Transient Receptor Potential Canonical (TRPC)/Orai1-dependent Store-operated Ca2+ Channels

Store-operated Ca2+ entry (SOCE) has emerged as an important mechanism in cardiac pathology. However, the signals that up-regulate SOCE in the heart remain unexplored. Clinical trials have emphasized the beneficial role of mineralocorticoid receptor (MR) signaling blockade in heart failure and associated arrhythmias. Accumulated evidence suggests that the mineralocorticoid hormone aldosterone, through activation of its receptor, MR, might be a key regulator of Ca2+ influx in cardiomyocytes. We thus assessed whether and how SOCE involving transient receptor potential canonical (TRPC) and Orai1 channels are regulated by aldosterone/MR in neonatal rat ventricular cardiomyocytes. Molecular screening using qRT-PCR and Western blotting demonstrated that aldosterone treatment for 24 h specifically increased the mRNA and/or protein levels of Orai1, TRPC1, -C4, -C5, and stromal interaction molecule 1 through MR activation. These effects were correlated with a specific enhancement of SOCE activities sensitive to store-operated channel inhibitors (SKF-96365 and BTP2) and to a potent Orai1 blocker (S66) and were prevented by TRPC1, -C4, and Orai1 dominant negative mutants or TRPC5 siRNA. A mechanistic approach showed that up-regulation of serum- and glucocorticoid-regulated kinase 1 mRNA expression by aldosterone is involved in enhanced SOCE. Functionally, 24-h aldosterone-enhanced SOCE is associated with increased diastolic [Ca2+]i, which is blunted by store-operated channel inhibitors. Our study provides the first evidence that aldosterone promotes TRPC1-, -C4-, -C5-, and Orai1-mediated SOCE in cardiomyocytes through an MR and serum- and glucocorticoid-regulated kinase 1 pathway.

The last discovered Ca 2ϩ channel family, transient receptor potential canonical (TRPC) 2 and Orai1 channels, is widely expressed, but the role of these channels and the modulation of their expression are not completely understood. Notably, the dysregulation of these important mediators of Ca 2ϩ -dependent signal transduction has been involved in a broad range of human diseases such as adenocarcinomas, type 2 diabetes and diabetic nephropathy, human proteinuric kidney disease focal and segmental glomerulosclerosis, severe immunodeficiency, congenital myopathy, chronic pulmonary disease, and ectodermal dysplasia (1,2). Although predominantly thought of in the context of non-excitable cells, store-operated Ca 2ϩ entry (SOCE) by store-operated channels (SOCs) has recently emerged as a potential mechanism to alter Ca 2ϩ in the diseased cardiomyocyte. Although still under debate, the prime candidate proteins for SOCs encompass the TRPC channel(s) as non-selective cation channels as well as Orai1, the Ca 2ϩ selective pore-forming unit of the Ca 2ϩ release-activated Ca 2ϩ channel (3). Stromal interaction molecule 1 (STIM1) serves as a Ca 2ϩ sensor in the endoplasmic reticulum/sarcoplasmic reticulum (SR), clustering proximal to the plasma membrane to activate Orai1 (4) and TRPC channel(s) (5) when Ca 2ϩ stores are depleted. The seven isoforms of the TRPC family (TRPC1-7) have been detected in heart (6) as have Orai1 and STIM1 (7). Functionally, activation of Ca 2ϩ influx after the depletion of intracellular Ca 2ϩ stores was first identified in embryonic and neonatal murine cardiomyocytes (8,9) and later in adult ventricular myocytes (10,11).
However, although it has been shown that SOCE machinery plays a role in modulating SR Ca 2ϩ load and cytosolic Ca 2ϩ levels in cardiac ventricular myocytes and in HL-1 cells (9,12,13), the contribution of SOCE to normal cardiac function is still discussed. Nonetheless, SOCE appears to play an important role in cardiac pathological processes such as hypertrophy and arrhythmia. An increase of TRPC5 and -C6 expression is observed in the failing human heart (14,15). In several animal models of cardiac hypertrophy and heart failure (HF), up-regulation of TRPC1 (16), -C3 (14), -C6 (17), -C7 (18), Orai1 (19), and STIM1 (20) has been reported. Combining different in vitro and in vivo approaches, including pharmacological and molecular silencing or overexpression, it has been shown that SOCE, through TRPC1, -C3, -C4, and -C6 channels, Orai1, and STIM1 are necessary for hypertrophy development and HF by activating calcineurin/NFAT signaling (11, 14 -16, 21-26). Furthermore, SOCE potentially has proarrhythmic effects (27)(28)(29)(30). SKF-96365, a common inhibitor of SOCs, decreased the number of ventricular tachycardia and early afterdepolarization in mouse hearts expressing constitutively active G␣ q with increased TRPC3 and -C6 expression (27). Excessive activation of TRPC3 in a cardiac overexpression mouse model is associated with arrhythmogenesis (31). The Orai activator 2-aminoethoxydiphenyl borate induced an SKF-96365-sensitive automatic activity in rat left ventricular papillary muscles (32). However, although the role of SOCE up-regulation in cardiac pathology is acknowledged, the signaling pathways that govern this up-regulation remain elusive.
Clinical trials have emphasized the high beneficial role of mineralocorticoid receptor (MR) signaling blockade in HF and associated arrhythmias (33). Accumulated evidence suggests that modulation of Ca 2ϩ flux is a central factor in the cardiac MR action and might be involved in triggering after depolarization-related fatal ventricular tachyarrhythmia (34 -39). In addition, aldosterone-mediated cardiac action may cause various responses associated with elevation of [Ca 2ϩ ] i , including the direct or indirect increase of the Ca 2ϩ -dependent phosphatase 2B calcineurin and its downstream transcription factor NFATc3 in the development of cardiac hypertrophy (40), the negative feedback on cAMP-response element-binding protein via protein phosphatase 2A in promoting apoptosis (41), and more recently the activation of the multifunctional Ca 2ϩ / calmodulin-dependent protein kinase II associated with poor outcomes after myocardial infarction (42).
Altogether, these data draw a parallel between the detrimental cardiac effect of aldosterone/MR and SOCE. We thus assessed whether and how SOCE-involved TRPC and Orai1 channels are regulated by cardiac aldosterone/MR signaling in neonatal rat ventricular cardiomyocytes (NRVMs). We showed that aldosterone promotes an MR-specific enhancement of STIM1-dependent SOCE activities, which correlated with the increased expression and activity of TRPC1, -C4, -C5, and Orai1 channels. This effect was dependent on the up-regulation of serum-and glucocorticoid-inducible kinase 1 (SGK1). This is the first demonstration of the cardiac SOCE activation by mineralocorticoid signaling, an effect that might contribute to the benefit of MR blockade in cardiac diseases.

Experimental Procedures
Cell Isolation and Culture-All experiments were carried out according to the ethical principles laid down by the French Ministry of Agriculture and to conform to the guidelines from Directive 2010/63/EU of the European Parliament on the protection of animals. NRVMs were isolated from 1-2-day-old Wistar rats. The heads were cut off, hearts were removed, ventricles were pooled, and ventricular cells were dispersed by successive enzymatic digestion with collagenase A (Roche Applied Science) and pancreatin (Sigma-Aldrich). The cell suspension was purified by centrifugation through a discontinuous Percoll gradient to obtain myocardial cell cultures with 99% myocytes. After seeding on glass coverslips coated with poly-D-lysine (Sigma-Aldrich) in 30-mm plastic wells, NRVMs were cultured in Dulbecco's modified Eagle's medium (DMEM)/Medium 199 (4:1) supplemented with 10% horse serum, 5% fetal bovine serum, 1% glutamine, and antibiotics (plating medium) for 24 h. Approximately 95% of the cells displayed spontaneous contractile activity in culture. Then the cells were cultured in serum-free media for 48 h and treated for 24 h in the presence or absence of various agents.
The mineralocorticoid hormone aldosterone (Aldo), the L-type Ca 2ϩ channel inhibitor nifedipine (NIF), the general TRPC channel inhibitor SKF-96365, and the irreversible sarco/ endoplasmic reticulum Ca 2ϩ -ATPase inhibitor thapsigargin (Tg) were from Sigma-Aldrich. The ryanodine receptor activator caffeine (caf) and the store-operated channel inhibitor BTP2 were purchased from Merck. The Orai1 inhibitor S66 was a kind gift from Richard Foster and David J. Beech (University of Leeds, UK). The Na ϩ /Ca 2ϩ exchanger inhibitor KB-R7943 and the selective antagonist of MR RU-28318 (RU) were purchased from Tocris Bioscience (Bristol, UK). The SGK1 inhibitor GSK650394 (GSK) was from Santa Cruz Biotechnology (Dallas, TX). All compounds were dissolved in DMSO except Aldo and RU, which were diluted in ethanol and water, respectively. All vehicles were diluted at least 1:1000, which had no effect on NRVM function by themselves. Of note, we did not observe cellular hypertrophy measured by planimetry after 24-h 1 M aldosterone treatment (untreated NRVMs, 788 Ϯ 21 m 2 , n ϭ 380; aldosterone-treated NRVMs, 785 Ϯ 21 m 2 , n ϭ 304).
Cell Transfections-NRVMs were transiently transfected with plasmids or siRNA using Lipofectamine 2000 (Life Technologies) according to the manufacturer's instructions. pMAX-GFP vector was obtained from Amaxa Biosystems (Lonza, Basel, Switzerland). The dominant negative mutants of TRPC1 (F562A) and TRPC4 (E647K/E648K) were a kind gift from Shmuel Muallem (National Institutes of Health, Bethesda, MD) (43), and Orai1-YFP (E106A mutant) construct was from Stephane Koening (University of Geneva, Switzerland). The mutant TRPC3 was a kind gift from Klaus Groschner (University of Graz, Austria) (44). Plasmids YFP-WT-STIM1 (19754) and YFP-STIM1-⌬K (18861) were purchased from Addgene (45). 0.5-2 g of plasmid or a 200 nM concentration of a specific pool of four different siRNA duplexes (siGENOME SMARTpool siRNA from Life Technologies) against either TRPC5 (M-094467-01-0005) or SGK1 (M-090762-01-0005) or a nontargeting siRNA (D-001210-01-05) used as a negative control and 5 l of Lipofectamine transfection reagent were mixed in 800 l of serum-free medium (Opti-MEM). The mixture was incubated for 30 min at 37°C and added drop by drop in each cell culture dish. After 12-24 h, the Opti-MEM was replaced by the maintenance medium. The cells were kept in culture for a further 48 h until expression of the YFP or GFP fusion protein (as a marker of successful transfection in the case of plasmid transfection) was detectable in the cells. Empty pMAXGFP or pcDNA3-YFP vectors were used as controls to verify whether transfection itself could affect the Ca 2ϩ response.
Co-immunoprecipitation-After NRVM lysis, 2 g of antibody was added to a volume of lysate containing 300 g of proteins diluted with 600 l of NET solubilization buffer (50 mM Tris-HCl, pH 7.4, 150 mM NaCl, 5 mM EDTA, 0.05% Nonidet P-40 (v/v)) and incubated at 4°C overnight with constant mixing. Then the incubation continued for 1 h at 4°C with constant mixing with 50 l of Protein A/G magnetic beads (Millipore). The immune complexes were collected with a magnetic stand and washed three times in NET solubilization buffer. After denaturing, samples were subjected to SDS-PAGE. The negative control was performed using lysates with beads (ctrl(Ϫ)) without any antibody. For technical reasons, it was not possible to perform immunoblotting of Orai1 because its molecular weight is approximately the same as the molecular weight of IgG antibodies.
Immunostaining-NRVMs were fixed and permeabilized in cold 100% methanol for 5 min or with 4% paraformaldehyde and 0.5% Triton X-100 (20 and 5 min, respectively) and washed in 1ϫ TBS containing in 20 mM Tris-HCl, 154 mM NaCl, 2 mM EGTA, 2 mM MgCl 2 , pH 7.4. A saturation step was executed with TBS plus 1% BSA for 10 min. Samples were incubated for 1 h with primary antibody against STIM1 diluted 1:100 or Orai1, TRPC1, or -C4 diluted 1:25 in TBS plus 1% BSA. After washing in TBS, cells were incubated for 1 h in TBS plus 1% BSA with Alexa Fluor 594-or 555-conjugated anti-mouse or -rab-bit IgG (Life Technologies). The cells were mounted using Mowiol mounting medium. Images were acquired using a Leica TCS SP5 confocal microscope.
Quantitative Real Time PCR Analysis-Total RNA from NRVMs was isolated using the TRIzol procedure (Life Technologies) according to the manufacturer's instructions. For cDNA synthesis, 1 g of total RNA was reverse transcribed using an iScript cDNA synthesis kit (Bio-Rad) according to the manufacturer's instructions in a total volume of 20 l. The cDNA was then subjected to five successive dilutions (from 10 to 0.65 ng/l). Then cDNA was used as a template for real time PCR with SYBR Green Supermix (Bio-Rad) and a 0.5 M concentration of each primer in a total reaction volume of 15 l. Cycling conditions were as follows: 30 s at 95°C followed by 40 cycles of 95°C for 2 s and 60°C for 5 s. A dissociation curve was performed with the following steps: 60 -95°C for 5 s with an increment of 0.5°C. Each sample was tested in duplicate. Quantitative determination of the different mRNA expression levels was performed with a CFX96 Touch TM Real-Time PCR Detection System (Bio-Rad) with either gene-specific primers or primers for YWHAZ, RPL32, and TBP as endogenous controls (Table 1) from Eurofins MWG Operon (France). mRNA levels were normalized to housekeeping genes and were expressed as -fold change of that determined in untreated NRVMs for each culture.
Measurement of Cytosolic Ca 2ϩ Changes-NRVMs were rinsed with a physiological solution containing 135 mM NaCl, 5 mM KCl, 1.8 mM CaCl 2 , 1 mM MgCl 2 , 10 mM HEPES, 10 mM D-glucose, pH 7.4 with NaOH) and then incubated at 37°C for 30 min in darkness in the same solution supplemented with 2 M Fura-2/AM dissolved in DMSO plus 20% Pluronic acid (Life Technologies). Loaded cells were washed twice with the physiological solution before imaging. Ratiometric Fura-2 images of Ca 2ϩ signal were obtained using an inverted microscope (Axio Observer.D1, Zeiss) equipped with a Polychrome V fast monochromator system (Till Photonics), which alternately changed the excitation wavelength between 340 and 380 nm. Emission at 510 nm was collected by a cooled, 16-bit electron-multiplying charge-coupled device camera (Evolve, Photometrics) coupled to the microscope (ϫ40 oil immersion fluorescence objective). Image acquisition in selected cells and analysis were performed with Metafluor 7.8 imaging software (Molecular Devices).
To study SOCE in NRVMs, the cells were depleted with 2 M Tg and 10 mM caffeine in a Ca 2ϩ -free solution containing 0.1 mM EGTA in the presence of 10 M NIF. Subsequently, 1.8 mM Ca 2ϩ ϩ NIF were re-added to the medium, and the peak amplitude of the fluorometric signal corresponding to the response to reintroduction of external Ca 2ϩ was determined with ⌬F (where F is the fluorescence ratio).
The cation influx, was measured as described previously (29). All experiments were done at a controlled 37°C.
The 340/380 ratios were calibrated using the equation according to Grynkiewicz et al. (75): where R is the fluorescence ratio recorded at the two excitation wavelengths (F 340 and F 380 ), K D represents the dissociation constant, R min and R max are the fluorescence ratios under Ca 2ϩ -free and Ca 2ϩ -saturating conditions, and ␤ ϭ F 340 , zero Ca 2ϩ solution/F 380 , saturating Ca 2ϩ solution.
For in situ measurements of R min and R max , Fura-2-loaded cells were first exposed to Ca 2ϩ -free solution with 10 mM caffeine ϩ 2 M thapsigargin for 5 min to empty SR Ca 2ϩ stores. Bath solution was then switched to K-buffer (10 mM NaCl, 130 mM KCl, 1 mM MgCl 2 , 10 mM HEPES, pH 7.2, ionic strength 0.142, 37°C) containing 10 mM EGTA. To obtain R min , the cells were incubated for 30 -45 min with 20 M non-fluorescent Ca 2ϩ ionophore ionomycin (Merck) and 10 M carbonyl cyanide p-trifluoromethoxyphenylhydrazone (Sigma-Aldrich) in K-buffer Ca 2ϩ -free solution, and measurements were taken at both wavelengths after the fluorescence reached stable values. Then R max was obtained by saturating the indicator with 10 or 20 mM CaCl 2 in the presence of 10 mM EGTA.
To calculate the K D , a dose response (0, 1.0, 2.0, 3.0, 4.0, 5.0, 6.0, 7.0, 8.0, 9.0, and 10.0 mM CaCl 2 and EGTA (free Ca 2ϩ ranging from 0 to 37 M) was performed in K-buffer. As a double log plot, the Ca 2ϩ response of the indicator is linear with the x intercept being equal to the log of the apparent K D of the indicator. We calculated the following values for NRVMs: R min ϭ 0.56, R max ϭ 1.83, and ␤ ϭ 1.55 with K D ϭ 377 nM. Of note, aldosterone treatment did not change any parameters.
The Fluo-4 signal was recorded with a laser scanning confocal microscope (Leica SP5) equipped with an ϫ40 water immersion objective. Fluo-4 was excited at 490 nm with a white light laser, and emission was collected at Ͼ510 nm. The NRVMs, continuously perfused with physiological saline solution, were paced at 1 Hz. [Ca 2ϩ ] i transients were recorded by scanning the cells in XY line mode with a resonant scanner (8000 Hz); the window size used allowed rates of one image/15 ms. The cytosolic Ca 2ϩ variation was normalized by dividing the peak fluorescence intensity (F) by the average resting fluorescence intensity (F 0 ) to generate an F/F 0 image. [Ca 2ϩ ] i transient properties as well as the amplitude (maximum F/F 0 ) or differences in diastolic fluorescence (F D /F 0 ), the time to peak (in ms), and the time constant of decay (; in ms) were calculated by fitting the decay portion of the fluorescence trace to a monoexponential function. The SOCE measured in NRVMs was recorded by scanning the cells in XY line mode with 700-Hz speed every 387 ms.
Electrophysiological Recordings-The whole-cell patch clamp currents were recorded at room temperature (22 Ϯ 2°C) using an Axopatch 200B amplifier, filtered at 2 kHz, and digitized at 10 kHz using pClamp 10 software. Borosilicate glass pipettes (Harvard Apparatus) were pulled with a Sutter puller and fire-polished and had a resistance of 3-5 megaohms. The pipette solution contained 135 mM CsCl, 4 mM MgCl 2 , 5 mM EGTA, 10 mM HEPES, 5 mM Na 2 ATP, and 3 mM Na 2 creatine phosphate, pH 7.2 with LiOH, and cells were perfused with a solution containing 132 mM NaCl, 1.1 mM MgCl 2 , 4 mM CsCl, 10 mM CaCl 2 , 10 mM glucose, 0.01 mM NIF, 0.05 mM KB-R7943, and 10 mM HEPES, pH 7.4 with LiOH. Series resistance was electronically compensated up to 50% and was continually monitored during the experiment. Membrane capacitances (9 -15 pF) were measured for each cell with pClamp 10 software. The SR Ca 2ϩ stores were depleted by 2 M Tg and 10 mM caf, and the SOC currents (I SOC ) were elicited by a 1.2-s voltage ramp from Ϫ100 to ϩ 60 mV preceded by a 0-to ϩ40-mV prepulse (400 ms) to inactivate the voltage-dependent Na ϩ channels.
The currents were recorded before and after 5 M BTP2 cell perfusion and then normalized to membrane capacitance to account for cell size variations. The slope conductance was calculated according to the formula: ⌬I/⌬V m ϭ I Ϫ90 mV Ϫ I ϩ60 mV )/(Ϫ90 Ϫ 60 mV).
Statistical Analysis-All analysis was performed with GraphPad Prism 6 software. All values are reported as mean Ϯ S.E. For all experiments, the difference between two groups was assessed with two-tailed unpaired Student's t test and among at

Results
Aldosterone Increases TRPC1, -C4, -C5, and Orai1 Expression-We first investigated the effect of 24-h aldosterone treatment on TRPC and Orai channel expression on NRVMs. The mRNA expression of TRPC1, -C3-C6, Orai1, and Orai3 in aldosterone-treated NRVMs was compared with untreated NRVMs using qRT-PCR. TRPC2 and -C7 isoforms were undetectable by classical PCR in our cells. As shown in Fig. 1A (from left to right panels), the mRNA levels of TRPC1, Orai1, and TRPC5 were significantly increased by 100 nM and 1 M aldosterone treatment, whereas lower doses (1 and 10 nM) had no effect. The effects were totally prevented when NRVMs were co-incubated for 24 h with the selective MR antagonist (10 M RU; Fig. 1A). By contrast, the mRNA levels of TRPC3, -C4, -C6, and Orai3 were unchanged after aldosterone treatment (Fig.  1B). These results were correlated to the increased expression of TRPC1, Orai1, and TRPC5 protein levels (Fig. 1C). Of inter- est, we observed an increased expression of TRPC4 at the protein level (Fig. 1C, right panel) in aldosterone-treated cells, whereas TRPC3, -C6, and Orai3 proteins were not affected (data not shown).
Aldosterone Increases SOCE Sensitive to BTP2 and SKF-96365 via MR Activation-Because TRPC1, -C4, -C5, and Orai1 form functional SOCs in a variety of cell types, we then investigated the effect of 24-h aldosterone treatment on SOCE in NRVMs. Variation of [Ca 2ϩ ] i following the depletion of intracellular Ca 2ϩ stores was measured on Fluo-4-loaded NRVMs. As shown in Fig. 2A, left panel, store depletion in Ca 2ϩ -free solution with 2 M Tg and 10 mM caf in the presence of 10 M NIF induced a large Ca 2ϩ release from the SR in NRVMs. The amplitude of the Ca 2ϩ release was not modified after aldosterone treatment, suggesting that the SR Ca 2ϩ load is unaltered as observed in adult cardiomyocytes (39). When Ca 2ϩ was reintroduced (in the presence of NIF), a moderate SOCE was recorded in untreated NRVMs (ctrl, light gray trace), whereas a large SOCE was observed in the aldosterone-treated NRVMs (100 nM, dark gray trace; 1 M, red trace). On average, chronic treatment with 100 nM and 1 M aldosterone significantly increased SOCE by up to 1.8-fold ( Fig. 2A, right panel). These effects were prevented with 24-h exposure to the MR antagonist RU (Fig. 2A). In addition, a lower aldosterone concentration (10 nM; Fig. 2A) or acute 1 M aldosterone exposure for 5 min (data not shown) did not affect SOCE in NRVMs. A pharmacological strategy using 5 M BTP2 or 40 M SKF-96365, widely used SOC inhibitors, decreased SOCE in control cells but also strongly prevented the enhanced SOCE induced by aldosterone treatment (Fig. 2B). Of note, 5 M KB-R7943, a Na ϩ /Ca 2ϩ exchanger blocker, did not alter SOCE in NRVMs treated or not with aldosterone (data not shown). To obtain direct evidence that exacerbated SOCE induced by aldosterone is closely linked to SOC activity, we measured the Mn 2ϩ -induced quenching of Fura-2 fluorescence as an indicator of Ca 2ϩ influx. Fig. 2C, left panel, shows representative traces of the progressive decay phase of Fura-2 fluorescence after 500 M Mn 2ϩ addition to NRVMs cultured in the presence (red trace) or absence (light gray trace) of 1 M aldosterone. On average, we observed a 2.4-fold increased cation entry through SOCs with aldosterone treatment (Fig. 2C, right panel).
This dominant negative strategy has the advantage to functionally block the target channel without affecting its endogenous expression, which might induce compensatory regulation of other channels. In a first attempt, we also tried dn-TRPC5 (49); however, in our hands, this construct was not expressed in NRVMs and thus we used an siRNA strategy. The knocked down expression of TRPC5 by a specific siRNA pool (four different siRNA duplexes against TRPC5), which reduced the TRPC5 mRNA level by 50% (data not shown) and protein level by 35% (Fig. 3C, left panel), significantly reduced SOCE in untreated and aldosterone-treated cells compared with scrambled siRNA-transfected NRVMs (Fig. 3C, right panel). These results suggested that TRPC1, -C4, and -C5 together with Orai1 form functional heterotetramer SOCs as proposed in other cell types (50 -52).
Aldosterone Increases I SOC Involving an Assembly of TRPC1, -C4, -C5, and Orai1-To further explore the nature of TRPC/ Orai1-dependent SOCs in untreated and aldosterone-treated NRVMs, we measured the store-dependent current using patch clamp in a whole-cell configuration under conditions in which major protagonists controlling the increase in intracellular Ca 2ϩ concentration (L-type Ca 2ϩ channel, Na ϩ /Ca 2ϩ exchanger, and sarco/endoplasmic reticulum Ca 2ϩ -ATPase pump) were blocked. Following SR Ca 2ϩ depletion (induced by  (Fig. 4A). On average (10 cells), the Tg/caf-induced current densities were Ϫ2.4 Ϯ 0.5 and 3.3 Ϯ 0.4 pA/pF at Ϫ80 and ϩ60 mV, respectively. Perfusion with 5 M BTP2 slightly reduced both inward and outward currents. As a result, the BTP2-sensitive current was very low in control cells (on average Ϫ0.9 Ϯ 0.4 and 1.3 Ϯ 0.3 pA/pF at Ϫ80 and ϩ60 mV, respectively). In cells incubated in the presence of 1 M aldosterone for 24 h (Fig. 4B), we observed robust increases in both inward and outward currents (on average (nine cells) Ϫ4.2 Ϯ 0.6 and 5.9 Ϯ 0.8 pA/pF at Ϫ80 and ϩ 60 mV, respectively; p Ͻ 0.05 versus ctrl) compared with untreated NRVMs. The aldosterone-enhanced I SOC was virtually abolished in the presence of BTP2, resulting in a much steeper BTP2-sensitive current than that found in untreated NRVMs (Ϫ3.0 Ϯ 0.5 and 3.5 Ϯ 0.2 pA/pF at Ϫ80 mV and ϩ60 mV, respectively; p Ͻ 0.05 versus ctrl). The slope conductance of the BTP2-sensitive current was significantly enhanced in the aldosterone condition (Fig. 4C). The nature of these aldosterone-enhanced I SOC induced by SR depletion that was reversed around 0 mV and presented slight rectification properties suggested the activation of non-selective cationic channels sensitive to SOC blocker. As shown in Fig. 4, D and E, we observed that the I SOC (dn-Orai1, Ϫ1.07 Ϯ 0.35 and 2.6 Ϯ 0.5 pA/pF at Ϫ80 mV and ϩ60 mV, respectively; GFP, Ϫ3.82 Ϯ 0.66 and 6.1 Ϯ 1.2 pA/pF at Ϫ80 mV and ϩ60 mV, respectively; p Ͻ 0.05 versus GFP) and the slope conductance were significantly reduced in aldosterone-treated NRVMs expressing dn-Orai1 in comparison with those expressing control plasmid. These results suggested a mixed nature of I SOC carried by TRPC and Orai1 channels.
To further investigate whether TRPC1-, -C4-, and -C5-Orai1-STIM1 could function as channel complexes driving SOCE, we performed co-immunoprecipitation experiments. In depleted NRVMs (induced by Tg and caf for 10 min), TRPC1, -C4, -C5, Orai1, and STIM1 all interacted together (Fig. 4F), suggesting that all these proteins might form a macromolecular complex and constitute one functional channel. Of note, TRPC3 only interacts with STIM1. These results are consistent with the idea that SOC function requires Orai1 and that TRPC and Orai1 might contribute to the same channel.
Aldosterone Regulates the SR Ca 2ϩ -sensing Protein STIM1-Multiple studies have shown that STIM1 and to a lesser extent STIM2 are central regulators of SOCE. We thus investigated whether aldosterone regulate STIM expression and activity in NRVMs. As shown in Fig. 5A, the mRNA levels of STIM1 and STIM2 were unchanged after 24-h aldosterone treatment. However, increased protein expression was observed for STIM1 after aldosterone treatment, whereas the STIM2 protein level was unaffected (Fig. 5B). Moreover, immunocytochemical staining showed higher labeling of endogenous STIM1 in aldosterone-treated NRVMs compared with untreated NRVMs (ctrl) with the STIM1 pattern becoming more punctum-like (Fig. 5C, top panels). Images from NRVMs transfected with YFP-WT-STIM1 also illustrated the punctum pattern of STIM1 in the aldosterone condition (Fig. 5C, bottom  panels). To assess the requirement for STIM1 signaling in the aldosterone-enhanced SOCE, NRVMs were transiently transfected, 24 h before treatment, either with YFP-WT-STIM1 or with the dominant negative construct for STIM1 deleted in its Lys-rich domain (YFP-STIM1-⌬K) (53), preventing an appropriate anchoring of STIM1 in the plasma membrane and activation of TRPC channel(s) but not Orai1 channels. In untreated cells (ctrl) expressing WT-STIM1, SOCE was drastically increased compared with empty pMAX-GFP-transfected cells. Interestingly, in aldosterone-treated cells, WT-STIM1 overexpression had no additional effect on SOCE, suggesting that the STIM1-dependent SOCE machinery is already entirely activated by aldosterone treatment (Fig. 5D). In addition, in cells expressing STIM1-⌬K, the aldosterone increase in SOCE was slightly smaller. Aldosterone-enhanced I SOC in STIM1-⌬K-expressing cells confirmed the slight/absent role of the Lys-rich domain of STIM1 in activation of I SOC (Fig. 5, E and F). SGK1 Regulates Orai1 Expression and SOCE-Because aldosterone-induced up-regulation of SOCs was observed only for a high dose (Figs. 1 and 2), we suspected the involvement of an intermediate factor. Among the classical aldosterone-induced proteins, SGK1, which is expressed in the heart, constitutes a novel and powerful modulator of SOCE, in particular of Orai1 (54). We thus investigated whether SGK1 might be involved in aldosterone-enhanced SOCE in NRVMs. Compared with untreated NRVMs, aldosterone-treated NRVMs showed a dose-dependent increased SGK1 expression at the mRNA level from low doses of aldosterone (Fig. 6A). Co-incubation for 24 h with the SGK1 inhibitor (1 M GSK) slightly decreased SOCE in control cells and strongly reduced the aldosterone-increased SOCE (Fig. 6B). In addition, NRVMs transfected with specific SGK1 siRNA, which reduced the SGK1 mRNA level by 50% (data not shown) and protein level by 51% (Fig. 6B, top panel), reduced SOCE in aldosterone-treated cells compared with scrambled siRNA-transfected or non-transfected NRVMs (Fig.  6B, bottom panel). GSK co-incubation with aldosterone did not affect the increased mRNA expression of TRPC1 and -C5 but prevented the increase of Orai1 mRNA levels (Fig. 6C), suggesting that SGK1 might constitute a key regulator of the cardiac SOCmediated Ca 2ϩ signaling dependent, at least, on Orai1.
Aldosterone  (a and c), aldosterone treatment led to [Ca 2ϩ ] i transients that failed to completely relax (Fig. 7A, aЈ, bЈ, and cЈ, bottom, right panel). This behavior was consistently observed for all aldosterone-treated NRVMs and eventually induced a brief period of excitability loss (data not shown). Analysis of the fluorescence signal revealed a striking increase in the diastolic [Ca 2ϩ ] i , maintained over time during stimulation (data not shown), expressed as the ratio F D /F 0 (where F D is fluorescence level at the diastolic period during 1-Hz stimulation and F 0 is the basal fluorescence before stimulation) with aldosterone treatment (Fig. 7B). We also observed that the resting [Ca 2ϩ ] i , measured by the ratiometric Ca 2ϩ indicator Fura-2, was significantly increased in aldosterone-treated cells (141.8 Ϯ 0.8 nM, n ϭ 674 versus 98.4 Ϯ 0.4 nM in control cells, n ϭ 793; p Ͻ 0.001). Interestingly, we observed that this aldosterone-increased resting [Ca 2ϩ ] i was reduced in cells transfected with dn-Orai1 (98.4 Ϯ 1.6 nM, n ϭ 74; p Ͻ 0.001 versus Aldo), dn-TRPC1 (106 Ϯ 1.5 nM, n ϭ 73; p Ͻ 0.001 versus Aldo), or dn-TRPC4 (107.4 Ϯ 2.6 nM, n ϭ 45; p Ͻ 0.001 versus Aldo) and with siRNA against TRPC5 (103 Ϯ 1.6 nM, n ϭ 120; p Ͻ 0.001 versus Aldo). Consistently, exposure of aldosteronetreated NRVMs to either SOC inhibitor (3 M BTP2) (Fig. 7C) or Orai1 blocker (1 M S66) for 2-3 min (Fig. 7D) significantly reduced the diastolic [Ca 2ϩ ] i increase during stimulation. In addition, the co-incubation of aldosterone with GSK for 24 h significantly reduced the diastolic [Ca 2ϩ ] i increase (Fig. 7E), showing that SGK1 might be a regulator of the cardiac SOCmediated Ca 2ϩ signaling.

Discussion
Accumulated evidence has shown that aldosterone/MR signaling plays a pivotal role in impairing cardiac Ca 2ϩ handling as a trigger of ventricular arrhythmias and sudden cardiac death in HF. However, to date, the cellular mechanism of action remains to be more precisely defined. In this context, it is imperative to identify and characterize new cardiac MR-specific downstream targets. Our study provides the first quantitative description of aldosterone/MR effects upon the SOCs in the heart. After 24-h treatment of NRVMs with aldosterone, we observed an MRspecific enhancement of SOCE, which is correlated to the increased expression and activity of the Ca 2ϩ sensor STIM1 and TRPC1, -C4, -C5, and Orai1 channels. Further mechanistic data showed that aldosterone/MR activation of Orai1-dependent SOCE involves SGK1 up-regulation.
Different reports have demonstrated the expression and activity of SOCs in the heart. SOCE appears to be developmen-tally dependent with greater amplitude in embryonic and neonatal cardiomyocytes but variable and limited in adult cardiomyocytes (7). Consistent with these studies, using a combination of functional Ca 2ϩ imaging, electrophysiological, and biochemical approaches, we showed that SOCE is active in NRVMs and is associated with the co-existence of all TRPC isoforms (except TRPC2 and -C7), STIM1/2, and Orai1/3.
We found that chronic aldosterone treatment significantly increased store depletion-induced Ca 2ϩ influx in cardiomyocytes. This is consistent with the finding of aldosterone-dependent up-regulation of SOCE in two other cell types (A7r5 rat aortic smooth muscle cells and metabolic syndrome chromaffin cells (55,56)). Pharmacological characterization and quenching of Mn 2ϩ by a Fura-2 fluorescence technique provide a fingerprint for SOCE in this effect. We showed that the aldosterone-enhanced cytosolic [Ca 2ϩ ] elevation following SR depletion is blunted by widely used SOC blockers. The remaining Ca 2ϩ entry in control or aldosterone-treated cells that is insensitive to SOC inhibitors suggests the involvement of other transient receptor potential-dependent Ca 2ϩ channels. The use of a selective antagonist of MR abolished the enhanced SOCE, suggesting that aldosterone stimulates SOCs via MR activation. Of note, acute aldosterone did not affect SOCE activated by Ca 2ϩ store depletion, strongly suggesting a genomic mechanism of aldosterone/MR on SOCs. Molecular screening using qRT-PCR and Western blotting demonstrated that aldosterone treatment for 24 h specifically increased mRNA and protein levels of Orai1, TRPC1, -C4, and -C5 through MR activation.
Using dominant negative pore mutants for TRPC1 and Orai1, TRPC4 mutant, or knockdown of TRPC5 by siRNA strategies, we confirmed that the aldosterone-increased SOCE is mediated by TRPC1, -C4, -C5, and Orai1 proteins. Our results echoed a previous study in metabolic syndrome adrenal chromaffin cells in which aldosterone increases TRPC1 and -C5 expression (55). A comparative analysis of the cardiac transcriptome of aldosterone-treated mice with cardiomyocytetargeted MR overexpression (MR-Cardio mice) and untreated MR-Cardio mice showed that aldosterone strongly regulates TRPC4 channel(s) (57). More recently, it has been shown that aldosterone significantly elevated the Orai1 transcript level and increased SOCE in UMR106 cells (58). Our study shows for the first time that aldosterone affects SOCE in cardiac myocytes. Importantly, blocking one of these molecular candidates is sufficient to blunt the aldosterone-induced increase of SOCE. This is probably due to the fact that the dn-Orai1 or dn-TRPC was shown to inhibit the function of all Orais or TRPC due to heteromultimerization. Similarly, we found that the co-overexpression of dn-TRPC1 and dn-Orai1, dn-TRPC4 and dn-Orai1 or dn-TRPC1, and dn-TRPC4 mutants did not have an addi- tional effect on SOCE, suggesting that Orai1 might function together with TRPC1 and -C4 and supporting the idea of the heteromultimerization process.
Orai1 has been shown to contribute to the ion permeability of Ca 2ϩ release-activated Ca 2ϩ channels, which has been reported to be independent of TRPC proteins (59,60). However, recent studies have proposed that Orai1 proteins might interact with TRPC proteins to form diverse SOCs in a vari-ety of cell types (61). For example, in human atrial myocytes, the SOCs are likely formed by TRPC1, STIM1, and Orai1 (51). It also has been established that activation of STIM1dependent TRPC1 and -C4 channels, via dynamic interaction, requires functional Orai1. Orai1 interacts with both TRPC4 and -C1 upon endoplasmic reticulum Ca 2ϩ store depletion, influencing the Ca 2ϩ selectivity of the TRPC1 and -C4 channels (52). A TRPC5/STIM1/Orai1 association in a stoichiometric manner to enhance SOCE also has been demonstrated (62).
We observed that NRVMs exhibit an endogenous BTP2-sensitive current activated by Ca 2ϩ store depletion (I SOC ) with a reversal potential near 0 mV and have current properties distinct from Ca 2ϩ release-activated Ca 2ϩ channels. Indeed, Ca 2ϩ release-activated Ca 2ϩ channels exhibit a highly Ca 2ϩ -selective inward current with a reversal potential near the equilibrium potential for Ca 2ϩ . Importantly, I SOC is significantly increased upon aldosterone treatment, confirming our Ca 2ϩ imaging data. The characteristics of SOCs in these cells are similar to those described before in other cell types (63,64) or in adult cardiomyocytes (65,66). Importantly, non-functional Orai1 mutant significantly reduced aldosterone-enhanced I SOC . This suggests that SOC function requires Orai1 and that TRPC and Orai1 might contribute to the same channel or distinct channels whereby the Orai1 function somehow regulates TRPC channel(s), conferring store-dependent activation of these channels. Co-immunoprecipitation experiments and I SOC properties are consistent with the first possibility because we found that TRPC1, -C4, -C5, Orai1, and STIM1 form a macromolecular complex following Ca 2ϩ store depletion, providing strong evidence that there is a close association between TRPC proteins and Orai1. These data show that dynamic assembly of TRPC-STIM1-Orai1 is involved in generating SOCE and concertedly regulates SOCE in NRVMs.
We further investigated the role of the SR-localized Ca 2ϩsensing protein STIM in activating SOCE. We observed that aldosterone increased protein expression of STIM1 and induced punctum-like formation. This suggests that aldosterone stimulates STIM1 oligomer accumulation into punctum structures at SR-plasmalemmal junctions. Untreated NRVMs expressing the native YFP-WT-STIM1 displayed a high level of SOCE compared with cells transfected with empty plasmid. By contrast, overexpression of native STIM1 in aldosteronetreated cells had no additional effect on SOCE, suggesting a maximal activation of the STIM1-dependent SOCE machinery after aldosterone treatment. Recent studies showed that gating of TRPC channel(s) by STIM1 is mediated by intermolecular electrostatic interaction between repeated Lys-rich domain of STIM1 and the TRPC C terminus (53). Expression of dominant negative STIM1-⌬K significantly decreased SOCE in untreated cells, indicating a requirement of STIM1 and its lysine-rich region to activate SOCE via TRPC channel(s) in cardiomyocytes. The inhibitory effect on SOCE of TRPC4 mutant, with a point mutation in the sequence essential for interaction with STIM1, strengthened the idea of a key role of STIM1 in SOCdependent signaling. By contrast, in aldosterone-treated cells, we observed a very small effect in enhanced SOCE measured with Ca 2ϩ imaging and a non-significant effect on I SOC. These data suggest that this domain is less involved in SOCE in the aldosterone condition. In addition, the fact that dn-Orai1 completely abolished aldosterone-dependent I SOC suggests that upregulation of Orai1 expression is of major importance in aldosterone-enhanced SOCE.
Orai1/3 as heteropentamers can act as arachidonate-regulated Ca 2ϩ channels, which are store-independent Orai channels (67). The fact that Orai3 gene was not affected by aldoster-one treatment strengthens the idea of the specific nature of the store-dependent Ca 2ϩ entry carried by TRPC1, -C4, -C5, and Orai1 channels. Of note, TRPC3, like receptor-operated channels, did not interact with TRPC1, -C4, -C5, and Orai1 and did not affect the aldosterone-enhanced SOCE, strengthening this idea.
Only high concentrations of aldosterone up-regulate SOCs, indicating an indirect effect of aldosterone/MR signaling in modulating these channels. We then inquired for a possible underlying mechanism. SGK1 was chosen as a plausible candidate due to its previous implication in the regulation of various ion channels, notably STIM1/Orai1 (54). Moreover, SGK1 expression is stimulated by mineralocorticoids (68). Recently, it has been shown that cardiac SGK1 is activated in human and murine HF and that cardiac-specific activation of SGK1 in mice increased mortality, cardiac dysfunction, and ventricular arrhythmias. Conversely, cardiac-specific inhibition of SGK1 protected mice after hemodynamic stress from fibrosis, HF, and induced Na ϩ channel alterations (69). In addition, cardiac SGK1 is activated early after pressure overload induced by transverse aortic, and acute activation promotes cardiomyocyte survival (70). Of interest, SGK1 expression (71) and activity (72) are stimulated by increased cytosolic Ca 2ϩ activity. Thus, SGK1 could serve as an amplifier of Ca 2ϩ entry. Our results demonstrated an up-regulation of SGK1 mRNA expression in NRVMs after aldosterone treatment. In addition, pharmacological inhibition of SGK1 by GSK650394 or silencing by specific SGK1 siRNA strongly reduced the enhanced SOCE. Interestingly, blockade of SGK1 prevented exclusively the increased Orai1 expression induced by aldosterone but not the up-regulation of TRPC1 and -C5. Similarly, recent observations uncovered that SGK1 up-regulates SOCE in both mast cells and platelets through regulation of Orai1 and STIM1 gene expression (54,58,73). Thus, these results support the idea of direct targets of MR for TRPC1 and -C5 by the presence of several glucocorticoid response element half-sites in the promoter region (55) and indirect targets of MR for Orai1 via SGK1.
Ca 2ϩ is an important regulator in many cardiomyocyte functions such as electrophysiological processes, excitation-contraction coupling, regulation of contractile protein activity, energy metabolism, and transcriptional regulation (74). Consequently, perturbation of its homeostasis leads to life-threatening disease, including hypertrophic cardiac remodeling, arrhythmias, and cell death. We were the first to show that the modulation of voltage-gated Ca 2ϩ channels at the plasma membrane (L-and T-type) and of the ryanodine receptor, the Ca 2ϩ release channel of the SR, is a central factor in the cardiac aldosterone/MR action and might be involved in triggered afterdepolarization-related fatal ventricular tachyarrhythmia (35)(36)(37)(38)(39). SOCE has also emerged as a potential mechanism to alter Ca 2ϩ in the cardiomyocyte. It is greatly recognized that Ca 2ϩ entry through TRPC1, -C3, -C4, and -C6 channels and STIM1, by changing the fetal gene program governed by calcineurin/NFAT signaling, which is a characteristic of stressed cardiomyocytes, contributes to the pathogenesis of hypertrophy and HF (11,14,15,(21)(22)(23)(24)(25)(26). Additionally, many studies speculate the potential deleterious effect of SOCs in atrial and ventricular arrhythmias (27,29,31,32). Interestingly, we and others have demonstrated that TRPC dysfunction could be responsible for the occurrence of ventricular tachycardia (27,28). In the current study, we showed a diastolic Ca 2ϩ increase during field stimulation in NRVMs induced by chronic aldosterone treatment. This increase is totally suppressed when SOCs are inhibited by BTP2 and when Orai1 is specifically blocked by S66 and partially reduced in the presence of the SGK1 inhibitor GSK, suggesting that SOCs contribute to the Ca 2ϩ handling at rest through an SGK1-dependent mechanism. Our findings provide direct experimental support for the prediction that SOCs play a key role in regulating cardiac diastolic Ca 2ϩ homeostasis.
Taken together, the current study underscores the importance of aldosterone/MR signaling as new regulatory element to exert its influence on TRPC1-, -C4-, -C5-, and Orai1-mediated SOCE in ventricular cardiomyocytes.