Role of caveolin-1 in endothelial BK_Ca channel regulation of vasoreactivity

Abstract

A novel vasodilatory influence of endothelial cell (EC) large-conductance Ca²⁺-activated K⁺ (BK_Ca) channels is present following in vivo exposure to chronic hypoxia (CH) and may exist in other pathological states. However, the mechanism of channel activation that results in altered vasoreactivity is unknown. We tested the hypothesis that CH removes an inhibitory effect of the scaffolding domain of caveolin-1 (Cav-1) on EC BK_Ca channels to permit activation, thereby affecting vasoreactivity. Experiments were performed on gracilis resistance arteries and ECs from control and CH-exposed (380 mmHg barometric pressure for 48 h) rats. EC membrane potential was hyperpolarized in arteries from CH-exposed rats and arteries treated with the cholesterol-depleting agent methyl-β-cyclodextrin (MBCD) compared with controls. Hyperpolarization was reversed by the BK_Ca channel antagonist iberiotoxin (IBTX) or by a scaffolding domain peptide of Cav-1 (AP-CAV). Patch-clamp experiments documented an IBTX-sensitive current in ECs from CH-exposed rats and in MBCD-treated cells that was not present in controls. This current was enhanced by the BK_Ca channel activator NS-1619 and blocked by AP-CAV or cholesterol supplementation. EC BK_Ca channels displayed similar unitary conductance but greater Ca²⁺ sensitivity than BK_Ca channels from vascular smooth muscle. Immunofluorescence imaging demonstrated greater association of BK_Ca α-subunits with Cav-1 in control arteries than in arteries from CH-exposed rats, although fluorescence intensity for each protein did not differ between groups. Finally, AP-CAV restored myogenic and phenylephrine-induced constriction in arteries from CH-exposed rats without affecting controls. AP-CAV similarly restored diminished reactivity to phenylephrine in control arteries pretreated with MBCD. We conclude that CH unmasks EC BK_Ca channel activity by removing an inhibitory action of the Cav-1 scaffolding domain that may depend on cellular cholesterol levels.

large-conductance (BK) Ca²⁺-activated K⁺ (BK_Ca) channels are important regulators of arterial tone via their hyperpolarizing influence on vascular smooth muscle (VSM). BK_Ca channels in VSM are regulated by localized increases in Ca²⁺ due to sparks from ryanodine-sensitive stores (18). Although most studies concerning the role of BK channels in vascular control focus on VSM, there is increasing evidence that endothelial cells (ECs) also express these channels, although their expression and physiological significance in native cells have been questioned (12). It has been suggested that although these channels may be quiescent in some vascular beds, their activity is unmasked in various pathological conditions or in cell culture (32). Recent experiments suggest that endothelial BK_Ca channels are active following in vivo exposure to chronic hypoxia (CH) and that they participate in regulation of vasoreactivity (16).

Systemic hypoxemia occurs in pathological conditions that impair pulmonary gas exchange or as a result of residence at high altitude. Patients exposed to CH demonstrate systemic vasodilation (4) and blunted reflex vasoconstriction to lower body negative pressure challenges (15). Diminished vasoconstrictor reactivity is not reversed by restoration of normoxia in these patients (4) or in experimental animals subjected to prolonged hypoxia (3, 6), suggesting that CH elicits long-term adaptations in vascular function.

CH causes a generalized reduction in vasoconstrictor reactivity, evidenced by similar blunting of myogenic and agonist-induced reactivity (8, 11, 19, 28). This impairment of vasoconstriction is endothelium-dependent and is associated with VSM hyperpolarization (8, 16). Inhibition of heme oxygenase (8, 26) or BK_Ca channels (16, 26) normalizes membrane potential (E_m) and restores vasoreactivity in arterioles from CH-exposed rats. Interestingly, although heme oxygenase-mediated VSM hyperpolarization and relaxation are BK_Ca channel-dependent in these arteries, it is unaffected by inhibition of pathways linked to VSM BK_Ca channels, such as ryanodine receptors and soluble guanylate cyclase activity (26), supporting a role for endothelial BK_Ca channels. This possibility was recently confirmed in studies specifically targeting endothelial BK_Ca channels with luminal administration of the specific inhibitor iberiotoxin (IBTX) in arteries from CH-exposed rats, which restored vasoconstrictor responsiveness and normalized EC and VSM E_m to control levels (16). Whole cell patch-clamp experiments demonstrated an IBTX-sensitive current in ECs from CH-exposed rats that was not observed in controls. Interestingly, BK_Ca channels were detected by immunofluorescence in ECs from control and CH-exposed rats, and BK_Ca current could be elicited in control cells with mild cholesterol depletion (16). These results suggest that although endothelial BK_Ca channels are expressed in both groups, their activity is normally inhibited in a cholesterol-dependent fashion and that CH exposure allows their activation.

Caveolae are cholesterol-rich plasmalemmal microdomains characterized by their flask-like structure and by the presence of the scaffolding protein caveolin-1 (Cav-1). Cav-1 has a high affinity for cholesterol and an intracellular binding domain that regulates the activity of various signaling molecules. Importantly, Cav-1 inhibits BK_Ca channel activity in cultured ECs (35) through its scaffolding domain (amino acids 82–101: DGIWKASFTTFTVTKYWFYR); however, the role of Cav-1 in BK_Ca channel function in native ECs has not been examined, and Cav-1 regulates BK_Ca channels differently in other cell types. Cav-1 knockout mice develop cardiac hypertrophy, pulmonary hypertension, and systemic hypotension (24). Additionally, Cav-1 knockout mice demonstrate augmented endothelium-dependent relaxation (7, 36) and decreased myogenic (1) and agonist-induced (7) vasoconstriction, which parallels observations in CH-exposed rats (9, 11, 16). Thus it is possible that altered vascular control in settings such as CH may be related to dysfunction of endothelial Cav-1.

The goal of the present study was to test the hypothesis that CH enhances EC BK_Ca channel activity and alters vasoreactivity via loss of an inhibitory effect of Cav-1. To examine this hypothesis in the most physiologically relevant manner possible, experiments were performed on intact arteries and on cells freshly dispersed from rats exposed to CH or control conditions.

METHODS

Animals

Experiments were performed on male Sprague-Dawley rats (Harlan Industries). All procedures were approved by the Institutional Animal Care and Use Committee of the University of New Mexico Health Sciences Center.

Hypoxic Exposure

CH rats were exposed to hypobaric hypoxia at a barometric pressure of 380 mmHg for 48 h, whereas normoxic control rats were housed in identical cages at ambient pressure (∼630 mmHg barometric pressure). In previous studies, we determined that 48 h of CH results in vasoconstrictor hyporeactivity indistinguishable from more prolonged (4 wk) exposure and that the effects of CH do not reverse for ≥96 h after exposure (19, 28).

Endothelial E_m in Intact Arteries

Rats were anesthetized with pentobarbital sodium (50 mg ip), and hindlimbs were removed and placed in ice-cold HEPES-buffered physiological saline solution (PSS; mmol/l: 150 NaCl, 6 KCl, 1 MgCl₂, 1.5 CaCl₂, 10 HEPES, and 10 glucose, with pH adjusted to 7.4 with NaOH). Gracilis artery branches (passive ID at 100 mmHg = 150–200 μm) were carefully dissected, cut into vascular strips, and superfused (10 ml/min) with warmed (37°C) HEPES-buffered PSS. All experiments were performed under normoxic conditions to examine sustained alterations in vascular control, rather than acute responses to hypoxia. For measurement of EC E_m, vessel strips were impaled with sharp electrodes backfilled with Lucifer yellow for subsequent confirmation of endothelial identity (10, 16, 25).

Patch-Clamp Studies on Isolated ECs

ECs were dispersed for electrophysiological study from control and CH rat aortas, as previously described (16). Previous studies demonstrated that ECs from aorta display the same characteristics as cells from gracilis arteries in terms of their response to CH (16). One to two drops of the resulting cell suspension were seeded on a glass coverslip mounted on an inverted fluorescence microscope (model IX71, Olympus) for 30 min prior to superfusion. Single ECs were identified by selective uptake of the fluorescently (Dil) labeled acylated LDL with a rhodamine filter (29) prior to each electrophysiological experiment. Freshly dispersed ECs were superfused under constant flow (2 ml/min) at room temperature (22–23°C) in an extracellular solution (ECS; mmol/l: 141 NaCl, 4.0 KCl, 1 MgCl₂, 1 CaCl₂, 10 HEPES, and 10 glucose, with pH adjusted to 7.4 with NaOH). Whole cell current data were generated using an Axopatch 200B amplifier (Axon Instruments). For experiments utilizing conventional whole cell patch-clamp configuration, biophysical criteria (>1 GΩ seal resistance, <25 MΩ series resistance) were checked following membrane rupture and monitored throughout the course of the experiment. Cells were held at −60 mV and dialyzed for 5 min with an intracellular solution (ICS; mmol/l: 140 KCl, 0.5 MgCl₂, 5 Mg₂ATP, 10 HEPES, and 1 EGTA, with pH adjusted to 7.2 with KOH). CaCl₂ was added to yield a free Ca²⁺ concentration of 1 μM, as calculated using WinMAXC chelator software. For experiments utilizing the perforated patch technique, 4- to 6-MΩ patch electrodes were backfilled with amphotericin B. After gigaseals were obtained, the series resistance fell over a 10- to 15-min period to 15–20 MΩ and remained stable for up to 1 h. ECs with stable series resistances <25 MΩ were used for experiments.

Single-Channel Recordings

For single-channel experiments, cell-attached and inside-out patch-clamp configurations were used. Patch pipettes had resistances ranging from 6 to 8 MΩ. ECS contained (mmol/l) 130 NaCl, 5 KCl, 1.2 MgCl₂, 10 HEPES, 10 glucose, and 1.2 CaCl₂, with pH adjusted to 7.3 with NaOH. ICS contained (mmol/l) 130 KCl, 2 Na₂ATP, 3 MgCl₂, 10 HEPES, and 1 EGTA, with pH adjusted to 7.30 with KOH. For inside-out patch experiments, ECS contained (mmol/l) 140 KCl, 10 HEPES, 1 CaCl₂, and 1 MgCl₂, with pH adjusted to 7.3 with KOH. For experiments investigating the effect of Ca²⁺ concentration on single-channel opening probability (NP_o) on inside-out patches, EGTA and free Ca²⁺ (CaCl₂) were varied to provide 10 nM, 50 nM, 100 nM, 1 μM, and 10 μM free Ca²⁺ solutions. Patches were held at −60, 0, and 40 mV test potentials for voltage sensitivity experiments. Single-channel unitary conductance was determined from the slope of currents measured in response to voltage steps applied from −100 to 150 mV in 10-mV increments from a holding potential of −60 mV. EC BK_Ca unitary conductance in symmetrical K⁺ and asymmetrical K⁺ were determined. All data were filtered at 5 kHz (−3 dB) and digitized at 10 kHz.

Whole cell currents were measured in response to voltage steps applied from −60 to +150 mV in 10-mV increments from a holding potential of −60 mV. Cell capacitance was monitored, and transmembrane currents were expressed in terms of current density (pA/pF).

Single-Channel Recordings From VSM Cells

For single-channel experiments on freshly isolated smooth muscle cells, aortas were digested as described above. After gentle trituration, freshly isolated VSM cells were placed on Sylgard-coated coverslips. VSM cells were identified by a spindle shape and an absence of Dil labeling of acetylated LDL. All experimental conditions were identical to EC recordings (see above), and Ca²⁺ sensitivity of VSM BK_Ca channels was compared with that of EC BK_Ca channels.

Immunofluorescence in Sectioned Arteries

Gracilis arteries from control and CH-exposed rats were collected following transcardial perfusion with PSS containing 10 mg of papaverine. The gracilis muscle was removed and frozen in optimal cutting temperature compound in liquid N₂ and isopentane. Sections (10 μm) were adhered to Superfrost slides (Fisher) and fixed in ice-cold methanol (100%) for 10 min. After fixation, cross sections were blocked in 4% donkey serum in PBS for 1 h at room temperature. Cross sections were incubated with primary antibodies for BK_Ca α-subunit (Alamone; 1:100 dilution) and Cav-1 (BD Biosciences; 1:50 dilution). Primary antibodies for BK_Ca α-subunit were detected with Cy5-conjugated donkey anti-rabbit secondary antibodies. Cav-1 was detected with a Cy3-conjugated donkey anti-mouse secondary antibody (1:500 dilution for all secondary antibodies; Jackson Labs). The nuclear stain SYTOX (1:10,000 dilution; Molecular Probes) was then applied. Confocal microscope images were obtained using a ×63 oil immersion differential interference contrast objective (numerical aperture 1.4) at a resolution of 524 × 524 and an optical slice of 0.7 μm. For colocalization analysis, z stacks (10 sections, 1 μm) were analyzed for fluorescence overlap of BK_Ca α-subunit and Cav-1. Images underwent nearest-neighbor deconvolution, and individual channels were thresholded to normalize intensity between channels. Thresholded channels were made into masks, and ECs were traced within the lumen of arteriolar cross sections made into a mask for analysis. Individual channel masks and the EC mask were combined to examine endothelial-specific colocalization as defined by correlation coefficient values of Manders et al. (22). Values range from 0 to 1, with values close to 0 indicating nonoverlapping images and values close to 1 reflecting colocalization. Relative fluorescence for Cav-1 or BK_Ca α-subunit in CH and control animals was determined and normalized to fluorescent standards for Cy3 or Cy5, respectively. Fluorescent standards for Cy3 or Cy5 determined maximal fluorescence (0–100 arbitrary units). Relative fluorescence intensity of Cav-1 or BK_Ca α-subunit was then determined for each group on the basis of values obtained from the fluorescent standards.

Study of Isolated Resistance Arteries

Hindlimbs from control or CH rats were removed and placed in ice-cold PSS (mmol/l: 119 NaCl, 4.7 KCl, 25 NaHCO₃, 1.18 KH₂PO₄, 1.17 MgSO₄, 0.026 K₂-EDTA, 5.5 glucose, and 2.5 CaCl₂). Gracilis arterioles (passive ID at 60 mmHg = 140–200 μm) were dissected free, cannulated on glass pipettes, and mounted in an arteriograph (model CH-1, Living Systems). Arteries were slowly pressurized to 60 mmHg with PSS using a servo-controlled peristaltic pump (Living Systems) and superfused (10 ml/min) with warmed (37°C) PSS aerated with a normoxic gas mixture (21% O₂-6% CO₂-73% N₂). Arteries were required to hold a steady pressure upon termination of the servo-control function to confirm the absence of a leak. Any vessels with apparent leaks were discarded. The vessel chamber was then transferred to the stage of a Nikon Eclipse TS100 microscope, and the preparation was superfused with PSS. Bright-field images of vessels were obtained with an IonOptix CCD100M camera, and dimensional analysis was performed by IonOptix Sarclen software to measure inner diameter, as described in previous studies from our laboratory (11). All experiments were performed under normoxic conditions to examine sustained alterations in vascular control, rather than acute vasodilatory responses to hypoxia. Arteriole exposure to normoxic conditions for 2.5–3 h throughout the course of the experiment does not reverse sustained alterations elicited by CH.

Experimental Protocols

Because of the sustained nature of the effect of CH, all experiments on tissue from CH and control rats were performed under normoxic conditions.

Effect of CH or cholesterol depletion on basal endothelial E_m.

EC E_m was measured using sharp electrodes in gracilis arteries from control and CH-exposed rats. To test whether CH-induced changes in E_m could be mimicked by caveolar disruption, some arteries from control animals were treated with the cholesterol-depleting agent methyl-β-cyclodextrin (MBCD; Sigma) at a concentration of 100 μmol/l, which is 100-fold less than the concentration commonly used in vascular preparations, and we previously showed that it does not affect structure or number of caveolae in this preparation (16).

IBTX sensitivity of E_m.

To examine the role of BK_Ca channels in endothelial E_m, sharp electrode measurements were made in control, CH-exposed, and MBCD-treated gracilis arteries in the presence of the BK_Ca channel-specific inhibitor IBTX (100 nmol/l; Sigma) or its vehicle.

Effect of Cav-1 scaffolding domain peptide on basal endothelial E_m.

EC E_m was measured in arteries from control and CH rats and control arteries treated with MBCD in the presence of the Cav-1 scaffolding domain peptide AP-CAV (10 μmol/l; Twenty-First Century Biochemicals) or a scrambled control peptide.

Effect of CH or cholesterol depletion on isolated EC transmembrane currents.

Currents were measured in aortic ECs freshly dispersed from control and CH rats. Previous studies demonstrated that ECs from aorta display electrophysiological characteristics identical to those of cells from gracilis arteries in terms of their response to CH and cholesterol depletion (16). In perforated patch-clamp mode, whole cell currents were measured in response to a series of voltage steps. After initial recording, one group of cells from control rats was superfused with MBCD for 15 min before voltage steps were applied and currents reexamined.

Effect of BK_Ca channel inhibition and activation on isolated EC transmembrane currents.

In perforated patch-clamp mode, recordings were taken before and 5 min after superfusion with IBTX (100 nmol/l) or the BK_Ca channel activator NS-1619 {1,3-dihydro-1-[2-hydroxy-5-(trifluoromethyl)phenyl]-5-(trifluoromethyl)-2H-benzimidazol-2-one, 10 μmol/l; Sigma} in cells from control and CH rats. Additional measurements were taken in a group of control cells after treatment with MBCD (100 μmol/l).

Effect of cholesterol repletion on EC BK_Ca currents.

Whole cell currents were first measured in cells from control rats in perforated patch-clamp mode following incubation with vehicle. After the initial recording, cells were superfused with MBCD for 15 min before voltage steps were applied and currents recorded, whereas cells from CH-exposed animals remained under vehicle conditions for 15 min. A mixture of MBCD and cholesterol (8:1 MBCD-ChL), which has been shown to replete membrane cholesterol and restore caveolar function (21, 31), was then administered to each group, and transmembrane currents were reexamined.

Effect of AP-CAV on whole cell currents.

Currents were assessed using conventional whole cell patch-clamp mode in aortic ECs from control and CH rats. One group of control cells was also treated with MBCD as described above. Measurements were first made following cellular rupture under vehicle conditions. Next, AP-CAV or the scrambled control peptide was backfilled into patch pipettes and dialyzed into the clamped cells for 5 min. Voltage steps were again applied, and currents were recorded following dialysis of AP-CAV or scrambled peptide. A limited number of experiments (n = 3 in each group) were conducted to test the effect of prior saturation of AP-CAV with cholesterol (10 μM AP-CAV incubated with 8:1 MBCD-ChL) on its inhibitory action on ionic conductance in cholesterol-depleted control cells.

Determination of unitary conductance of EC BK_Ca channels.

Unitary conductance was determined in symmetrical K⁺ in inside-out patches of aortic ECs from CH rats and control ECs pretreated with MBCD. Assessments were also made in whole cell-cell attached mode to determine conductance in asymmetrical K⁺ relevant to the in situ environment.

Determination of EC BK_Ca channel Ca²⁺ and voltage sensitivity.

Inside-out patches from ECs of CH rats were held at varying test potentials and intracellular Ca²⁺ levels to investigate channel NP_o. Parallel experiments were performed on freshly isolated VSM to compare Ca²⁺ sensitivity of BK_Ca channels between the two cell types.

Sensitivity of EC BK_Ca channels to IBTX.

Inside-out patches from aortic ECs from CH rats were held at varying test potentials in 1 μmol/l free Ca²⁺ solution. NP_o was assessed in the presence or absence of intrapipette IBTX.

Myogenic responses.

Experiments were performed on cannulated, pressurized gracilis arterioles. Active and passive (Ca²⁺-free) pressure-diameter relationships were determined, as described previously (11, 16), over intraluminal pressure steps between 20 and 160 mmHg. Vessel inner diameter was monitored using video microscopy and edge-detection software (IonOptix Sarclen). Endothelial integrity was assessed by dilation with ACh (10 μmol/l) prior to Ca²⁺-free superfusion. Pressure-induced vasoconstrictor responses were determined in the presence of AP-CAV or scrambled control peptide. AP-CAV or the scrambled peptide was administered intraluminally in an effort to specifically target the endothelium.

Vasoconstrictive reactivity to phenylephrine.

Arteriole inner diameter was assessed in response to increasing phenylephrine (PE) concentrations (from 10⁻⁹ to 10⁻⁵ μmol/l) in the presence of intraluminal AP-CAV or scrambled control peptide. Reversal of constriction with 10 μM ACh at the termination of the PE concentration-response curve was used to verify endothelial integrity.

Vasoconstrictive reactivity to PE in cholesterol-depleted arteries from control rats.

Arteries were pretreated luminally with MBCD (100 μM) for 30 min at 37°C before washout of the dextrin and administration of AP-CAV or the scrambled control peptide. Arteriole inner diameter was then assessed in response to increasing PE concentrations. Reversal of constriction with 10 μM ACh at the termination of the PE concentration-response curve was used to verify endothelial integrity at the termination of each experiment.

Calculations and Statistics

Values are means ± SE; n represents the number of animals in each group, except for patch-clamp studies, where n represents the number of cells. Data were analyzed by repeated-measures ANOVA and by Bonferroni's modified unpaired Student's t-test for multiple comparisons when differences were indicated. Unpaired t-tests were used for single comparisons between groups. P ≤ 0.05 was accepted as indicating a statistically significant difference.

RESULTS

Effect of CH or Cholesterol Depletion on Basal Endothelial E_m

Consistent with an earlier report (16), EC E_m was hyperpolarized in gracilis arteries from CH rats compared with controls (Fig. 1A). Mild cholesterol depletion with MBCD (100 μmol/l) elicited a similar degree of hyperpolarization (Fig. 1A). As observed earlier (16, 33), resting E_m was more depolarized in these arteries than in some other beds, likely due to the lack of myoendothelial gap junctions (33). Using techniques identical to those used in the present study, we observed significantly more polarized E_m in other beds (29).

Fig. 1.

Endothelial membrane potential (E_m) and outward K⁺ current. A: E_m in arterial strips from control (C) and chronically hypoxic (H) rats and control strips treated with methyl-β-cyclodextrin (MBCD; M). B: effect of iberiotoxin (IBTX) to reverse hyperpolarization in chronic hypoxia (CH)-exposed and MBCD-treated vessels. C: reversal of hyperpolarization by caveolin-1 scaffolding domain peptide (AP-CAV). D: outward current measured in perforated patch configuration in cells from each group. E: mean current density at +60 mV in cells from each group. Outward currents were significantly greater in cells from CH rats and MBCD-treated cells than controls (at −40 to +150 mV test potentials). Number of animals is shown in bars in A–C. *P ≤ 0.05 vs. control. #P ≤ 0.05 vs. CH vehicle or scrambled. †P ≤ 0.05 vs. MBCD vehicle or scrambled.

IBTX Sensitivity of E_m

IBTX significantly depolarized E_m to baseline vehicle-treated control level in ECs from CH rats and in MBCD-treated arteries from control rats but was without effect in untreated controls (Fig. 1B). These results support the hypothesis that BK_Ca channel activity causes endothelial hyperpolarization following MBCD treatment and CH exposure.

Effect of AP-CAV on Basal Endothelial E_m

AP-CAV had no effect on endothelial E_m in arteries from control rats compared with arteries treated with the scrambled peptide (Fig. 1C). In contrast, AP-CAV depolarized the endothelium of arteries from CH rats and control arteries treated with MBCD compared with control (scrambled) levels (Fig. 1C). These results suggest that the depolarizing influences of CH exposure and cholesterol depletion are caused by loss of an inhibitory effect of the Cav-1 scaffolding domain on basal BK_Ca channel activity.

Effect of CH or Cholesterol Depletion on EC Transmembrane Currents

Whole cell outward currents were greater in ECs from CH rats and MBCD-treated cells than untreated controls at each test potential from −50 to +150 mV (Fig. 1D). Cholesterol depletion and CH exposure resulted in similar outward currents that were not different (Fig. 1, D and E).

Measurement of EC BK_Ca Unitary Conductance and Channel Sensitivity to Voltage and Ca²⁺

EC BK_Ca unitary conductance in patches from CH rats (Fig. 2A) was 221 ± 12 pS in symmetrical K⁺. Conductance measured in asymmetrical K⁺ in cell-attached whole cell mode was 145 ± 9 pS. Unitary conductances did not vary between cells from CH rats and control cells treated with MBCD (Fig. 2A). EC BK_Ca channels displayed enhanced Ca²⁺ sensitivity compared with VSM channels (Fig. 2B), consistent with our evidence for tonic activity of these channels at basal intracellular Ca²⁺ concentration ([Ca²⁺]_i) in ECs from CH-exposed rats. EC channels from CH rats also displayed voltage sensitivity (Fig. 2C).

Fig. 2.

Endothelial large-conductance Ca²⁺-activated K⁺ (BK_Ca) channel characteristics in cell-attached and inside-out patches. A: unitary conductance in symmetrical (inside-out) and asymmetrical (cell-attached) K⁺ in endothelial cells (ECs) from CH rats and controls treated with MBCD (n = 7–14 patches per group). There were no differences between treatments. B: Ca²⁺ sensitivity of open probability (NP_o) in inside-out patches of ECs (n = 12) and vascular smooth muscle (VSM) cells (n = 8) from CH rats. *P ≤ 0.05 vs. VSM. C: voltage sensitivity of NP_o at 100 nmol/l intracellular Ca²⁺ in inside-out patches from CH rats (n = 11). #P ≤ 0.05 vs. −60 mV.

Effect of BK_Ca Channel Inhibition and Activation on EC Transmembrane Currents

IBTX significantly reduced whole cell currents in ECs from CH rats (at test potentials from −50 to +150 mV; Fig. 3, B and D) but had no effect in controls (Fig. 3, A and D), as previously described in conventional whole cell studies at much higher [Ca²⁺]_i (16) than in the present perforated patch experiments. Similarly, IBTX normalized outward current in cholesterol-depleted cells to levels of controls (at test potentials from −40 to +150 mV; Fig. 3, C and D), demonstrating active BK_Ca channels following caveolar disruption with MBCD. Consistent with these whole cell measurements, IBTX applied via the pipette to inside-out patches effectively inhibited channel opening (Fig. 3E).

Fig. 3.

IBTX sensitivity of currents. A–C: transmembrane currents in the presence of IBTX or its vehicle. IBTX-sensitive currents were observed in CH-exposed and MBCD-treated groups, but not controls. D: mean current density at +40 mV. E: effect of IBTX on NP_o in inside-out patches from cells from CH rats. *P ≤ 0.05 vs. vehicle control. #P ≤ 0.05 vs. CH vehicle. †P ≤ 0.05 vs. MBCD + vehicle.

The BK_Ca channel activator NS-1619 had no effect in cells from control rats (Fig. 4, A and D) but further increased outward currents in cells from CH rats at test potentials between −40 and +150 mV (Fig. 4, B and D). NS-1619 also effectively enhanced current in control cholesterol-depleted cells (from −30 to +150 mV; Fig. 4, C and D).

Fig. 4.

Effect of the large-conductance Ca²⁺-activated K⁺ (BK_Ca) channel activator NS-1619 on transmembrane currents. NS-1619 significantly enhanced current in CH-exposed (−40 to +150 mV; B) and MBCD-treated (−30 to +150 mV; C) cells but did not affect controls (A). *P ≤ 0.05 vs. control vehicle. #P ≤ 0.05 vs. CH vehicle. †P ≤ 0.05 vs. MBCD.

Effect of Cholesterol Repletion on Whole Cell Currents

Administration of MBCD-ChL to restore Cav-1 function significantly reduced outward currents in cells from CH rats (Fig. 5, B and D) and cells treated with MBCD (Fig. 5, B and D) but had no effect in controls (Fig. 5, A and D). These results demonstrate the dynamic regulation of EC BK_Ca channels by cellular cholesterol and, presumably, Cav-1.

Fig. 5.

Effect of cholesterol repletion (ChL) on transmembrane currents. ChL diminished outward current in CH-exposed (−40 to +150 mV; B) and MBCD-treated (−30 to +150 mV; C) cells but did not affect controls (A). *P ≤ 0.05 vs. control vehicle. #P ≤ 0.05 vs. CH vehicle. †P ≤ 0.05 vs. MBCD.

Effect of AP-CAV on Whole Cell Currents

AP-CAV potently inhibited BK_Ca channel-dependent current in ECs from CH rats (Fig. 6, B and D) and control MBCD-treated ECs from control rats (Fig. 6, C and D), demonstrating an inhibitory effect of the Cav-1 scaffolding domain on channel activity in these settings. AP-CAV had no effect on current in untreated controls (Fig. 6, A and D). Currents in cells treated with the scrambled peptide did not differ from currents in vehicle-treated cells in any of the groups (data not shown). In addition, preincubation of AP-CAV with cholesterol did not affect the peptide's inhibitory action on outward current in MBCD-treated cells from control animals (n = 3).

Fig. 6.

Effect of AP-CAV on transmembrane currents. AP-CAV reduced current to the level of scrambled control in CH-exposed (−50 to +150 mV; B) and MBCD-treated (−40 to + 150 mV; C) cells but did not affect controls (A). *P ≤ 0.05 vs. control vehicle. #P ≤ 0.05 vs. CH vehicle. †P ≤ 0.05 vs. MBCD.

Colocalization of Cav-1 and BK_Ca α-Subunit in Gracilis Arterioles

Staining for BK_Ca α-subunit (green) was positive in VSM cells and ECs (Fig. 7A) in control and CH rats. Cav-1 (red) was also found in VSM cells and ECs (Fig. 7A) in control and CH rats. However, overlap of Cav-1 and BK_Ca α-subunit was significantly less in arteries from CH-exposed rats than controls, as assessed by the correlation coefficient of Manders et al. (Fig. 7B). However, there were no differences in fluorescence intensity for Cav-1 or BK_Ca α-subunit between groups (Fig. 7C), suggesting that association, rather than expression, is affected by CH.

Fig. 7.

Colocalization of BK_Ca α-subunit (BK; green) and caveolin-1 (Cav-1; red) in endothelium (EC) and VSM of gracilis resistance arterioles from control (C) and CH rats (A). Nuclei are stained white. Overlap of BK_Ca α-subunit and Cav-1 was significantly less in CH-exposed ECs than controls, as determined by the correlation coefficient of Manders et al. (B); however, fluorescence intensity did not differ for Cav-1 or BK_Ca α-subunit between groups (C). IEL, internal elastic lamina.

Effect of AP-CAV on Myogenic and PE-Induced Tone in Gracilis Arteries From Control and CH Rats

Myogenic tone was less in scrambled peptide-treated gracilis arteries from CH rats than controls (Fig. 8), consistent with earlier reports from untreated vessels (16). Luminal application of AP-CAV to target and inhibit EC BK_Ca channels restored myogenic tone in arteries from CH rats, similar to luminal IBTX or endothelial disruption in earlier studies (8, 11, 16), but was without effect in controls. AP-CAV similarly restored PE-induced vasoconstrictor responses in arteries from CH rats to levels of scrambled peptide-treated controls without affecting arteries from normoxic rats (Fig. 9A).

Fig. 8.

AP-CAV restores pressure-induced myogenic reactivity in arteries from CH rats to levels of scrambled controls. *P ≤ 0.05 vs. control scrambled. Starting vessel diameter was 145 ± 7 μm and did not differ between experimental groups.

Fig. 9.

Restoration of phenylephrine (PE)-induced vasoconstrictor reactivity by AP-CAV in arteries from CH rats (A) and control arteries pretreated with MBCD (B). *P < 0.05 vs. control scrambled. #P < 0.05 vs. MBCD scrambled. Starting vessel diameter was 154 ± 6 μm and did not differ between experimental groups.

Effect of Cholesterol Depletion and AP-CAV on PE-Induced Tone in Gracilis Arteries From Control Rats

MBCD reduced PE-induced tone similar to CH exposure. AP-CAV restored PE constriction (Fig. 9B), while the scrambled peptide had no effect. Effects of vehicle and scrambled peptide were not different in MBCD-treated arteries from control animals.

DISCUSSION

The major findings of the present study are as follows. 1) CH exposure or cholesterol depletion removes Cav-1 inhibition of BK_Ca channels, leading to persistent channel activity and hyperpolarization of ECs. 2) Cholesterol repletion or introduction of the scaffolding domain of Cav-1 potently inhibits EC BK_Ca channel activity and normalizes outward K⁺ currents to levels of controls. 3) In EC BK_Ca channels, unitary conductance is similar to that in VSM cells, but Ca²⁺ sensitivity is greater. 4) Colocalization of endothelial BK_Ca channel and Cav-1 is reduced following CH, although Cav-1 expression appears unchanged. 5) Diminished myogenic and agonist-induced tone following CH or acute cholesterol depletion of control arteries can be restored by endothelial introduction of the scaffolding domain of Cav-1. These results provide evidence for a novel mode of vascular control elicited by CH involving loss of an inhibitory effect of Cav-1 on EC BK_Ca channels.

Consistent with an earlier report (16), we observed EC hyperpolarization in arteries from CH rats compared with controls. Interestingly, mild cholesterol depletion with MBCD elicited a similar degree of hyperpolarization in control arteries. Neither treatment results in discernable alteration in gross structure or number of endothelial caveolae (16), suggesting that modest interference of Cav-1 function or impaired cholesterol homeostasis has functional consequences on EC E_m. Hyperpolarization of the endothelium can be conveyed to the underlying VSM through several postulated mechanisms to promote vasodilation. VSM hyperpolarization can occur by conduction of charge through low-resistance myoendothelial gap junctions (14); however, the femoral circulation does not appear to have these structures (33). This is evident from observations by us and others of the relatively depolarized EC E_m in the femoral bed, in contrast with the mesenteric circulation, which possesses direct myoendothelial communication (16, 33). Another possible mechanism of VSM hyperpolarization is elevated extracellular K⁺ concentration due to activation of EC K⁺ channels, which in turn activate VSM inwardly rectifying K⁺ channels and Na⁺-K⁺-ATPase (17). Regardless of the initiating event, VSM hyperpolarization results in inhibition of voltage-gated Ca²⁺ channels and, hence, dilation. Persistent hyperpolarization of the vascular wall, as seen with CH (8, 11, 16), opposes vasoconstriction and, thus, dampens vasoconstrictor reactivity.

Interestingly, EC hyperpolarization following CH exposure or MBCD treatment appears to be due to tonic activity of BK_Ca channels not seen in controls. We previously showed that this basal hyperpolarization leads to more profound hyperpolarization responses to such agonists as ACh, thereby amplifying endothelium-dependent vasodilation (16). However, BK_Ca channels in VSM require high local Ca²⁺ concentration from sparks to elicit activity (18), which led us to hypothesize that EC channels may exhibit enhanced Ca²⁺ sensitivity. Previous experiments from our laboratory examined IBTX-sensitive currents in conventional whole cell configuration with [Ca²⁺]_i maintained at 1 μmol/l (16). This Ca²⁺ concentration maximally activates the channel and, thus, provides little insight into channel activity at endogenous [Ca²⁺]_i. In the present study, we employed perforated patch-clamp configuration to maintain more physiological [Ca²⁺]_i and still observed BK_Ca currents in cells from CH rats and cells treated with MBCD. Although EC BK_Ca channels exhibited unitary conductances consistent with those from VSM, the Ca²⁺ sensitivity of EC channels was enhanced, as evidenced by significantly greater channel opening at 10 and 100 nmol/l Ca²⁺ than in VSM patches. Thus, within the endothelium, basal Ca²⁺ levels seem sufficient to maintain BK_Ca channel activity. This enhanced Ca²⁺ sensitivity does not appear to be related to activity of the β₁-accessory subunit of the BK_Ca channel, since the β₁-activator tamoxifen (5) inhibited, rather than enhanced, outward current in ECs from CH rats, whereas tamoxifen predictably stimulated current in VSM (data not shown).

Previous studies on cultured ECs suggest that the scaffolding domain of Cav-1 exerts an inhibitory effect on BK_Ca channels that can be eliminated by cholesterol depletion (35). Our data support this inhibitory role of Cav-1 and, for the first time, demonstrate that an in vivo stimulus can tonically activate endothelial BK_Ca channels by affecting this mode of regulation. In addition, consistent with cultured cell studies, cholesterol depletion with MBCD unmasked BK_Ca currents in ECs from control animals. The presence of BK_Ca channels in ECs from control and CH rats was confirmed by immunofluorescence (Fig. 7). We additionally demonstrated that the apparent association between BK_Ca channels and Cav-1 was less in arteries from CH rats than controls, as assessed by pixel overlap analysis, although relative fluorescence for each protein did not differ. These results suggest that reduced association between the two proteins, rather than diminished Cav-1 expression, is responsible for enhanced channel activity following CH. The importance of the scaffolding domain of Cav-1 in regulating BK_Ca channel activity was demonstrated by the effectiveness of AP-CAV to inhibit BK_Ca currents in cells from CH rats and cholesterol-depleted control ECs and to reverse EC hyperpolarization in intact arteries from these groups. Importantly, luminal AP-CAV also restored myogenic and PE-induced vasoconstrictor responsiveness in arteries from CH rats without affecting control arteries, in a pattern identical to the effects of intraluminal IBTX or endothelial disruption (8, 11, 16). Similarly, AP-CAV restored reactivity to PE in control arteries treated with MBCD (Fig. 9B). We attribute the effects of cholesterol depletion with MBCD in arteries from control rats to selective actions on the endothelium, rather than on the underlying VSM. We employed luminal administration of a low concentration (100 μM) of MBCD in the intact artery experiments to localize the effects of cholesterol depletion to the endothelium. Furthermore, in additional experiments (data not shown), we observed no effect of 100 μM MBCD on transmembrane currents in isolated VSM cells, whereas this concentration enhanced currents in ECs from control rats (Fig. 1). Our observation that preincubation of AP-CAV with cholesterol does not affect the peptide's ability to inhibit BK_Ca currents suggests that the restorative effects of the Cav-1 scaffolding domain on vascular tone and E_m may reflect direct interaction of AP-CAV with the channel independent of the peptide's affinity to cholesterol. However, further investigation into the specific mechanisms involved in AP-CAV-mediated restoration of vascular function is warranted. These results clearly demonstrate the central role of altered EC Cav-1 function in altered vascular control following CH.

In the present studies, cholesterol supplementation inhibited BK_Ca currents in cells from CH rats and MBCD-treated ECs, suggesting a key role of cholesterol in channel regulation. However, it is unclear if the cholesterol repletion protocol acts similarly to administration of the scaffolding domain of Cav-1. Cholesterol directly binds to Cav-1 in vitro (23) and in vivo (34). Cav-1 localization at the plasma membrane is closely associated with the presence of free cholesterol (30); decreases in cellular cholesterol levels significantly reduce Cav-1 transport and membrane association. Interestingly, CH reduces de novo cholesterol synthesis by inhibiting 3-hydroxy-3-methylglutaryl (HMG)-CoA reductase activity in cultured cells (27). HMG-CoA reductase is the rate-limiting enzyme in de novo synthesized cellular cholesterol; thus CH could limit cholesterol production in the present model. It is noteworthy that cultured rat aortic ECs exposed to hypoxia demonstrate reduced cellular cholesterol and cholesterol esters (2). In addition, inhibition of de novo cholesterol with cerivastatin enhances EC BK_Ca channel activity and results in significant membrane hyperpolarization in human umbilical vein ECs (20). These actions of the HMG-CoA reductase inhibitor on EC BK_Ca channel activity were rapidly reversed by the addition of mevalonate, suggesting that endothelial BK_Ca channel activity is regulated by this agent and other nonsteroidal isoprenoids (20). Isoprenoids such as geranyl pyrophosphate and farnesyl pyrophosphate are required for membrane insertion, caveolar targeting, and localization of multiple proteins (13). Thus changes in EC free cholesterol following hypoxemia could possibly underlie the changes in vascular function in the CH model and have functional implication in other pathologies associated with altered EC cholesterol homeostasis.

In conclusion, the present experiments describe a novel mode of regulation of vascular tone associated with Cav-1 regulation of EC BK_Ca channels. Although these studies center on a model of hypoxemic disorders, i.e., CH, the establishment of the central role of cholesterol in regulating this pathway suggests that these observations may be relevant in a variety of pathophysiological and therapeutic settings.

GRANTS

This work was supported by National Heart, Lung, and Blood Institute Grants HL-58124, HL-63207, HL-95640, and HL-07736 (B. R. Walker) and American Heart Association South Central Affiliate Predoctoral Fellowship 09PRE2261215 (M. A. Riddle).

DISCLAIMER

The contents present here are solely the responsibility of the authors and do not necessarily represent the official views of the National Heart, Lung, and Blood Institute or American Heart Association.

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

M.A.R., J.M.H., and B.R.W. are responsible for conception and design of the research; M.A.R. and J.M.H. performed the experiments; M.A.R. and J.M.H. analyzed the data; M.A.R., J.M.H., and B.R.W. interpreted the results of the experiments; M.A.R. and B.R.W. prepared the figures; M.A.R. drafted the manuscript; M.A.R., J.M.H., and B.R.W. approved the final version of the manuscript; J.M.H. and B.R.W. edited and revised the manuscript.