteries	+		+	+	+	+		Flemming et al., 2003
Rabbit portal vein myocyte						+		Albert and Large, 2003
Human lower oesophageal sphinctor smooth muscle	+		+	+	+	+		Wang et al., 2003
Human coronary & cerebral artery	+		+	+	+	+		Yip et al., 2004
Human pulmonary artery	+							Golovina et al., 2001
Human internal mammary artery	+					+		Bergdahl et al., 2005
Pig trachea smooth muscle			+	+		+		Ay et al., 2004
Murine & canine smooth muscle				+		+	+	Walker et al., 2001
Rat prostate smooth muscle cell line			+			+		Thebault et al., 2005
Lamb fetal pulmonary smooth muscle cells	+		+		+	+		Resnik et al., 2007

Abbreviation: TRPC, transient receptor potential family.

Remark: ‘+'-expression.

Empty cells refers to ‘particular TRPC is not expressed in the tissue'.

^aIndicates that particular TRPC may involve SOCC activity. TRPC2 is a pseudogene in man.

Some studies favour a role for TRPC proteins in smooth muscle function. For example, TRPC1 partially mediates SOCE in smooth muscle (Inoue et al., 2001). In arterioles, the application of an antibody against an extracellular epitope of TRPC1 (T1E3) reduces the thapsigargin-induced reduction in [Ca²⁺] by 25%, suggesting that part of thapsigargin-evoked SOCC activity is likely to be mediated by TRPC1 (Xu and Beech, 2001). A T1E3 antibody was also found to cause a 50% reduction in the SOCE-mediated contraction of rat cerebral arteries (Bergdahl et al., 2005), thus, supporting the notion that additional TRPC subunits are likely to play a positive role in thapsigargin-induced activation of SOCCs. More recently, Xu et al. (2006) showed that E3-targeted externally acting anti-TRPC5 blocking antibody (T5E3) suppressed Ca²⁺ entry in arterioles only after activation in store-operated mechanism triggered by thapsigargin in the absence of extracellular Ca²⁺, while T5E3 pre-adsorbed to its antigenic peptide had no effect. Collectively, these findings suggest that Ca²⁺ entry caused by passive store-depletion in arteriolar VSMCs may involve TRPC1 and TRPC5 (Xu et al., 2006).

Golovina et al. (2001) provided evidence for an upregulation of SOCC activity in proliferating VSMCs, suggesting that the increased TRPC1 mRNA may underlie SOCC-dependent rises in [Ca²⁺]_i during VSMC proliferation. In addition, they also demonstrated that human pulmonary artery myocytes treated with antisense oligonucleotides to cleave mRNA for TRPC1 have a low expression of TRPC1, a reduced amplitude of CPA-evoked currents and a decreased cell growth rate (Golovina et al., 2001). The phenylephrine- and CPA-evoked non-selective cation channel activation mediating tonic constrictions in rabbit vena cava is associated with oscillations of [Ca²⁺]_i generated by SOCE that may be specifically encoded by genes for TRPC1 (Liu et al., 2000; Lee et al., 2001). However, there is not always such a clear association between the expression of mRNA or channel proteins and their physiological significance in native cells. Despite a prominent role for TRPC1 as the main candidate for SOCCs in smooth muscle, other evidence showed that TRPC4 may be also related to SOCCs. The mRNA for TRPC4 was found to be the most abundant among all TRPCs detected in murine and canine smooth muscle cells (Walker et al., 2001). In addition, other studies also support a positive role of TRPC6 in mediating noradrenaline-evoked cation current (I_cat, ROC) in VSMCs (Inoue et al., 2001).

The newly identified STIMs may mediate SOCE. Roos et al. (2005) proposed that STIM1 is an essential and conserved component of SOCC. STIM1 contains a functional EF-hand domain for Ca²⁺ binding and can act as a Ca²⁺ sensor to monitor the Ca²⁺-loading levels inside the stores (reviewed by Draber and Draberova, 2005; Putney, 2007). After store depletion, STIM1 is transolocated to the plasma membrane to activate SOCCs by the following proposed mechanisms: (i) interaction with a putative TRP pore-forming subunit; (ii) activation of Ca²⁺ entry by means of conformational coupling to its coiled domain and (iii) assembly with additional STIM1 monomers and other components to form a unique functional Ca²⁺ channel (Liou et al., 2005; Zhang et al., 2005b; Spassova et al., 2006).

Pathophysiological importance of SOCE

SOCE and cardiovascular diseases

Some cardiovascular diseases are specifically associated with a failure or malfunction of SOCE. The Ca²⁺-handling capability of the SR is defective in pulmonary hypertension and hyperglycaemia.

Prolonged exposure to alveolar hypoxia causes pulmonary hypertension with profound vascular remodelling and alterations in the Ca²⁺ homeostasis in pulmonary artery VSMCs (Shimoda et al., 2000). Several studies provide evidence that pulmonary hypertension (including chronic hypoxic pulmonary hypertension and chronic intrauterine pulmonary hypertension) is related to SOCE. Firstly, store-operated channels of pulmonary artery VSMCs are upregulated by chronic hypoxia, and increased SOCC activity contributes to the enhanced vascular tone in hypoxic pulmonary hypertension. Small interfering RNA knockdown of TRPC1 and TRPC6 specifically inhibits the thapsigargin- and 1-oleoyl-2-acetyo-sn-glycerol-induced cation entry. Removal of extracellular Ca²⁺ or inhibition of SOCE by La³⁺ and SKF-96365 prevent the elevated levels of [Ca²⁺]_i in pulmonary artery VSMCs and inhibit the augmented vascular tone in pulmonary arteries of chronic hypoxic rats. In contrast, nifedipine has a negligible effect (Lin et al., 2004). Secondly, hypoxia-related pulmonary vasoconstriction requires SOCE in isolated rat lungs, since the enhanced vascular tone can be inhibited by SK&F-96365, NiCl₂ or LaCl₃ (Weigand et al., 2005). Thirdly, the platelet-derived growth factor-mediated proliferation of pulmonary arterial smooth muscle cells is associated with c-Jun/STAT3-induced upregulation of the TRPC6 expression. The resultant increase in SOCE raises [Ca²⁺]_i, which facilitates the return of Ca²⁺ to the SR and augments pulmonary artery VSMC growth (Yu et al., 2003). Lastly, chronic intrauterine pulmonary hypertension increases SOCE and acute normoxia normally diminishes SOCE in fetal lamb pulmonary artery VSMCs. Under normoxic conditions, the expression of TRPC1, 3, 5 and 6 is greater in normotensive than hypertensive VSMCs (Resnik et al., 2007). The SOCCs may thus represent an additional target for therapeutic intervention.

Hyperglycaemia also attenuates SOCE in VSMCs following depletion of intracellular Ca²⁺ stores (Rivera et al., 1995); the underlying mechanisms may be related to altered PKC activity as PKC-dependent phosphorylation was thought to be contributory to the inactivation of SOCE (Parekh and Penner, 1995). Reduced SOCE in the retinal arterioles from streptozotocin-treated rats may also be related to the channel modulation by PKC_β (Curtis et al., 2003).

Organ culture and SOCE

Blood vessel preparations kept in organ culture are viable for several days and maintain their contractility with little evidence of a significant change in phenotype from the contractile to the synthetic state when vessels are cultured in the absence of supplementary growth factors (Hellstrand, 1998; Lindqvist et al., 1999). On the other hand, organ culture of human saphenous vein and porcine aorta leads to the formation of a neointima, a lesion that is characteristic of restenosis after angioplasty (Soyombo et al., 1990; Koo and Gotlieb, 1991). The synthetic phenotype of smooth muscle cells is of importance in clinical situations where growth and proliferation of VSMCs, such as neointima formation, is part of the atherosclerotic process (Ross, 1993).

Since the distinction between contractile and synthetic smooth muscle phenotypes was made largely based upon ultrastructure (Owen, 1995; Thyberg et al., 1996), the issue arises as to what roles alterations of ion channel properties and intracellular Ca²⁺ stores may play in this process. Changes in Ca²⁺ handling capacity in intact arteries, similar to those observed in cultured VSMCs are likely to influence cell excitability and SOCC pharmacology, making SOCC a potential target for the prevention or even treatment of hypertension and atherosclerosis.

In rat basilar and tail arteries, the intracellular Ca²⁺ release upon depletion of SR stores is increased after 3–4 days in serum-free organ culture (Dreja et al., 2001), which is not associated with any increase in basal [Ca²⁺]_i. The voltage-dependent Ca²⁺ currents are reduced but SOCE is augmented in organ-cultured basilar arteries. SOCE is enhanced and vascular contractility is maintained in rat cerebral arteries kept in culture for several days. Under these conditions, there is also a 50% increase in the nifedipine-insensitive currents that are activated by store depletion. The mRNA levels for TRPC1 and TRPC6 increase significantly after organ culture, and a polyclonal TRPC1 antibody against an extracellular epitope (T1E3 antibody) inhibits contractility by 50% (Bergdahl et al., 2005). In segments of the human internal mammary, artery kept in organ culture for 24 h and then exposed to balloon dilatation in vitro, followed by further culture for 48 h, the mRNA expression of TRPC1 and TRPC6 mRNA is higher as compared with the undilated segments (Bergdahl et al., 2005), suggesting that organ culture or mechanical injury could impact on the plasticity of TRPC and cellular Ca²⁺ handling in blood vessels.

NO and SOCE

The endothelial production of NO accounts for endothelium-dependent vasodilatation in response to dilator agonists. Several mechanisms have been proposed to explain NO-dependent modulation of [Ca²⁺]_i in VSMCs, which is the primary regulator of vascular tone. NO affects voltage-dependent activation of smooth muscle L-type Ca²⁺ channels either directly or indirectly through opening K⁺ channels and ensuing hyperpolarization (Robertson et al., 1993; Archer et al., 1994; Bolotina et al., 1994). While other studies indicate that NO is capable of modulating the activity of voltage-independent SOCCs. NO induces a rapid decrease in [Ca²⁺]_i by accelerating sequestration of Ca²⁺ into intracellular stores via SERCA; as a result, refilled Ca²⁺ stores inhibit SOCE and reduce vascular tone (Cohen et al., 1999). NO-mediated inhibition of SOCCs and ROCCs is impaired in chronic hypoxia-induced pulmonary hypertension (Jernigan et al., 2006). Considering that a rise in [Ca²⁺]_i is the major stimulus for contraction, gene expression and proliferation of VSMCs, such impaired NO signalling may have important implications in the regulation of not only pulmonary vascular tone but also arterial wall remodelling; both are jointly involved in the development of pulmonary hypertension associated with chronic hypoxia. In contrast, excessive NO interferes with the release of Ca²⁺ from thapsigargin-sensitive stores and reduces SOCE into VSMCs subsequent to depletion of Ca²⁺ stores in the mesenteric vascular bed of bile duct-ligated rats (a model of liver cirrhosis). This mechanism may mediate the reduced pressor response reported in cirrhosis (Atucha et al., 2005). It appears that NO modulates the SOCC activity via a cGMP-dependent pathway because sodium nitroprusside, which releases intracellular cGMP by activating guanylyl cyclase, was found to inhibit non-selective cation currents activated by store depletion in the mouse anococcygeus smooth muscle cells (Wayman et al., 1996b). Moreover, sodium nitroprusside, S-nitroso-N-acetyl-DL-penicillamine and 8-bromo-cGMP inhibit a Ca²⁺-permeable, non-selective cation current activated by endothelin-1 in rat aortic smooth muscle cells (Minowa et al., 1997). Therefore, regulation of the agonist-activated SOCE is another mechanism responsible for NO-mediated vasodilatation. More importantly, the Ca²⁺-permeable non-selective cation channels are important targets for nitrovasodilators.

Conclusions

Capacitative Ca²⁺ entry via SOCCs, along with VOCCs and ROCCs, plays an important role in the regulation of smooth muscle tone. The relative contribution of SOCE to excitation/contraction coupling depends on the smooth muscle type and appears to be greatest in tonic smooth muscle. However, many questions remain unanswered and need full resolution. The most important areas of future research are (i) identification of selective activators/inhibitors that act directly on SOCCs for further pharmacological characterization of these channels and their potential therapeutic value; (ii) elucidation of cellular mechanisms that link depletion of the SR to the opening of SOCCs in the plasma membrane and (iii) elucidation of a complete molecular structure, a crucial step towards full understanding of the channel function. It is important to determine not only the expression of TRP proteins in VSMCs but also their tendency to heteromultimerise in native cells. SOCCs in arterioles have a distinct pharmacological profile. Knowledge of this profile provides support for the hypothesis that there exist multiple types of SOCCs in smooth muscle and will facilitate comparisons with heterologously expressed genes that encode putative subunits of SOCCs. The results presented in this review favour the proposal that arteriolar SOCCs are likely to be TRPC1 and TRPC5. Finally, are there mechanisms of store-operated entry involving additional means of activation (excluding Stim1) and other store-operated channels (excluding Orais)? The existence of distinct SOCCs in VSMCs may aid in developing target-specific novel therapeutic drugs.

Abbreviations

8-bromo-cyclic GMP

8-bromo-guanosine-3′,5′-cyclicmonophosphate

CaM

calmodulin

CIF

calcium influx factor

CPA

cyclopiazonic acid

DAG

diacylglycerol

DES

diethylstilbestrol

IP₃

inositol 1,4,5-trisphosphate

NO

nitric oxide

ROCC

receptor-operated Ca²⁺ channels

SERCA

sarcoplasmic/endoplasmic reticulum Ca²⁺ ATPases

SOCC

store-operated Ca²⁺ channel

SOCE

store-operated calcium entry

SR

sarcoplasmic reticulum

STIM1

stromal-interacting molecule 1

TRP

transient receptor potential

VOCC

voltage-operated Ca²⁺ channel

VSMCs

vascular smooth muscle cells

Conflict of interest

The authors state no conflict of interest.