Capacitative calcium entry as a pulmonary specific vasoconstrictor mechanism in small muscular arteries of the rat

Abstract

1. The effect of induction of capacitative Ca²⁺ entry (CCE) upon tone in small (i.d. 200–500 μm) intrapulmonary (IPA), mesenteric (MA), renal (RA), femoral (FA), and coronary arteries (CA) of the rat was examined.

2. Following incubation of IPA with 100 nM thapsigargin (Thg) in Ca²⁺-free physiological salt solution (PSS), a sustained contraction was observed upon reintroduction of 1.8 mM Ca²⁺, which was unaffected by either diltiazem (10 μM) or the reverse mode Na⁺/Ca²⁺ antiport inhibitor KB-R7943 (10 μM). An identical protocol failed to elicit contraction in MA, RA, or CA, while a small transient contraction was sometimes observed in FA.

3. The effect of this protocol on the intracellular Ca²⁺ concentration ([Ca²⁺]_i) was assessed using Fura PE3-loaded IPA, MA, and FA. Reintroduction of Ca²⁺ into the bath solution following Thg treatment in Ca²⁺-free PSS caused a large, rapid, and sustained increase in [Ca²⁺]_i in all the three types of artery.

4. 100 nM Thg induced a slowly developing noisy inward current in smooth muscle cells (SMC) isolated from IPA, which was due to an increase in the activity of single channels with a conductance of ∼30 pS. The current had a reversal potential near 0 mV in normal PSS, and persisted when Ca²⁺-dependent K⁺ and Cl⁻ currents were blocked; it was greatly inhibited by 1 μM La³⁺, 1 μM Gd³⁺, and the IP₃ receptor antagonist 2-APB (75 μM), and by replacement of extracellular cations by NMDG⁺.

5. In conclusion, depletion of intracellular Ca²⁺ stores with Thg caused capacitative Ca²⁺ entry in rat small muscular IPA, MA, and FA. However, a corresponding contraction was observed only in IPA. CCE in IPA was associated with the development of a small La³⁺- and Gd³⁺-sensitive current, and an increased Mn²⁺ quench of Fura PE-3 fluorescence. These results suggest that although CCE occurs in a number of types of small arteries, its coupling to contraction appears to be of particular importance in pulmonary arteries.

Introduction

Capacitative Ca²⁺ entry (CCE) is activated upon depletion of intracellular Ca²⁺ stores and is demonstrable in a variety of nonexcitable cell types (e.g. Putney, 1990; Berridge, 1995). This pathway is thought to be the key regulatory mechanism in the maintenance of Ca²⁺ loading of intracellular stores within the sarcoplasmic/endoplasmic reticulum (SR/ER; for a review see Berridge, 1995). However, in recent years it has become apparent that CCE may also be of importance in the regulation of a number of diverse cellular functions such as apoptosis, secretion, and gene transcription (for review see Parekh & Penner, 1997). Although the precise molecular identity of the ion channels involved in CCE is not known, it has been suggested that CCE in vascular smooth muscle is mediated by one or more of a family of seven proteins homologous to the transient receptor potential (TRP) protein of Drosophila (Hardie & Minke, 1993; Birnbaumer et al., 1996; Inoue et al., 2001; Xu & Beech, 2001).

A number of reports have provided evidence for the involvement of CCE in the regulation of smooth muscle tone in a variety of preparations, including mouse anococcygeus (Wallace et al., 1999), guinea-pig and cat gastric fundus (Petkov & Boev, 1996a, 1996b), and rat ileum (Ohta et al., 1995), urinary bladder (Munro & Wendt, 1994), and spleen (Burt et al., 1995; reviewed by McFadzean & Gibson, 2002). Activation of CCE, usually via the application of the selective SR/ER ATPase (SERCA) inhibitors thapsigargin (Thg) or cyclopiazonic acid, has also been shown to modulate tone in vascular smooth muscle preparations. The inhibition of SERCA, and subsequent induction of CCE, has been reported to result in constriction of several conduit arteries, including rat pulmonary (de la Fuente et al., 1995), femoral (Nomura et al., 1997), and carotid arteries (Sekiguchi et al., 1996) and also isolated rat aorta (Kwan et al., 1994; Tepel et al., 1994; Low et al., 1996).

We have previously presented preliminary evidence that activation of CCE in smooth muscle cells (SMC) isolated from the rat intrapulmonary artery (IPA) results in the activation of a Ca²⁺-permeable cation channel with a conductance of ∼30 pS (Snetkov et al., 2001). Similar results have also been reported by Ng & Gurney (2001), who found that SMC isolated from rat main pulmonary artery possess a small inward current associated with Ca²⁺ influx that is activated upon inhibition of SERCA. McDaniel et al. (2001) demonstrated a much larger CCE-associated current in cells cultured from the same preparation. These currents appeared to differ not only in amplitude, but also in their sensitivity to La³⁺, which blocked the current described in the former report with an IC₅₀ of 600 μM, but virtually abolished the current in the cultured PA cells at a concentration of 50 μM.

Little information exists as to the functional effects of CCE in the resistance vasculature, although Flemming et al. (2002) have recently shown that store depletion causes Ca²⁺ influx, but no constriction in rabbit pial arterioles. Curtis & Scholfield (2001) have also demonstrated the presence of a nifedipine-sensitive CCE-associated current in rabbit choroidal arteriolar myocytes, but did not assess whether this current contributed directly to contraction. In the present study, we have examined the effects of induction of CCE upon tone in small intrapulmonary (IPA), mesenteric (MA), renal (RA), femoral (FA), and coronary (CA) arteries of the rat. We report that although induction of CCE is uniformly associated with a rise in [Ca²⁺]_i, it elicits contraction only in the small pulmonary arteries. CCE in these arteries is associated with the activation of a Na⁺- and Ca²⁺-permeable channel with a conductance of ∼30 pS. In contrast to the work of Ng & Gurney (2001), we observe that both the contraction and channel activity associated with CCE are potently blocked by La³⁺.

Methods

Tension measurement in small muscular arteries

Male Wistar rats (250–350 g) were anaesthetised with sodium pentobarbital (55 mg kg⁻¹ IP) and killed by cervical dislocation as approved by the local Home Office Inspector. The heart, lungs, intestines, kidneys and hind limbs were excised and placed in cold physiological salt solution (PSS, see below). Small (150–600 μm internal diameter (i.d.), median ∼300 μm) IPA, CA, MA, RA, and FA were dissected free of adventitia, mounted in a temperature-controlled myograph at 37°C (Cambustion AM10, Cambustion Ltd, Cambridge, U.K.) or dual channel Mulvany–Halpern myograph, (Danish Myo Technology, Aarhus, Denmark), and gassed continuously with 95% air/5% CO₂ (pH 7.35). After 30 min, a length–tension curve was obtained, and the artery stretched to 90% of the diameter that is equivalent to a transmural pressure of 30 mmHg for pulmonary and 100 mmHg for systemic arteries as previously described (Leach et al., 1994). The viability of the preparation was assessed by measuring the response to 80 mM K⁺ PSS (KPSS; isotonic replacement of NaCl by KCl), and was deemed suitable for use in experiments only when they produced at least 4 and 8 mN mm⁻¹ length for pulmonary and systemic arteries, respectively. Prior to beginning the experiments, arteries were equilibrated with four 2 min exposures to KPSS. Tension recording from small arteries has previously been described in detail (Leach et al., 1994; Robertson et al., 1995).

Measurement of intracellular [Ca²⁺]_i in isolated arteries

In some experiments, [Ca²⁺]_i was measured simultaneously with tension. In this case, following mounting, stretching, and assessment of viability, an artery was incubated with PSS containing 2 μM Fura PE-3/AM for 90 min at room temperature, followed by a further 30 min at 37°C to facilitate intracellular hydrolytic cleavage of the dye. The preparation was then washed several times with fresh PSS and stimulated repeatedly with high KPSS till a stable tension response was achieved. The myograph was mounted on an inverted microscope equipped with spectrofluorimeter (Cairn Research Ltd, Faversham, U.K.) and illuminated alternately with a UV light source at 340 and 380 nm, while measuring the intensities of emitted light using a photomultiplier tube. The ratio of intensities of emitted light (at >510 nm) corresponding to excitation wavelengths of 340 and 380 nm (R_340/380) was taken as a measure of [Ca²⁺]_i.

Protocol for induction of CCE with Thg

Following the equilibration procedure, the myograph bath solution was exchanged for nominally Ca²⁺-free PSS for a period of 10 min. Thg (100 nM) was then added to the myograph bath for a period of 10 min, after which 1.8- mM Ca²⁺ was then reintroduced into the bath solution for 10 min. The myograph bath solution was then replaced again with Ca²⁺-free PSS. As shown in Figure 1, CCE-related contraction stabilized at the third reintroduction of Ca²⁺; therefore, in all experiments only the third response was included in the analysis.

Figure 1.

Thg induces sustained constriction in rat small isolated IPA. (a) The trace shows a typical response of rat 232 μm IPA to 100 nM Thg (applied in Ca²⁺-free PSS). Upon the reintroduction of Ca²⁺ to the myograph bath, a sustained constriction was observed, which was reversed upon removal of Ca²⁺ from the myograph bath. The response increased with repeated Ca²⁺ exposures, yet was reproducible subsequent to the third exposure. This response was unaffected by the voltage-gated channel antagonist diltiazem (10 μM), but was completely abolished by 1 μM La³⁺. (b) shows the mean response from 10 to 26 IPA.

To prevent precipitation, experiments involving La³⁺ were performed with HEPES-buffered PSS gassed with air, containing (mM): NaCl 130, MgCl₂ 1, glucose 5.56, CaCl₂ 1.8, KCl 4, and HEPES 10, with pH adjusted to 7.4 with NaOH.

Cell isolation for electrophysiology

Dissected small arteries were cut open and placed in 2 ml of Ca²⁺-free HEPES-buffered PSS, containing (mM): NaCl 135, KCl 5, MgCl₂ 2, glucose 11, and HEPES 10, pH adjusted to 7.4 with NaOH at 37°C for 20 min. Tissues were then placed in the same solution containing in addition 2 mg ml⁻¹ collagenase (Sigma type XI), 1 mg ml⁻¹ papain, 1 mg ml⁻¹ soyabean trypsin inhibitor, and 1 mM dithiothreitol for 40 min. The tissue was then washed with an enzyme-free dispersion solution four times and left at 37°C for another 20 min. The solution volume was then reduced to ∼0.5 ml and the tissue was gently triturated with a fire-polished Pasteur pipette to release single cells. These were stored in a dispersion solution at 4°C, and then allowed to settle onto a coverslip, which formed the base of the recording chamber, prior to application of the HEPES-buffered solution used for current recording.

Current recording

Single SMC were studied with standard patch-clamp techniques using an Axopatch-1C amplifier (Axon Instruments Inc., Foster City, CA, U.S.A.). The experiments were performed at room temperature. The bath (∼0.2 ml volume) was continuously superfused (1–2 ml min⁻¹) with HEPES-buffered PSS. It should be noted that EGTA was not added to external solutions containing La³⁺ and Gd³⁺ in order to avoid chelation. Agonists and drugs were applied via the perfusion system. Whole-cell currents were filtered with −3 dB, low-pass 80 dB dec⁻¹, Bessel-type built-in filter, and digitized using a CED 1401 interface and PATCH v.6 software (Cambridge Electronic Design, Cambridge, U.K.) as appropriate. Whole-cell current–voltage relationships were obtained using a voltage ramp protocol, where the holding potential was maintained at −60 mV, and a 0.5 s ramp from −100 to +100 mV was applied every 5 s. All traces shown are representative of at least four similar experiments.

Measurement of Fura PE-3 quench in SMC isolated from IPA

Single SMC isolated from IPA as above were loaded with Fura-PE3 via incubation of the cells with Fura PE-3/AM (1 μM) for 30 min at room temperature. A drop of cell suspension was placed on the bottom of the chamber mounted on an inverted microscope (Nikon Diaphot, Nikon U.K. Ltd, Kingston, U.K.) equipped with a spectrofluorimeter. SMC were allowed to attach to the bottom of the chamber, washed to remove excess Fura PE-3/AM, and left for another 20 min. Fluorescence recordings were made at room temperature in HEPES-buffered solution with excitation wavelengths of 340, 360, and 380 nm. Changes in [Ca²⁺]_i were assessed by calculating the ratio of the light intensities emitted (>510 nm) when the vessel was illuminated at 340 and 380 nm, respectively. Fluorescence quench was recorded at 360 nm using either 1 mM Mn²⁺ to study the basal influx or 0.1 mM Mn²⁺ for Thg-induced influx, where the quench rate was much higher.

Solutions

PSS used for tension recording contained (mM): NaCl 118, NaHCO₃ 24, KCl 4, CaCl₂ 1.8, MgSO₄ 1, NaH₂PO₄ 0.434, and glucose 5.56. HEPES-buffered PSS used for tension recording in the presence of lanthanides contained (mM): NaCl 130, KCl 4, CaCl₂ 1.8, MgCl₂ 1, glucose 5.56, and HEPES 10, pH adjusted to 7.4 with NaOH. Low Na⁺ PSS contained: NMDG 118, choline bicarbonate 24, CaCl₂ 1.8, MgSO₄ 1, NaH₂PO₄ 0.434, and glucose 5.56, pH adjusted to 7.4 with HCl.

Electrophysiology

The HEPES-buffered bath solution contained (mM): NaCl 135, KCl 5, CaCl₂ 2, MgCl₂ 2, glucose 5, and HEPES 10, pH adjusted to 7.4 with NaOH. In experiments designed to study cation permeability, the latter solution was modified by replacing 140 mM Na⁺ and 5 mM K⁺ with 140 mM K⁺, 140 mM Cs⁺, or 80 mM Ca²⁺, respectively. Ca²⁺-free solution had Ca²⁺ omitted; no EGTA was used except when 0.1 mM EGTA was added to remove La³⁺ or Gd³⁺. Ca²⁺-, Mg²⁺-free solution had no Ca²⁺ or Mg²⁺ and contained 1 mM EDTA. The pipette solution contained (mM): 140 mM KCl, 2 mM MgCl₂, 0.1 mM EGTA, and 10 mM HEPES, pH adjusted to 7.2 with KOH. In experiments designed to minimize K⁺ and Cl⁻ currents, and with strong intracellular Ca²⁺ buffering, the pipette solution contained (mM): Cs methanesulphonate 120, BAPTA 10, and HEPES 10, pH adjusted to 7.2 with CsOH. NMDG⁺, La³⁺, Gd³⁺, and Mn²⁺ were used as chloride salts.

Tension is presented as a percentage of the maximum tension (T_K) obtained to the final exposure to KPSS during the equilibration procedure. Mean changes in [Ca²⁺ ]_i are expressed as a percentage of the change in R_340/380 induced by KPSS; although not linearly related to [Ca²⁺]_i, this provides a reasonable qualitative index. Results are expressed as mean±s.e.m., and means were compared using paired or unpaired Student's t-test (SigmaStat, SPSS Inc., Chicago, U.S.A.). A difference was deemed significant if P<0.05.

Results

Comparison of the effects of Thg in rat small arteries

Following incubation of rat small IPA with 100 nM Thg in Ca²⁺-free PSS, reintroduction of Ca²⁺ resulted in a significant sustained contraction (23.7±4.0% T_K, n=26). The size of this response increased with repeated 10 min exposures to Ca²⁺ (10 min apart), typically stabilizing after the third exposure to Ca²⁺ (Figure 1; second Ca²⁺ exposure=33.5±4.5% T_K, n=24; third Ca²⁺ exposure=38.6±4.4% T_K, n=18; fourth Ca²⁺ exposure=38.4±5.7% T_K, n=10). In all experiments described subsequently, IPAs were given three 10 min exposures to 1.8 mM Ca²⁺, 10 min apart, prior to experimental protocols, and time-matched controls were performed in every experiment. In accordance with our previous findings (Robertson et al., 2000), this constrictor response was completely blocked by preincubation with the trivalent cation La³⁺ at 1 μM (Figure 1b; Thg response to 1.8 mM Ca²⁺ in the presence of 1 μM La³⁺=0.6±0.2% T_K, n=8, P<0.0001). This response, however, was unaffected by the voltage-gated Ca²⁺ channel inhibitor diltiazem (10 μM, Figure 1a, n=4).

As illustrated in Figure 2, no sustained contractile response could be observed to Thg in any of the ‘systemic' arteries examined in this study. However, in about 1/3 of the FA examined, a small transient contraction that decayed to baseline within 2–3 min was observed (peak response ∼9% T_K, see Figure 4 and Figure 5). Removal of the endothelium had no effect on the response of either IPA or MA studied using this protocol (data not shown).

Figure 2.

CCE-induced constriction is not observed in rat small MA, RA, CA, or FA. (a, b, c, d) traces show typical responses to 100- nM Thg and subsequent reintroduction of Ca²⁺ in rat MA (210 μm i.d.), RA (303 μm i.d.), CA (187 μm i.d.), and FA (324 μm i.d.), respectively, to 100 nM Thg. No constriction was ever observed in MA, RA, or CA upon reintroduction of Ca²⁺ to the myograph bath (n=8–12). A small transient contraction was observed in four of the 12 FA examined (see Figure 4).

Figure 4.

Simultaneous recording of tension development and [Ca²⁺]_i during application of KPSS and Thg in IPA (upper traces), MA (middle traces), and FA (lower traces). Following addition of KPSS, arteries were exposed to a nominally Ca²⁺-free solution for 10 min. Subsequent reapplication of Ca²⁺ caused a small transient increase in [Ca²⁺]_i. Arteries were then exposed to Ca²⁺-free solution, and 100 nM Thg was applied. Reapplication of Ca²⁺ in the presence of Thg caused a large and sustained increase in [Ca²⁺]_i, which was stable when the cycle of Ca²⁺ removal and replacement was repeated. This increases in [Ca²⁺]_i was associated with a sustained contraction in IPA, with no contraction in MA. A transient contraction was sometimes observed in FA, but tended to wane with further cycles of Ca²⁺ removal and replacement.

Figure 5.

Summary of CCE-related tension development and [Ca²⁺]_i in IPA, MA, and FA. Effects of reapplication of extracellular Ca²⁺ following Thg treatment on tension development and [Ca²⁺]_i obtained in individual IPA (n=5), MA (n=6), and FA (n=4) during experiments similar to those shown in Figure 4. The mean±s.e.m. of these responses are shown, with the solid bars representing tension (left axis) and the open bars representing the increase in [Ca²⁺]_i, expressed in terms of the raw R_340/380 values (right axis).

The above experiments do not rule out the possibility that the apparent absence of CCE-related contraction in the systemic arteries was due to a less effective Ca²⁺ store depletion by Thg in these arteries compared to IPA. In order to address this possibility, we used the procedure illustrated in Figure 3a. Arteries were preincubated for 90 s in Ca²⁺-free solution containing 0.5 mM EGTA, to remove extracellular Ca²⁺, and then challenged with 10 μM PE (maximally effective concentration). This caused a transient contraction, dependent on Ca²⁺ release, which provided an index of the amount of releasable Ca²⁺ in the store. Arteries were then placed in normal PSS for 10 min. This led to a large initial transient contraction, which was presumably caused by the brief simultaneous presence of PE and Ca²⁺ in the bath, since it was absent if PE was removed 2 min before Ca²⁺ was reapplied (n=8, data not shown). Thg (100 nM) was then applied for 10 min, after which the arteries were again incubated in Ca²⁺-free PSS before being exposed to a second PE challenge. The arteries, still in Ca²⁺-free PSS, were then exposed to 10 mM caffeine for 2 min, before a third and last PE challenge, after which normal Ca²⁺-containing PSS was reapplied (Figure 3a).

Figure 3.

Contraction after depletion of cellular Ca²⁺ store with Thg (100 nM), caffeine (10 mM), and phenylephrine (PE, 10 μM). (a) The trace shows three responses to PE, administered 90 s after the removal of Ca²⁺ from the solution (with addition of 0.5 mM EGTA), in a rat mesenteric resistance artery. After a typical transient control PE contraction (left), stores were refilled by returning Ca²⁺ to the solution, and Thg was applied. This almost abolished the subsequent response to PE in Ca²⁺-free solution. After the application and removal of caffeine and then PE in the absence of Ca²⁺ to ensure further Ca²⁺ store depletion, readdition of Ca²⁺ to the solution (at the end of the long open bar) did not cause contraction. (b) A protocol identical to that described above was carried out in a rat IPA. Here, Ca²⁺ reapplication after store depletion caused a large contraction (right). (c) The mean±s.e.m. amplitude of the three PE contractions were recorded as described above in small FA, MA, RA, and CA. n=4; *P<0.05; **P<0.01.

Thg itself had no effect on tension in MA, RA, FA, or CA, but the tension induced by the subsequent PE challenge was markedly reduced, reflecting a substantially reduced content of the PE-releasable Ca²⁺ store (Figure 3a, c, solid and open columns). In all types of arteries studied, subsequent treatment with caffeine induced no increase in tension, and the following PE challenge was also without significant effect (Figure 3c). These data strongly suggest that the releasable Ca²⁺ stores were essentially empty at this point. Nevertheless, in MA, RA, FA, and CA, reapplication of Ca²⁺-containing PSS caused no increase in tension whatsoever (Figure 3a), whereas in IPA it caused a rapid and powerful contraction (Figure 3b). These experiments show firstly that Thg was effective in depleting intracellular stores of Ca²⁺ in all the five types of resistance artery, and secondly that the difference between IPA and the systemic arteries persisted even when store depletion was further assured by treating the arteries with PE and caffeine in addition to Thg.

It should be noted that consistent with previous reports (Priest et al., 1997), PE caused only a very small transient contraction in IPA in the absence of extracellular Ca²⁺, although this was also reduced following application of Thg. Additionally, in IPA, Thg caused a gradually developing contraction in normal PSS (Figure 3b) in contrast to the systemic arteries where it had no effect on tension.

Effect of Thg on [Ca²⁺]_i in isolated arteries

Effects of Thg treatment on [Ca²⁺]_i were investigated in IPA, FA, and MA. Figure 4 illustrates the simultaneous records of tension and [Ca²⁺]_i obtained in an IPA (top), MA (middle), and FA (bottom). In each case, readmission of Ca²⁺, following treatment of the artery with Thg in Ca²⁺-free PSS, caused an increase in R_340/380, which was comparable to (and in the case of FA, much larger than) the increase observed during application of KPSS. In the FA, the rise in [Ca²⁺]_i in response to KPSS was consistently smaller than in the IPA or MA. Conversely, as described above, tension development upon Ca²⁺ readmission was negligible in the MA, substantial in the IPA, and small and transient, or absent, in the FA. Pooled measurements from a number of arteries of the CCE-mediated increases in [Ca²⁺]_i and tension are presented in Figure 5, with the bar for tension development in the FA representing the peak of the transient response. These results indicate that CCE occurs in each type of artery, but that it appears to be differentially coupled to tension development.

Effect of Thg upon inward currents in SMC isolated from small IPA

The effect of Thg on the membrane current was studied in cells isolated from the IPA. The high input resistance of these cells meant that the holding current needed to maintain the membrane potential of the whole cell at −80 mV was very small, and occasionally spontaneous single-channel events could be detected in whole-cell mode as inward currents of ∼3 pA (Figure 6). Thg (100 nM) in Ca²⁺-free solution caused a progressive increase in the frequency of openings, the overlapping of which resulted in a strongly fluctuating inward current with amplitude of 20–30 pA within 3 min (Figure 6). The development of the inward current in the presence of Thg occurred both with standard KCl-based intracellular solution and with a Cs methanesulphonate pipette solution containing 10- mM BAPTA to minimize the contribution of Ca²⁺-dependent K⁺ and Cl⁻ currents (not shown).

Figure 6.

Thg causes an inward current in SMC isolated from rat small IPA. Below, the typical trace of the noisy inward current developing in SMC isolated from small IPA in response to 100 nM Thg. Above, the expanded sections of the same trace showing the dramatic increase of single-channel activity. Holding potential −80 mV, intracellular solution – Cs methanesulphonate, 10 -mM BAPTA. Lines indicate zero current level. Traces shown are typical of five to seven experiments.

The trivalent cations La³⁺ or Gd³⁺ (both 1 μM) were found to block the Thg-induced current rapidly (Figure 7a); for example, in four cells La³⁺ inhibited the current by 84±4%. Their effect could be reversed by washing with Ca²⁺-free extracellular solution containing 0.1 mM EGTA, which strongly chelates lanthanides (Halaszovich et al., 2000). The replacement of all extracellular cations by NMDG⁺ also inhibited the Thg-induced current by more than 90% (Figures 7b, c).

Figure 7.

Block of Thg-induced inward current. (a) Trivalent cations La³⁺ and Gd³⁺ block Thg-induced inward current; this block could be reversed by chelating lanthanides with EGTA. (b) 2-APB caused partial and reversible block of the Thg-induced inward current, while replacement of all extracellular permeable cations by NMDG⁺ abolished the current completely. (c) Single channels underlying Thg-induced whole-cell current were revealed in the presence of 2-APB, which reduced the opening frequency without affecting conductance. No activity could be observed in NMDG⁺-based solution. Holding potential −80 mV, intracellular solution – Cs methanesulphonate, 10 mM BAPTA. Lines indicate zero current level. Traces shown are typical of six to experiments.

Thg-induced current was not significantly different in Ca²⁺-free extracellular solution (n=8). However, the current increased considerably when both Ca²⁺ and Mg²⁺ were removed using a solution containing 1 mM EDTA to chelate any traces of divalent cations (Figure 8a). Both K⁺ and Cs⁺ supported the current to a certain extent and a current of ∼40% of the normal amplitude persisted when 80 mM Ca²⁺ was the only extracellular cation (Figure 8b). This would be consistent with the channel having a somewhat smaller permeability to Ca²⁺ than to Na⁺.

Figure 8.

Ion permeability of Thg-induced inward current. (a) Removal of divalent cations with 1 mM EDTA considerably enhanced Thg-induced current. (b) 140 mM K⁺, 140 mM Cs⁺, and 80 mM Ca²⁺ supported the inward current to some extent, although it was smaller than with 140 mM Na⁺ (control). The lines indicate zero current level. Traces shown are typical of six to eight experiments. (c) When the membrane potential was ramped from −100 to +100 mV in the absence of both intra- and extracellular K⁺, Thg treatment (1 μM, 3 min) caused the development of nonrectifying current. (d) After treatment with Thg, the replacement of extracellular Na⁺ for NMDG⁺ suppressed the inward current. Holding potential −80 mV, intracellular solution – Cs methanesulphonate, 10 mM BAPTA. Typical traces are from four to eight experiments.

When cells were ramped from –100 to +100 mV in the absence of both intra- and extracellular K⁺, Thg treatment caused the development of a current which reversed at ∼0 mV (Figure 8c). This current was strongly inhibited at negative potentials by replacement of extracellular Na⁺ with NMDG⁺ (Figure 8d). In five cells, the inward current at –80 mV was reduced by 76.5±5.6%, whereas the outward current at +80 mV was reduced by only 19.2±7.1%, when NMDG⁺ replaced Na⁺.

Effect of 2-aminoethoxydiphenyl borate (2-APB) and KB-R7943

The putative membrane-permeable IP₃ receptor antagonist 2-APB has been reported to block both endogenous store-operated channels and heterologously expressed TRP3 channels in HEK293 cells (Ma et al., 2000). We examined the effect of 2-APB on both IPA constriction and Thg-induced inward currents. 2-APB (75 μM) was found to reduce the inward current (Figure 7b), and in doing so revealed individual channel openings (Figure 7c) similar in amplitude to those recorded spontaneously, and following Thg treatment. 2-APB (75 μM) also completely abolished the Thg-induced contraction in small IPA (n=14), and caused a small but significant inhibition of the KPSS contractile response (KPSS in the presence of 75- μM 2-APB=78.4±11.8% T_K, P<0.05, n=5).

Since the current observed after Thg treatment was predominantly due to Na⁺ influx, it is possible that a subsarcolemmal increase in Na⁺ could cause an increase in intracellular [Ca²⁺] and therefore contraction, by activating reverse mode Na⁺/Ca²⁺ exchange, as suggested by Lee et al. (2001) for rabbit inferior vena cava. However, the inhibitor of reverse mode Na⁺/Ca²⁺ exchange, KB-R7943 (10 μM), did not significantly reduce the Thg-induced contraction in IPA (contraction in the presence of KB-R7943=85±5% of control response, n=5, P>0.05). Moreover, the rise in [Ca²⁺]_i induced by reintroduction of extracellular Ca²⁺ was unchanged in low Na⁺ PSS (data not shown; n=4). However, KB-R7943 was found to inhibit the KPSS-induced contraction (contraction in the presence of KB-R7943=31.7±4.6% of control response, n=5, P<0.05), which is mediated by voltage-gated Ca²⁺ channels. This suggests that this compound is not selective for the reverse mode Na⁺/Ca²⁺ exchanger in these arteries.

Basal Mn²⁺ permeability of SMC from IPA

In SMC freshly isolated from IPA and loaded with Fura-PE3, application of extracellular Mn²⁺ (1 mM) caused a relatively fast quenching of fluorescence under basal conditions. Application of 1 μM La³⁺ did not affect the rate of quenching (106±8% of control, n=6); however, 10 μM La³⁺ caused a significant reduction in the rate of quenching (25±4%, n=6), and 10 μM Gd³⁺ suppressed it entirely (2±1%, n=6). Diltiazem (10 μM) had no effect (98±8%, n=4), while 2-APB (75 μM) markedly blocked basal quenching (25±4%, n=5) (Figure 9a). These results indicate that IPA SMC possess a constitutively active Ca²⁺ influx pathway permeable to Mn²⁺. Pretreatment of cells with Thg (100 nM) increased the rate of quenching to 625±96% (n=8, P<0.001) of control, and this increase was abolished by 75 μM 2-APB (n=5) and by 1 μM La³⁺ (n=4) (Figure 9b).

Figure 9.

SMC isolated from IPA possess a Mn²⁺-permeable, La³⁺-, Gd³⁺-, and 2-APB-sensitive basal and Thg-induced influx pathways. (a) Relative rates of basal Fura-PE3 fluorescence quenching with 1 mM Mn²⁺ in unstimulated SMC freshly isolated from IPA. La³⁺ (10 μM, but not 1 μM) caused a significant reduction in the rate of quenching and 10 μM Gd³⁺ suppressed it entirely. Diltiazem (10 μM) had no effect, while 2-APB (75 μM) attenuated the basal quenching. (b) Thg (100 nM) increased the rate of quenching (measured with 0.1 mM Mn²⁺) more than six-fold, and this increase was abolished by 75 μM 2-APB or 1 μM La³⁺. n=10–16, ***P<0.001.

Discussion

CCE subsequent to inhibition of the SERCA has been proposed as a ubiquitous mechanism in smooth muscle controlling not only the Ca²⁺ load of stores within the ER/SR, but also modulating tone (for review see McFadzean & Gibson, 2002). We have recently reported that the transient portion of the vasoconstrictor response to hypoxia in rat-isolated IPA may be mediated in part via an activation of CCE (Robertson et al., 2000). In the present study, we applied Thg in the absence of Ca²⁺ to deplete Ca²⁺ stores, and then restored Ca²⁺ to the solution to examine CCE. Among the five types of small arteries which we studied, Ca²⁺ restoration caused a substantial sustained contraction in the IPA, sometimes caused a small transient contraction in the FA, and was not associated with contraction in the MA, RA, or CA. Ca²⁺ restoration was, however, associated with substantial increases in [Ca²⁺]_i in MA, FA, and IPA, each of the three arteries representing a different pattern of tension development.

CCE and arterial contraction

The majority of investigations into the possible role of CCE in regulating vascular tone have been performed in large conduit arteries. Induction of CCE, via inhibitors of the SERCA or Ca²⁺-releasing agonists, has consistently been shown to induce a prompt and marked contraction in the rat main pulmonary artery (de la Fuente et al., 1995; McDaniel et al., 2001; Ng & Gurney, 2001), which is associated with a rise in [Ca²⁺]_i (McDaniel et al., 2001; Ng & Gurney, 2001). However, agents that induce CCE tend to cause inconsistent, small, and/or more slowly developing contractions in systemic conduit arteries, including rat aorta (Noguera et al., 1998; Shima & Blaustein, 1992), tail artery (Kwan et al., 1994), carotid artery (Sekiguchi et al., 1996), and FA (Nomura et al., 1997), and mouse aorta (Cohen et al., 1999), in spite of the fact that, when measured, [Ca²⁺]_i has been shown to rise substantially in these arteries (Sekiguchi et al., 1996; Tosun et al., 1998; Cohen et al., 1999). The few analogous studies in smaller systemic arteries suggest that the rise in [Ca²⁺]_i is completely dissociated from contraction. Naganobu & Ito (1994) observed that cyclopiazonic acid caused a sustained rise in [Ca²⁺]_i with no contraction in rat mesenteric resistance arteries, and Flemming et al. (2002) reported a similar effect in rabbit pial arterioles. The present study confirms and extends these findings, showing firstly that CCE is associated with a marked and sustained contraction in smaller pulmonary arteries from rats, and secondly that CCE causes no or a minimal transient contraction in rat small MA, FA, RA, and CA.

The observations of Naganobu & Ito (1994) and Flemming et al. (2002), as well as the results shown in Figure 4 and Figure 5, indicate, however, that Thg-induced store depletion causes a marked and sustained increase in [Ca²⁺]_i in small systemic as well as pulmonary arteries. The basis of the difference between IPA and the other arteries remains obscure. The Thg-induced current appeared to be mainly carried by Na⁺ (see below), a finding that is consistent with the report of Arnon et al. (2000) that store release greatly elevated [Na⁺]_i in rat cultured mesenteric arterial myocytes. It therefore seemed possible that Na⁺ influx through store-operated channels could cause an increase in intracellular [Ca²⁺] and contraction by activating reverse mode Na⁺/Ca²⁺ exchange (Lee et al., 2001) in IPA but not the other arteries, especially because our preliminary observations suggest that Na⁺/Ca²⁺ exchange is prominent in the IPA, but not in the MA (Moir & Aaronson, unpublished observations). However, we found no evidence for this mechanism, as KB-R7943, an inhibitor of reverse mode Na⁺/Ca²⁺ exchange, did not significantly reduce the Thg-induced contraction, and removal of extracellular Na⁺ did not affect the CCE-associated increase in [Ca²⁺]_i. It should be noted that KB-R7943 inhibited KPSS-induced contractions in IPA, which we have previously found are completely abolished by voltage-gated Ca²⁺ channel blockers (Robertson et al., 2000). This suggests that KB-R7943 is not selective for reverse-mode Na⁺/Ca²⁺ exchange in these arteries, and is consistent with its ability to block a variety of voltage-gated ion channels (Tanaka et al., 2002).

Although IPA responded to CCE with a contraction, it is apparent from Figure 4 that even in these arteries, the CCE-mediated contraction was much smaller than that evoked by high K⁺, even though both stimuli were associated with comparable increases in [Ca²⁺]_i. This disproportion between increases in [Ca²⁺]_i and tension, an extreme form of which apparently occurs in small systemic arteries and arterioles (Figure 4, see also Naganobu & Ito, 1994; Flemming et al., 2002), was initially highlighted in the rat aorta by Tosun et al. (1998), who proposed that Ca²⁺ entering during CCE is confined to a compartment that has limited access to the contractile apparatus. The case of the FA is particularly interesting, since here CCE caused an increase in [Ca²⁺]_i that was much larger than that caused by high K⁺ depolarization, even though it caused a contraction that was negligible compared to the high K⁺ response. It seems that even in the absence of Ca²⁺ uptake into the sarcoplasmic reticulum, SMC retain the ability to selectively direct Ca²⁺ entering the cell via different pathways either towards or away from the contractile apparatus.

Properties of CCE in rat IPAs

Treatment of isolated IPA SMC with Thg caused the development of a noisy inward current in cells held at –80 mV. The current amplitude was ∼30 pA in cells held at, or ramped to, this potential. At negative potentials, the current was not affected when Ca²⁺ was removed in the presence of a normal [Na⁺], but was greatly reduced when Na⁺ was replaced by NMDG⁺, suggesting that it was mainly carried by Na⁺ ions. A current of reduced amplitude was, however, observed when Na⁺ was replaced by 80 mM Ca²⁺, indicating that the channel is also Ca²⁺-permeable. The current was also supported by other monovalent cations, and reversed near 0 mV, suggesting that it is mediated by a nonselective cation channel.

During the onset of the current, and following its block by 2-APB, it was possible to observe individual channel openings of ∼3 pA at –80 mV, indicating a channel conductance of ∼30 pS. Although we did not further characterize the properties of these channels, they appear to differ from the store-operated channels in rabbit portal vein, which were recently described in greater detail by Albert & Large (2002), as the latter demonstrated a conductance of ∼2 pS and reversed at a more positive potential under similar conditions.

The current observed in IPA appears to be similar, however, to that previously reported in rat main pulmonary artery by Ng & Gurney (2001), with the difference that the current in IPA is much more potently inhibited by La³⁺. It seems likely, however, that the relative La³⁺ insensitivity (IC₅₀∼600 μM) reported by Ng & Gurney (2001) can be explained by their use of a phosphate-containing buffer system. Since lanthanum phosphate is known to be insoluble (Moeller, 1961), the use of phosphate-containing buffers in experiments where La³⁺ is used would result in the formation of lanthanum phosphate, thereby greatly decreasing the concentration of free La³⁺ in solution.

It has been reported that La³⁺ can directly prevent the inhibition of SERCA by Thg (Papp et al., 1991). It is therefore possible that the inhibitory effects of La³⁺ upon Thg-induced contraction, and channel activation, in pulmonary arteries may be due to stabilization of SERCA by La³⁺. However, our results suggest that the blockade of the CCE-associated current by La³⁺ is extracellular since the pipette solution contained EGTA, which would strongly chelate any La³⁺ entering the cells. The inclusion of EGTA in the pipette solution would also appear to preclude the possibility that La³⁺ may block CCE in pulmonary arteries via an action on the cytosolic side of the plasma membrane, as reported for the blockade of TRP3 channels by La³⁺ in Chinese hamster ovary cells (Halaszovich et al., 2000).

In marked contrast to our results, McDaniel et al. (2001) reported a store release-activated current of over 300 pA in pulmonary artery SMC. The most likely explanation for this discrepancy is that McDaniel et al. (2001) utilized cells that had been maintained in primary culture, a procedure that is known to enhance CCE greatly (Golovina, 1999; Dreja et al., 2001).

It is interesting to note that infrequent openings of channels apparently similar to those underlying the Thg-induced current could be recorded prior to Thg exposure (see Figure 6a). It is possible that cell dialysis with high concentrations of Ca²⁺-chelating agents, such as EGTA or BAPTA, could itself cause depletion of intracellular stores and subsequent activation of CCE (Albert & Large, 2002). However, in nondialysed cells a Mn²⁺-permeable influx pathway was observed under basal conditions (Figure 9a), which was blocked by 10 μM La³⁺ but not by 1 μM La³⁺. Thg preincubation elicited an approximately six-fold increase in Mn²⁺ quenching of the fluorescence signal, which was abolished by 1 μM La³⁺ (Figure 9b). There was, therefore, a difference in the La³⁺ sensitivity of the basal and Thg-associated influx pathways. This suggests the involvement of separate channels, although this requires additional confirmation.

2-APB was effective in abolishing the Thg-induced increases in both inward currents, and constrictor response in rat-isolated small IPA. This agent, which blocks the IP₃ receptor (Maruyama et al., 1997), was initially used to assess the possible role that this receptor could play in the regulation of CCE (e.g. Ma et al., 2000). Subsequent observations suggested, however, that 2-APB possesses effects upon CCE that are independent of any action upon the IP₃ receptor (Dobrydneva & Blackmore, 2001; Iwasaki et al., 2001; Prakriya & Lewis, 2001; Ma et al., 2002). The present results are consistent with these more recent findings. Irrespective of the mechanism by which 2-APB is acting to inhibit CCE, its effect on the Thg-induced current and contraction in IPA provides additional evidence linking these effects, although the minor inhibition of the contraction to KPSS by 2-APB illustrates its lack of complete selectivity for CCE.

In summary, CCE-induced vasoconstriction appears to be especially prominent in pulmonary arteries, at least at the level of small muscular arteries. The results suggest that the extent to which CCE contributes to agonist-induced vasoconstriction may vary widely between different vascular beds, and underline the importance of this pathway in the pulmonary vasculature. The CCE-associated current in rat IPA involves a nonselective cation channel with a conductance of ∼30 pS, and is sensitive to relatively low concentrations (1 μM) of the trivalent cations La³⁺ and Gd³⁺. Further studies are required to determine the basis of the heterogeneity of the contractile response of various arteries to Ca²⁺ store release.

Abbreviations

2-APB

2-aminoethoxydiphenyl borate

CA

coronary artery

CCE

capacitative Ca²⁺ entry

FA

femoral artery

IPA

intrapulmonary artery

MA

mesenteric artery

PSS

physiological salt solution

RA

renal artery

SERCA

sarcoplasmic/endoplasmic reticulum Ca²⁺ ATPase

SMC

smooth muscle cell

SR/ER

sarco/endoplasmic reticulum

Thg

thapsigargin

TRP

transient receptor potential protein