STIM1 Positively Regulates the Ca2+ Release Activity of the Inositol 1,4,5-Trisphosphate Receptor in Bovine Aortic Endothelial Cells

The endothelium is actively involved in many functions of the cardiovascular system, such as the modulation of arterial pressure and the maintenance of blood flow. These functions require a great versatility of the intracellular Ca2+ signaling that resides in the fact that different signals can be encoded by varying the frequency and the amplitude of the Ca2+ response. Cells use both extracellular and intracellular Ca2+ pools to modulate the intracellular Ca2+ concentration. In non-excitable cells, the inositol 1,4,5-trisphosphate receptor (IP3R), located on the endoplasmic reticulum (ER), is responsible for the release of Ca2+ from the intracellular store. The proteins STIM1 and STIM2 are also located on the ER and they are involved in the activation of a store-operated Ca2+ entry (SOCE). Due to their Ca2+ sensor property and their close proximity with IP3Rs on the ER, STIMs could modulate the activity of IP3R. In this study, we showed that STIM1 and STIM2 are expressed in bovine aortic endothelial cells and they both interact with IP3R. While STIM2 appears to play a minor role, STIM1 plays an important role in the regulation of agonist-induced Ca2+ mobilization in BAECs by a positive effect on both the SOCE and the IP3R-dependent Ca2+ release.


Introduction
The endothelium is the monolayer of cells lining the interior of blood vessels. The endothelial cells that constitute this monolayer are actively involved in many

Cell cultures
Bovine thoracic aortas from calves 8-10 months of age were obtained from a nearby slaughterhouse (Viandes Giroux, East Angus, QC). BAECs were isolated and characterized as previously described [16]. The cells were maintained in lowglucose DMEM containing 2 mM L-glutamine, 10% FBS, 100 U/ml penicillin, and 100 mg/ml streptomycin at 37˚C in a humidified atmosphere containing 5% CO 2 . They were used between the 5th and 20th passages. Experiments were approved by the ''Comité facultaire de protection des animaux'' of the Faculty of Medicine and Health Sciences of the Université de Sherbrooke.

Immunoprecipitation and Western blotting
Cells were washed twice with phosphate-buffered saline (137 mM NaCl, 2.8 mM KCl, 1.5 mM KH 2 PO 4 , 8 mM Na 2 HPO 4, pH 7.4) and solubilized for 30 min on ice in lysis buffer (50 mM Tris-HCl, pH 7.4, 150 mM NaCl, 1% Triton X-100, 1 mM EDTA, and Complete protease inhibitor cocktail). The lysates were clarified by centrifugation at 10 0006 g for 10 min. For the immunoprecipitation studies, identical amounts of protein from each sample were incubated overnight at 4˚C with 5 mg/ml of a specific antibody. The immune complexes were collected by incubating the mixtures with 50 ml (50% suspension) of protein A/G-agarose beads. Nonspecifically bound proteins were removed by washing the beads three times with 1 ml of lysis buffer, and bound material was solubilized in 50 ml of 26 Laemmli sample buffer [17], boiled for 5 min, and resolved by SDS-PAGE. The proteins were transferred onto polyvinylidene difluoride membranes, which were blocked for 1 h at room temperature with TBST buffer (20 mM Tris-HCl, pH 7.5, 147 mM NaCl, 0.1% Tween 20) containing 5% nonfat dried milk, and incubated with primary antibody overnight at 4˚C. The membranes were then incubated with horseradish peroxidase-conjugated anti-mouse or anti-rabbit secondary antibodies, and the immunoreactive proteins were visualized with an ECL detection system.

Immunofluorescence staining
Endothelial cells were seeded on 25 mm cover glasses in 6-wells plates and maintained in culture until they reached 60% confluence. Cells were washed with PBS and fixed with 100% methanol for 10 min at 220˚C. Non-specific sites were blocked with 2% BSA in PBS for 1 h at room temperature. After being washed, cells were incubated overnight at 4˚C with primary anti-STIM1 and anti-IP 3 R-1 antibodies (20 mg/ml) prepared in PBS. After three washes with PBS, cells were incubated for 1 h at room temperature with secondary AlexaFluor 488-conjugated goat antibodies against rabbit IgG (1:300 dilution) or AlexaFluor 594-conjugated goat antibodies against mouse IgG (1:300 dilution) (Molecular Probes, Eugene, OR). After extensive washing with PBS, cover glasses were mounted on microscope slides using Vectashield (Vector Laboratories, Burlingame, CA, USA) and examined on a Zeiss Axio Observer microscope (406 oil immersion objective). Pictures were obtained with a Zeiss Axiocam MRm camera using AxioVision LE software. In control experiments performed in parallel, no specific immunofluorescent staining was observed when primary antibodies were omitted.

Transfection
Six-well plates of BAECs were cultured to 70% of confluence. BAECs were transfected with 40 nM of siRNA using 0.2% of LipofectAMINE 2000 following the protocol provided by the manufacturer (Invitrogen). The cells were maintained in DMEM 10% FBS without antibiotics. The sequences of the sense and anti-sense small interfering RNAs against STIM1 are 59-CCAAGGAGCACAUGAAGAAdTdT-39 and 59-GGUUCCUCGUGUACUUCUUdTdT-39; and those against STIM2 are 59-UGACAAAGAUGGUGGAAUCUUdTdT-39 and 59-UUACUGUUUCUACCACCUUAGdTdT-39. Fresh medium was added 5 h posttransfection, and experiments were conducted 48 h after transfection.

Quantitative PCR
Total RNA was extracted from BAECs, transfected or not with appropriate siRNA oligonucleotide, using TRIzol. RNA quality and presence of contaminating genomic DNA was verified as previously described [18]. RNA integrity was assessed with an Agilent 2100 Bioanalyzer (Agilent Technologies, Mississauga, ON). Reverse transcription was performed on 2.2 mg total RNA with Transcriptor reverse transcriptase, random hexamers, dNTPs (Roche Diagnostics, Laval, QC), and 10 units of RNAseOUT (Invitrogen) following the manufacturer's protocol in a total volume of 20 ml. All forward and reverse primers were individually resuspended to 20-100 mM stock solution in Tris-EDTA buffer (IDT, Coralville, IA) and diluted as a primer pair to 1 mM in RNase DNase-free water (IDT). Quantitative PCR (qPCR) reactions were performed in 10 ml in 96 well plates on a CFX-96 thermocycler (BioRad, Mississauga, ON) with 5 ml of 26 iTaq Universal SYBR Green Supermix (BioRad), 10 ng (3 ml) cDNA, and 200 nM final (2 ml) primer pair solutions. The following cycling conditions were used: 3 min at 95˚C; 50 cycles: 15 sec at 95˚C, 30 sec at 60˚C, 30 sec at 72˚C. Relative expression levels were calculated using the qBASE framework [19] and the housekeeping genes ubiquitin (UBC), beta-2-microglobulin (B2M) and Ribosomal protein (RPLP0) for bovine cDNA. Primer design and validation was evaluated as described elsewhere [18]. In every qPCR run, a no-template control was performed for each primer pair and these were consistently negative. All primer sequences are shown in Table 1.

Dynamic video imaging of cytosolic Ca 2+
BAECs grown on glass coverslips were washed twice with HBSS (120 mM NaCl . Fluorescence from isolated fura-2-loaded cells was monitored by videomicroscopy using 340 nm and 387 nm excitatory wavelengths, and emitted fluorescence was recorded at 510 nm. Images were digitized and analyzed with the MetaFluor software (Universal Imaging Corporation, Downingtown, PA). The data are expressed as the intracellular free Ca 2+ concentration (nM) calculated from the 340/387 fluorescence ratio according to [20]. All experiments were performed at room temperature.

Data analysis
All experiments were performed at least three times. Results are expressed as mean ¡ standard deviation (SD). When needed, the data were analyzed using an analysis of variance (ANOVA), and pairwise comparisons were performed using Dunnet's test. In all cases, results were considered statistically significant when p,0.05 (*). The concentration response curve fitting were done using the R statistical software 2.14.1 (R Development Core Team, Vienna, Austria) and the EC50s and maximal responses were expressed as best fit ¡ standard error (SE).

STIM1 and STIM2 are expressed in BAECs and participate to SOCE
The qPCR analysis shown in Fig. 1D (left) reveals that the relative level of encoding STIM1 mRNA was efficiently lowered in siSTIM1 cells compared to siCtrl cells (0.27¡0.09 and 1.00¡0.08, respectively) without affecting the levels of STIM2 mRNA (1.03¡0.03 in siSTIM1 cells and 1.00¡0.09 in siCtrl cells). The relative level of STIM2 mRNA (Fig. 1D, right) was also efficiently decreased in siSTIM2 cells compared to siCtrl cells (0.37¡0.01 and 1.00¡0.09 respectively). The level of STIM1 mRNA was also slightly reduced in siSTIM2 cells (0.78¡0.06 in siSTIM2 cells and 1.00¡0.08 in siCtrl cells). Western blots shown in Fig. 1A indicate that under our experimental conditions, the proteins STIM1 (top panel) and STIM2 (bottom panel) are expressed in BAECs. Fig. 1A also shows that the level of STIM1 and STIM2 expression were efficiently (.90%) lowered by their respective siRNAs. Not targeting siRNA (siCtrl) did not alter STIM1 and STIM2 expression. To determine the functional consequence of STIM1 and STIM2 knockdown, the IP 3 -sensitive Ca 2+ pool content and the activation of SOCE were evaluated in intact BAECs. BAECs were bathed in a nominally Ca 2+ free medium and treated with 1 mM thapsigargin (a SERCA pump inhibitor). Thapsigargin increased the intracellular Ca 2+ to a similar level in BAECs transfected with siCtrl, siSTIM1 or siSTIM2 (Fig. 1B). The average peak amplitudes were 70.0¡5.2 nM, 76.0¡1.6 nM and 72.2¡7.7 nM, respectively (Fig. 1C, left). The subsequent addition of 1.8 mM extracellular Ca 2+ revealed that the SOCE was attenuated in cells transfected with siSTIM1 or siSTIM2, as compared to cells transfected with siCtrl (Fig. 1B). The average peak amplitude was 103.9¡9.1 nM in cells transfected with siCtrl and was significantly reduced to 78.6¡9.6 nM in cells transfected with siSTIM2 and nearly abolished to 11.3¡2.2 nM in cells transfected with siSTIM1 ( Fig. 1C, right). It is important to mention that under each condition, the basal intracellular Ca 2+ concentration was similar. The moderate reduction of STIM1 mRNA expression in siSTIM2 cells presumably contributed to reduce the SOCE in these cells. These results revealed that the knockdown of STIM1 or STIM2 did not alter the content of the IP 3 -sensitive Ca 2+ pool in BAECs but moderately (for STIM2) or strongly (for STIM1) affected their SOCE activity.

STIM1 and STIM2 co-immunoprecipitate with IP 3 Rs
To verify whether STIMs could functionally interact with IP 3 R under basal conditions, we first examined if their intracellular localization made this possible in BAECs. Fig. 2 shows the immunostaining obtained with anti-STIM1 and anti-IP 3 R-1 antibodies in untransfected and unstimulated BAECs. Using the anti-STIM1 antibody, the fluorescence was widely distributed throughout the cell with a higher intensity in the perinuclear region corresponding to the endoplasmic reticulum. The outer limits of the cell were not clearly defined, which indicates that the plasma membrane was not stained (left panel). Similar results were obtained with the anti-IP 3 R-1 antibody (central panel). The overlay image of the two staining clearly shows that STIM1 and IP 3 R-1 were mostly present in the same region of the endoplasmic reticulum and that their physical interaction was possible in a wide part of the cell (right panel). A co-immunoprecipitation approach was used to further verify whether these two proteins interact together. Isoform specific antibodies were used to precipitate the IP 3 R-1 from BAECs lysates and the presence of STIM1 and STIM2 in the resulting immune complex was verified with isoform specific antibodies.
The Western blots showed that both STIM1 and STIM2 interact with IP 3 R-1 (Fig. 3A). Considering the high level of STIM1 and STIM2 detected in the small fraction of BAECs lysates (3% v/v), and the relatively low level of STIM1 and STIM2 detected in the immune complex from the whole lysates, it must be concluded that a very small proportion of STIMs are implicated in these interactions. Nevertheless these results suggest that STIM1 and STIM2 physically interact with IP 3 R-1. To further confirm the presence of a physical interaction between STIMs and IP 3 R-1, BAECs lysates were immunoprecipitated with anti-STIM1 or anti-STIM2 antibodies and IP 3 R-1 was detected in these immunoprecipitates (Fig. 3C).

The knockdown of STIM1 dampens the IP 3 R-dependent intracellular Ca 2+ release in BAECs
We verified whether STIM1 and STIM2 influence the IP 3 R-dependent intracellular Ca 2+ release in intact BAECs. A videomicroscopic approach was used to monitor the intracellular Ca 2+ concentration following stimulation with ATP or bradykinin (BK), two well known Ca 2+ -mobilizing agonists in BAECs [21][22]. To focus exclusively on IP 3 R-dependent Ca 2+ release, the experiments were done  (Fig. 4B). These results show that the knockdown of STIM1 significantly reduced the peak amplitude of IP 3 R-dependent Ca 2+ release whereas the knockdown of STIM2 did not significantly alter IP 3 R-dependent Ca 2+ release in BAECs.
We repeated these experiments using increasing concentrations of ATP and reported graphically the mean peak amplitude obtained with each concentration, as shown in Fig. 4C. Nonlinear regression analysis furnished the concentrationresponse curve that best fitted these data, as shown in Fig. 4C. The curves clearly indicate that over the range of concentrations used, the cells transfected with siSTIM2 exhibited an IP 3 R-dependent Ca 2+ release similar to that of cells transfected with siCtrl. Actually, the two curves are nearly superimposable. However, cells transfected with siSTIM1 showed significantly lower Ca 2+ responses upon stimulation with high concentrations of ATP (.100 nM). The peak Ca 2+ response obtained with a maximal concentration of ATP was 206¡5 nM Ca 2+ in cells transfected with siCtrl, 205¡5 nM Ca 2+ in cells transfected with siSTIM2 and 138¡4 nM Ca 2+ (67% of control response) in cells transfected with siSTIM1. Concentration-response curves were also obtained using BK (Fig. 4D). As observed with ATP, cells transfected with siSTIM2 responded similarly to cells transfected with siCtrl, whereas cells transfected with siSTIM1 had a significantly lower Ca 2+ response upon stimulation with high concentrations of BK. The peak Ca 2+ response obtained with a maximal concentration of BK was 131¡4 nM Ca 2+ in cells transfected with siCtrl, 129¡6 nM Ca 2+ in cells transfected with siSTIM2 and 80¡5 nM Ca 2+ (62% of control response) in cells transfected with siSTIM1. These results also show that BK is less efficient than ATP to mobilize Ca 2+ . Indeed, in control cells, the maximal response obtained with BK corresponds to only 64% of the maximal response obtained with ATP. Interestingly, while the maximal response obtained with BK is 36% lower than that obtained with ATP, the reduction of the maximal response of cells transfected with siSTIM1 is similar with both hormones (about 35% lower than that of control cells). To illustrate the effect of the knockdown of STIM1 and STIM2 on the apparent affinities of both agonists, the data shown in Fig.4C and Fig. 4D were expressed as a function of the maximal response obtained under each condition. Fig. 4E and Fig. 4F show that the concentration-response curves nearly superimposed, indicating that the apparent agonist affinities were not affected by the knockdown of STIM1 or STIM2. ATP showed an EC50 of 65¡6 nM in cells transfected with siCtrl, of 69¡9 nM in cells transfected with siSTIM1 and of 64¡6 nM in cells transfected with siSTIM2. BK showed an EC50 of 1.0¡0.1 nM in cells transfected with siCtrl, of 1.0¡0.2 nM in cells transfected with siSTIM1 and of 1.2¡0.2 nM in cells transfected with siSTIM2. These results indicate that the knockdown of STIM1, but not that of STIM2, dampens the IP 3 R-dependent agonist-induced intracellular Ca 2+ release in BAECs without affecting the apparent affinity of the Ca 2+ -mobilizing agonists.

Discussion
In endothelial cells, both IP 3 R-dependent Ca 2+ release and SOCE contribute to shape the agonist-induced Ca 2+ response [15]. However, most of the work toward the characterization of the role of STIMs in endothelial cells has focused exclusively on Ca 2+ entry [9][10][11][12][13][14]. In the present study, we evaluated the contribution of STIMs on IP 3 R-dependent Ca 2+ release. We showed that STIM1 and STIM2 are expressed in BAECs, which is also the case in most cellular types including endothelial cells [9-10, 13-14, 23-24]. We further showed that, without affecting the amount of Ca 2+ available in the ER, the knockdown of STIM1 nearly abolished the SOCE while the knockdown of STIM2 resulted in a minor reduction of the SOCE. A strong abolition of the SOCE induced by the knockdown of STIM1 has also been reported in human umbilical vein endothelial cells (HUVEC and EA.hy926 cells), in porcine aortic endothelial cells (PAEC) and in mouse lung endothelial cells, thus confirming the crucial role of STIM1 in endothelial SOCE [9][10][11]14]. To the best of our knowledge, no quantification of the contribution of STIM2 to endothelial SOCE has been reported, probably because of the strong contribution of STIM1 that may mask the weak contribution of STIM2. However, we showed here that STIM2 contributes to a small fraction of SOCE in BAECs in the presence of native STIM1. These results are in accordance with numerous studies addressing the differential roles of STIM1 and STIM2 which, with the exception of rare specific cases, point toward a major role of STIM1 in SOCE (for a review on the subject see [6] and [25]. Some studies reported a role for STIM2 in the regulation of the basal cytosolic Ca 2+ concentration in different cell types including HUVEC [25][26]. In the specific case of HUVEC, the basal cytosolic Ca 2+ concentration was lowered by about 3 nM after silencing STIM2 [26]. In BAECs, we did not observe any significant change in the basal cytosolic Ca 2+ level after silencing STIM1 or STIM2.
The activity of IP 3 R is known to be finely tuned by a variety of interacting molecules and by the luminal or cytosolic Ca 2+ concentration [5,29]. Since the available Ca 2+ in the ER and the basal cytosolic Ca 2+ concentration did not change when STIM1 or STIM2 were silenced, the only way they can modulate the IP 3 Rdependent Ca 2+ release is via an interaction with a component that leads directly to Ca 2+ release via IP 3 R.
We showed that IP 3 R-1 and STIM1 share a similar pattern of distribution throughout BAECs, which is a sign of proximity between them. The wide distribution of these two proteins could allow the regulation of the Ca 2+ release at a cellular scale rather than only in focalized regions of the cells. Using coimmunoprecipitation studies, we demonstrated that STIM1 and STIM2 interact with IP 3 R-1. Another example of proximity between STIMs and IP 3 Rs has been provided in HEK293 cells where the formation of a protein complex composed of Orai1, STIM1, TRPC3, RACK3 and IP 3 R-1 is formed upon activation with ATP or Carbachol [27]. Also, in MDCK cells, an interaction of STIM1 with IP 3 R-3 has been observed upon activation with ATP [28]. More specifically in endothelial cells, STIM1 has been shown to be actively transported by microtubules to the front edge of migrating HUVEC were IP 3 R-mediated Ca 2+ release are also more frequently observed [29]. Together, these observations suggest that STIMs may interact with IP 3 R.
Our results also revealed that the knockdown of STIM1, but not that of STIM2, dampens the IP 3 R-dependent intracellular Ca 2+ release induced by ATP and BK in intact BAECs. Similar results were observed in HEK293 cells where the formation of a protein complex composed of Orai1, STIM1, TRPC3, RACK3 and IP 3 R-1 was shown to increase the IP 3 R-dependent Ca 2+ release [27]. Isolated from STIM1knockout mice, mast cells stimulated with IgE, embryonic fibroblasts stimulated with ATP and platelets stimulated with ADP, thrombin, or thromboxane A 2 showed a reduced intracellular Ca 2+ release [30][31][32]. These studies from independent laboratories suggest a role for STIM1 in the agonist-induced Ca 2+ release via IP 3 R.
Using concentration-response analyses, we demonstrated that the knockdown of STIM1, but not that of STIM2, reduced the maximal IP 3 R-dependent intracellular Ca 2+ release in BAECs stimulated with ATP or BK without affecting their apparent affinity. These results suggest that STIM1 exerts a direct effect on IP 3 R functionality. Recently, Santoso et al. [28] suggested a mechanism in which STIM1 binds to IP 3 R-3 to prevent the binding of polycystin-2 in MDCK cells. Located on the ER, polycystin-2 is a Ca 2+ channel activated by Ca 2+ that enhances the IP 3 -dependent Ca 2+ release when interacting with the IP 3 R [28,[33][34][35]. By a similar mechanism, STIM1 could also prevent the inactivation of IP 3 Rs by other molecules. An example of such a molecule is CIB1, a widely expressed Ca 2+ binding protein, that has been shown to bind and inactivate IP 3 R, thus reducing the number of channels available for activation by IP 3 [36]. Alternatively, STIM1 may also directly potentiate IP 3 R activity by stabilizing a conformation allowing an increased flux of Ca 2+ upon activation. Some interacting proteins or kinases have been shown to directly modulate IP 3 R gating by affecting its open probability (for a review see [2]). STIM1 localization and activity were suggested as key features to maintain the spatial and temporal dynamics of the Ca 2+ signal necessary to promote HUVEC migration [29]. A positive regulatory role of STIM1 on IP 3 R activity is compatible with such a mechanism. Further investigations are needed to establish precisely how STIM1 induces a positive effect on IP 3 R functionality in endothelial cells.
In conclusion, we showed that STIM1 and STIM2 are expressed in BAECs and that SOCE strongly depends on STIM1 and partially on STIM2. We also identified STIM1 and STIM2 as interacting partners for IP 3 R-1, which is a sign of proximity between STIMs and IP 3 R populations. Moreover, we demonstrated that STIM1, but not STIM2, is a positive regulator of IP 3 R in BAECs. The mechanism does not involve a change in the sensitivity of IP 3 R for IP 3, but the results rather suggest that STIM1 increases the efficacy of IP 3 R. Therefore, while the role of STIM2 appears to be minor, STIM1 plays an important role in the regulation of agonistinduced Ca 2+ mobilization in BAECs by a positive effect on both the SOCE and the IP 3 R-dependent Ca 2+ release.