Plasma membrane large-conductance Ca2+-activated K+ (BKCa) channels and sarcoplasmic reticulum inositol 1,4,5-trisphosphate (IP3) receptors (IP3Rs) are expressed in a wide variety of cell types, including arterial smooth muscle cells. Here, we studied BKCa channel regulation by IP3 and IP3Rs in rat and mouse cerebral artery smooth muscle cells. IP3 activated BKCa channels both in intact cells and in excised inside-out membrane patches. IP3 caused concentration-dependent BKCa channel activation with an apparent dissociation constant (Kd) of ∼4 µM at physiological voltage (−40 mV) and intracellular Ca2+ concentration ([Ca2+]i; 10 µM). IP3 also caused a leftward-shift in BKCa channel apparent Ca2+ sensitivity and reduced the Kd for free [Ca2+]i from ∼20 to 12 µM, but did not alter the slope or maximal Po. BAPTA, a fast Ca2+ buffer, or an elevation in extracellular Ca2+ concentration did not alter IP3-induced BKCa channel activation. Heparin, an IP3R inhibitor, and a monoclonal type 1 IP3R (IP3R1) antibody blocked IP3-induced BKCa channel activation. Adenophostin A, an IP3R agonist, also activated BKCa channels. IP3 activated BKCa channels in inside-out patches from wild-type (IP3R1+/+) mouse arterial smooth muscle cells, but had no effect on BKCa channels of IP3R1-deficient (IP3R1−/−) mice. Immunofluorescence resonance energy transfer microscopy indicated that IP3R1 is located in close spatial proximity to BKCa α subunits. The IP3R1 monoclonal antibody coimmunoprecipitated IP3R1 and BKCa channel α and β1 subunits from cerebral arteries. In summary, data indicate that IP3R1 activation elevates BKCa channel apparent Ca2+ sensitivity through local molecular coupling in arterial smooth muscle cells.
Large-conductance Ca2+-activated K+ (BKCa) channels are expressed in a wide variety of cell types and consist of pore-forming α subunits and auxiliary β1–4 subunits that modify channel function (Lu et al., 2006). In arterial smooth muscle cells, β1 subunits are the principal molecular and functional BKCa channel auxiliary subunit isoform (Brenner et al., 2000; Plüger et al., 2000). Arterial smooth muscle cell BKCa channel activation causes membrane hyperpolarization, leading to a reduction in voltage-dependent Ca2+ channel activity, a decrease in intracellular Ca2+ concentration ([Ca2+]i), and vasodilation (Davis and Hill, 1999; Jaggar et al., 2000). In contrast, BKCa channel inhibition causes membrane depolarization, which activates voltage-dependent Ca2+ channels, leading to an [Ca2+]i elevation and vasoconstriction.
Global cytosolic [Ca2+]i is typically 100–300 nM, whereas BKCa channels are sensitive to micromolar [Ca2+]i (Jaggar et al., 2000; Pérez et al., 2001). In arterial smooth muscle cells, BKCa channels are activated by localized micromolar [Ca2+]i transients termed Ca2+ sparks (Nelson et al., 1995; Jaggar et al., 2000). Ca2+ sparks are generated by the concerted opening of several SR ryanodine-sensitive Ca2+ release (RYR) channels (Nelson et al., 1995; Jaggar et al., 2000). Ca2+ spark–induced BKCa currents induce membrane hyperpolarization and vasodilation.
SR inositol 1,4,5-trisphosphate (IP3)-gated Ca2+ release channels are also expressed in many different cell types, including arterial smooth muscle (Thrower et al., 2001; Morel et al., 2003; Foskett et al., 2007; Zhao et al., 2008; Zhou et al., 2008). In native and cultured vascular smooth muscle cells, IP3 receptor (IP3R) activation stimulates propagating intracellular Ca2+ waves and elevates global [Ca2+]i (Lee et al., 2002; Lamont and Wier, 2004; Wilkerson et al., 2006; Zhao et al., 2008). Three different IP3R isoforms (1–3) have been identified, each of which is encoded by a different gene (Ross et al., 1992; Blondel et al., 1993). Type 1 IP3Rs are the principal molecular and functional IP3R isoform mediating agonist and IP3-induced intracellular Ca2+ signals in aortic and cerebral artery smooth muscle cells (Zhao et al., 2008; Zhou et al., 2008). IP3R2 also contributes to acetylcholine-induced Ca2+ oscillations in cultured portal vein smooth muscle cells (Morel et al., 2003). IP3R1 activation stimulates propagating intracellular Ca2+ waves and causes an increase in global [Ca2+]i (Zhao et al., 2008). IP3R1 activation also stimulates a nonselective cation current (ICat) via an SR Ca2+ release–independent mechanism in cerebral artery smooth muscle cells (Xi et al., 2008; Zhao et al., 2008). This cation current occurs due to physical coupling of IP3R1 to TRPC3 channels, is primarily due to Na+ influx, and leads to membrane depolarization, voltage-dependent Ca2+ channel activation, and a global [Ca2+]i elevation (Xi et al., 2008; Adebiyi et al., 2010). IP3R1-mediated SR Ca2+ release and TRPC3 channel activation both elevate [Ca2+]i, leading to vasoconstriction (Xi et al., 2008; Zhao et al., 2008).
IP3R-mediated SR Ca2+ release activates BKCa channels in basilar artery smooth muscle cells (Kim et al., 1998). Given that IP3Rs directly activate nearby TRPC3 channels in arterial smooth muscle cells (Xi et al., 2008; Zhao et al., 2008; Adebiyi et al., 2010), we studied the mechanisms by which IP3 and IP3Rs modulate BKCa channels. We tested the hypothesis that IP3 activates BKCa channels via an SR Ca2+ release–independent mechanism. Our data indicate that IP3 activates BKCa channels both in intact cells and excised membrane patches where SR Ca2+ release cannot occur. IP3 elevated BKCa channel apparent Ca2+ sensitivity, required IP3R1 activation, and was absent in IP3R1-deficient (IP3R1−/−) cells. Our data also indicate that IP3R1 is located in close spatial proximity to BKCa channels and coimmunoprecipitates with BKCa channel α and β1 subunits. This study identifies a novel signaling mechanism whereby IP3R1 activation stimulates nearby BKCa channels. Since IP3Rs and BKCa channels are broadly expressed, this coupling mechanism may exist in a wide variety of different cell types.
MATERIALS AND METHODS
Tissue preparation and cell isolation
All animal protocols were reviewed and approved by the University of Tennessee Animal Care and Use Committee. Male Sprague-Dawley rats (6–8 wk) and wild-type and IP3R1−/− mice (3 wk) were euthanized by intraperitoneal injection of 150 mg/kg sodium pentobarbital. Generation of the IP3R1 knockout mice will be described in a future publication. In brief, exon 5 of IP3R1 was flanked by two lox P sites, and IP3R1-floxed mice were generated by homologous recombination. IP3R1-floxed mice were subsequently crossed with protamine Cre mice (O’Gorman et al., 1997) to generate conventional IP3R1 knockout mice in which exon 5 is deleted globally and no IP3R1 protein can be detected (see Fig. 5 A). The phenotype of the global knockout IP3R1 mice generated is the same as that previously published for global loss of function of IP3R1 mice (Matsumoto et al., 1996). Brains were removed after the rats and mice were euthanized. Aorta was also collected from mice. Tissues were placed into ice-cold (4°C), oxygenated (21% O2, 5% CO2), physiological saline solution containing (in mM): 119 NaCl, 4.7 KCl, 24 NaHCO3, 1.2 KH2PO4, 1.6 CaCl2, 1.2 MgSO4, 0.023 EDTA, and 11 glucose. Posterior cerebral and cerebellar arteries (∼50–200 µm in diameter) were dissected from the brains and cleaned of connective tissue. Arterial smooth muscle cells were enzymatically dissociated from cerebral arteries as described previously, maintained at 4°C, and used within 8 h (Cheranov and Jaggar, 2006). Mouse aorta was used only for Western blotting experiments.
Patch clamp electrophysiology
Single BKCa channel currents were recorded in isolated cerebral artery smooth muscle cells using either the cell-attached or inside-out patch clamp configuration (Axopatch 200B and Clampex 8.2; MDS Analytical Technologies). For cell-attached patch, the pipette and bath solution contained (in mM): 130 KCl, 10 HEPES, 1 MgCl2, 5 EGTA, 1.6 HEDTA, and 10 µM free Ca2+, pH 7.2 with KOH. For inside-out patches, the same pipette and bath solutions were used, except for experiments measuring BKCa channel Ca2+ sensitivity, where free Ca2+ concentration was adjusted to between 1 and 300 µM by the addition of CaCl2 and free Mg2+ maintained at 1 mM by adjustment of MgCl2. Where indicated, equimolar EGTA was substituted for BAPTA, a fast Ca2+ chelator, in both the pipette and bath solutions. Free Ca2+ concentration in solutions was measured using a Ca2+-sensitive (no. 476041; Corning) and reference (no. 476370; Corning) electrode. Cell-attached and inside-out patch experiments were performed at membrane voltages of +60 and −40 mV, respectively. BKCa currents were filtered at 1 kHz and digitized at 5 kHz. Analysis was performed offline using Clampfit 9.2 (MDS Analytical Technologies).
Mouse aorta or rat cerebral artery proteins were separated using 7.5% SDS-polyacrylamide gels and transferred onto nitrocellulose membranes. Membranes were cut so that the same lysate could be probed for several different molecular weight proteins. Membranes were incubated with mouse monoclonal anti-IP3R1 (NeuroMab), mouse monoclonal anti–BKCa α (NeuroMab), or rabbit polyclonal anti–BKCa β1 (Abcam) primary antibodies overnight at 4°C in Tris-buffered solution (TBS) containing 0.1% Tween 20 (TBS-T) and 5% nonfat dry milk. After washing with TBS-T, membranes were incubated with horseradish peroxidase–conjugated secondary antibodies, followed by washing with TBS-T. Membranes were then developed using enhanced chemiluminescence (GE Healthcare), and digital images were obtained using a Kodak FX Pro imaging system.
Immunofluorescence resonance energy transfer (immuno-FRET)
Isolated cells were allowed to adhere to poly-L-lysine–coated coverslips. Cells were then fixed with 3.7% paraformaldehyde, permeabilized with 0.1% Triton X-100, and treated with the following primary antibodies: mouse monoclonal anti-IP3R1 (clone L24/18; NeuroMab) and either rabbit polyclonal anti-BKCaα (Abcam) or rabbit polyclonal anti-TRPM4 (Thermo Fisher Scientific), each at a dilution of 1:100. After washing, cells were incubated with the following secondary antibodies: Cy3-conjugated donkey anti–mouse for IP3R1 (Jackson ImmunoResearch Laboratories, Inc.) and Cy2-conjugated goat anti–rabbit (Jackson ImmunoResearch Laboratories, Inc.) for BKCaα or TRPM4. After washing, coverslips were dried and mounted onto glass slides. Fluorescence images were acquired using a laser-scanning confocal microscope (LSM Pascal; Carl Zeiss, Inc.). Cy2 and Cy3 were excited at 488 and 543 nm, and emission was collected at 505–530 and ≥560 nm, respectively. Negative controls were prepared by omitting primary antibodies. Images were background-subtracted, and N-FRET was calculated on a pixel-by-pixel basis for the entire image and in regions of interest (within the boundaries of the cell) using the Xia method (Xia and Liu, 2001) and LSM FRET Macro tool (v2.5; Carl Zeiss, Inc.).
Arterial lysate was harvested from cerebral arteries pooled from ∼15 rats using ice-cold lysis buffer (Thermo Fisher Scientific), giving ∼1.5 mg of total protein. coIP was done using the Thermo Fisher Scientific Co-Immunoprecipitation kit. The IP3R1 antibody was first immobilized for 2 h using coupling resin (AminoLink Plus; Thermo Fisher Scientific). The resin was then washed and incubated with arterial lysate overnight. After incubation, the resin was again washed and protein was eluted using elution buffer. Non-denaturing sample buffer (Thermo Fisher Scientific) was added to the eluate and boiled. Negative controls received the same treatment, except that the coupling resin was replaced with control agarose resin that is not amine-reactive. Samples were analyzed using Western blotting with mouse monoclonal anti-IP3R1, mouse monoclonal anti–BKCa α, or rabbit polyclonal anti–BKCa β1 primary antibodies, and horseradish peroxidase–conjugated secondary antibodies.
BKCa channel activity (NPo) was calculated from continuous gap-free data using the following equation: NPo = Σ (t1 + t2…ti), where ti is the relative open time (time open/total time) for each channel level. Open probability (Po) was calculated by dividing NPo by channel number (N). The total number of channels in inside-out patches was determined by introducing 1 mM of free Ca2+ into the bath solution at the end of each experiment. BKCa channel IP3 sensitivity data and relationships between BKCa channel open probability (Po) and free Ca2+ concentration were fit with a Boltzmann function: Y = Pomin + [(Pomax−Pomin)/(1+exp[(Kd-X)/slope])]. Values are expressed as mean ± SE. Student’s t test and repeated measures analysis of variance with Student-Newman-Keuls post-hoc test were used for comparing paired or unpaired data and multiple datasets, as appropriate. P < 0.05 was considered significant.
Online supplemental material
Data in Fig. S1 demonstrate that antigenic peptides abolish fluorescent labeling by BKCa channel α and TRPM4 channel antibodies. Also shown are differential interference contrast images of the same cells imaged for fluorescence. Fig. S1 is available at http://www.jgp.org/cgi/content/full/jgp.201010453/DC1.
IP3 activates BKCa channels in cerebral artery smooth muscle cells
BKCa channel regulation by IP3 was first measured in intact arterial smooth muscle cells using the cell-attached configuration of the patch clamp technique. Bt-IP3, a membrane-permeant IP3 analogue, at concentrations of 10 and 50 µM, increased mean BKCa channel activity (NPo) ∼1.8- and 1.9-fold, respectively, at +60 mV (Fig. 1, A and B).
Next, we investigated whether IP3 has effects on BKCa channels that are independent of plasma membrane Ca2+ influx or intracellular Ca2+ release. BKCa channel regulation was studied in excised inside-out membrane patches at a physiological steady membrane voltage of −40 mV, which is similar to that of cerebral arteries pressurized to 60 mmHg (Knot and Nelson, 1998). In inside-out patches exposed to symmetrical 10 µM of free [Ca2+]i, 10 µM IP3 increased mean BKCa channel open probability (Po) from ∼0.24 to 0.37, or 1.54-fold (Fig. 2, A and B). In contrast, IP3 did not alter single BKCa channel amplitude (pA: control, 8.5 ± 0.47; IP3, 8.0 ± 0.36; n = 5; P > 0.05; Fig. 2 A). IP3 also caused concentration-dependent BKCa channel activation (Fig. 2 B). Fitting an IP3 concentration–response curve with a Boltzmann function indicated that IP3 increased BKCa channel Po with an apparent Kd of ∼4.1 µM, a slope of ∼2.8, and a maximal Po of ∼0.38 (Fig. 2 B). IP3 activated BKCa channels in excised patches for as long as the seal could be maintained (>30 min), indicating that SR Ca2+ release was unlikely to be responsible for channel activation. In support of this concept, equimolar substitution of bath and pipette EGTA for BAPTA, a fast Ca2+ chelator, did not alter IP3-induced BKCa channel activation (Fig. 2 C). Elevating pipette free Ca2+ from 10 µM to 2 mM also did not alter IP3-induced BKCa channel activation, indicating that plasma membrane Ca2+ influx was not responsible for channel activation (Fig. 2 C). Collectively, these data indicate that IP3 activates BKCa channels via an SR Ca2+ release–independent mechanism.
IP3 elevates BKCa channel apparent Ca2+ sensitivity
IP3 regulation of BKCa channel apparent Ca2+ sensitivity was examined at a steady membrane voltage of −40 mV. In control, mean apparent Kd for Ca2+ was ∼20 µM, with a slope of ∼1.2, and a maximum Po of ∼0.82 (Fig. 3 A). In the same membrane patches, 10 µM IP3 decreased the mean Kd for Ca2+ to ∼12 µM, but did not alter the slope or the maximum Po (Fig. 3 A). Relative activation by IP3 increased considerably between 1 and 10 µM Ca2+ (Fig. 3 B). For example, with 1 µM of free Ca2+, IP3 increased mean BKCa channel Po 1.14-fold, whereas with 10 µM of free Ca2+, IP3 increased mean Po 1.48-fold (Fig. 3 B). At free [Ca2+] >10 µM, relative activation by IP3 became smaller because channel activity was approaching maximal. These data suggest that IP3 elevates BKCa channel apparent Ca2+ sensitivity in arterial smooth muscle cells.
IP3R1 activation mediates IP3-induced BKCa channel activation
To identify mechanisms by which IP3 activates BKCa channels, we tested the hypothesis that IP3Rs mediate IP3-induced BKCa channel activation. In inside-out patches, 1 mg/ml heparin, an IP3R blocker, reversed BKCa channel activation by IP3 (Fig. 4 A). In contrast, 1 mg/ml heparin alone did not alter mean BKCa channel Po (98.9 ± 12.5% of control; n = 5; P > 0.05). These data indicate that IP3-induced IP3R activation stimulates BKCa channels.
Type 1 IP3Rs are the principal molecular isoform expressed in cerebral artery and aortic smooth muscle cells (Zhao et al., 2008; Zhou et al., 2008; Adebiyi et al., 2010). Therefore, we studied whether IP3R1 mediates IP3-induced BKCa channel activation. In inside-out patches, a monoclonal IP3R1 antibody (1:100) reversed IP3-induced BKCa channel activation (Fig. 4, B and C). In contrast, the IP3R1 antibody (1:100) did not alter mean BKCa channel activity when applied alone (94.2 ± 4.9% of control; n = 4; P > 0.05). The addition of heparin in the presence of IP3 plus IP3R1 antibody did not cause any further reduction in BKCa channel Po (Fig. 4, B and C). Boiled (95°C for 15 min) IP3R1 antibody did not alter IP3-induced BKCa channel activation (Fig. 4 C). To further examine BKCa channel regulation by IP3Rs, we used adenophostin A as an alternate IP3R agonist. 1 µM adenophostin A increased BKCa channel activity ∼1.60-fold (Fig. 4 D). These data indicate that IP3R1 activation is necessary for IP3-mediated BKCa channel activation in arterial smooth muscle cells.
To further investigate the necessity for IP3R1 and to determine whether IP3-induced BKCa channel activation occurs in another species, we studied BKCa channel regulation in mouse wild-type (IP3R1+/+) and IP3R1 knockout (IP3R1−/−) cerebral artery smooth muscle cells. Western blotting confirmed that IP3R1 was present in IP3R1+/+ mouse aorta, but absent in IP3R1−/− mouse aorta (Fig. 5 A). BKCa channel Po (IP3R1+/+, 0.30 ± 0.08; IP3R1−/−, 0.27 ± 0.03; n = 5) and amplitude (pA: IP3R1+/+, 10.2 ± 0.8; IP3R1−/−, 11.4 ± 0.8; n = 5) were similar in inside-out patches from IP3R1+/+ and IP3R1−/− cerebral artery smooth muscle cells exposed to 10 µM Ca2+ (P > 0.05 for each). IP3 increased mean BKCa channel Po to ∼142% of control in inside-out patches from mouse IP3R1+/+ cells (Fig. 5 B). Furthermore, in patches from IP3R1+/+ cells, IP3-induced BKCa channel activation was reversed by heparin (Fig. 5 B). In contrast, IP3 or IP3 plus heparin did not alter BKCa channel Po in excised patches from IP3R1−/− arterial smooth muscle cells (Fig. 5 B). These data indicate that IP3R1 is essential for IP3-induced BKCa channel activation in cerebral artery smooth muscle cells.
BKCa channel α subunits colocalize with IP3R1
Our data indicate that IP3R1 and BKCa channels functionally interact. Therefore, spatial localization of these proteins was studied using immuno-FRET microscopy. Cy2- and Cy3-tagged secondary antibodies targeting primary antibodies bound to IP3R1 and BKCa channel α subunits, respectively, generated whole cell N-FRET of 20.4 ± 1.3% (n = 11; Fig. 6 A). In contrast, whole cell N-FRET between IP3R1 and TRPM4 channels, which do not colocalize in cerebral artery smooth muscle cells (Adebiyi et al., 2010), was significantly lower at 7.4 ± 1.0% (n = 10; P > 0.05; Fig. 6 A). N-FRET between IP3R1 and BKCa channels was observed both at the plasma membrane and intracellularly. Antigenic peptides abolished fluorescent labeling by BKCa and TRPM4 channel antibodies, respectively (Fig. S1). An antigenic peptide was not available for the monoclonal IP3R1 antibody, but this antibody detects only IP3R1 protein in a Western blot, indicating selectivity (Adebiyi et al., 2010). These data indicate that IP3R1 is located in close proximity to plasma membrane BKCa channels in arterial smooth muscle cells.
BKCa channel α and β1 subunits coimmunoprecipitate with IP3R1
coIP was performed to test the hypothesis that IP3R1 and BKCa channels are contained within the same macromolecular protein complex. Due to the small size of the resistance (100–200-µm diameter), cerebral arteries used in this study, arteries collected from ∼15 rats, were required for each coIP experiment. The monoclonal IP3R1 antibody coimmunoprecipitated IP3R1 with BKCa channel α and β1 subunits from cerebral artery lysate (Fig. 6 B). These data indicate that IP3R1 and BKCa channels are located within the same macromolecular complex in arterial smooth muscle cells.
Here, we demonstrate that IP3-induced IP3R1 activation stimulates BKCa channels via a local SR Ca2+ release–independent coupling mechanism in cerebral artery smooth muscle cells. Novel findings are that: (a) IP3 activates BKCa channels in excised membrane patches removed from cytosolic signaling pathways; (b) IP3 elevates BKCa channel apparent Ca2+ sensitivity; (c) IP3R1 expression and activation are required for IP3 to stimulate BKCa channels; and (d) IP3R1 and BKCa channel subunits are located in close spatial proximity and coimmunoprecipitate. These data identify a novel signaling mechanism whereby IP3R1 channels activate plasma membrane BKCa channels via an SR Ca2+ release–independent local coupling mechanism in arterial smooth muscle cells. Given that both IP3Rs and KCa channels are widely expressed, such communication may occur in other cell types.
Agonist binding to PLC-coupled receptors leads to phosphatidylinositol 4,5-bisphosphate cleavage and an elevation in both diacyglycerol and IP3 in smooth muscle cells. Diacyglycerol remains membrane bound and stimulates PKC, which phosphorylates a wide variety of target proteins that regulate contractility, including ion channels and the contractile apparatus (Davis and Hill, 1999; Jaggar et al., 2000). Relevant to this study, PKC inhibits arterial smooth muscle cell BKCa channels, leading to membrane depolarization and vasoconstriction (Davis and Hill, 1999; Jaggar et al., 2000). Arteries undergo steady-state changes in membrane potential, with a dynamic range between ∼−60 and −20 mV (Knot and Nelson, 1998; Davis and Hill, 1999). Here, BKCa channel regulation was studied primarily at a steady voltage of −40 mV, which is the membrane potential of cerebral arteries at a physiological intravascular pressure of 60 mmHg (Knot and Nelson, 1998). IP3 activated BKCa channels in both intact cells and in excised membrane patches. In excised patches, IP3 increased BKCa channel activity with an apparent Kd of ∼4 µM. Previous studies have estimated global intracellular IP3 concentration ([IP3]i) to be 0.1–3 µM in unstimulated cells and 1–20 µM in agonist-stimulated cells (Finch and Augustine, 1998; Luzzi et al., 1998; Takechi et al., 1998; Patel et al., 1999). Recent studies using fluorescent IP3 biosensors suggested that receptor agonists elevate global [IP3]i to ∼30 nM in cardiac myocytes and up to 700 nM in COS-7 and HSY-EA1 cells (Remus et al., 2006; Tanimura et al., 2009). In arterial myocytes, global [IP3]i is unclear, but local IP3 gradients higher than global [IP3]i should exist, particularly within the immediate vicinity of PLC where IP3 is generated. The [IP3]i nearby a target protein such as an IP3R will depend on many factors, including PLC activity, proximity of IP3Rs to PLC, and IP3 metabolism. KCa channel IP3 sensitivity determined here may indicate IP3 concentrations that would occur nearby IP3Rs that are located in close proximity to plasma membrane BKCa channels.
IP3 acts primarily by relieving Ca2+ inhibition of IP3Rs, thereby permitting Ca2+-induced channel activation (Thrower et al., 2001; Foskett et al., 2007). Physiological micromolar [Ca2+]i concentrations used here to study KCa channel activity would be expected to attenuate IP3-induced IP3R activation and may provide one explanation for the micromolar IP3 sensitivity of BKCa channel activation. IP3R IP3 sensitivity can also vary widely from nanomolar to micromolar depending on multiple factors in addition to [Ca2+]i, including IP3R isoform, splice variation, which can occur in many regions of the protein including the IP3-binding domain, and potentially through isoform heterotetramerization (Thrower et al., 2001; Foskett et al., 2007). For example, canine cerebellar IP3R1 is sensitive to IP3 over a broad concentration range and exhibits high (nM) and low (10 µM) affinity IP3-binding sites. The IP3 and Ca2+ sensitivity of arterial smooth muscle cell IP3Rs has not been determined, nor have these channels been cloned. Therefore, deriving mechanistic details regarding IP3R communication with KCa channels is difficult until detailed knowledge of smooth muscle cell IP3Rs is available, particularly IP3 and Ca2+ sensitivity. Our data indicate that under physiological voltage and local [Ca2+]i, micromolar [IP3]i is required for IP3Rs to activate BKCa channels.
Data indicate that IP3 does not directly activate BKCa channels. Rather, IP3R1 mediates IP3-induced BKCa channel activation. This conclusion is supported by our observation that heparin, an IP3R1 antibody, and IP3R1 ablation all blocked IP3-induced BKCa channel activation. These data also indicate that functional IP3R1 protein is excised together with BKCa channels in inside-out patches. The Förster distance between Cy2 and Cy3 is 5–6 nm. If Cy2 and Cy3 are located in such local proximity, nonradiative dipole–dipole coupling between excited Cy2 (donor) and Cy3 (acceptor) leads to Cy3 emission. Thus, for FRET to occur, IP3R1 must be located in very close spatial proximity to BKCa channels. Supporting local interaction between these proteins, coIP data indicated that IP3R1 and BKCa channel α and β1 subunits were located in the same macromolecular complex. In contrast to FRET data indicating close spatial proximity of IP3R1 and BKCa channels, immuno-FRET data indicated that IP3R1 and TRPM4 channels are not spatially localized in arterial smooth muscle cells, in agreement with a recent report (Adebiyi et al., 2010). BKCa channels are also located nearby IP3Rs in cultured glioma cells, but in contrast to the observations made here, glioma cell IP3Rs activate BKCa channels via Ca2+ signaling and direct molecular coupling is absent (Weaver et al., 2007). Electron microscopy studies have shown that the SR and plasma membranes can be located in very close proximity (∼20 nm) in arterial smooth muscle cells (Devine et al., 1972). Conceivably, IP3R1 and BKCa channels may be present within macromolecular complexes that bridge the SR and plasma membranes, allowing local molecular communication between these proteins. Immuno-FRET between IP3R1 and BKCa channel α subunits was observed both at the plasma membrane and intracellularly. The physiological function of close localization between IP3R1 and intracellular BKCa channels is unclear. Conceivably, BKCa channels may be contained within the SR or Golgi before membrane trafficking, and intracellular FRET may reflect the close proximity of these intracellular BKCa channels to IP3R1 located on the SR membrane. Alternatively, IP3R1 and BKCa channels may form into a protein complex before membrane trafficking.
Intact SR may have been excised within membrane patches, although SR Ca2+ would be depleted due to the recording conditions used (no ATP in bath solution). Furthermore, IP3 activated BKCa channels in patches that had been excised for >30 min and in solutions that contained BAPTA, which would rapidly buffer any Ca2+ released by SR. Therefore, in inside-out patches studied here, IP3R-mediated SR Ca2+ release cannot underlie IP3-induced BKCa channel activation. Recently, we demonstrated that IP3R1 physically and functionally couples to TRPC3, but not TRPC6, channels in cerebral artery smooth muscle cells (Adebiyi et al., 2010). When considering previous observations and those made here, IP3R1, TRPC3, and BKCa channels may coassemble within a macromolecular complex that regulates Ca2+ signaling, membrane potential, and arterial contractility. Conceivably, IP3R-induced Ca2+ influx through plasma membrane TRPC3 channels could elevate submembrane [Ca2+]i and activate nearby BKCa channels (Adebiyi et al., 2010). Several observations indicate that such a mechanism does not underlie IP3-induced BKCa channel activation reported here: (a) the driving force for Ca2+ influx (symmetrical 10 µM Ca2+) is weak and unlikely to sufficiently elevate [Ca2+]i to activate BKCa channels; (b) BKCa channel activation occurred in the presence of BAPTA, which would rapidly buffer any entering Ca2+; and (c) elevating pipette free [Ca2+] to 2 mM did not alter IP3-induced BKCa channel activation. In intact cells with physiological Ca2+ buffers and cation gradients (i.e., 2 mM [Ca2+]o and 100–300 nM [Ca2+]i), Ca2+ influx through TRPC3 channels may contribute to IP3R1-induced BKCa channel activation. However, TRPC3 channels are also Na+ permeant, and the larger driving force for Na+ influx would limit Ca2+ influx, reducing any potential Ca2+ signal to BKCa channels.
BKCa channels exhibited a micromolar Kd for Ca2+ similar to that previously determined in rat cerebral artery smooth muscle cells at the same voltage (Pérez et al., 2001). IP3 caused an elevation in BKCa channel apparent Ca2+ sensitivity with activation most prominent between 1 and 10 µM [Ca2+]i. These data indicate that IP3 will increase BKCa channel sensitivity to local micromolar Ca2+ gradients, but will not shift BKCa channel Ca2+ sensitivity into the global nanomolar [Ca2+]i range. Data here and in a previous study indicate that IP3R1 activation both stimulates SR Ca2+ release and amplifies the sensitivity of nearby BKCa channels to micromolar [Ca2+]i (Zhao et al., 2008). Therefore, the molecular coupling mechanism likely sensitizes BKCa channels to SR Ca2+ released from nearby IP3Rs.
When considering the physiological function of the local coupling mechanism between IP3R1 and BKCa channels, it is important to consider that PLC activation not only elevates IP3, but also stimulates PKC. PKC inhibits arterial smooth muscle cell BKCa channels both by reducing Ca2+ spark frequency and by direct BKCa channel inhibition (Jaggar and Nelson, 2000; Jaggar et al., 2000). Thus, PKC reduces BKCa activation by RYR channels that generate Ca2+ sparks (Jaggar et al., 2000). However, vasoconstrictors also activate Ca2+ waves and elevate global [Ca2+]i (Jaggar and Nelson, 2000; Mauban et al., 2001; Zhao et al., 2008; Adebiyi et al., 2010). IP3R1 activation is essential for vasoconstrictor-induced Ca2+ waves and contributes to the global [Ca2+]i elevation (Zhao et al., 2008). Collectively, these studies suggest that PLC-coupled receptor agonists shift control of BKCa channel activity from RYR channels to IP3Rs. IP3-induced BKCa channel activation would limit PKC-induced BKCa channel inhibition and attenuate the membrane depolarization and vasoconstriction.
The signaling mechanism by which IP3R1 elevates BKCa channel apparent Ca2+ sensitivity was not studied here. Several possibilities exist, including that an IP3-induced conformational change in IP3R1 may lead to direct interaction with either the BKCa channel α and/or β1 subunit. IP3R1 may also activate BKCa channels via an indirect interaction through intermediate proteins, including caveolin-1 (Cheng and Jaggar, 2006; Alioua et al., 2008). A deeper understanding of the molecular mechanisms by which IP3Rs stimulate BKCa channels will require further investigation.
In summary, data indicate that IP3-induced IP3R1 activation elevates BKCa channel apparent Ca2+ sensitivity through localized molecular coupling in arterial smooth muscle cells. This negative feedback mechanism would limit IP3-induced vasoconstriction in cerebral arteries. These data also raise the possibility that local communication between IP3Rs and BKCa channels may occur in a wide variety of cell types that express both of these proteins.
We thank Drs. Alejandro M. Dopico and Damodaran Narayanan for helpful comments on the manuscript.
This study was supported by grants from the National Institutes of Health (HL67061, HL077678, and HL094378 to J.H. Jaggar; HL096411 to A. Adebiyi; and HL080101 to J. Chen).
Lawrence G. Palmer served as editor.
G. Zhao and Z.P. Neeb contributed equally to this paper.