Sigma-1 receptor (Sig-1R) is an intracellular chaperone protein with many ligands, located at the endoplasmic reticulum (ER). Binding of cocaine to Sig-1R has previously been found to modulate endothelial functions. In the present study, we show that cocaine dramatically inhibits store-operated Ca2+ entry (SOCE), a Ca2+ influx mechanism promoted by depletion of intracellular Ca2+ stores, in rat brain microvascular endothelial cells (RBMVEC). Using either Sig-1R shRNA or pharmacological inhibition with the unrelated Sig-1R antagonists BD-1063 and NE-100, we show that cocaine-induced SOCE inhibition is dependent on Sig-1R. In addition to revealing new insight into fundamental mechanisms of cocaine-induced changes in endothelial function, these studies indicate an unprecedented role for Sig-1R as a SOCE inhibitor.
Sigma-1 receptors (Sig-1Rs) are chaperone proteins residing at the interface between endoplasmic reticulum (ER) and mitochondria [1,2]; upon stimulation, they translocate to other cell regions [3,4]. Sig-1Rs interact with various targets, including ion channels, G protein-coupled receptors, tyrosine kinase receptors and kinases, to regulate cellular functions [3,5]. Stimulating factors include Sig-1R agonists and ER stress . Moreover, activated Sig-1Rs typically associate with inositol 1,4,5-trisphosphate receptor type 3, favouring an increase in cytosolic Ca2+ concentration [2,6].
Store-operated Ca2+ entry (SOCE) [7,8] is an important mechanism for Ca2+ influx, described in a variety of cells, including the endothelium [8,9]. Physiologically, SOCE is activated by depletion of intracellular Ca2+ pools following stimulation of plasma membrane receptors that couple to phosphoinositide hydrolysis and inositol 1,4,5-trisphosphate generation. These decreases in ER Ca2+ content are sensed by the ER Ca2+ sensor, stromal interaction molecule 1 (STIM1) [8,10], resulting in activation of Ca2+ channels in the plasma membrane termed Orai1, also known as Ca2+ release-activated Ca2+ channels [8,11].
All chemicals were from Sigma–Aldrich, unless otherwise mentioned. Cocaine hydrochloride was generously supplied by NIDA; NE-100 hydrochloride was from Santa Cruz Biotechnology.
RBMVEC from Cell Applications, Inc. were cultured in rat brain endothelial basal medium and rat brain endothelial growth supplement, according to the manufacturer's instructions (Cell Applications, Inc.). Cells were grown in T75 flasks coated with attachment factor (Cell Applications, Inc.) until 80% confluent. Cells were plated on round coverslips of 25 mm diameter coated with human fibronectin (Discovery Labware) for measurements of cytosolic Ca2+ and membrane potential and on six-well plates for Western blot analysis.
RBMVEC were transfected 24 h after plating with four unique 29mer shRNA constructs Oprs1 (Sig-1R) rat, with the following sequences: TF711128A (GCCATTCGGGACGATACTGGGCTGAGATT), TF711128B (ATCATCTCTGGCACTTTCCACCAGTGGAG), TF711128C (CCTGTTTCTGACTATTGTGGCGGTGCTGA), TF711128D (TTGCACGCCTCGCTGTCTGAGTACGTGCT), in pRFP-C-RS vector or scrambled negative control non-effective shRNA in pRFP-C-RS (OriGene Technologies), using Cytofect endothelial cell transfection kit (Cell Applications, Inc.). Throughout the manuscript, the Sig-1R shRNA constructs are abbreviated Sig-1R shRNA (A), Sig-1R shRNA (B), Sig-1R shRNA (C) and Sig-1R shRNA (D).
Western blot analysis
Whole-cell lysates were separated on Mini-PROTEAN TGX 4–20% gels (Bio-Rad) by SDS/PAGE followed by immunoblotting. Proteins were transferred to an Odyssey nitrocellulose membrane (Li-Cor Biosciences). After blocking with Odyssey blocking buffer, the membranes were incubated overnight with primary antibody against Sig-1R (OPRS1, rabbit polyclonal, 1:1000, OriGene Technologies). An antibody against β-actin (mouse monoclonal, 1: 10,000; Sigma–Aldrich) was used to confirm equal protein loading. Membranes were washed with TBS-Tween 20 (TBST) and incubated with the secondary antibodies: IRDye 800CW conjugated goat anti-rabbit IgG and IRDye 680 conjugated goat anti-mouse IgG (1 : 10000, 1 h at 21°C). Membranes were then washed in TBST and scanned using an Li-Cor Odyssey Infrared Imager and analysed using Odyssey software.
Cytosolic Ca2+ measurement
Intracellular Ca2+ levels were evaluated as previously described [15,16]. Briefly, cells were incubated with 5 μM Fura-2 AM (Molecular Probes) in Hanks’ balanced salt solution at 21°C for 1 h and washed with dye-free Hanks’ balanced salt solution. Coverslips were mounted in an open bath chamber (QR-40LP; Warner Instruments) on the stage of an inverted microscope Nikon Eclipse TiE (Nikon Inc.), equipped with a Perfect Focus System and a Photometrics CoolSnap HQ2 charge-coupled device camera (Photometrics). Ca2+ levels were determined ratiometrically using Fura-2 AM fluorescence (emission 510 nm), following alternate excitation at 340 and 380 nm. This was acquired at a frequency of 0.25 Hz, and the relative 340 nm/380 nm fluorescence ratio was further plotted as a function of time. Images were acquired/analysed using NIS-Elements AR software (Nikon, Inc.).
Measurement of membrane potential
The relative changes in membrane potential of single neurons were evaluated using bis-(1,3-dibutylbarbituric acid) trimethine oxonol, DiBAC4(3) (Invitrogen), a slow response voltage-sensitive dye, as previously described [15,16]. Upon membrane hyperpolarization, the dye concentrates in the cell membrane, leading to a decrease in fluorescence intensity, whereas depolarization induces the sequestration of the dye into the cytosol, resulting in an increase in the fluorescence intensity. RBMVEC were incubated for 30 min in HBSS containing 0.5 mM DiBAC4(3) and the fluorescence monitored at 0.17 Hz, excitation/emission: 480 nm/540 nm. Calibration of DiBAC4(3) fluorescence following background subtraction was performed using the Na+–K+ ionophore gramicidin in Na+-free physiological solution and various concentrations of K+ (to alter membrane potential) and N-methylglucamine (to maintain osmolarity). Under these conditions, the membrane potential was approximately equal to the K+ equilibrium potential determined by the Nernst equation. The intracellular K+ and Na+ concentration were assumed to be 130 and 10 mM, respectively.
Data are expressed as mean ± S.E.M. One-way ANOVA followed by post hoc analysis using Bonferroni and Tukey tests were used to evaluate significant differences between groups; P<0.05 was considered statistically significant.
RESULTS AND DISCUSSION
To determine the effect of cocaine on stored-operated Ca2+ entry, we used a classical protocol to induce SOCE. Briefly, ER Ca2+ stores were depleted in Fura-2-loaded cells using the sarcoplasmic/ER Ca2+ ATPase inhibitor thapsigargin (1 μM) in the absence of extracellular Ca2+ (Figure 1A). No differences in the resultant rise in cytosolic Ca2+ were observed in the presence or absence of cocaine (3 μM, 15 min) indicating no effect of cocaine on ER Ca2+ content. SOCE was then stimulated by the addition of Ca2+ (1 mM). Surprisingly, a ∼50% decrease in SOCE was observed in cocaine-pretreated cells (Figures 1A and 1B).
Cocaine inhibits SOCE in brain microvascular endothelial cells: pharmacological evidence of Sig-1R involvement
Since cocaine is a Sig-1R agonist , we assessed the possibility that Sig-1R mediates cocaine-induced SOCE inhibition. Pretreatment with the selective Sig-1R antagonists BD-1063 (10 μM, 20 min)  or NE-100 (3 μM, 20 min) [18,19], completely eliminated cocaine-induced SOCE inhibition (Figures 1A and 1B).
To exclude any bias due to potential effects of cocaine on membrane potential, we examined the effect of cocaine on membrane potential in RBMVEC loaded with a voltage-sensitive dye, DiBAC4(3) . We observed no changes in membrane potential in response to cocaine application (Figure 1C).
Finally, RBMVECs were transfected with four different Sig-1R shRNA constructs. Western analysis revealed that Sig-1R expression was decreased by 22.5±2.5% (n=3 independent experiments) after transfection with Sig-1R shRNA (B) (ATCATCTCTGGCACTT TCCACCAGTGGAG) and 28±4.3% (n=3) after transfection with Sig-1R shRNA (D) (TTGCACGCCTCGCTGTCTGAGTACGTGCT). Sig-1R shRNA (C) was less efficacious in knocking down Sig-1R, and (A) was ineffective; these constructs were not used in the subsequent functional studies. Fluorescence microscopic analysis of red fluorescent protein (RFP) expression revealed ∼60% transfection efficiency. Since the 22.5±2.5% and 28±4.3% knockdowns observed by Western analysis were based on whole populations, the actual knockdowns in transfected cells (selectively measured in SOCE assay) were 45.6% for Sig-1R shRNA (B) and 56.8% for Sig-1R shRNA (D). As depicted in Figures 2B and 2C, under these conditions, cocaine again failed to inhibit SOCE. Considered collectively, these findings reveal that cocaine inhibits SOCE in cerebral microvascular endothelial cells via Sig-1R activation.
Knockdown of Sig-1Rs prevents cocaine-induced SOCE inhibition
Cocaine is well defined as a modulator of endothelial function and has been shown to increase blood–brain barrier permeability [13,20] and facilitate transmigration of inflammatory leucocytes into the brain [21–23]. Involvement of Sig-1R in this effect has also been established [13,14]. What was not previously recognized, however, is the possibility that these effects of cocaine and Sig-1R on endothelial permeability and activation could be Ca2+-dependent. However, changes in Ca2+ concentration are well known to modulate a wide variety of endothelial functions  including barrier permeability , endothelial proliferation [9,26] and expression of adhesion molecules responsible for inflammatory cell recruitment . Hence, the current investigation provides a new role for SOCE modulation in cocaine- and Sig-1R-induced changes in endothelial function.
Based on the present findings, Sig-1Rs can be added to the growing list of endogenous molecules with the capacity to modulate SOCE . Importantly, Sig-1Rs appear to mediate SOCE termination, the mechanisms of which have remained elusive.
Although we have not defined this mechanism in the current study, there are several possibilities ranging from interfering with STIM–Orai interaction to kinase modulation. Indeed, Sig-1R has been shown to modulate the functions of p38 mitogen-activated protein kinase (MAPK) , extracellular-signal-regulated kinases (ERK)1/2  and calcineurin , all of which regulate STIM1 phosphorylation [32–34]. Moreover, both STIM1 [33,35] and Orai1 [36,37] are known to be negatively regulated by phosphorylation. Future investigations may provide further insight into the novel biochemical mechanism.
Cristina Brailoiu, Elena Deliu, Linda Console-Bram, Jonathan Soboloff and Eugen Brailoiu performed experiments and analysed data. Cristina Brailoiu and Elena Deliu drafted the manuscript. Mary Abood, Ellen Unterwald and Jonathan Soboloff analysed data and revised the manuscript critically for important intellectual content. Eugen Brailoiu designed the study. All authors reviewed the results, read and approved the final version of the manuscript.
This work was supported by the NIH [grant numbers DA035926 (to M.E.A.) and P30DA013429 (to E.M.U.)].