Genetic ablation of caveolin-1 modifies Ca²⁺ spark coupling in murine arterial smooth muscle cells

Abstract

L-type, voltage-dependent calcium (Ca²⁺) channels, ryanodine-sensitive Ca²⁺ release (RyR) channels, and large-conductance Ca²⁺-activated potassium (K_Ca) channels comprise a functional unit that regulates smooth muscle contractility. Here, we investigated whether genetic ablation of caveolin-1 (cav-1), a caveolae protein, alters Ca²⁺ spark to K_Ca channel coupling and Ca²⁺ spark regulation by voltage-dependent Ca²⁺ channels in murine cerebral artery smooth muscle cells. Caveolae were abundant in the sarcolemma of control (cav-^+/+) cells but were not observed in cav-1-deficient (cav-1^−/−) cells. Ca²⁺ spark and transient K_Ca current frequency were approximately twofold higher in cav-1^−/− than in cav-1^+/+ cells. Although voltage-dependent Ca²⁺ current density was similar in cav-1^+/+ and cav-1^−/− cells, diltiazem and Cd²⁺, voltage-dependent Ca²⁺ channel blockers, reduced transient K_Ca current frequency to ∼55% of control in cav-1^+/+ cells but did not alter transient K_Ca current frequency in cav-1^−/− cells. Furthermore, although K_Ca channel density was elevated in cav-1^−/− cells, transient K_Ca current amplitude was similar to that in cav-1^+/+ cells. Higher Ca²⁺ spark frequency in cav-1^−/− cells was not due to elevated intracellular Ca²⁺ concentration, sarcoplasmic reticulum Ca²⁺ load, or nitric oxide synthase activity. Similarly, Ca²⁺ spark amplitude and spread, the percentage of Ca²⁺ sparks that activated a transient K_Ca current, the amplitude relationship between sparks and transient K_Ca currents, and K_Ca channel conductance and apparent Ca²⁺ sensitivity were similar in cav-1^+/+ and cav-1^−/− cells. In summary, cav-1 ablation elevates Ca²⁺ spark and transient K_Ca current frequency, attenuates the coupling relationship between voltage-dependent Ca²⁺ channels and RyR channels that generate Ca²⁺ sparks, and elevates K_Ca channel density but does not alter transient K_Ca current activation by Ca²⁺ sparks. These findings indicate that cav-1 is required for physiological Ca²⁺ spark and transient K_Ca current regulation in cerebral artery smooth muscle cells.

calcium (Ca²⁺) is a signaling messenger that regulates a wide variety of cellular functions, including secretion, proliferation, and contraction (2). To regulate specific Ca²⁺-dependent functions, cells can generate a variety of intracellular Ca²⁺ signals with distinct frequency, amplitude, and spatial-temporal characteristics.

Smooth muscle cells generate several different modes of Ca²⁺ signaling (15, 20). The global intracellular Ca²⁺ concentration ([Ca²⁺]_i) arises because of extracellular Ca²⁺ influx and intracellular Ca²⁺ release. An elevation in global [Ca²⁺]_i stimulates contraction, whereas a reduction in global [Ca²⁺]_i results in relaxation. Localized [Ca²⁺]_i transients, termed “Ca²⁺ sparks,” also occur in smooth muscle cells (20, 27). Ca²⁺ sparks occur due to the opening of several ryanodine-sensitive Ca²⁺ release (RyR) channels on the sarcoplasmic reticulum (SR) (15, 27). In smooth muscle cells, a Ca²⁺ spark does not directly elevate global Ca²⁺ but activates a number of nearby large-conductance Ca²⁺-activated potassium (K_Ca) channels, resulting in a transient K_Ca current. Membrane hyperpolarization induced by asynchronous transient K_Ca currents reduces voltage-dependent Ca²⁺ channel activity, leading to a decrease in global [Ca²⁺]_i and relaxation. An elevation in [Ca²⁺]_i activates Ca²⁺ sparks and transient K_Ca currents, forming a negative-feedback loop that limits Ca²⁺ entry through voltage-dependent Ca²⁺ channels (15, 27). Thus voltage-dependent Ca²⁺ channels, RyR channels, and K_Ca channels comprise a functional unit that regulates smooth muscle contractility (17). The differential regulation of contractility by local and global Ca²⁺ signaling is also effective because of differences in the frequency and amplitude of the Ca²⁺ signals and the Ca²⁺ sensitivities of the downstream targets for each signal mode (15). For local Ca²⁺ signaling mechanisms to operate, downstream targets must also be located in the proximity of the Ca²⁺ source. In smooth muscle cells, whether membrane proteins maintain spatial organization of voltage-dependent Ca²⁺ channels, K_Ca channels, and RyR channels to permit Ca²⁺ signaling between these proteins is unclear.

Compartmentalization of signaling molecules can occur in small, cholesterol-enriched, flask-shaped membrane invaginations termed “caveolae” (11, 13, 34, 39). Caveolins, of which three isoforms have been cloned (cav 1–3), are structural components required for caveolae formation (11, 13, 34, 39). Although all three caveolins have been identified in vascular smooth muscle cells, caveolin-1 is the primary isoform (9, 18, 28, 34, 39). Supporting a role for caveolae in the regulation of arterial smooth muscle local Ca²⁺ signaling is evidence that acute cholesterol depletion with dextrin inhibits Ca²⁺ sparks (23). Similarly, transient K_Ca current frequency is reduced in cerebral artery smooth muscle cells of cav-1-deficient (cav-1^−/−) mice, when compared with wild-type controls (10). In ureter smooth muscle cells, K_Ca channels are found in buoyant membrane fractions, and in cultured myometrial cells K_Ca channels are localized to caveolae (1, 3). Thus cav-1 and caveolae may regulate Ca²⁺ sparks and K_Ca channels in smooth muscle.

Here, the communication between voltage-dependent Ca²⁺ channels and RyR channels that generate Ca²⁺ sparks and signaling between Ca²⁺ sparks and K_Ca channels was investigated in cerebral artery smooth muscle cells of cav-1^+/+ and cav-1^−/− mice. Data indicate that cav-1 ablation abolishes caveolae, elevates Ca²⁺ spark frequency, and attenuates Ca²⁺ spark regulation by voltage-dependent Ca²⁺ channels. In contrast, transient K_Ca current activation by Ca²⁺ sparks is unaltered in cav-1^−/− cells, even though there is an increase in K_Ca channel density.

MATERIALS AND METHODS

Tissue preparation

Animal procedures used were reviewed and approved by the Animal Care and Use Committee policies at the University of Tennessee. Age-matched (4–6 wk) cav-1^−/− mice (Cav1^tm1Mls, stock no. 004585; see Ref. 33) or wild-type control (cav-1^+/+) mice were purchased from Jackson Laboratories (Bar Harbor, ME). Mice were killed by peritoneal injection of a pentobarbital sodium overdose (130 mg/kg). The brain was removed and placed into ice-cold (4°C) physiological saline solution (PSS) containing (in mM) 112 NaCl, 4.8 KCl, 26 NaHCO₃, 1.8 CaCl₂, 1.2 MgSO₄, 1.2 KH₂PO₄, and 10 glucose and gassed with 74% N₂-21% O₂-5% CO₂ (pH 7.4). Posterior cerebral, cerebellar, and middle cerebral arteries (∼150 μm in diameter) were removed, cleaned of connective tissue, and maintained in ice-cold Ca²⁺-free buffer (solution A) containing (in mM) 55 NaCl, 80 Na glutamate, 5.6 KCl, 2 MgCl₂, 10 HEPES, and 10 glucose (pH 7.3 with NaOH). For single cell isolation, smooth muscle cells were dissociated from cerebral arteries with the use of enzymes, as previously described (14).

Electron microscopy

The brain was fixed in PSS containing 2.5% glutaraldehyde for 1 h. Cerebral arteries were dissected from the fixed brain, cleaned of connective tissue, and postfixed in 1% osmium tetroxide in PSS for 4 h. Arteries were then rinsed briefly in deionized water and en bloc stained with 2% uranyl acetate in 0.85% sodium chloride overnight at 4°C. Arteries were dehydrated in graded solutions of ethanol, from 30% through 100% for 1 h each, infiltrated with 50% Spurr in 100% ethanol overnight at room temperature, 100% Spurr over an 8-h period involving at least three changes of Spurr, and then cured at 60°C for 2 days. Transverse sections (75 nm) were cut with a Reichert Ultracut E microtome and poststained with uranyl acetate and lead citrate. Cells were observed and photographed with a JEOL 2000EX TEM located in the Electron Microscope Facility at the University of Tennessee Health Science Center.

Patch-clamp electrophysiology

Potassium currents were measured by using the conventional whole cell, perforated-patch or inside-out patch-clamp configurations with an Axopatch 200B amplifier and Clampex 8.2 (Axon Instruments, Union City, CA). For transient K_Ca current measurement, the bath solution (solution B) contained (in mM) 134 NaCl, 6 KCl, 2 CaCl₂, 1 MgCl₂, 10 HEPES, and 10 glucose (pH 7.4 with NaOH), and the pipette solution contained (in mM) 110 potassium aspartate, 30 KCl, 10 NaCl, 1 MgCl₂, 10 HEPES, and 0.05 EGTA (pH 7.2 with KOH). Amphotericin B (Sigma) was dissolved in DMSO and diluted into the pipette solution to give a final concentration of 200 μg/ml. For whole cell K⁺ current measurement, the bath solution was solution B and the pipette solution contained (in mM) 135 KCl, 5 EGTA, 1.0 BAPTA, 3.5 MgCl₂, 2.5 Na₂ATP, and 10 HEPES (pH 7.2 with KOH). For inside-out patch-clamp experiments to measure single K_Ca channel currents, the pipette and bath solution both contained (in mM) 140 KCl, 10 HEPES, 5 EGTA, and 1.6 1-N-(2-hydroxyethyl)ethylenediamine-N,N′,N′-triacetic acid (pH 7.2 with KOH). MgCl₂ and CaCl₂ were supplemented to bath solution to provide 1 mM free Mg²⁺ and the required free Ca²⁺ concentration (1–100 μM). Free Ca²⁺ concentration in the pipette solution was 10 μM. Free Ca²⁺ concentrations were measured by using a Ca²⁺-sensitive and reference electrode (Corning). For voltage-dependent Ba²⁺ current measurements, the pipette solution contained (in mM) 125 CsCl, 20 tetraethylammonium chloride (TEACl), 1.0 MgCl₂, 0.5 EGTA, 10 HEPES, and 10 glucose (pH 7.2 with CsOH), and the bath solution contained (in mM) 60 NaCl, 60 TEACl, 1.0 MgCl₂, 20 BaCl₂, 10 glucose, and 10 HEPES (pH 7.4 with NaOH). Pipette resistance was measured by applying a 5-mV pulse using the seal test function of pClamp 8.2. Transient K_Ca currents were measured at a steady membrane potential of −40 mV. Voltage-dependent Ba²⁺ currents were activated from a holding potential of −80 mV by applying 300-ms voltage steps to voltages between −50 and +60 mV in increments of 10 mV. Whole cell K⁺ (I_k) currents were activated from a holding potential of −80 mV by applying 250-ms voltage steps to voltages between −70 and +80 mV in 10-mV increments. In inside-out patches, single K_Ca channel currents were measured at steady voltages of −40 or +40 mV. Transient K_Ca currents were filtered at 1 kHz and digitized at 5 kHz. Voltage-dependent Ba²⁺ currents were filtered at 1 kHz and digitized at 4 kHz. Other current measurements were filtered at 2 kHz and digitized at 10 kHz. Transient K_Ca current analysis was performed off-line using methodology described elsewhere (6). A transient K_Ca current was defined as the simultaneous opening of three or more K_Ca channels. Open probability (P_o) was calculated from the following equation: P_o = (∑t_ii)/nT, where t_i is the time at each channel level i, n is the number of channels in the patch, and T is the total time of analysis. The total number of K_Ca channels in an inside-out patch was determined at a voltage of +40 mV with 100 μM free Ca²⁺ in the bath solution. In each patch under each condition, at least 5 min of continuous data were analyzed to calculate transient K_Ca current frequency and amplitude, and 2–5 min were analyzed to determine single K_Ca channel open probability.

Confocal Ca²⁺ imaging

Isolated smooth muscle cells were incubated in solution A containing fluo-4 AM (10 μM) for 20 min at room temperature, followed by a 30-min wash to allow indicator deesterification. Smooth muscle cells were imaged with the use of a Noran Oz laser scanning confocal microscope (Noran Instruments, Middleton, WI) and a ×60 water immersion objective (1.2 numerical aperture) attached to a Nikon TE300 microscope. Fluo-4 was excited by using the 488 nm line of a krypton-argon laser, and emitted light >500 nm was captured. Images (56.3 × 52.8 μm) were recorded every 8.3 ms (120 images/s). The laser intensity used did not alter transient K_Ca current frequency or amplitude. Current and fluorescence measurements were synchronized by using a light-emitting diode placed above the recording chamber that was triggered during acquisition. Each cell was imaged for 15 s. Ca²⁺ sparks in smooth muscle cells were analyzed with the use of software kindly provided by Dr. M. T. Nelson (University of Vermont). Detection of Ca²⁺ sparks was performed by dividing an area 1.54 μm (7 pixels) × 1.54 μm (7 pixels) (i.e., 2.37 μm²) in each image (F) by a baseline (F₀) that was determined by averaging 10 images without Ca²⁺ spark activity. The entire area of each image was analyzed to detect Ca²⁺ sparks. A Ca²⁺ spark was defined as a local increase in F/F₀ that was >1.2.

Fura-2 imaging

Cerebral arteries were incubated with the ratio-metric fluorescent Ca²⁺ indicator fura-2 AM (5 μM) and 0.05% pluronic F-127 for 20 min, followed by a 15-min wash. All experiments were performed using solution B (composition described in Patch-clamp electrophysiology). Fura-2 was alternately excited at 340 or 380 nm using a PC-driven hyperswitch (Ionoptix, Milton, MA). Background corrected ratios were collected every 1 s at 510 nm using a photomultiplier tube (Ionoptix). SR Ca²⁺ load ([Ca²⁺]_SR) was estimated by rapidly applying a high concentration of caffeine (10 mM), an RyR channel activator, and measuring the amplitude of [Ca²⁺]_i transients (i.e., Δ[Ca²⁺]_i). Intracellular Ca²⁺ concentrations were calculated by using the following equation (12):

$$[{\mathrm{Ca}}^{2+}{]}_{\mathrm{i}}=K_{\mathrm{d}}\frac{(\mathrm{R}-{\mathrm{R}}_{\min })}{({\mathrm{R}}_{\max }-\mathrm{R})}\cdot \frac{\left({\mathrm{S}}_{\mathrm{f}2}\right)}{\left({\mathrm{S}}_{\mathrm{b}2}\right)}$$

where R is the 340/380 nm ratio; R_min and R_max are the minimum and maximum ratios determined in Ca²⁺-free and saturating Ca²⁺ solutions, respectively, S_f2/S_b2 is the ratio of Ca²⁺-free to Ca²⁺-replete emissions at 380 nm excitation, and K_d is the dissociation constant for fura-2 (282 nM) (19). R_min, R_max, S_f2, and S_b2 were determined at the end of experiments and in separate experiments by increasing the Ca²⁺ permeability of smooth muscle cells with ionomycin (10 μM) and by perfusing cells with a high-Ca²⁺ (10 mM) or Ca²⁺-free (no added Ca²⁺, 5 mM EGTA) solution.

Measurement of cellular dimensions

Isolated smooth muscle cells were allowed to settle on a glass coverslip in a chamber with Ca²⁺-free, HEPES-containing patch-clamp bath solution (composition described in Patch-clamp electrophysiology). Cell images were acquired with the use of a Zeiss LSM5 confocal microscope. Cell length and width were calculated using Pascal software (Zeiss).

Statistics

Values are expressed as means ± SE. Student's t-test and Student-Newman-Keuls test were used for comparing paired or unpaired data and multiple data sets, respectively. Simultaneous Ca²⁺ spark and transient K_Ca current amplitude data were fit with a linear regression function, and the slope ± SE of each fit was compared using Student's t-test and Graphpad Prizm (San Diego, CA). P < 0.05 was considered significant.

RESULTS

Caveolae are absent in cerebral artery smooth muscle cells of cav-1^−/− mice

Electron microscopy revealed that abundant caveolae were present in the sarcolemma of smooth muscle cells of small-diameter (∼150 μm) cerebral arteries of cav-1^+/+ mice (Fig. 1, A and B). In contrast, caveolae were not observed in cerebral artery smooth muscle cells of cav-1^−/− mice (Fig. 1, C and D). These data indicate that cav-1 is necessary for caveolae formation in smooth muscle cells of small cerebral arteries.

Fig. 1.

Genetic ablation of caveolin-1 (cav-1) abolishes caveolae in cerebral artery smooth muscle cells. A: electron micrograph illustrates sarcolemma of smooth muscle cell in cerebral artery from cav-1^+/+ mouse. Abundant caveolae are present within sarcolemma. B: magnification of area illustrated by black box in A showing caveolae. C: caveolae were absent in sarcolemma of cav-1^−/− cerebral artery smooth muscle cells. D: magnification of area highlighted by black box in C.

Transient K_Ca current frequency is higher in cav-1^−/− than in cav-1^+/+ cerebral artery smooth muscle cells

To investigate the effects of cav-1 ablation on Ca²⁺ spark-to-K_Ca channel signaling, transient K_Ca currents were measured in cerebral artery smooth muscle cells of cav-1^+/+ and cav-1^−/− mice. Transient K_Ca currents were measured at −40 mV, which is the membrane potential of cerebral arteries pressurized to 60 mmHg (19). Mean transient K_Ca current frequency was 1.11 ± 0.14 Hz in cav-1^+/+ cells (n = 35) and 2.20 ± 0.30 Hz in cav-1^−/− cells (n = 31), or approximately twofold higher (Fig. 2, A and B). In contrast, transient K_Ca current amplitude was similar in cav-1^+/+ and cav-1^−/− cells (Fig. 2, A and C).

Fig. 2.

Cav-1 ablation elevates transient large-conductance Ca²⁺-activated potassium (K_Ca) current frequency in cerebral artery smooth muscle cells. A: original traces illustrate transient K_Ca currents in a cav-1^+/+ (top) and cav-1^−/− (bottom) cell at −40 mV. B: mean transient K_Ca current frequency was higher in cav-1^−/− (n = 31) when compared with cav-1^+/+ (n = 35) cells. C: mean transient K_Ca current amplitude was similar in cav-1^+/+ (n = 35) and cav-1^−/− (n = 31) cells. D: mean cell capacitance of cav-1^−/− cells (n = 49) was smaller than for cav-1^+/+ cells (n = 55). E: thapsigargin (100 nM) abolishes transient K_Ca currents in a cav-1^−/− cell voltage-clamped at −40 mV. Similar results were obtained in 4 cells. *P < 0.05 when compared with cav-1^+/+ cells.

Cell capacitance of cav-1^−/− cells (∼8.4 pF) was smaller than for cav-1^+/+ cells (∼12.1 pF; Fig. 2D). To investigate whether cell size underlies the difference in cav-1^+/+ and cav-1^−/− cell capacitance, cellular dimensions were measured. Cells were observed in a Ca²⁺-free bath solution to induce maximal relaxation. Cav-1^+/+ and cav-1^−/− cell length [in μm: cav-1^+/+, 37.1 ± 1.6 (n = 32); cav-1^−/−, 39.4 ± 2.6 (n = 33)] and width [in μm: cav-1^+/+, 9.6 ± 0.3 (n = 32); cav-1^−/−, 9.4 ± 0.3 (n = 33)] were not different (P > 0.05 for each). These data suggest that cav-1 ablation reduces the cell surface area but does not alter the dimensions of cerebral artery smooth muscle cells.

In arterial smooth muscle cells, transient K_Ca currents occur due to SR Ca²⁺ release (38). We sought to determine mechanisms that activate transient K_Ca currents in cav-1^−/− cells. In cav-1^−/− cells, thapsigargin (100 nM), an SR Ca²⁺ ATPase inhibitor, abolished transient K_Ca currents, indicating that these events occur due to SR Ca²⁺ release (Fig. 2E; n = 4).

Ca²⁺ spark frequency is elevated in cav-1^−/− cells when compared with cav-1^+/+ cells, but the amplitude relationship between Ca²⁺ sparks and transient K_Ca currents is similar

Elevated transient K_Ca current frequency in cav-1^−/− cells could occur because of an increase in Ca²⁺ spark frequency or an increase in the percentage of Ca²⁺ sparks that activate a transient K_Ca current (i.e., percent coupling). To investigate these possibilities and to compare Ca²⁺ spark properties in cav-1^+/+ and cav-1^−/− cells, simultaneous measurements of Ca²⁺ sparks and evoked transient K_Ca currents were obtained by performing confocal Ca²⁺ imaging in combination with patch-clamp electrophysiology.

At −40 mV, Ca²⁺ spark frequency was ∼1.8-fold higher in cav-1^−/− cells than in cav-1^+/+ cells (Fig. 3, A and B). In contrast, Ca²⁺ spark amplitude (Fig. 3C) and spatial spread [full width at half-maximal amplitude; cav-1^+/+, 1.82 ± 0.13 μm (n = 40); cav-1^−/−, 1.94 ± 0.11 μm (n = 59)] were similar in cav-1^+/+ and cav-1^−/− cells (P > 0.05 for each). The percentage of Ca²⁺ sparks that activated a transient K_Ca current (cav-1^+/+, 67 ± 8%; cav-1^−/−, 71 ± 4%) and the amplitude relationship between Ca²⁺ sparks and transient K_Ca currents were also similar (Fig. 3, A and D). Taken together, these data indicate that genetic ablation of cav-1^−/− elevates Ca²⁺ spark frequency, leading to an increase in transient K_Ca current frequency.

Fig. 3.

Ca²⁺ spark frequency is higher in cav-1^−/− cells than in cav-1^+/+ cells. A: simultaneous measurements of Ca²⁺ sparks (bottom colored traces) and transient K_Ca currents (top black traces) at −40 mV in a cav-1^+/+ cell (left) and cav-1^−/− cell (right). Inset: peak of Ca²⁺ spark, labeled by asterisk in A, with yellow box in image indicating region from where F/F₀ trace was obtained. B: mean Ca²⁺ spark frequency in cav-1^+/+ (n = 9) and cav-1^−/− (n = 7) cells. C: mean Ca²⁺ spark amplitude in cav-1^+/+ (n = 88) and cav-1^−/− (n = 124) cells. D: amplitude correlation of Ca²⁺ sparks and evoked transient K_Ca currents in cav-1^+/+ (black squares) and cav-1^−/− cells (red circles) with linear fit and 95% confidence bands to each data set [slope ± SE: cav-1^+/+, 14.2 ± 1.4 (r = 0.25, n = 62, P < 0.0001); cav-1^−/−, 16.0 ± 1.1 (r = 0.32, n = 91, P = 0.01)]. Amplitude relationship was not significantly different in cav-1^+/+ and cav-1^−/− cells (P > 0.05). *P < 0.05 when compared with cav-1^+/+ cells.

Cytosolic [Ca²⁺]_i and [Ca²⁺]_SR are similar in cav-1^+/+ and cav-1^−/− cells

Cytosolic [Ca²⁺]_i and [Ca²⁺]_SR regulate Ca²⁺ sparks in arterial smooth muscle cells (6, 38). However, [Ca²⁺]_i and [Ca²⁺]_SR, as determined by caffeine (10 mM)-induced Ca²⁺ transients, were not different in cav-1^+/+ and cav-1^−/− cerebral arteries (Fig. 4, A–C). Thus the higher Ca²⁺ spark frequency in cav-1^−/− cells is not because of elevated cytosolic or SR Ca²⁺ concentration.

Fig. 4.

Intracellular Ca²⁺ concentration ([Ca²⁺]_i) and sarcoplasmic reticulum Ca²⁺ load are similar in cav-1^+/+ and cav-1^−/− cerebral arteries. A: [Ca²⁺]_i and caffeine (10 mM)-induced Ca²⁺ transients were measured in cav-1^+/+ and cav-1^−/− cerebral arteries using fura-2. B: [Ca²⁺]_i was similar in cav-1^+/+ (n = 7) and cav-1^−/− (n = 9) cerebral arteries. C: caffeine-induced Ca²⁺ transients were similar in cav-1^+/+ (n = 7) and cav-1^−/− (n = 9).

Nω-nitro-l-arginine does not inhibit transient K_Ca currents in cav-1^+/+ or cav-1^−/− cells

Cav-1 ablation leads to nitric oxide (NO) synthase (NOS) activation and an increase in NO generation (10, 34). NO donors activate Ca²⁺ sparks in arterial smooth muscle cells (31). Therefore, we investigated whether Ca²⁺ spark frequency is higher in cav-1^−/− cells because of elevated NOS activity. N^ω-nitro-l-arginine (1 mM), a NOS blocker, did not alter transient K_Ca current frequency or amplitude in either cav-1^+/+ [%control; frequency, 122 ± 16; amplitude, 107 ± 21 (n = 6)] or cav-1^−/− [frequency, 106 ± 13; amplitude, 90 ± 17 (n = 5)] cells (P > 0.05 for each). These data suggest that the Ca²⁺ spark frequency elevation in cav-1^−/− cells does not occur through NOS activation.

Transient K_Ca current regulation by voltage-dependent Ca²⁺ channels is abolished in cav-1^−/− cells

In smooth muscle cells, Ca²⁺ sparks are activated by Ca²⁺ entering through sarcolemma voltage-dependent Ca²⁺ channels (14, 16, 20). We investigated whether Ca²⁺ sparks are more frequent in cav-1^−/− cells because of enhanced coupling between voltage-dependent Ca²⁺ channels and RyR channels. In cav-1^+/+ cells, Cd²⁺ (250 μM), a voltage-dependent Ca²⁺ channel blocker, or diltiazem (50 μM), an L-type Ca²⁺ channel blocker, reduced transient K_Ca current frequency to ∼51 and ∼57% of control, respectively, but did not alter transient K_Ca current amplitude (Fig. 5, A–C). In contrast, Cd²⁺ and diltiazem did not alter transient K_Ca current frequency or amplitude in cav-1^−/− cells (Fig. 5, A–C). Thus Ca²⁺ spark regulation by voltage-dependent Ca²⁺ channels is abolished in cav-1^−/− cells, indicating that higher Ca²⁺ spark frequency is not through enhanced coupling between voltage-dependent Ca²⁺ channels and RyR channels.

Fig. 5.

Voltage-dependent Ca²⁺ channel blockers reduce transient K_Ca current frequency in cav-1^+/+ cells but not in cav-1^−/− cells. A: original traces illustrating effect of CdCl₂ (250 μM) on transient K_Ca currents in a cav-1^+/+ and cav-1^−/− cell at −40 mV. B: mean effects of CdCl₂ on transient K_Ca current frequency and amplitude in cav-1^+/+ (n = 5) and cav-1^−/− (n = 6) cells. C: mean effect of diltiazem (50 μM) on transient K_Ca currents in cav-1^+/+ (n = 5) and cav-1^−/− (n = 4) cells. *P < 0.05 when compared with control obtained in same cells before Cd²⁺ or diltiazem application; #P < 0.05 when compared with cav-1^+/+ cells.

Voltage-dependent Ba²⁺ current density is similar in cav-1^+/+ and cav-1^−/− cells

In cav-1^−/− cells, voltage-dependent Ca²⁺ channel density may be reduced or voltage-dependent Ca²⁺ channels may be insensitive to blockers. To test these hypotheses, voltage-dependent Ca²⁺ current density relationships were compared in cav-1^+/+ and cav-1^−/− cells using Ba²⁺ as the charge carrier. Voltage-dependent Ba²⁺ current density was similar, and Cd²⁺ abolished voltage-dependent Ba²⁺ currents in both cav-1^+/+ and cav-1^−/− cells (Fig. 6, A and B). Thus the absence of Ca²⁺ spark regulation by voltage-dependent Ca²⁺ entry in cav-1^−/− cells cannot be explained by alterations in sarcolemma voltage-dependent Ca²⁺ current density or Cd²⁺ sensitivity.

Fig. 6.

Voltage-dependent Ca²⁺ current density is similar in cav-1^+/+ and cav-1^−/− cells. A: original traces illustrating voltage-dependent Ba²⁺ currents (I_Ba) elicited by a voltage step from a holding potential of −80 to +20 mV in cav-1^+/+ and cav-1^−/− cerebral artery smooth muscle cells. CdCl₂ (250 μM) abolished (I_Ba) in cav-1^+/+ and cav-1^−/− cells. B: current density-voltage relationship of CdCl₂ (250 μM)-sensitive currents in cav-1^+/+ (n = 9) and cav-1^−/− (n = 9) cells. P > 0.05 for each voltage when compared with cav-1^+/+.

K_Ca and K_V current density is higher in cav-1^−/− than in cav-1^+/+ cells

In cardiac myocytes, coupling between voltage-gated Ca²⁺ channels and RyR channels is tight, whereas in smooth muscle cells, this communication is loose (5, 20, 24, 36). Conceivably, Ca²⁺ spark regulation by voltage-dependent Ca²⁺ channels may be abolished in cav-1^−/− cells because of an increase in the distance between the sarcolemma and the SR. Therefore, we sought to study the distance between Ca²⁺ spark sites and a sarcolemmal Ca²⁺ spark target in cav-1^+/+ and cav-1^−/− cells. To do this, K_Ca channel properties were measured, thereby allowing further investigation of Ca²⁺ spark to K_Ca channel coupling.

Transient K_Ca current amplitude is dependent on K_Ca channel conductance and Ca²⁺ sensitivity (4). In excised inside-out patches, single K_Ca channel slope conductance between −40 and +40 mV [cav-1^+/+, 268 ± 3 pS(n = 9); cav-1^−/−, 260 ± 3 pS (n = 7); P > 0.05] and K_Ca channel-apparent Ca²⁺ sensitivity were similar in cav-1^+/+ and cav-1^−/− cells (Fig. 7, A and B). Transient K_Ca current amplitude also depends on the number of K_Ca channels activated by a Ca²⁺ spark. On average, inside-out patches pulled from cav-1^−/− cells contained ∼1.5-fold more K_Ca channels than patches obtained from cav-1^+/+ cells (Fig. 7C). This finding was not due to differences in the size of patch pipettes used for these experiments [cav-1^+/+, 28.6 ± 1.2 MΩ (n = 17); cav-1^−/−, 27.1 ± 0.7 MΩ (n = 22); P > 0.05]. These data suggest that sarcolemma K_Ca channel density is higher in cav-1^−/− cells than in cav-1^+/+ cells.

Fig. 7.

Single K_Ca channel properties in cav-1^+/+ and cav-1^−/− cells. A: original records illustrate single K_Ca channel openings in 10 μM Ca²⁺ at −40 and +40 mV in inside-out patches from a cav-1^+/+ and cav-1^−/− cell. B: at −40 and +40 mV, apparent Ca²⁺ sensitivity of K_Ca channels was similar for cav-1^+/+ (n = 8–9 for each [Ca²⁺]) and cav-1^−/− (n = 7 for each [Ca²⁺]) cells (P > 0.05 for each). P_o, open probability. C: inside-out patches from cav-1^−/− cells (n = 22) contained more K_Ca channels than patches from cav-1^+/+ cells (n = 17). C, closed; O, open. *P < 0.05.

To further examine K_Ca channel properties, whole cell K⁺ current density was measured in cav-1^+/+ and cav-1^−/− cells. K⁺ currents were measured by using a strongly buffered Ca²⁺-free pipette solution to abolish Ca²⁺-dependent K_Ca channel activation. Whole cell K⁺ current density was larger in cav-1^−/− cells than in cav-1^+/+ cells (Fig. 8, A and D). Paxilline (300 nM), a selective K_Ca channel blocker (21, 35), partially inhibited K⁺ currents in both cav-1^+/+ and cav-1^−/− cells (Fig. 8B). Paxilline-insensitive current density, which arises because of K_V channel activation, was larger in cav-1^−/− cells than in cav-1^+/+ cells (Fig. 8, B and E). Similarly, paxilline-sensitive current density, which occurs through K_Ca channel activation, was larger in cav-1^−/− cells than in cav-1^+/+ cells (Fig. 8, C and F). These data suggest that cav-1 ablation elevates the sarcolemmal current density of both K_Ca and K_V channels.

Fig. 8.

Whole cell K⁺, K_Ca, and K_V current density is elevated in cav-1^−/− cells. A: original recordings of whole cell K⁺ currents activated by depolarizing voltage steps in a cav-1^+/+ and cav-1^−/− cell. B: paxilline (300 nM) inhibits K⁺ currents in the same cells illustrated in A. C: paxilline-sensitive K⁺ currents. D: mean whole cell K⁺ current density (pA/pF) in cav-1^+/+ (n = 11) and cav-1^−/− (n = 9) cells. E: paxilline-insensitive K⁺ current density. F: paxilline-sensitive K⁺ current density. *P < 0.05 when compared with cav-1^+/+ cells.

DISCUSSION

The impact of cav-1 ablation on the regulation of Ca²⁺ sparks by voltage-dependent Ca²⁺ channels and the communication between Ca²⁺ sparks and K_Ca channels were investigated in arterial smooth muscle cells. Caveolae were not observed in cav-1^−/− cerebral artery smooth muscle cells, consistent with a role for this protein in caveolae formation (10, 34, 39). Cav-1 ablation attenuated coupling between voltage-dependent Ca²⁺ channels and RyR channels that generate Ca²⁺ sparks. Cav-1 ablation also elevated Ca²⁺ spark frequency, although this was not due to an increase in voltage-dependent Ca²⁺ current density, cytosolic or SR Ca²⁺ concentration, or NOS activity. Although there was an increase in K_Ca channel density in cav-1^−/− cells, transient K_Ca current amplitude was similar to that in cav-1^+/+ cells. Taken together, the data suggest that cav-1 abolishment leads to uncoupling between voltage-dependent Ca²⁺ channels and RyR channels that generate Ca²⁺ sparks but does not alter transient K_Ca current activation by Ca²⁺ sparks.

In smooth muscle cells, coupling between voltage-dependent Ca²⁺ channels and RyR channels is “loos” (20). Although structural and molecular mechanisms that establish this coupling process are unclear, loss of voltage-dependent Ca²⁺ channel to RyR channel communication in cav-1^−/− cells likely occurs because of spatial separation of these proteins. Supporting this hypothesis is evidence that although voltage-dependent Ca²⁺ current density was similar in cav-1^+/+ and cav-1^−/− cells, regulation of transient K_Ca currents by voltage-dependent Ca²⁺ channel blockers was abolished in cav-1^−/− cells. In addition, although K_Ca channel density was elevated in cav-1^−/− cells, mean transient K_Ca current amplitude was similar to that in cav-1^+/+ cells. K_Ca channel conductance, Ca²⁺ spark amplitude and spatial spread, the effective coupling of Ca²⁺ sparks to K_Ca channels, and K_Ca channel-apparent Ca²⁺ sensitivity were similar in cav-1^+/+ and cav-1^−/− cells. Caveolae are structural membrane invaginations that may reduce the sarcolemma to SR distance and physically localize RyR channels nearby voltage-dependent Ca²⁺ channels and K_Ca channels in cav-1^+/+ cells. A flattening of the sarcolemma in cav-1^−/− cells may explain the proposed increase in the signaling distance between RyR channels and voltage-dependent Ca²⁺ channels and K_Ca channels. In cav-1^−/− cells, a distant spark would impact a smaller membrane area and induce a lower subsarcolemmal [Ca²⁺]_i elevation. Consistent with our observations, higher K_Ca channel density would be required for distant Ca²⁺ sparks to activate transient K_Ca currents of similar amplitude to those in cav-1^+/+ cells.

Another explanation for the proposed increase in signaling distance is that cav-1 abolishment leads to delocalization of voltage-dependent Ca²⁺ channels and K_Ca channels from within the vicinity of the spark site. Several lines of evidence support such a proposal. In smooth muscle cells, caveolae compartmentalize voltage-dependent Ca²⁺ channels and K_Ca channels (3, 8). In arterial smooth muscle cells, K_Ca channels are proposed not to cluster above Ca²⁺ sparks sites, although in Bufo marinus stomach smooth muscle cells, such clustering may occur (30, 42). In guinea pig bladder smooth muscle, peripheral SR and caveolae are in close proximity, and L-type Ca²⁺ channels located in caveolae strips may be closely opposed to SR RyR channels (25). It is not clear what causes the elevation in K_Ca channel density in cav-1^−/− cells, but in cultured human myometrial smooth muscle cells, K_Ca channels associate with caveolins 1 and 2, and cholesterol depletion with cyclodextrin leads to an increase in whole cell K_Ca current density (3), consistent with the results here. Furthermore, in bovine endothelial cells, cav-1 physically interacts with and inhibits K_Ca channels and this effect can be removed by cholesterol depletion, leading to channel activation (37).

Investigating the effects of chronic caveolae deficiency on smooth muscle Ca²⁺ signaling can provide insights into physiological regulation by these membrane organelles in cav-1^+/+ cells and potential changes that occur during pathophysiology and ontogeny (11, 29, 32, 34, 39). In contrast to the observations here, acute cholesterol depletion with dextrin inhibited arterial smooth muscle Ca²⁺ sparks (23). Acute cholesterol depletion and chronic cav-1 ablation may differentially regulate Ca²⁺ spark frequency through opposing effects on signaling pathways that regulate these events, and in the case of the genetic model, through developmental changes that occur in the sustained absence of cav-1 (10, 39). However, dextrin treatment also induces a change in membrane fluidity that may inhibit Ca²⁺ sparks (26). In agreement, both dextrin and cav-1 ablation abolished coupling between voltage-dependent Ca²⁺ channels and RyR channels that generate sparks (present data and Ref. 23). Whereas dextrin treatment disrupts both caveolar and noncaveolar lipid rafts, in the absence of cav-1, noncaveolar lipid rafts would remain (see Ref. 40). Thus our data suggest that caveolae-localized voltage-dependent Ca²⁺ channels regulate Ca²⁺ sparks in arterial smooth muscle cells. In another previous study, transient K_Ca current frequency was reduced in arterial smooth muscle cells of cav-1^−/− mice (10). It is unclear why data here are contradictory to those observations, but in the previous study, effects of cav-1 deficiency on Ca²⁺ sparks were not measured (10).

Through its amino-terminal scaffolding domain, cav-1 inhibits the activity of many binding partners, including NOS, protein kinase C, and adenylyl cyclase (28, 34). Cav-1 is found in the cytosol (22) and, conceivably, may interact with and inhibit RyR channels. Cav-1 is also found in mitochondria, which regulate Ca²⁺ sparks in cerebral artery smooth muscle cells (7, 41). Cav-1 abolishment may also activate RyR channels indirectly by modulating the activity of signaling pathways that regulate Ca²⁺ sparks (e.g., see Refs. 15 and 41). Data here indicate that cav-1 ablation does not activate Ca²⁺ sparks through NOS activation (31, 33). Because cav-1 ablation may activate RyR channels directly and indirectly, future studies will be required to determine the specific underlying mechanisms. The physiological impact of the transient K_Ca current frequency elevation in cav-1^−/− cells would be a membrane hyperpolarization that reduces voltage-dependent Ca²⁺ channel activity, leading to a decrease in global [Ca²⁺]_i and vasodilation (15). Cav-1^−/− aortas contract less in response to vasoconstrictors and relax more in response to acetylcholine, a vasodilator, when compared with cav-1^+/+ arteries (10, 33), supporting upregulation of a vasodilatory pathway.

In summary, data show that genetic ablation of cav-1 leads to loss of caveolae, an increase in Ca²⁺ spark frequency, abolishment of Ca²⁺ spark regulation by voltage-dependent Ca²⁺ channels, and an increase in K_Ca channel density in cerebral artery smooth muscle cells. In contrast, cav-1 ablation does not attenuate transient K_Ca current activation by Ca²⁺ sparks.