Mechanism of asynchronous Ca²⁺ waves underlying agonist-induced contraction in the rat basilar artery

Abstract

Background and purpose:

Uridine 5'-triphosphate (UTP) is a potent vasoconstrictor of cerebral arteries and induces Ca²⁺ waves in vascular smooth muscle cells (VSMCs). This study aimed to determine the mechanisms underlying UTP-induced Ca²⁺ waves in VSMCs of the rat basilar artery.

Experimental approach:

Isometric force and intracellular Ca²⁺ ([Ca²⁺]_i) were measured in endothelium-denuded rat basilar artery using wire myography and confocal microscopy respectively.

Key results:

Uridine 5'-triphosphate (0.1–1000 µmol·L⁻¹) concentration-dependently induced tonic contraction (pEC₅₀ = 4.34 ± 0.13), associated with sustained repetitive oscillations in [Ca²⁺]_i propagating along the length of the VSMCs as asynchronized Ca²⁺ waves. Inhibition of Ca²⁺ reuptake in sarcoplasmic reticulum (SR) by cyclopiazonic acid abolished the Ca²⁺ waves and resulted in a dramatic drop in tonic contraction. Nifedipine reduced the frequency of Ca²⁺ waves by 40% and tonic contraction by 52%, and the nifedipine-insensitive component was abolished by SKF-96365, an inhibitor of receptor- and store-operated channels, and KB-R7943, an inhibitor of reverse-mode Na⁺/Ca²⁺ exchange. Ongoing Ca²⁺ waves and tonic contraction were also abolished after blockade of inositol-1,4,5-triphosphate-sensitive receptors by 2-aminoethoxydiphenylborate, but not by high concentrations of ryanodine or tetracaine. However, depletion of ryanodine-sensitive SR Ca²⁺ stores prior to UTP stimulation prevented Ca²⁺ waves.

Conclusions and implications:

Uridine 5'-triphosphate-induced Ca²⁺ waves may underlie tonic contraction and appear to be produced by repetitive cycles of regenerative Ca²⁺ release from the SR through inositol-1,4,5-triphosphate-sensitive receptors. Maintenance of Ca²⁺ waves requires SR Ca²⁺ reuptake from Ca²⁺ entry across the plasma membrane via L-type Ca²⁺ channels, receptor- and store-operated channels, and reverse-mode Na⁺/Ca²⁺ exchange.

Introduction

Uridine 5'-triphosphate (UTP) is a potent constrictor of cerebral arteries which exerts its effects through purinergic P2Y receptors and the phospholipase C pathway (Urquilla, 1978; Strobaek et al., 1996; Horiuchi et al., 2001). Brain tissue is especially rich in UTP and cerebral vessels have greater reactivity to UTP compared with other vessels (Shirasawa et al., 1983; Hardebo et al., 1987). UTP may be involved in the regulation of cerebrovascular tone under both physiological conditions and pathophysiological reactions in disease states such as subarachnoid haemorrhage or migraine (Debdi et al., 1992; Boarder and Hourani, 1998; Burnstock, 1998).

Smooth muscle contraction is initiated by an increase of intracellular Ca²⁺ ([Ca²⁺]_i) from resting levels of ∼100 nmol·L⁻¹ to values up to 1 µmol·L⁻¹. In general, the [Ca²⁺]_i profile following stimulation is biphasic, consisting of a rapid transient rise in [Ca²⁺]_i from sarcoplasmic reticulum (SR) Ca²⁺ release followed by a plateau phase, which is mediated by Ca²⁺ entry from voltage-gated Ca²⁺ channels and store/receptor-operated channels (van Breemen et al., 1978; Bolton, 1979; Streb et al., 1983; Putney, 1986). The advent of confocal microscopy has allowed the employment of physiological preparations to examine the Ca²⁺ signals in individual in situ vascular smooth muscle cells (VSMCs) of intact blood vessels. It has since become apparent that the average arterial wall [Ca²⁺]_i observed previously is not representative of the Ca²⁺ signalling events within individual VSMCs, which are capable of generating Ca²⁺ signals with varying spatial and temporal patterns (see Lee et al., 2002). Among these signals are Ca²⁺ waves, which are manifested as changes in [Ca²⁺]_i which travel the length of VSMCs, and constitute a specialized form of agonist-induced Ca²⁺ signalling which appears to be involved in contractile regulation. Since 1994 when they were first described, Ca²⁺ waves have been observed in the smooth muscle fibres of a variety of intact blood vessels, including cerebral vessels (Iino et al., 1994; Asada et al., 1999; Miriel et al., 1999; Jaggar, 2001; Lee et al., 2001; Peng et al., 2001). Although there are likely to be underlying physiological reasons for signalling with Ca²⁺ waves (as opposed to steady state [Ca²⁺]_i elevations), the mechanisms behind how Ca²⁺ waves within individual VSMCs signal for contraction remain poorly understood.

Ca²⁺ waves in cerebral arteries can be induced by a variety of stimuli, including vasoconstrictor agonists such as UTP, pressure and alkaline pH (Jaggar and Nelson, 2000; Jaggar, 2001; Heppner et al., 2002). In cerebral arteries, UTP stimulation shifts Ca²⁺ sparks to Ca²⁺ waves through the differential regulation of inositol-1,4,5-triphosphate receptors (IP₃Rs) and ryanodine receptors (RyRs) (Jaggar and Nelson, 2000). From studies in cultured basilar artery smooth muscle cells, it is generally accepted that UTP induces vasoconstriction by a combination of stimulated plasma membrane Ca²⁺ entry and SR Ca²⁺ release (Sima et al., 1997). However, little is known about the mechanism underlying between agonist-induced Ca²⁺ waves and their relationship to vasoconstriction in the cerebral vasculature.

In our present study, we investigated the mechanism of UTP-induced Ca²⁺ waves in the rat basilar artery, focusing on the mode(s) of Ca²⁺ entry involved in sustaining the UTP-induced cyclical release of Ca²⁺ by identifying the Ca²⁺ transport molecules involved in the generation and maintenance of UTP-induced Ca²⁺ waves.

Methods

Experimental animals and tissue preparation

All animal experiments and procedures were carried out in accordance with the guidelines of the University of British Columbia. Male Sprague-Dawley rats (250–350 g) were obtained from Charles River and housed in the institutional animal facility (University of British Columbia, Child and Family Research Institute) under standard animal room conditions (12 h light–12 h dark, at 25°C, two animals in a cage). Rats were anesthetized with a mixture of ketamine hydrochloride (70 mg·kg⁻¹) and xylazine hydrochloride (5 mg·kg⁻¹) given intraperitoneally. The brain was quickly removed and placed in ice-cold, oxygenated (95% O₂–5% CO₂) Krebs solution. The basilar artery (180–280 µm in diameter) was removed, carefully cleaned and cut into 2 mm segments. Endothelial denudation was achieved by gently rubbing the inside of the vessel with a 40 µm tungsten wire.

Measurement of [Ca²⁺]_i

The arterial rings were loaded with Fluo-4AM (5 µmol·L⁻¹ with 5 µmol·L⁻¹ Pluronic F-127, 1 h at 37°C) and isometrically mounted, followed by a 30 min washout time in HEPES-buffered physiological saline solution (PSS). Sustained Ca²⁺ waves were induced by 100 µmol·L⁻¹ UTP, and all mechanistic studies were carried out at this concentration. Images were acquired on an upright Olympus BX50WI microscope with a 60× water-dipping objective (NA 0.9) and equipped with an Ultraview confocal imaging system (Perkin-Elmer). The tissue was illuminated using the 488 nm line of an Argon-Krypton laser and a high-gain photomultiplier tube collected the emission at wavelengths between 505 and 550 nm. The scanned regions correspond to a 91.685 × 66.68 µm area (or 248 × 328 pixels). The representative fluorescence traces shown reflect the averaged fluorescence signals from a region of 3 × 3 pixels (1.69 µm²) of the VSMC. The rate of image acquisition was three frames·s⁻¹. The frequency of Ca²⁺ waves was determined by counting the number of waves occurring within 40 s. The measured changes in Fluo-4 fluorescence level are proportional to the relative changes in [Ca²⁺]_i. All parameters (laser intensity, gain, etc.) were maintained constant during the experiment. The confocal images were analysed off-line with the Ultraview 4.0 Software (Perkin-Elmer). Fluorescence traces were extracted from the movies to exclude nuclear regions and traces were normalized to initial fluorescence values.

Measurement of isometric force

Basilar artery segments were mounted isometrically in a small vessel wire myograph (A/S Danish Myotechnology, Aarhus N, Denmark) using two 40 µm tungsten wires, for measuring generated force. The chambers were kept at 37°C and bubbled continuously with 95% O₂–5% CO₂ in Krebs solution. Optimal tension (3 mN) was determined in preliminary experiments by subjecting arterial segments to different resting tensions and stimulating with 60 mmol·L⁻¹ KCl. The vessels were stretched to the optimal tension for 60 min and challenged twice with 60 mmol·L⁻¹ KCl before experiments were continued. The percent of contraction compared with the second 60 mmol·L⁻¹ KCl-induced contraction was recorded at different concentrations of UTP and concentration-response curves were constructed. Tonic contraction was induced by 100 µmol·L⁻¹ UTP and all mechanistic studies carried out at this concentration. The negative logarithm (pD₂) of the concentration of UTP giving half-maximum response (EC₅₀) was assessed by linear interpolation on the semilogarithm concentration-response curve [pD₂ = −log(EC₅₀)].

Statistics

Values are expressed as mean ± standard error (SE) from at least six independent experiments. Statistical analysis and construction of concentration-response curves were performed using GraphPad Prism 4.0 software (San Diego, CA, USA). Differences between groups were analysed by Student's two-tailed t-test. Statistical significance was defined as P values <0.05.

Drugs, solutions and chemicals

HEPES-PSS containing (in mmol·L⁻¹) NaCl 140, glucose 10, KCl 5, HEPES 5, CaCl₂ 1.5 and MgCl₂ 1 (pH 7.4) was used for all confocal studies. High-K⁺ (60 mmol·L⁻¹ extracellular K⁺) PSS was identical in composition to normal PSS with the exception of (in mmol·L⁻¹) NaCl 85 and KCl 60. Zero-Ca²⁺ PSS was prepared in the same way as normal PSS, but CaCl₂ was replaced with 1 mmol·L⁻¹ EGTA. Krebs solution containing (in mmol·L⁻¹) NaCl 119, glucose 11.1, KCl 4.7, CaCl₂ 1.6, KH₂PO₄ 1.18, MgSO₄ 1.17 and EDTA 0.023 (pH 7.4) was used for all isometric contraction studies.

Uridine 5'-triphosphate, cyclopiazonic acid (CPA), 2-aminoethoxydiphenylborate (2-APB), nifedipine, SKF-96365, tetracaine and pluronic F-127 were obtained from Sigma-Aldrich (Oakville, Ontario, Canada). Ryanodine and KB-R7943 were obtained from Calbiochem (Gibbstown, NJ, USA). Fluo-4AM was purchased from Molecular Probes (Eugene, OR, USA). All drug and molecular target nomenclature conforms to the British Journal of Pharmacology's Guide to Receptors and Channels (Alexander et al., 2008).

Results

Relation between UTP-induced tonic contraction and UTP-induced Ca²⁺ waves

Uridine 5'-triphosphate produced tonic contraction in a concentration-dependent manner, with a pEC₅₀ of 4.34 ± 0.13 and maximal response (E_max) of 105.5 ± 7.3% (n = 9 animals, Figure 1A,B). At 100 µmol·L⁻¹ UTP, the average contraction was 70.3 ± 4.5% (n = 12 animals, normalized to 60 mmol·L⁻¹ KCl). In parallel experiments, confocal microscopy was used to observe changes in [Ca²⁺]_i within the VSMCs following UTP stimulation. In the absence of UTP, asynchronous Ca²⁺ waves of low amplitude were observed in a small percentage (<10%) of the cells, similar to the ‘Ca²⁺ ripples’ described previously in rat tail artery (Asada et al., 1999). Application of UTP induced a large transient Ca²⁺ response which was followed by sustained repetitive oscillations in intracellular [Ca²⁺]_i which propagated along the length of the VSMC as Ca²⁺ waves (Figures 1C and 2). The frequency of Ca²⁺ waves increased in a concentration-dependent manner, closely paralleling the development of force (Figure 1D, pEC₅₀ = 4.74 ± 0.14, maximum frequency = 0.089 ± 0.007 Hz, 109 cells from 12 animals). At 100 µmol·L⁻¹ UTP, the average frequency of the Ca²⁺ waves was 0.082 ± 0.005 Hz (n = 48 cells from eight animals). The number of cells displaying Ca²⁺ waves was also concentration-dependent; at 100 µmol·L⁻¹ UTP, 91 ± 2.5% of cells displayed at least one Ca²⁺ wave (n = 48 cells from eight animals, Figure 1E). The velocity of wave propagation, illustrated in Figure 1F, also shows a strong dependence on concentration of UTP. At the highest concentrations, wave propagation speeds reached 67.5 ± 3.6 µm·s⁻¹ (n = 10 cells from four animals). The Ca²⁺ waves originated from distinct intracellular foci and propagated down the longitudinal axis of the individual VSMCs (Figure 2). They did not appear to propagate intercellularly, and were sustained during the entire experimental period.

Figure 1.

Properties of uridine 5'-triphosphate (UTP)-induced Ca²⁺ waves underlying tonic contraction in rat basilar artery. (A) UTP (100 µmol·L⁻¹)-induced tonic contraction. Traces are representative of results from six animals. (B) Concentration-response curve for UTP-induced tonic contraction. pEC₅₀ = 4.24 ± 0.13 (n = 9 animals) (C) In parallel experiments, application of UTP (100 µmol·L⁻¹) produced sustained Ca²⁺ oscillations. Experimental Ca²⁺ traces are representative of results from 58 cells from six animals. (D) Concentration-response curve for frequency of UTP-induced Ca²⁺ waves pEC₅₀ = 4.74 ± 0.14 (n = 109 cells from 12 animals). (E) A greater percentage of VSMCs generated Ca²⁺ signals as UTP concentration increased. This recruitment occurred between 3 and 1000 µmol·L⁻¹, with maximal recruitment achieved at 300 µmol·L⁻¹ UTP (n = 90 cells from 10 animals). The number of cells firing is expressed as a percentage of cells responding to maximal concentration. (F) The apparent propagation speed of the Ca²⁺ waves was correlated to increasing UTP concentration (n = 87 cells from 11 animals).

Figure 2.

Uridine 5'-triphosphate (UTP)-induced Ca²⁺ waves in rat basilar artery. (A) [Ca²⁺]_i changes in two intracellular regions from two different smooth muscle cells upon UTP (100 µmol·L⁻¹) stimulation are depicted in the Ca²⁺ traces taken from the steady state of UTP-induced Ca²⁺ waves. It should be noted that the Ca²⁺ waves occurred at different frequencies. Experimental Ca²⁺ traces are representative of results from 58 cells in six animals. (B) Intact rat basilar artery smooth muscle cells challenged with UTP (100 µmol·L⁻¹) displayed Ca²⁺ waves which originated from distinct intracellular foci and propagated along the longitudinal axis of the smooth muscle cells (indicated by AOI1 and AOI2). The area of AOI is 3 × 3 pixels (1.69 µm²). Scale bar = 10 µm.

Dependence of UTP-induced Ca²⁺ waves on extracellular Ca²⁺ influx

There are two potential sources of Ca²⁺ that can contribute to the generation of UTP-induced Ca²⁺ waves: Ca²⁺ release from the intracellular stores and Ca²⁺ influx from the extracellular space. To determine the contribution of extracellular Ca²⁺ to the initiation and maintenance of UTP-induced Ca²⁺ waves, extracellular Ca²⁺ was removed prior to and during 100 µmol·L⁻¹ UTP stimulation respectively. Removal of extracellular Ca²⁺ immediately prior to UTP stimulation reduced the Ca²⁺ signal to only a few transient Ca²⁺ waves (n = 39 cells from six animals, Figure 3A), while UTP-induced Ca²⁺ waves were completely abolished in the absence of extracellular Ca²⁺ within 1 min of treatment (n = 34 cells from five animals) (Figure 3B), showing that extracellular Ca²⁺ influx was necessary for maintenance of Ca²⁺ waves.

Figure 3.

Extracellular Ca²⁺ influx is required for maintenance of uridine 5'-triphosphate (UTP)-induced Ca²⁺ waves. (A) Removal of extracellular Ca²⁺ during ongoing UTP (100 µmol·L⁻¹)-induced Ca²⁺ waves results in their disappearance within 1 min. Traces shown are representative of 39 cells from six animals. (B) Removal of extracellular Ca²⁺ immediately prior to UTP (100 µmol·L⁻¹) stimulation limits Ca²⁺ signalling to transient Ca²⁺ waves. Traces shown are representative of 34 cells from five animals.

To further define the Ca²⁺ entry pathways involved in maintaining UTP-induced Ca²⁺ waves, nifedipine, a selective inhibitor of L-type Ca²⁺ channels, and SKF-96365, an inhibitor of receptor- and store-operated channels, were used. Nifedipine (10 µmol·L⁻¹) reduced the frequency of 100 µmol·L⁻¹ UTP-induced Ca²⁺ waves to 59 ± 4% of control, while the combined application of nifedipine and SKF-96365 (50 µmol·L⁻¹) completely abolished the Ca²⁺ waves (P < 0.001, n = 42 cells from eight animals). In parallel, application of nifedipine (10 µmol·L⁻¹) significantly reduced tonic contraction to 52 ± 4% of control (P < 0.001, n = 6 animals), while the additional application of SKF-96365 (50 µmol·L⁻¹) decreased tonic contraction to 2.2 ± 1.1% of control (P < 0.001, n = 5 animals) (Figure 4A,B). It should also be noted that nifedipine (10 µmol·L⁻¹) completely abolished the contraction induced by 60 mmol·L⁻¹ KCl (data not shown).

Figure 4.

Effect of L-type Ca²⁺ channel antagonist nifedipine and receptor/store-operated channel antagonist SKF-96365 on uridine 5'-triphosphate (UTP) (100 µmol·L⁻¹)-induced Ca²⁺ waves and tonic contraction. (A) The frequency of Ca²⁺ waves is significantly reduced following nifedipine (10 µmol·L⁻¹) application, but not abolished. The nifedipine-insensitive component is completely abolished following addition of SKF-96365 (50 µmol·L⁻¹). The breaks in the record indicate a 2 min interval. (B) UTP (100 µmol·L⁻¹)-induced tonic contraction is significantly reduced by nifedipine (10 µmol·L⁻¹) and almost completely abolished after SKF-96365 (50 µmol·L⁻¹). ***P < 0.001.

In addition to the conventional plasmalemmal Ca²⁺ permeable channels, the Na⁺/Ca²⁺ exchanger operating in the reverse-mode is also an important pathway for Ca²⁺ entry in VSMCs (Lee et al., 2002; Poburko et al., 2006; Fameli et al., 2007). KB-R7943, an inhibitor of reverse-mode Na⁺/Ca²⁺ exchange at low (≤10 µmol·L⁻¹) concentrations, was used to examine whether reverse-mode Na⁺/Ca²⁺ exchange is involved in supporting nifedipine-insensitive Ca²⁺ waves (Iwamoto et al., 1996; Ladilov et al., 1999). The application of KB-R7943 (10 µmol·L⁻¹) abolished nifedipine-insensitive Ca²⁺ waves (P < 0.001, n = 34 cells from six animals) and inhibited tonic contraction to 3.6 ± 1.0% of control (P < 0.001, n = 7 animals) (Figure 5A,B). Application of KB-R7943 (10 µmol·L⁻¹) by itself also abolished UTP-induced Ca²⁺ waves (P < 0.001, n = 29 cells from four animals) and tonic contraction (P < 0.001, n = 4 animals), suggesting that Ca²⁺ entry through reverse-mode Na⁺/Ca²⁺ exchange plays an important role in maintenance of Ca²⁺ waves even when L-type Ca²⁺ channels are operative (Figure 5C,D). Although KB-R7943 (10 µmol·L⁻¹) reduced 60 mmol·L⁻¹ KCl-induced tonic contraction by 10.6% ± 3.0%, this effect was not significant (P = 0.08, n = 5 animals) (Figure 6A). KB-R7943 may also inhibit store-operated channels (Arakawa et al., 2000). To test for possible effects on store-operated channels, we used UTP (100 µmol·L⁻¹) to stimulate sustained Ca²⁺ waves and then applied CPA (10 µmol·L⁻¹), an inhibitor of the sarco(endo)plasmic reticulum Ca²⁺-ATPase, to inhibit SR Ca²⁺ reuptake. This resulted in an elevation of [Ca²⁺]_i, on which KB-R7943 had no effect, although the addition of SKF-96365 brought [Ca²⁺]_i to baseline (Figure 6B). This suggests that KB-R7943 did not abolish the Ca²⁺ waves through blockade of store/receptor-operated channels. It is also important to note that extracellular Na⁺-depletion with the use of zero-Na⁺ PSS also abolished the Ca²⁺ waves, which further supports the role of reverse-mode Na⁺/Ca²⁺ exchange (data not shown).

Figure 5.

Effect of the reverse-mode Na⁺/Ca²⁺ exchanger inhibitor KB-R7943 on uridine 5'-triphosphate (UTP)-induced Ca²⁺ waves and tonic contraction. (A) Blockade of reverse (Ca²⁺-entry) mode Na⁺/Ca²⁺ exchange using KB-R7943 (10 µmol·L⁻¹) abolished the nifedipine-resistant UTP (100 µmol·L⁻¹)-induced Ca²⁺ waves (P < 0.001, n = 34 cells from six animals) (B) Similarly, KB-R7943 (10 µmol·L⁻¹) also inhibited the nifedipine-insensitive tonic contraction to 3.6 ± 1.0% of control (P < 0.001, n = 6 animals) (C) Application of KB-R7943 (10 µmol·L⁻¹) alone reduced the frequency of UTP (100 µmol·L⁻¹)-induced Ca²⁺ waves, followed by complete abolition (P < 0.001, n = 29 cells from four animals) (D) KB-R7943 (10 µmol·L⁻¹) alone also abolished UTP (100 µmol·L⁻¹)-induced tonic contraction (P < 0.001, n = 5 animals).

Figure 6.

Effects of KB-R7943 on L-type Ca²⁺ channels and store/receptor-operated channels in rat basilar artery. (A) Application of KB-R7943 reduced tonic contraction induced by 60 mmol·L⁻¹ KCl by 10.6 ± 3.0% (P = 0.08, n = 5 animals). (B) Application of uridine 5'-triphosphate (UTP) (100 µmol·L⁻¹) followed by cyclopiazonic acid (CPA) (10 µmol·L⁻¹) resulted in a maintained elevation in Ca²⁺. Application of KB-R7943 (10 µmol·L⁻¹) did not affect this plateau response, whereas the addition of SKF-96365 (50 µmol·L⁻¹) abolished the maintained Ca²⁺ elevation and returned to pre-stimulation baseline level. Representative trace shown is typical of the responses obtained in 36 cells from four rats.

Dependence of UTP-induced Ca²⁺ waves on SR Ca²⁺ release

The all-or-none wave-like nature of Ca²⁺ signal in the rat basilar artery suggests regenerative Ca²⁺ release from the SR. If this is the case, blockade of SR Ca²⁺ uptake should completely inhibit the Ca²⁺ waves. The application of CPA (10 µmol·L⁻¹) to ongoing UTP-induced Ca²⁺ waves resulted in a broadening of the Ca²⁺ waves followed by their complete abolition, leaving a significant elevation in baseline [Ca²⁺]_i corresponding to 35 ± 4% of the peak [Ca²⁺] of the Ca²⁺ waves (P < 0.001, n = 38 cells from six animals) (Figure 7A). In parallel, CPA (10 µmol·L⁻¹) also produced a 79 ± 2% inhibition of the UTP-induced tonic contraction (Figure 7B, P < 0.001, n = 5 animals).

Figure 7.

Effect of blockade of the sarco(endo)plasmic reticulum Ca²⁺ ATPase by cyclopiazonic acid (CPA) on uridine 5'-triphosphate (UTP)-induced Ca²⁺ waves. (A) Addition of CPA (10 µmol·L⁻¹) to ongoing UTP (100 µmol·L⁻¹)-induced Ca²⁺ waves completely abolished the oscillations, leaving a small but significant elevation in baseline Ca²⁺ which corresponds to 35 ± 4% of the peak [Ca²⁺] of the asynchronous Ca²⁺ waves (P < 0.001, n = 28 cells from five animals) (B) Application of CPA (10 µmol·L⁻¹) produced a 79 ± 2% inhibition of the UTP (100 µmol·L⁻¹)-induced tonic contraction (P < 0.001, n = 5 animals).

Ca²⁺ release from the SR can be mediated through either the IP₃R and/or the RyR. 2-APB (100 µmol·L⁻¹), an inhibitor of IP₃Rs in VSMCs (Missiaen et al., 2001), was used to examine the role of IP₃Rs in UTP-induced Ca²⁺ waves. Addition of 2-APB (100 µmol·L⁻¹) to ongoing Ca²⁺ waves immediately abolished them (n = 31 cells from five animals), and inhibited tonic contraction to 3.4 ± 0.7% of the control level (P < 0.001, n = 6 animals) (Figure 8A,B). Furthermore, UTP (100 µmol·L⁻¹) failed to elicit a Ca²⁺ transient or contraction in basilar arteries pre-incubated for 30 min with 2-APB (100 µmol·L⁻¹; Figure 8C,D).

Figure 8.

Effect of 2-aminoethoxydiphenylborate (2-APB) on uridine 5'-triphosphate (UTP)-induced Ca²⁺ waves and tonic contraction. (A) The application of 2-APB (100 µmol·L⁻¹) immediately abolished UTP (100 µmol·L⁻¹)-induced Ca²⁺ waves (n = 31 cells from five animals) (B) 2-APB (100 µmol·L⁻¹) decreased UTP (100 µmol·L⁻¹)-induced tonic contraction to 3.4 ± 0.7% of the control level (P < 0.001, n = 6 animals). (C) UTP (100 µmol·L⁻¹) stimulation after pretreatment of vessels with 2-APB (100 µmol·L⁻¹) for 30 min failed to elicit a Ca²⁺ response (P < 0.0001, n = 24 cells from three animals). In contrast, control vessels, without 2-APB (100 µmol·L⁻¹) preincubation, displayed Ca²⁺ waves after UTP stimulation. (D) UTP (100 µmol·L⁻¹) stimulation after pretreatment of vessels with 2-APB (100 µmol·L⁻¹) for 30 min failed to induce contraction (P < 0.0001, n = 5 animals) compared with control vessels without 2-APB preincubation.

It therefore appears that the opening of the IP₃Rs is required for UTP-mediated vasoconstriction and Ca²⁺ waves. However, as the specificity of 2-APB has been questioned, it was important to examine the selectivity of 2-APB in our preparation, especially with regard to Ca²⁺ translocators such as RyRs, sarco(endo)plasmic reticulum Ca²⁺ ATPase, the store-operated channels and the L-type Ca²⁺ channels. As shown in Figure 9A, pretreatment with 2-APB (100 µmol·L⁻¹) did not significantly affect the peak amplitude of caffeine (25 mmol·L⁻¹)-induced Ca²⁺ release (103 ± 8% of the control, P = 0.66, n = 5 animals), and therefore appears to be inactive against RyRs. Furthermore, 2-APB marginally affected SR refilling, as the peak amplitude of the third caffeine-induced Ca²⁺ transient was decreased slightly, but not significantly by 13.4 ± 4.1% (P = 0.11, n = 6 animals).

Figure 9.

Effects of 2-aminoethoxydiphenylborate (2-APB) on the ryanodine-sensitive sarcoplasmic reticulum (SR) Ca²⁺ release channels, sarco(endo)plasmic reticulum Ca²⁺ ATPase, L-type Ca²⁺ channels and store-operated channels in rat basilar artery. (A) Left: Three pulses of caffeine (25 mmol·L⁻¹) were applied with a 5 min interval (breaks in record) between each pulse. Maximum amplitude of the caffeine-induced Ca²⁺ transient from the first pulse reflects control SR Ca²⁺ level as Ca²⁺ from the SR is released through the opened RyR channels. After the addition of 2-APB (100 µmol·L⁻¹), the second pulse of caffeine resulted in a single Ca²⁺ transient whose maximum amplitude is similar to the first pulse (P = 0.66, n = 6 animals). The third pulse of caffeine resulted in a Ca²⁺ transient whose maximum amplitude was slightly, but not significantly, diminished compared with the first pulse (P = 0.11, n = 6 animals). Right: Bar graph comparing the average maximum amplitude of the second and third caffeine pulses to the third pulse (n = 6 animals). (B) Application of 2-APB (100 µmol·L⁻¹) reduced tonic contraction induced by 60 mmol·L⁻¹ KCl by 14.8 ± 4.3% (P = 0.041, n = 6 animals). (C) Application of uridine 5'-triphosphate (UTP) (100 µmol·L⁻¹) followed by cyclopiazonic acid (CPA) (10 µmol·L⁻¹) resulted in a maintained elevation in Ca²⁺ (upper panel). Application of 2-APB (100 µmol·L⁻¹) did not affect this plateau response, whereas the addition of SKF-96365 (50 µmol·L⁻¹) abolished the maintained Ca²⁺ elevation and returned to pre-stimulation baseline level (lower panel). Representative trace shown is typical of the responses obtained in 30 cells from four rats. RyR, ryanodine receptor.

To test for direct effects on Ca²⁺ entry pathways, the effects of 2-APB on L-type Ca²⁺ channels and store-operated channels, two plasmalemmal channels important to UTP-mediated Ca²⁺ waves, were examined. 2-APB (100 µmol·L⁻¹) reduced 60 mmol·L⁻¹ KCl-induced tonic contraction by 14.8 ± 4.3% (Figure 9B, P = 0.041, n = 6 animals). However, this slight inhibition of L-type Ca²⁺ channels cannot account for the complete inhibition of UTP-induced tonic contraction by 2-APB, as blockade of L-type Ca²⁺ channels with nifedipine only reduced force by 52%.

In addition to L-type Ca²⁺ channels, store-operated channels are also important for maintaining the Ca²⁺ waves. 2-APB has been reported to have non-selective effects on store-operated channels (Bootman et al., 2002). The vessel was first stimulated with UTP (100 µmol·L⁻¹) to generate sustained Ca²⁺ waves, and then CPA (10 µmol·L⁻¹) was applied to inhibit SR Ca²⁺ reuptake, resulting in a maintained elevation of [Ca²⁺]_i and depletion of the SR (Figure 9C). The application of 2-APB (100 µmol·L⁻¹) did not affect the [Ca²⁺]_i plateau, although Ca²⁺ returned to baseline upon the subsequent addition of the store-operated channel blocker SKF-96365 (50 µmol·L⁻¹), indicating that in this preparation 2-APB does not inhibit the store-operated channels directly.

Although the opening of IP₃Rs are required for UTP-mediated Ca²⁺ waves and contraction, it does not rule out Ca²⁺ release through the RyR, another type of SR Ca²⁺ release channel which is functionally important in VSMCs. To assess the involvement of RyRs, high-concentration ryanodine (200 µmol·L⁻¹) and tetracaine (100 µmol·L⁻¹) were used to lock RyRs in their closed configuration. Neither ryanodine nor tetracaine had any effect on the frequency of the ongoing UTP-induced Ca²⁺ waves (P = 0.67, n = 33 cells from five animals; P = 0.71, n = 29 cells from four animals respectively) or tonic contraction (P = 0.64, n = 5 animals; P = 0.64, n = 6 animals) (Figure 10A,B,C,D). This supported the notion that Ca²⁺ release from the RyR-dependent SR store is not responsible for the generation of Ca²⁺ waves. To explore this issue further, RyRs were locked in the subconductance state by preincubation with ryanodine (50 µmol·L⁻¹) followed by a brief (1 min) exposure to caffeine (25 mmol·L⁻¹). The first caffeine exposure caused a transient Ca²⁺ response, whereas a second exposure elicited a much-reduced Ca²⁺ transient, and the third failed to elicit any Ca²⁺ transient at all (Figure 10E). We interpreted the final lack of Ca²⁺ transient in response to caffeine to indicate that release of Ca²⁺ through RyRs on the SR was no longer possible due to depletion of SR Ca²⁺ content and/or the locking of RyRs in an open state. UTP (100 µmol·L⁻¹) stimulation immediately after depletion of the RyR-sensitive store no longer elicited a Ca²⁺ response. These results suggest that IP₃Rs and RyRs have access to a common SR Ca²⁺ store, but that opening of RyRs do not appear to be critical for the maintenance of UTP-induced Ca²⁺ waves.

Figure 10.

Effect of ryanodine, tetracaine and caffeine-induced depletion of sarcoplasmic reticulum (SR) Ca²⁺ stores on uridine 5'-triphosphate (UTP)-induced Ca²⁺ waves and tonic contraction in rat basilar artery. (A) Application of a high concentration (200 µmol·L⁻¹) of ryanodine did not affect ongoing UTP-induced Ca²⁺ waves (P = 0.67, n = 33 cells from five animals) (B) Ryanodine (200 µmol·L⁻¹) also did not affect UTP-induced tonic contraction (P = 0.71, n = 5 animals). (C) High concentrations (100 µmol·L⁻¹) of tetracaine also did not affect ongoing UTP-induced Ca²⁺ waves (P = 0.71, n = 29 cells from four animals) (D) Tetracaine (100 µmol·L⁻¹) had no significant effect on UTP-induced tonic contraction (P = 0.64, n = 5 animals). (E) To determine effects of depletion of RyR-sensitive SR Ca²⁺ stores, the artery was exposed to three 1 min treatments of caffeine (25 mmol·L⁻¹) in the continuous presence of ryanodine (50 µmol·L⁻¹) resulted in depletion of SR Ca²⁺ stores. The second stimulation of caffeine produced a much reduced Ca²⁺ transient, while the third stimulation produced no Ca²⁺ response. After depletion of SR Ca²⁺ stores, stimulation with UTP (100 µmol·L⁻¹) in the presence of ryanodine failed to elicit a Ca²⁺ response. Breaks in the record indicate a 5 min interval. RyR, ryanodine receptor.

Discussion

The concept that tonic vasoconstriction is based on asynchronous Ca²⁺ waves in individual smooth muscle cells has had a profound impact on our views on the molecular mechanisms underlying Ca²⁺ signalling in smooth muscle. The presence of agonist-induced Ca²⁺ waves in cerebral arteries has been documented by various groups (Jaggar and Nelson, 2000; Jaggar, 2001; Heppner et al., 2002), but a detailed investigation of their underlying mechanisms has not yet been conducted. We have investigated the link between agonist-induced Ca²⁺ waves and tonic contraction using an in situ preparation of the rat basilar artery and have systematically studied the ionic mechanisms underlying these Ca²⁺ waves, which appear to be similar to those described in VSM from larger conduit blood vessels (Lee et al., 2002). Our studies of UTP-induced Ca²⁺ waves were performed in isometrically stretched arteries, which are similar to the conditions in which UTP-induced Ca²⁺ waves were first described (Jaggar and Nelson, 2000), and may shed new light on how wall tension may be regulated in the basilar artery.

The response to UTP is typified by repetitive transient elevations in Ca²⁺ which originate in distinct intracellular foci and then spread out as waves over the length of the smooth muscle cell. The cells respond independently of each other in that the Ca²⁺ waves are asynchronous, and the cells vary in their sensitivity to UTP, such that recruitment of responding cells increases with increasing UTP concentration. Furthermore, the propagation velocity and frequency also increases with increasing UTP concentration (Figure 1C,D). Functionally, it appears that Ca²⁺ waves underlie tonic contraction, as their inhibition with nifedipine, SKF-96365, KB-R7943, or 2-APB is associated with complete inhibition of force (Figures 4, 5 and 8). Finally, the lack of synchronicity between neighbouring VSMCs explains how summation of individual-cell Ca²⁺ waves can lead to tonic contraction, as the summation of Ca²⁺ signals in all the cells averages out to be a steady-state Ca²⁺ increase in whole vessels (Ruehlmann et al., 2000; Mauban et al., 2001). The apparent importance of Ca²⁺ waves for tonic contraction is further demonstrated when their abolition by CPA markedly reduces force by 80%, although the average [Ca²⁺]_i remains significantly elevated above baseline (Figure 7). This indicates a higher force-to-[Ca²⁺]_i ratio when VSMCs are activated with Ca²⁺ waves as compared with sustained [Ca²⁺]_i, suggesting that Ca²⁺ waves represent a more efficient method to deliver Ca²⁺ to activate myosin light-chain kinase, which is tethered to the contractile filaments (Lee et al., 2001; Wilson et al., 2002). However, ultimately contraction is determined by the level of phosphorylation of myosin light chain (MLC) which is both Ca²⁺-dependent and Ca²⁺-independent (Weber et al., 1999). The way in which Ca²⁺ waves, as opposed to a steady state elevation of Ca²⁺, might be related to this level of phosphorylation is as yet unknown. Further studies are needed to determine the underlying physiological explanation of how Ca²⁺ waves signal for contraction.

The UTP-induced Ca²⁺ waves appear to be propagated by regenerative Ca²⁺ release from the SR network, as depletion of SR Ca²⁺ stores with CPA abolishes the Ca²⁺ waves (Figure 7). Extracellular Ca²⁺ influx appears to be critical for the maintenance of Ca²⁺ waves, although the ability of Ca²⁺ waves to persist for a time in the absence of Ca²⁺ is likely to be due to several Ca²⁺ transport mechanisms. In smooth muscle, a proportion of the Ca²⁺ released by the SR is inevitably extruded to the extracellular space by the actions of the plasma membrane Ca²⁺ ATPase, and in a Ca²⁺-free medium all Ca²⁺ release from the SR is irreversibly lost to the extracellular space (Leijten and van Breemen, 1986). However, removal of Ca²⁺ towards the extracellular space is in competition with SR Ca²⁺ reuptake through the sarco(endo)plasmic reticulum Ca²⁺ ATPase, which allows the SR to continue releasing decreasing amounts of Ca²⁺ to sustain the Ca²⁺ waves. Finally, a third mechanism of Ca²⁺ unloading of the SR during Ca²⁺-free conditions is the transfer of Ca²⁺ release by the peripheral RyRs towards the forward-mode (Ca²⁺-extrusion) Na⁺/Ca²⁺ exchange, a mechanism which as been described in both VSMCs and endothelial cells (Nazer and van Breemen, 1998; Liang et al., 2004). Therefore, without refilling of the SR, all of the Ca²⁺ is eventually extruded resulting in the disappearance of the Ca²⁺ waves.

Influx of extracellular Ca²⁺ through L-type Ca²⁺ channels is central in the control of cerebrovascular arterial diameter (Nelson et al., 1990). However, it is interesting that the UTP-induced Ca²⁺ waves were not abolished by nifedipine, but only reduced in frequency. One mechanism through which frequency could be decreased is that the absence of stimulated Ca²⁺ influx through L-type Ca²⁺ channels may reduce the rate of refilling of the SR Ca²⁺ store. As SR luminal Ca²⁺ can regulate IP₃R channel opening probability, a reduced rate of SR Ca²⁺ refilling can result in a decreased frequency of SR Ca²⁺ release at the wave initiation site (Meldolesi and Pozzan, 1998). Similarly, blockade of L-type Ca²⁺ channels in pressurized mouse mesenteric arteries, which abolished myogenic tone, also reduced the frequency of phenylephrine-induced Ca²⁺ waves (Zacharia et al., 2007). However, pressure-induced Ca²⁺ waves in small rat cerebral arteries were completely abolished with diltiazem (Jaggar, 2001). Although the larger cerebral vessels, such as the basilar artery, share some properties of resistance vessels (Toyoda et al., 1996), these differing observations may be the result of tissue differences, with respect to relative involvement of the various Ca²⁺ entry mechanisms. Furthermore, these apparent mechanistic differences may also be attributed to different physiological conditions, for example pressurization versus tension. For example, the development of myogenic tone may influence the Ca²⁺ signal elicited by agonists (Zacharia et al., 2007). Consequently, comparisons between the mechanisms of Ca²⁺ waves must take differences in vascular beds and experimental preparations into consideration.

Another major finding in this study is that UTP-induced Ca²⁺ waves are abolished by KB-R7943, an inhibitor of reverse-mode Na⁺/Ca²⁺ exchanger-mediated Ca²⁺ entry across the plasma membrane (Iwamoto et al., 1996; Ladilov et al., 1999). The plasma membrane Na⁺/Ca²⁺ exchanger is a transmembrane protein that normally couples the influx of Na⁺ ions to the efflux of Ca²⁺ ions in a 3:1 ratio (Philipson and Nicoll, 2000). However, Na⁺ entry through receptor- and store-operated channels which are functionally coupled to the Na⁺/Ca²⁺ exchanger may influence the dynamics of Na⁺/Ca²⁺ exchange, as Na⁺ accumulates regionally in a restricted sub-plasmalemmal space between the superficial SR and the plasma membrane (Arnon et al., 2000; Poburko et al., 2004; Lemos et al., 2007). This build-up in subcellular Na⁺ was hypothesized to change the electrochemical gradient to favour Ca²⁺ influx through reversal of the Na⁺/Ca²⁺ exchanger, which in turn refills the SR Ca²⁺ stores (Lee et al., 2001). This would explain our findings that the nifedipine-resistant Ca²⁺ waves and tonic contraction are similarly sensitive to both SKF-96365, an inhibitor of store- and receptor-operated channels, and KB-R7943.

However, to achieve reversal of the Na⁺/Ca²⁺ exchanger, the subplasmalemmal Na⁺ concentration ([Na⁺]_subPM) must reach at least the level of the K_m of the exchanger. Although the [Na⁺]_subPM has not been measured in this preparation, we have predicted from our studies with rat aortic smooth muscle cells that reverse-mode Na⁺/Ca²⁺ exchange should occur when [Na⁺]_subPM exceeds 23–25 mmol·L⁻¹, assuming E_m = −60 mV, E_NCX = 3E_Na − 2E_Ca, [Ca²⁺]_o = 1.2 mmol·L⁻¹, [Ca²⁺]_subPM = 500 µmol·L⁻¹ and [Na⁺]_o = 145 mmol·L⁻¹ (where [Ca²⁺]_o = extracellular [Ca²⁺], [Ca²⁺]_subPM = sub-plasmalemmal [Ca²⁺], and [Na⁺]_o = extracellular [Na⁺]) (Poburko et al., 2006). Recently, Poburko et al. provided the first direct demonstration of localized subcellular increases in Na⁺ through receptor-operated/store-operated channels to ≥30 mmol·L⁻¹ (Poburko et al., 2007), which is consistent with estimates of Na⁺ ranging from 24 to 40 mmol·L⁻¹ in ventricular myocytes (Wendt-Gallitelli et al., 1993; Isenberg et al., 2003). In addition, the space constant for the sub-plasmalemmal Na⁺ gradient in ventricular myocytes is 28 nm, which is highly consistent with the intermembrane separation (∼20 nm) in PM-SR junctions in the rat basilar artery preparation (unpublished observations). Furthermore, given that the resting membrane potential in rat basilar artery is approximately −43 mV (Haddock and Hill, 2002), and that a more depolarized membrane decreases [Na⁺]_subPM required for reversal of Na⁺/Ca²⁺ exchange, it seems plausible that reverse-mode Na⁺/Ca²⁺ exchange is a physiological route of Ca²⁺ entry in cerebral arteries. This is especially relevant as our finding that KB-R7943 abolishes Ca²⁺ waves suggests that SR refilling is critically dependent on reverse-mode Na⁺/Ca²⁺ exchange during UTP stimulation (Figure 5C). Although it remains to be investigated, this may serve as an example of privileged delivery of Ca²⁺ from a transport site located in one membrane to a second Ca²⁺ transport site in an apposing membrane, a process which serves to circumvent free diffusion throughout the cytoplasm (Poburko et al., 2004; Fameli et al., 2007). In addition to its inhibition of reverse-mode Na⁺/Ca²⁺ exchange, KB-R7943 has been also been reported to have effects on L-type Ca²⁺ and store-operated channels, neuronal nicotinic acetylcholine receptors, the N-methyl-D-aspartic acid (NMDA) receptor and the noradrenaline transporter (Watano et al., 1996; Sobolevsky and Khodorov, 1999; Arakawa et al., 2000; Iwamoto, 2004). However, in our preparation, KB-R7943 does not significantly inhibit L-type Ca²⁺ channels or store-operated channels, which supported the notion that it is the reverse-mode Na⁺/Ca²⁺ exchange, which is important in refilling the SR to maintain Ca²⁺ waves.

Uridine 5'-triphosphate exerts its effects on metabotropic purinergic P2Y receptors, and has been shown to augment Ca²⁺ release via an increase in cytoplasmic inositol 1,4,5-trisphosphate [Ins(1,4,5)P₃] (Strobaek et al., 1996; Sima et al., 1997). To investigate the role of IP₃Rs, we used 2-APB, a small molecular weight membrane permeable modulator of the IP₃R (Missiaen et al., 2001). However, its use to block IP₃Rs has been criticized for its non-specific effects on other ion transport mechanisms, notably its inhibition of store-operated channels (Broad et al., 2001; Ma et al., 2001; Ratz and Berg, 2006). Importantly in our preparation, 2-APB immediately abolished ongoing Ca²⁺ waves and tonic contraction, and did not affect caffeine-releasable Ca²⁺ stores, which is consistent with an action of 2-APB to block IP₃Rs (Figure 9A). Furthermore, pre-incubation with 2-APB did not elicit a Ca²⁺ response or contraction. It also had only a minor non-significant effect on SR Ca²⁺ reuptake. 2-APB caused no significant inhibition of store-operated channels and, although it does affect L-type Ca²⁺ channels, the slight inhibition observed could not have accounted for the abolition of Ca²⁺ waves (Figure 9B,C). Therefore, our findings support the notion that opening of IP₃R channels is not only responsible for the initial Ca²⁺ release, but is also required for subsequent regenerative release of Ca²⁺ underlying the propagation of the Ca²⁺ waves. Acetylcholine-induced Ca²⁺ waves in rat portal vein myocytes are also dependent on activation of IP₃Rs, although interestingly the IP₃R2 subtype, which is most sensitive to Ca²⁺, appears to be most important for the propagation of Ca²⁺ waves (Morel et al., 2003; Fritz et al., 2008). It should also be noted that the Ca²⁺ waves are maintained by the intrinsic sensitivity of the IP₃R2 subtype to cytosolic [Ca²⁺]_i, and not due to oscillation of Ins(1,4,5)P₃ levels (Fritz et al., 2008). A possible scenario in the rat basilar artery is that UTP-induced Ca²⁺ wave begins with elevation of Ins(1,4,5)P₃. The Ins(1,4,5)P₃ sensitizes the IP₃R to Ca²⁺, and when Ca²⁺ reaches a threshold concentration, the release channels open (Streb et al., 1983; Ferris et al., 1992). As the concentration of UTP is raised, the concentrations of Ins(1,4,5)P₃ and basal [Ca²⁺]_i are also raised, which shortens the time required for Ca²⁺ to reach threshold value for the initiation of the next wave. The regenerative nature depends on the positive feedback of increasing Ca²⁺ on the Ins(1,4,5)P₃ sensitivity of IP₃R. This mechanism, combined with the fact that InsP₃ sensitizes the IP₃R to Ca²⁺, ensures that both the frequency and velocity increase with increasing UTP concentration. However, knowledge of the Ins(1,4,5)P₃ dynamics in our preparation is required before this mechanism can be established.

The observed effect of 2-APB indicates an essential role of IP₃Rs in the initiation and maintenance of UTP-induced Ca²⁺ waves, but does not exclude involvement of RyRs. Although there is general agreement that the initiation of oscillations and waves is a response to agonists acting on sarcolemmal receptors which releases Ca²⁺ from the SR via IP₃Rs, controversy remains whether or not Ca²⁺ release from the IP₃Rs then activates RyRs to generate further release by Ca²⁺-induced Ca²⁺-release and to propagate waves, or whether the entire release process arises from IP₃Rs without significant RyR involvement (Mccarron et al., 2003). The former proposal is supported by studies which showed that drugs which block RyRs often abolish Ca²⁺ oscillations initiated by Ins(1,4,5)P₃-generating agonists (Hyvelin et al., 1998; Boittin et al., 1999; Jaggar and Nelson, 2000). This is possibly due to co-localization of RyRs and IP₃Rs, which allows Ca²⁺ released locally by IP₃Rs to activate adjacent clusters of RyRs by Ca²⁺-induced Ca²⁺ release (Gordienko and Bolton, 2002). On the other hand, some preparations which lack a Ca²⁺-induced Ca²⁺ release mechanism still exhibit Ca²⁺ waves (DeLisle and Welsh, 1992; Lechleiter and Clapham, 1992). Furthermore, in pressurized rat mesenteric artery, RyRs do not appear to play a role in agonist-stimulated Ca²⁺ waves (Lamont and Wier, 2004). It is important to note here that many studies which utilize pharmacological tools to inhibit RyRs, such as the plant alkaloid ryanodine, are complicated by the concentration-dependent effects in different tissues. For example, low concentrations (<100 µmol·L⁻¹) of ryanodine cause persistent opening of the channels which may lead to store depletion (Rousseau et al., 1987, Kanmura et al., 1988, Xu et al., 1994), while higher concentrations are reported to lock RyRs in a closed state to inhibit Ca²⁺ release (Fill and Copello, 2002). Furthermore, the drugs may also block Ins(1,4,5)P₃-mediated Ca signals themselves (either directly or indirectly) without RyR involvement in Ca²⁺ increase.

In our preparation, depletion of the RyR-sensitive Ca²⁺ stores using a combination of caffeine and low concentration of ryanodine to lock the RyRs in a subconductance state eliminated the ability of UTP to induce Ca²⁺ waves (Figure 10). The concentration of ryanodine (50 µmol·L⁻¹) we used which is greater than the concentration which is known to lock RyRs in an open state in smooth muscle (Iino et al., 1988; Kanmura et al., 1988). This suggests that the IP₃R and RyR both access a common SR Ca²⁺ store such that the depletion of RyR stores prevents Ca²⁺ waves, as has been demonstrated (Lepretre and Mironneau, 1994; McCarron and Olson, 2008), but does not prove that RyRs participate in the propagation of Ca²⁺ waves. Therefore, we used a high concentration of ryanodine (200 µmol·L⁻¹) to lock the RyRs in the closed-configuration and found that UTP-induced Ca²⁺ waves were not affected (Figure 10). Additionally, we used tetracaine (200 µmol·L⁻¹), which is not dependent on the opening of RyRs to exert their effects, and also found that the Ca²⁺ waves were not affected (Györke et al., 1997). This is in contrast to the rat cerebral arteries, where ryanodine (10 µmol·L⁻¹) inhibited UTP-induced Ca²⁺ waves (Jaggar and Nelson, 2000). However, it should be noted that in the same preparation, ryanodine also inhibited Ca²⁺ sparks, most likely as a result of SR Ca²⁺ store depletion. Additionally, another possibility is that in our preparation and in others, the RyRs do not play a role because they are not localized near the IP₃Rs. However, well-controlled double-labeling of the IP₃Rs and RyRs at electron microscopic resolutions is required before such a conclusion can be made.

It is interesting to note that the mechanism of UTP-induced asynchronous Ca²⁺ waves elicited in this study shares some similarities to the mechanism of Ca²⁺ oscillations underlying spontaneous vasomotion, as they were not abolished by nifedipine, dependent on a functional SR, and were abolished by antagonists of IP₃ (Haddock and Hill, 2002; 2005). This is more significant in light of the fact that asynchronous Ca²⁺ waves often precede the rhythmic contraction of blood vessels, or vasomotion, which has been observed to occur spontaneously or in response to high concentrations of agonist stimulation, and may have physiological and pathophysiological importance (Gratton et al., 1998; Hudetz et al., 1998, Shimamura et al., 1999; Rücker et al., 2000). In agonist-stimulated vasomotion, asynchronous Ca²⁺ waves are first initiated without the generation of tension. In the presence of the endothelium, the periodic increases in [Ca²⁺]_i activate cGMP-dependent, Ca²⁺-sensitive Cl^- channels, which cause Cl^- currents which depolarize the membrane periodically. The depolarization spreads rapidly through neighbouring cells and activates L-type Ca²⁺ channels, facilitating Ca²⁺ influx which facilitates Ca²⁺-induced Ca²⁺-release to initiate the next Ca²⁺ wave, which will then occur simultaneously in all the nearby smooth muscle cells and generate oscillatory vasomotion (Peng et al., 2001; Rahman et al., 2005).

Similarly, with spontaneous vasomotion, the trigger for synchronicity of Ca²⁺ waves is thought to be due to the activation of a chloride-dependent Ca²⁺ channel (Haddock and Hill, 2002). The resulting depolarization then spreads quickly to neighbouring cells, such that L-type Ca²⁺ channels are simultaneously activated. The resulting Ca²⁺ influx then facilitates Ca²⁺-induced Ca²⁺ release to initiate a synchronous Ca²⁺ release and contraction. In the study by Haddock and Hill (2002), synchronized Ca²⁺ waves were abolished upon blockade of L-type Ca²⁺ channels with nifedipine, but asynchronous Ca²⁺ oscillations persisted in individual cells, which supports the hypothesis that the entrainment of L-type Ca²⁺ channels are important in synchronized Ca²⁺ oscillations. Synchronization of Ca²⁺ oscillations between VSMCs underlying vasomotion is critically dependent on the coordination of Ca²⁺ signals within individual VSMCs leading to synchronized Ca²⁺ responses and the development of simultaneous contractions along the vessel length (Christ et al., 1996). In small vessels, this coordination may be dependent on an intact endothelium, which may be one reason why we did not observe synchronized Ca²⁺ waves upon UTP stimulation in our preparation (Haddock and Hill, 2005).

In summary, the data presented in this article show that several different Ca²⁺ translocating proteins are involved in the generation of Ca²⁺ waves in UTP-stimulated smooth muscle cells of the rat basilar artery. These Ca²⁺ waves appear to be produced by repetitive cycles of SR Ca²⁺ release which are mediated by IP₃Rs, followed by sarco(endo)plasmic reticulum Ca²⁺-ATPase-mediated SR Ca²⁺ reuptake of Ca²⁺ entry involving L-type Ca²⁺ channels, receptor/store-operated channels and reverse-mode Na⁺/Ca²⁺ exchange. In general, the mechanisms of the Ca²⁺ waves in the basilar artery are similar to those in the large conduit vessels, which may indicate a common Ca²⁺ signalling mechanism which initiates and sustains Ca²⁺ waves in the vasculature.

Glossary

Abbreviations:

[Ca²⁺]_i

intracellular Ca²⁺

2-APB

2-aminoethoxydiphenylborate

CPA

cyclopiazonic acid

Ins(1,4,5)P₃

inositol 1,4,5-trisphosphate

IP₃R

inositol-1,4,5-triphosphate receptor

RyR

ryanodine receptor

SR

sarcoplasmic reticulum

UTP

uridine 5'-triphosphate

VSMCs

vascular smooth muscle cells

Conflict of interest

None.