Circulating hormones stimulate the phospholipase Cβ (PLC)/Ca2+ influx pathway to regulate numerous cell functions, including vascular tone. It was proposed previously that Ca2+-independent phospholipase A2 (iPLA2)-dependent store-operated Ca2+ influx channels mediate hormone-induced contractions in isolated arteries, because bromoenol lactone (BEL), a potent irreversible inhibitor of iPLA2, inhibited such contractions. However, the effects of BEL on other channels implicated in mediating hormone-induced vessel contractions, specifically voltage-gated Ca2+ (CaV1.2) and transient receptor potential canonical (TRPC) channels, have not been defined clearly. Using isometric tension measurements, we found that thapsigargin-induced contractions were ∼34% of those evoked by phenylephrine or KCl. BEL completely inhibited not only thapsigargin- but also phenylephrine- and KCl-induced ring contractions, suggesting that CaV1.2 and receptor-operated TRPC channels also may be sensitive to BEL. Therefore, we investigated the effects of BEL on heterologously expressed CaV1.2 and TRPC channels in human embryonic kidney cells, a model system that allows probing of individual protein function without interference from other signaling elements of native cells. We found that low micromolar concentrations of BEL inhibited CaV1.2, TRPC5, TRPC6, and heteromeric TRPC1–TRPC5 channels in an iPLA2-independent manner. BEL also attenuated PLC activity, suggesting that the compound may inhibit TRPC channel activity in part by interfering with an initial PLC-dependent step required for TRPC channel activation. Conversely, BEL did not affect endogenous voltage-gated K+ channels in human embryonic kidney cells. Our findings support the hypothesis that iPLA2-dependent store-operated Ca2+ influx channels and iPLA2-independent hormone-operated TRPC channels can serve as smooth muscle depolarization triggers to activate CaV1.2 channels and to regulate vascular tone.
Circulating hormones, such as angiotensin II, histamine, endothelin, and catecholamines, regulate vascular tone. An excessive plasma concentration of these hormones has been associated with chronically elevated blood pressure (Sitter et al., 2004; Harris et al., 2008), a risk factor for stroke, kidney failure, and heart failure. In vascular smooth muscle cells (Fig. 1A), circulating hormones activate Gq/11 protein-coupled receptors that, in turn, stimulate phospholipase Cβ (PLC) activity. Activated PLC hydrolyzes phosphatidylinositol 4,5-bisphosphate (PIP2) into diacylglycerol and inositol trisphosphate (IP3). Whereas diacylglycerol stimulates protein kinase C, IP3 acts on the IP3 receptor in the endoplasmic reticulum, an intracellular Ca2+ store, and stimulates release of the stored Ca2+. Upon Ca2+ store depletion, a plasma membrane store-operated Ca2+ influx (SOC) channel is activated. In addition, a debated signal downstream of the PLC pathway stimulates receptor-operated Ca2+-permeable transient receptor potential canonical (TRPC) channels (Hofmann et al., 1999; Clapham, 2003; Beech, 2005; Ramsey et al., 2006). Cation influx via receptor- and store-operated channels depolarizes smooth muscle cells. Smooth muscle cell depolarization, in turn, activates dihydropyridine-sensitive L-type voltage-gated Ca2+ (CaV1.2) channels (Catterall, 2000; Moosmang et al., 2003) that provide further Ca2+ entry into the cells, thus resulting in smooth muscle cell contraction.
The role of CaV1.2 channels in regulating vascular tone is widely accepted, and inhibitors of CaV1.2 channels have been used as antihypertensive drugs for decades. However, less is known about the contributions of SOC and TRPC channels to hormone-activated Ca2+ influx in vascular smooth muscle cells. SOC channels are formed by the Orai proteins (Orai1–Orai3) (Hogan et al., 2010). Such SOC channels are highly Ca2+ selective (Dietrich et al., 2010a). However, the existence of a nonselective SOC channel in vascular smooth muscle cells also has been described, suggesting some heterogeneity of vascular SOC channels (Bolotina and Csutora, 2005; Li et al., 2008). Receptor-operated TRPC channels are highly homologous to the Drosophila transient receptor potential (TRP) channels that play a role in phototransduction, a PLC-dependent process (Liu et al., 2007). There are seven members in the TRPC subfamily, which are subdivided further into TRPC1/4/5 and TRPC3/6/7 subgroups (Clapham, 2003; Ramsey et al., 2006) on the basis of sequence homology. Smooth muscle cells predominantly express TRPC1 and TRPC6 channels (Albert et al., 2009; Dietrich et al., 2010a), and up-regulation of TRPC6 channel expression has been implicated in the pathogenesis of some types of hypertension (Yu et al., 2004).
The lack of selective antagonists for SOC and receptor-operated TRPC channels has slowed the progress of identifying the role of these channels in the hormone-activated contractions of blood vessels. The findings that Ca2+-independent phospholipase A2 (iPLA2) is activated upon store depletion and plays a key role during SOC channel activation (Fig. 1A) (Wolf and Gross, 1996; Wolf et al., 1997; Seegers et al., 2002) prompted the use of a specific inhibitor of iPLA2, E-6-(bromoethylene)tetrahydro-3-(1-naphthyl)-2H-pyran-2-one (bromoenol lactone, BEL; Fig. 1A), for defining SOC channel functions in various physiological and pathophysiological systems. BEL first was described as a selective mechanism-based inhibitor of iPLA2 (Balsinde et al., 1995). BEL is a “suicide” substrate of this enzyme, because its cleavage product covalently binds to a cysteine residue within the enzyme's catalytic pocket to irreversibly inhibit iPLA2 activity. BEL exhibits a >1000-fold selectivity for iPLA2 over the Ca2+-dependent cytosolic PLA2 (cPLA2) (Hazen et al., 1991). Because complete inhibition of iPLA2 was achieved only at a concentration of ∼10 μM (Balsinde et al., 1995), BEL is used commonly at concentrations of 3 to 25 μM or even higher (Jörs et al., 2006; Park et al., 2008; Johnson et al., 2009; McElroy et al., 2009; Ratz et al., 2009; Wong et al., 2010).
Phenylephrine is a potent vasoconstrictor and acts on the Gq/11 protein-coupled α1 adrenoceptor in aortic, mesenteric, cerebral, and carotid smooth muscle cells to stimulate PLC-dependent Ca2+ influx. It has been proposed that phenylephrine-induced Ca2+ influx is mediated solely by iPLA2-dependent SOC channels, because BEL pretreatment totally eliminated phenylephrine-induced contractions in mesenteric, cerebral, and carotid arteries (Park et al., 2008). Conversely, another report indicated that BEL did not affect phenylephrine contractions but inhibited acetylcholine-induced endothelium-dependent relaxation (Seegers et al., 2002). To clarify these discrepancies and establish if BEL inhibits other proteins involved in Ca2+ homeostasis, we assessed the effect of BEL on phenylephrine-, thapsigargin-, and KCl-induced responses in isolated rat aortic rings and investigated the effects of BEL on CaV1.2 and TRPC channels heterologously expressed in human embryonic kidney (HEK) cells. Our data indicate that BEL inhibits CaV1.2 channels and interferes with the receptor-operated mode of TRPC channel activation by inhibiting PLC activity.
Materials and Methods
Isometric Tension Measurements.
The animal protocol was approved by the Institutional Animal Care and Use Committee at the Indiana University School of Medicine. All of the experiments were carried out in accordance with the Declaration of Helsinki and/or with the Guide for the Care and Use of Laboratory Animals as adopted by the National Institutes of Health (Institute of Laboratory Animal Resources, 1996). Sprague-Dawley rats (3–6 weeks old) were purchased from Taconic Farms (Germantown, NY). The thoracic aorta was isolated quickly and immersed in ice-cold phosphate-buffered saline (Sigma-Aldrich, St. Louis, MO). After the removal of the adventitia, aortic segments were cut into 3-mm rings and mounted in organ baths (Emka, Falls Church, VA) containing oxygenated Krebs buffer maintained at 37°C. Both denuded and intact rings were used during the experiments. The composition of the Krebs buffer was as follows: 131.5 mM NaCl, 5 mM KCl, 2.5 mM CaCl2, 1.2 mM NaH2PO4, 1.2 mM MgCl2, 25 mM NaHCO3, and 10 mM glucose. Optimal length was determined by assessing contraction with 60 mM KCl added to the bath. Passive tension was increased in gram increments until there was <10% change in active tension developed in response to KCl. All of the drugs were added directly to the organ baths.
HEK Cell Culture and Transfections.
HEK cells were purchased from American Type Culture Collection (Manassas, VA) and cultured in Eagle's minimum essential medium supplemented with 10% fetal bovine serum. Cells were split one to two times per week. For patch-clamp experiments, HEK cells were seeded on poly-l-lysine-coated glass coverslips and transfected using the Lipofectamine LTX reagent (Invitrogen, Carlsbad, CA) in accordance with the manufacturer's recommendations. The following cation channel clones were studied: TRPC1 (GenBank accession number NM_011643), TRPC5 (GenBank accession number NM_009428.2), TRPC6 (GenBank accession number FJ205713), and CaV1.2 (GenBank accession number NM_001136522). CaV1.2 channels were coexpressed with β1a and α2-δ subunits (GenBank accession numbers: NM_001082279 and 012919, respectively).
In all of the electrophysiological experiments, the whole cell patch-clamp technique was used. Currents were recorded using an Optopatch amplifier (Cairn Research, Kent, UK) and digitized at a sampling rate of 1 kHz. The pCLAMP 10 software package was used for data analyses. The standard external solution contained: 145 mM NaCl, 2.5 mM KCl, 2 mM CaCl2, 1 mM MgCl2, 10 mM HEPES, and 5.5 mM glucose (pH 7.2). The intracellular solution contained the following: 125 mM CsMeSO3, 3.77 mM CaCl2, 2 mM MgCl2, 10 mM EGTA (yielding free [Ca2+] = 100 nM), and 10 mM HEPES (pH 7.2). The standard intracellular solution was supplemented with 2 mM ATP and did not contain CaCl2 for recording currents via voltage-gated Ca2+ channels. The NMDG+ solution contained: 150 mM NMDG-Cl, 10 mM HEPES, and 5.5 mM glucose (pH 7.2). The composition of the Ba2+ extracellular solution was as follows: 135 mM N-methyl-d-glucamine (NMDG)-Cl, 10 mM BaCl2, 10 mM HEPES, and 5.5 mM glucose (pH 7.2). We used a 100-μl perfusion chamber equipped with a manifold allowing the application of various compound-containing extracellular solutions directly into the bath. In this chamber, a full exchange of the bath solution can be achieved within ∼10 to 20 s at a perfusion rate of 3 ml/min.
In CaV1.2 channel-transfected HEK cells, voltage-gated calcium channels were stimulated by 200-ms voltage ramps from −100 to +100 mV from a holding potential of −60 mV. Ba2+ was used as a surrogate for Ca2+ to prevent Ca2+-induced Ca2+ channel inactivation, a characteristic property of CaV1.2 channels. To determine the effect of BEL on CaV1.2 channels, we compared the peak amplitudes of Ba2+ currents recorded during the voltage ramps in BEL- or dimethyl sulfoxide (DMSO)-pretreated CaV1.2 channel-expressing HEK cells. The IBa density was calculated by dividing the peak Ba2+ current by the cell capacitance. To determine the voltage dependence of the steady-state inactivation for CaV1.2 channels, we used a standard protocol consisting of a 5-s voltage prepulse to various potentials from −80 to +30 mV with 10-mV increments, followed by a 5-ms repolarization to −80 mV, and a 50-ms test pulse to +10 mV with a 15-s interval between episodes (Fig. 4A). We fitted the data to a Boltzmann-charge-voltage function f(V) = C + Imax/(1 + e[(V0.5 − V)/VC]), where V0.5 is the voltage of half-maximum inactivation) using the Levenberg-Marquardt search method and the sum of squared errors as the minimization approach (pCLAMP 10 software). All of the electrophysiological experiments were performed at room temperature (22–23°C).
A monochromator-based imaging system (TILL-Photonics, Martinsreid, Germany) equipped with a DU885 charge-coupled device camera (Andor Technology plc, South Windsor, CT) was used to monitor the fluorescence of green fluorescent protein (GFP) and fura-2. The GFP fluorescence was excited at 488 nm, whereas fura-2 fluorescence was excited at 345 and 380 nm. Emitted light was collected with a 510-nm long-pass filter. Data were analyzed using TILLvisION software.
PLC Activity Assay.
HEK cells were seeded in 12-well plates, grown to 60% confluence, and transiently transfected (2 μg per well) with the histamine receptor type 1 cDNA using Lipofectamine 2000 (Invitrogen) according to manufacturer's instructions. Cells were incubated overnight with 5 μCi/ml [myo-3H]inositol (American Radiolabeled Chemicals, St. Louis, MO) to label the membrane inositol lipids. The cells were washed three times with the standard external solution and incubated for 30 min in the presence of 10 mM LiCl (Jenkinson et al., 1994) to block inositol monophosphate hydrolysis, with or without the indicated concentrations of BEL or 25 μM 1-[6-[[17β-methoxyestra-1,3,5(10)-trien-17-yl]amino]hexyl]-1H-pyrrole-2,5-dione (U73122), a well established inhibitor of PLC (Calbiochem, San Diego, CA). The cells then were challenged with 40 μM histamine for 30 min, and the reaction was terminated by buffer aspiration and addition of ice-cold tricholoracetic acid (0.5 M). Water-soluble [3H]inositol-containing components then were extracted by addition of tri-n-octylamine/1,1,2-trichlorofluoroethane (1:1 ratio), and [3H]inositol phosphates were separated by ion exchange chromatography (Berridge et al., 1982) using Dowex resin (Sigma-Aldrich) in the formate form. Glycerophospholipids were removed by elution with 40 ml of 0.1 M ammonium formate/0.1 M formic acid. The inositol phosphate species then were eluted with 10 ml of 1.2 M ammonium formate/0.1 M formic acid. Ultima-Flo (PerkinElmer Life and Analytical Sciences, Waltham, MA) was added to the eluate, and 3H content was determined using liquid scintillation counting. Data are expressed as the percentage of histamine response in the absence of inhibitors after the subtraction of basal counts.
Before BEL (a racemic mixture; Sigma-Aldrich) pretreatment, rat aortic rings or HEK cells were washed three times with the standard extracellular/bath solution to remove any residues of the serum. BEL was dissolved in DMSO. The stock solution of BEL was stored in small aliquots at −80°C. A fresh aliquot was used for each experiment. BEL was diluted in an appropriate volume of the standard extracellular/bath solution just before each treatment. For vehicle control experiments, the same volume of pure DMSO was added to the standard extracellular solutions. All of the pretreatments were carried out with 100 nM to 100 μM BEL at 37°C for 30 min in the dark. After BEL pretreatment, aortic rings or HEK cells were washed three times with the standard extracellular/bath solution and used immediately for isometric tension or patch-clamp experiments. In some experiments, BEL was added acutely to the bath during a recording.
To determine the IC50 values, the normalized data were fitted to a four-parameter logistic function: F(x) = min + (max − min)/(1 + 10(logIC50 − x) × Hill slope), where min indicates the minimum of F(x), max indicates the maximum of F(x), IC50 indicates median inhibitory concentration, and Hill slope characterizes the slope of the curve at its midpoint.
Data sets were compared using the Student's t test followed by the Mann-Whitney rank sum test or one-way analysis of variance followed by Dunn's post hoc test. The significance level was set to 0.05. Data were expressed as mean ± S.E.M.
BEL Inhibits Thapsigargin-, Phenylephrine-, and KCl-Induced Contractions.
Using isometric tension measurements, we determined that KCl and phenylephrine stimulated strong contractions (1.9 ± 0.4 and 2.1 ± 0.5 g, respectively; Fig. 1, B and C) in intact and denuded rat aortic rings. Isradipine (10 μM; Sigma-Aldrich), a dihydropyridine derivative that potently inhibits voltage-gated CaV1.2 channels, significantly decreased phenylephrine-induced contractions by 67% (Fig. 1C), indicating that voltage-gated Ca2+ channels mediate a significant portion of Ca2+ entry necessary for generating aortic ring contractions. After rat aortic rings were incubated with 25 to 30 μM BEL for 30 min, KCl- and phenylephrine-induced contractions were decreased drastically (0.1 ± 0.1 and 0.03 ± 0.01 g, respectively; Fig. 1, D–F). Figure 1F shows the dose-response curves for phenylephrine-induced contractions in rat aortic rings pretreated with various concentrations of BEL. BEL (25 μM) markedly and irreversibly decreased phenylephrine-induced contractions at all of the tested phenylephrine concentrations. To quantify the effects of various BEL concentrations on 10 μM phenylephrine-induced contractions, the data were fitted to the four-parameter logistic function (under Materials and Methods). The fit yielded an IC50 value of 7.0 ± 0.1 μM with a Hill slope of −1.4 ± 0.1 (Fig. 1G).
Thapsigargin, an irreversible inhibitor of the sarcoplasmic/endoplasmic reticulum Ca2+ ATPase pump that causes Ca2+ store depletion and vascular SOC channel activation, also evoked contractions of rat aortic rings that were 34% of those observed with phenylephrine in denuded rings (0.6 ± 0.6 g; Fig. 1C). Notably, thapsigargin-induced contractions developed more slowly (Fig. 1E, inset) than phenylephrine- or KCl-induced contractions. In agreement with previous reports (Park et al., 2008), BEL completely and irreversibly blocked thapsigargin-induced contractions (Fig. 1, D and E).
BEL Inhibits CaV1.2 Channels.
Thus, we found that the isradipine-sensitive voltage-gated Ca2+ channels mediated a significant portion of phenylephrine-induced contractions (see above and Fig. 1C). KCl depolarizes smooth muscle cells in the vascular wall and stimulates voltage-gated Ca2+ channels in a hormone-independent manner. KCl-induced (60 mM) ring contractions were potently inhibited in BEL-pretreated rings (Fig. 1E). We determined that BEL inhibited KCl-induced contractions with an IC50 value of 16.3 ± 1.3 μM and a Hill slope of −1.7 ± 0.2 (Fig. 2). Because the removal of BEL from the bath did not restore the KCl-induced contractions in BEL-treated rat aortic rings, we concluded that BEL irreversibly inhibited KCl-induced contractions (Fig. 2C). In the presence of 30 μM BEL, the residual KCl-induced contractions were not sustained, as was seen in control rings, but decayed slowly to the baseline (Fig. 2C).
We then hypothesized that dihydropyridine-sensitive, voltage-gated Ca2+ channels may be a target of BEL. Vascular smooth muscle cells express CaV1.2 channels (under Introduction). To investigate whether BEL affects CaV1.2 channels, we heterologously coexpressed the CaV1.2 (α1c-b) channel together with β1a and α2-δ auxiliary subunits in HEK cells and investigated the channel's sensitivity to BEL pretreatments. In these experiments, we used Ba2+ instead of Ca2+ in the extracellular solution to reduce the Ca2+-dependent inactivation of CaV1.2/β1a/α2-δ channels (the Ba2+ extracellular solution, under Materials and Methods). Voltage-gated Ba2+ currents were recorded during a single 200-ms voltage ramp recorded 2 min after establishing the whole-cell configuration. No voltage-activated Ba2+ currents were observed in nontransfected HEK cells (Fig. 3A). Conversely, the CaV1.2/β1a/α2-δ channel-transfected cells exhibited large voltage-gated Ba2+ currents with a peak current density of −34.4 ± 4.6 pA/pF activated during voltage ramps from −100 to +100 mV in DMSO (vehicle)-pretreated cells (Fig. 3). The current-voltage relationships of the Ba2+ currents reached a maximum amplitude at 6.1 ± 2.0 mV (Fig. 3A). BEL (1 μM) pretreatment reduced the Ba2+ currents through CaV1.2/β1a/α2-δ channels (−26.0 ± 4.6 pA/pF), and 25 μM BEL pretreatment almost completely eliminated the Ba2+ currents through CaV1.2/β1a/α2-δ channels (−4.3 ± 1.2 pA/pF; Fig. 3, A and B). In the CaV1.2/β1a/α2-δ channel-expressing HEK cells pretreated with 25 μM BEL, the current-voltage relationships of the Ba2+ currents reached a maximum amplitude at 15.8 ± 2.2 mV (Fig. 3A). A fit of the BEL dose-inhibition curve to the four-parameter logistic function yielded an IC50 value of 7.6 ± 0.3 μM with a Hill slope of −1.4 ± 0.1 (Fig. 3B).
We next investigated whether BEL affected the voltage dependence of the steady-state inactivation of CaV1.2/β1a/α2-δ channels. Figure 4 shows the Ba2+ currents in DMSO- or 1 μM BEL-pretreated cells recorded using the standard protocol to assess the steady-state inactivation. Fits of the Ba2+ current decay data to the standard two-exponential function yielded the following time constants: 1) the τfast constants were 237.5 ± 48.5 ms (DMSO) and 88.4 ± 4.6 ms (BEL); and 2) the τslow constants were 1428.4 ± 314.4 ms (DMSO) and 406.0 ± 29.2 ms (BEL). Thus, the BEL pretreatment significantly accelerated both the fast and the slow components of the Ba2+ currents (Fig. 4A). The fact that the Ba2+ currents decayed much faster in the BEL-pretreated cells is consistent with our finding that BEL pretreatment facilitated the rate of ring relaxation in KCl-precontracted aortic rings (Fig. 2, B and C). The steady-state inactivation curves were fitted to the Boltzmann charge-voltage function as described under Materials and Methods. The fits yielded the following values for the voltages of half-maximum inactivation: −21.5 ± 0.6 mV in DMSO-pretreated CaV1.2/β1a/α2-δ channel-expressing cells (n = 4) and −39.2 ± 1.6 mV in BEL-pretreated CaV1.2/β1a/α2-δ channel-expressing cells (n = 3; Fig. 4B).
BEL Inhibits TRPC Channels.
Smooth muscle cells also express TRPC1, TRPC5, and TRPC6 channels (Albert et al., 2009; Edwards et al., 2010), which have been implicated in mediating receptor-operated contractions of blood vessels. Because BEL inhibits hormone-induced vessel contractions, we hypothesized that the drug also may block hormone-activated TRPC channels. To test this hypothesis, we expressed TRPC channels in HEK cells and assessed the ability of BEL to modulate the channel activity. No histamine-operated currents were observed in untransfected HEK cells (Supplemental Fig. 1) (Schaefer et al., 2000; Obukhov and Nowycky, 2004, 2008). In HEK cells expressing TRPC6 channel and the H1 histamine receptor (Fig. 5A), histamine activated robust inward currents exhibiting a mean current density of −24.8 ± 17.2 pA/pF with an S-shaped current-voltage relationship, a feature of TRPC6 channels (Hofmann et al., 1999). These currents were abolished in the solution containing bulky NMDG+ as the only cation, indicating that they are carried by small cations (Fig. 5A). Pretreatment with BEL at 25 and 100 μM concentrations dose-dependently inhibited histamine-elicited currents in TRPC6 channel-expressing HEK cells (−4.5 ± 8.7 and −0.5 ± 0.6 pA/pF, respectively) measured at a holding potential of −60 mV (Fig. 5). A fit of the dose-inhibition curve to the four-parameter logistic function yielded an IC50 value of 8.1 ± 1.6 μM with a Hill slope of −1.6 ± 0.6 (Fig. 5B, inset). Transiently expressed TRPC6 channels exhibited slight spontaneous activity (Fig. 5A, indicated as SA), which was noted as small currents with S-shaped TRPC6 channel-like current-voltage relationships in the absence of histamine. Interestingly, spontaneous activity of TRPC6 channels (−1.7 ± 1.9 pA/pF) was not affected significantly by BEL pretreatments at either 25 or 100 μM concentrations (Fig. 5B). This suggests that an upstream molecule involved in TRPC6 channel activation rather than the channel itself most likely was inhibited by BEL pretreatment.
We also investigated whether homomeric TRPC5 and heteromeric TRPC1–TRPC5 channels (TRPC1 and TRPC5 channel cDNA were a gift from Dr. Michael Schaefer, Rudolf-Boehm-Institut für Pharmakologie und Toxikologie, Leipzig, Germany) are sensitive to BEL pretreatment. These channels also were found in some vascular smooth muscle cells (Albert et al., 2009). The current density of histamine-activated TRPC5 channel currents was much larger than that of TRPC1–TRPC5 channel currents (−80.7 ± 45.9 and −12.8 ± 7.1 pA/pF, respectively). TRPC5 channel currents exhibited S-shaped current-voltage relationships, whereas TRPC1–TRPC5 channel current-voltage relationships were outwardly rectifying. Both TRPC5 and TRPC1–TRPC5 channel currents also were abolished in the solution containing NMDG+ as the only cation, indicating that the currents are carried by small cations (Figs. 6 and 7). BEL (25–100 μM) pretreatments significantly inhibited histamine-activated TRPC5 and TRPC1–TRPC5 channel currents as summarized in Figs. 6B and 7B. Fits of the dose-inhibition curves to the four-parameter logistic function yielded IC50 values of 10.6 ± 0.5 and 7.2 ± 0.7 μM with Hill slopes of −2.4 ± 0.2 and −2.2 ± 0.4 for TRPC5 and TRPC6 channels, respectively (Figs. 6B and 7B, insets). The Hill coefficient value of ∼2 suggests that there may be some cooperativity of BEL effects. As with TRPC6 channels, the spontaneous activity of TRPC5 channels was not affected by 25 μM BEL, yet, unlike TRPC6 channels, a higher concentration of BEL (100 μM) slightly potentiated the spontaneous activity of TRPC5 channels (Fig. 6B). Spontaneous currents through the TRPC1–TRPC5 heteromeric channel were inhibited significantly by 100 μM BEL (Fig. 7B).
Potent inhibition of TRPC channel currents was observed after a 30-min preincubation with BEL. We also examined whether an acute application of 100 μM BEL affects TRPC channel currents (Fig. 8). We found that acutely applied 100 μM BEL facilitated the rate of natural decay of the histamine-activated TRPC5 and TRPC6 channel currents. TRPC channel currents decayed to the steady-state level within 1 to 2 min in the presence of BEL. To determine the rate of solution exchange in our perfusion chamber, we determined how fast removing the current carriers from the bath, by substitution of the NMDG+ solution for the standard extracellular solution, would eliminate the histamine-activated inward currents in TRPC channel-expressing HEK cells. We found that the histamine-activated inward currents were eliminated completely in NMDG+ solution in much less time (∼10–20 s; Figs. 5A, 6A, and 8) than we observed during the acute application of 100 μM BEL (Fig. 8). This suggests that BEL may inhibit TRPC channel activity within a 5-min interval.
BEL Inhibits TRPC Channels in an iPLA2-Independent Manner.
Because BEL is reported to be a specific iPLA2 inhibitor, we reasoned that PLA2 may be required for TRPC channel activation. This hypothesis was supported by the fact that lysophospholipids, the products of PLA2, were reported to stimulate some TRPC channels under certain conditions (Fig. 1) (Beech et al., 2009). Therefore, we next investigated whether other iPLA2 inhibitors affect TRPC channel activity. We used a potent, broad-spectrum cPLA2 and iPLA2 inhibitor, palmitoyl trifluoromethyl ketone (PACOCF3). To replicate the BEL protocol, we preincubated the TRPC channel-expressing cells with 100 μM PACOCF3 for 30 min at 37°C before obtaining whole-cell patch-clamp recordings and then applied histamine in the presence of 100 μM PACOCF3. In control experiments, we preincubated the TRPC channel-expressing cells with DMSO (vehicle) for 30 min at 37°C, and then the TRPC channel currents were stimulated with histamine in the presence of DMSO. Figure 9 shows that the amplitudes of histamine-activated TRPC channel currents recorded in the presence or absence of 100 μM PACOCF3 were not significantly different.
BEL Does Not Inhibit Voltage-Gated Potassium KV Channels in HEK Cells.
Previous studies demonstrated that BEL inhibits iPLA2-dependent SOC channels (Park et al., 2008). Data from the present investigation clearly show that BEL also blocks voltage-gated Ca2+ channels and TRPC channels. Thus, it is possible that BEL is a broad-spectrum cation channel inhibitor. To test this hypothesis, we investigated whether BEL inhibits activation of potassium KV channels in HEK cells. We found that 100 μM BEL did not significantly alter the amplitude of endogenous KV channels in HEK cells (Fig. 10). The lack of BEL effects on potassium channels suggests that BEL may be specific for Ca2+-permeable channels such as SOC, CaV1.2, and TRPC channels.
BEL Inhibits PLC Activity.
Spontaneous TRPC5 and TRPC6 channel currents were less sensitive to BEL than receptor-operated TRPC5 and TRPC6 channel currents, suggesting that a molecular element upstream of TRPC channel activation may be a target of BEL. Our finding that PACOCF3 does not affect histamine-activated TRPC channel currents suggests that iPLA2 is not required for the receptor-operated mode of TRPC channel activation under conditions that prevent store depletion. Another element that is activated upstream of TRPC channels in the presence of circulating hormones is PLC (Clapham, 2003; Ramsey et al., 2006). Importantly, PLC belongs to the same superfamily of phospholipases as iPLA2, a known target of BEL. Therefore, we reasoned that BEL also may inhibit PLC activity. To investigate the effect of BEL on PLC activity, we used a combination of fluorescence imaging techniques that allowed for the recording of PLC-dependent increases in intracellular Ca2+ levels and simultaneous tracking of histamine-activated translocation of a molecular PIP2/IP3 probe. The ratiometric Ca2+ imaging approach was used due to changes in HEK cell morphology in the presence of histamine. To track PIP2 turnover in live cells, we used the phospholipase Cδ PH domain-GFP fusion (GFP-PH) developed by Várnai and Balla (1998). The schematics in Fig. 11, A and B, show that GFP-PH binds PIP2 located in the plasma membrane of unstimulated HEK cells. Upon cleavage of PIP2 by PLC, GFP-PH translocates to the cytosol together with IP3, thus enabling the visualization of the PLC activity by tracking the distribution of GFP-PH fluorescence in live cells. We transfected HEK cells with the H1 histamine receptor and a 10 times lower amount of GFP-PH cDNA (a gift from Dr. Tamas Balla, National Institute of Child Health and Human Development, Bethesda, MD) and investigated whether BEL pretreatment alters the ability of GFP-PH to translocate to the cytosol in response to H1 histamine receptor stimulation. Figure 11, C and E, and Supplemental Fig. 2 show that histamine application induced robust translocation of GFP-PH to the cytosol in control HEK cells. Pretreatments with 25 μM BEL significantly inhibited GFP-PH translocation (Fig. 11G). Histamine-activated Ca2+ release from the intracellular Ca2+ store was not decreased significantly in this case (Fig. 11F). These data suggest that reduced IP3 production is still able to trigger IP3-dependent Ca2+ release from intracellular stores (Fig. 11F). We speculate that this increase in Ca2+ release is likely subsequently amplified by Ca2+-induced Ca2+ release. Figure 11, H and I, show that 100 μM BEL almost completely abolished both hormone-induced Ca2+ release and GFP-PH translocation. HEK cells pretreated with 100 μM BEL exhibited slightly elevated levels of intracellular Ca2+. This is consistent with our electrophysiological data indicating that HEK cells pretreated with 100 μM BEL had increased leak currents. The GFP-PH translocation data were used only for qualitative assessments of PLC-induced PIP2 hydrolysis, and the exact IC50 values for BEL were not calculated due to changes in cell morphology that resulted in brighter GFP fluorescence in images acquired during histamine application compared with that in control images.
Our fluorescence imaging experiments suggested that micromolar concentrations of BEL inhibit PLC activity. To exclude the possibility that BEL interferes with GFP-PH binding to PIP2 and to obtain a quantitative assessment of BEL effects on PLC activity, we used a direct assay of receptor-stimulated inositol phosphate generation. This biochemical assay of PLC activity measures the release of [3H]inositol phosphates into the cytosol of prelabeled HEK cells. Our data show that BEL inhibited histamine-induced inositol phosphate production in a concentration-dependent manner (Fig. 12). Data fit to the four-parameter logistic function yielded an IC50 value of 364.0 ± 34.2 nM and a Hill slope of −0.8 ± 1.2. Inhibition was partial, with maximum inhibition of 35% with 100 μM BEL. Partial inhibition (45%) also was observed with the classic PLC inhibitor U73122 in this assay. Thus, full inhibition of histamine-induced inositol phosphate formation was not achieved in this cell system. This may reflect the large receptor reserve due to the overexpressed histamine receptor (under Discussion). Taken together, these data demonstrate the ability of BEL to significantly inhibit PLC activity.
Our major finding is that BEL potently inhibits the recombinant voltage-gated CaV1.2/β1a/α2-δ channel and three TRPC channels (TRPC5, TRPC6, and TRPC1–TRPC5) but not voltage-gated potassium channels endogenously expressed in HEK cells. BEL inhibited TRPC channel activity in an iPLA2-independent manner. We were surprised to find that BEL inhibited only receptor-operated activity of TRPC channels but not the spontaneous activity of the channels. This suggests that the compound acts on a molecular target upstream of TRPC channels. While unraveling the mechanism of this effect, we found that BEL potently inhibits PLC (IC50 value of 384 nM), which is required for receptor-operated TRPC channel activation (Clapham, 2003; Ramsey et al., 2006).
Ca2+ influx is crucial for vessel constriction. Compelling evidence indicates that, in vascular smooth muscle cells, store depletion induces iPLA2 activity. This leads to the stimulation of store-operated Ca2+-permeable channels (Wolf and Gross, 1996; Wolf et al., 1997; Seegers et al., 2002; Smani et al., 2004; Bolotina and Csutora, 2005). Because BEL, a selective inhibitor of the cytosolic β-isoform of iPLA2 (Jenkins et al., 2003), totally inhibited store-operated channel activation, it has emerged as a vital tool for revealing the role of SOC channels during contractile responses in various vascular beds. BEL totally inhibited phenylephrine-induced constrictions in cerebral and mesenteric arteries. Therefore, it has been proposed that the iPLA2-dependent activation of the SOC channel is the major pathway mediating phenylephrine-induced Ca2+ influx in these arteries (Park et al., 2008). However, the precise role of iPLA2 in mediating vessel contractions is still debated, because isolated pressurized mesenteric arteries from iPLA2(−/−) knockout mice exhibited increased contractions to phenylephrine compared with the mesentery from the wild-type mice (Dietrich et al., 2010b). Conversely, it was found that endothelial SOC channel-dependent acetylcholine-induced dilation was reduced in the iPLA2 knockout model. Notably, the fact that iPLA2 is predominantly expressed in endothelial cells but not in smooth muscle cells (Wong et al., 2010) is consistent with the observation that the genetic ablation of iPLA2 does not inhibit phenylephrine-induced vessel contractions.
Our data also indicate that BEL inhibits SOC channels in rat aortic rings. However, we found that BEL also inhibits other receptor-operated channels that can mediate receptor-operated Ca2+ influx in aortic rings. In addition, submicromolar concentrations of BEL inhibited PLC activity (Fig. 12). Because both TRPC and SOC channels are PLC dependent, PLC inhibition may effectively prevent TRPC and SOC channel activation in intact vessels. However, we found that BEL inhibits TRPC channels with an IC50 value that is ∼17 times higher than that for PLC. One potential explanation for this, which also can explain the lower sensitivity of histamine-induced Ca2+ signals (Fig. 11, D–I), is that there is a large histamine receptor reserve in these cells. This would make it necessary to inhibit a very large proportion of PLC before there is a significant effect on the downstream signaling steps. Another potential consideration is that TRPC channel activation involves integration of multiple signaling pathways (Schaefer et al., 2000), some of which may be less sensitive to BEL. The discrepancy in IC50 values between assays therefore may arise from differences in BEL efficacies for individual PLC subtypes, signaling pathways, and the histamine receptor reserve.
The mechanism by which BEL inhibits PLC is not yet clear. It is possible that BEL may be a weak substrate of PLC, as is the case for iPLA2 (Song et al., 2006). Therefore, BEL may be cleaved in the PLC catalytic center. In such a case, a product of BEL cleavage may bind covalently to a residue within the catalytic center of PLC, thus inactivating the enzyme. Further studies will be required to test this hypothesis.
Our data indicate that BEL also inhibited KCl-induced contractions in the rat aorta. This finding contradicts the results of Park et al. (2008) and Guo et al. (2003), who found that BEL does not inhibit KCl-induced Ca2+ level increases and contractions of cerebral and mesenteric arteries. It is not clear why KCl-induced contractions of the rat aorta were more sensitive to BEL than cerebral and mesenteric arteries. However, our data on BEL inhibition of KCl responses in isometric tension aortic ring experiments (Figs. 1 and 2) are consistent with the fact that BEL potently inhibited heterologously expressed CaV1.2/β1a/α2-δ channels in HEK cells (Fig. 3). CaV1.2 channels are implicated in mediating Ca2+ influx in vascular smooth muscle cells. BEL pretreatment accelerated both fast and slow components of the CaV1.2 channel current decay. In addition, BEL pretreatment caused an 18-mV leftward shift of the steady-state inactivation curve compared with that of control cells (Fig. 4B), suggesting that the compound may decrease the steady-state availability of CaV1.2/β1a/α2-δ channels for activation at physiological potentials. At this time, it remains unclear whether BEL directly acts on the pore-forming CaV1.2 channel subunit or one of the CaV1.2 channel auxiliary subunits (β or α2-δ).
Our findings that BEL inhibits heterologously expressed TRPC channels are consistent with the results of Jörs et al. (2006), who found that 25 μM BEL inhibited Drosophila TRPγ heterologously expressed in HEK cells. TRPγ is highly homologous to mammalian TRPC channels and is activated by a signal downstream of PLC. However, Jörs et al. (2006) attributed the effect of BEL on TRPγ to the inhibition of iPLA2. In our experiments, we found that 100 μM PACOCF3, a potent, broad-spectrum cPLA2 and iPLA2 inhibitor, did not affect histamine-activated currents through TRPC5, TRPC6, and TRPC1–TRPC5 channels heterologously expressed in HEK cells (Fig. 7), suggesting that it is highly unlikely that PLA2 is required for TRPC channel activation. Data in this study showing that BEL inhibits PLC and TRPC channel activation argue against the proposal that SOC channels are the only Ca2+ influx channel involved in hormone-activated contractions in blood vessels. In fact, it is most likely that TRPC channels also contribute to hormone-induced vessel contractions.
BEL possesses a chiral center (see the BEL structure shown in Fig. 1A) and may exist as two enantiomeric isoforms, namely, S-BEL or R-BEL. We used a racemic mixture of BEL during the above-described experiments. However, BEL enantiomers may exhibit different potencies for their newly identified target channels. Future studies will be needed to determine whether individual BEL enantiomers can preferentially inhibit CaV1.2 channels, TRPC channels, or PLC activity.
In conclusion, we demonstrated that BEL potently inhibits two types of Ca2+-permeable cation channels, namely, CaV1.2 and TRPC channels, in an iPLA2-independent manner. The inhibitory effect of BEL on TRPC channels most likely is achieved by inhibition of PLC activity. We propose that, besides the iPLA2-dependent SOC channels, several iPLA2-independent hormone-operated TRPC channels also can serve as hormone-operated smooth muscle depolarization triggers to activate CaV1.2 channels and to regulate vascular tone.
Participated in research design: Thomas, Sturek, Tune, and Obukhov.
Conducted experiments: Chakraborty, Berwick, and Bartlett.
Contributed new reagents or analytic tools: Kumar.
Performed data analysis: Chakraborty, Berwick, and Bartlett.
Wrote or contributed to the writing of the manuscript: Thomas, Sturek, Tune, and Obukhov.
We thank Dr. Tamas Balla for the clone of the GFP-PH cDNA, Dr. Michael Schaefer for clones of TRPC1 and TRPC5 channels, Dr. Yasuo Mori for β1a and α2-δ cDNAs, Dr. Jean-Charles Schwartz for the histamine H1 receptor cDNA, and Drs. Franz Hofmann and Norbert Klugbauer for Cav1.2b (α1C-b) cDNA.
This work was supported by the National Institutes of Health National Heart, Lung, and Blood Institute [Grant HL083381].
Article, publication date, and citation information can be found at http://jpet.aspetjournals.org.
- phospholipase Cβ
- bromoenol lactone
- dihydropyridine-sensitive L-type voltage-gated Ca2+ channels
- Ca2+-dependent cytosolic phospholipase A2
- dimethyl sulfoxide
- green fluorescent protein
- the phospholipase Cδ PH domain-green fluorescent protein fusion
- human embryonic kidney
- inositol trisphosphate
- Ca2+-independent phospholipase A2
- palmitoyl trifluoromethyl ketone
- phosphatidylinositol 4,5-bisphosphate
- store-operated Ca2+ influx
- transient receptor potential
- transient receptor potential canonical
- Received May 6, 2011.
- Accepted July 26, 2011.
- Copyright © 2011 by The American Society for Pharmacology and Experimental Therapeutics