Transient Ca²⁺-Activated Cl⁻ Currents with Endothelin in Isolated Arteriolar Smooth Muscle Cells of the Choroid

Abstract

purpose. To characterize the effects of endothelin (ET)-1 on the Ca²⁺-activated Cl⁻ conductance of choroidal arteriolar smooth muscle.

methods. Microvascular smooth muscle cells were enzymatically isolated from choroidal arterioles from the eyes of freshly killed rabbits. Cells were voltage-clamped at −60 mV using the whole-cell perforated patch-clamp technique. Internal pipette solutions were K⁺ based and contained amphotericin B (200 μg/ml). The cells were bathed in a 20 mM tetraethyl–ammonium solution to block outward K⁺ currents.

results. Within 2 to 5 seconds of adding ET-1 (10 nM), inward current pulses were generated at a frequency of around 1 Hz. These evoked transient inward currents were blocked by niflumic acid (10 μM) or anthracene-9-carboxylic acid (1 mM). They were increased 2.4 ± 0.1-fold when Cl⁻ was replaced by I⁻ in the bathing medium and lost within 4 minutes when external Cl⁻ was reduced from 151.6 to 20 mM. The reversal potential was −1 ± 2 mV with 135 mM Cl⁻ in the recording pipette and with 54 mM Cl it was −18 ± 4 mV. When gramicidin D (100 μg/ml), which maintains [Cl⁻]_i, was used instead of amphotericin B, the reversal potential was −18 ± 1 mV. Ca²⁺ release by caffeine (10 mM) produced a single transient inward current. Endothelin-evoked transient inward currents were slowly reduced and eventually abolished in Ca²⁺-free solution (∼2 to 3 minutes) and were eliminated after ∼30 seconds by the sarcoplasmic reticulum Ca²⁺-uptake inhibitor cyclopiazonic acid (5 μM). The ET_A receptor antagonist BQ123 (1 μM) prevented an effect by endothelin but did not inhibit the current oscillations once they had been triggered.

conclusions. In choroidal arteriolar smooth muscle ET-1 evokes transient inward Ca²⁺-activated Cl⁻ currents induced through the cyclical release and re-uptake of Ca²⁺ from intracellular stores after ET_A receptor stimulation.

The choroid of the eye lies between the sclera and retina and is comprised mainly of arterioles. It is important for the supply of nutrients to the retina in both lower mammals (e.g., guinea pig and rabbit), in which the nutrients used by the retina are almost completely derived from the choroid, and in many higher mammals, including humans, in which the retina is supplied by both choroidal and retinal vessels. ¹ The choroidal circulation has a high blood flow, for example, the blood flow of monkey choroid is ∼2000 ml/100 grams per minute. ¹ This high blood flow through the choroid, as well as supplying nutrients, provides the eye with protection from thermal damage even under extreme conditions. ¹  

Although a large amount of data has been accrued on the cellular physiology of large vessel vascular smooth muscle, little is known about arteriolar smooth muscle cells, particularly those in the eye, and most of this work is derived from studies on intact vascular beds. Notwithstanding this, the innervation pattern of choroidal arterioles and their responses to stimulation have been examined. The guinea pig choroid is innervated by at least 3 different populations of nerves, adrenergic nerves that evoke excitatory responses, cholinergic nerves that evoke inhibitory responses, and a population of nerves that cause the release of nitric oxide (NO). ² Although the acetylcholine and NO–induced hyperpolarizing responses are thought to result from the opening of K⁺ channels, the ionic channels underlying the adrenergic depolarizing response have yet to be identified. In iris arterioles, activation of α-adrenoceptors leads to the release of intracellular calcium that activates Ca²⁺-activated Cl⁻ channels in the cell membrane leading to depolarization. ³  

Ca²⁺-activated Cl⁻ conductances, I_Cl(Ca), have been identified in several types of smooth muscle including arteries ^{4 5 6} and have been shown to be elicited by a wide range of agonists that raise[ Ca²⁺]_i. Because the Cl⁻ equilibrium potential is usually less negative than the resting membrane potential, these conductances cause membrane depolarization. Thus, these agonists produce their depolarizing effects, at least in part, through I_Cl(Ca), which causes further activation through the opening of L-type Ca²⁺ channels. I_Cl(Ca) can be either a steady or slowly inactivating current, or it may be present as a periodic spontaneous transient inward current. 

Endothelin (ET)-1 is a peptide, released from endothelial cells, which causes profound vasoconstriction. Secretion of ET-1 from endothelial cells is stimulated by a wide range of substances and is inhibited by some prostaglandins. ⁷ Choroidal blood vessels possess ET_A receptor binding sites, ⁸ and ET-1 is believed to play an important role in choroidal autoregulation by competing with locally produced NO and as yet an unknown neural dilator. ⁹  

Despite the importance of ET-1 in the regulation of the choroidal circulation the electrophysiological effects of ET receptor stimulation in choroidal arterioles has not been determined. In the present study, we show that in single choroidal arteriolar smooth muscle cells, ET-1 evokes transient inward currents that have properties accordant with their being mediated through I_Cl(Ca). 

Methods

Cell Isolation

The investigation conforms with the Guide for the Care and Use of Laboratory Animals published by the US National Institutes of Health (NIH Publication No. 85-23, revised 1996). New Zealand white rabbits (2–4 kg) of either sex were killed by 80 mg/kg sodium pentobarbital. The eyes were removed and placed in a Ca²⁺-free solution at 4°C. The choroids were removed as sheets and cut into pieces (2–4 mm²). After 30 minutes in Ca²⁺-free solution at 15°C, the tissue was incubated for 10 minutes at 34°C in a solution containing 20 μM Ca²⁺ and 0.05 mg/ml trypsin (Sigma, Poole, UK). It was then resuspended in a solution containing 20μ M Ca²⁺ and collagenase type 1A (0.4 mg/ml; Sigma) and incubated at 34°C for 25 to 30 minutes with continuous agitation by a magnetic stirring bar. Single microvascular smooth muscle cells were mechanically dispersed by trituration of the tissue pieces using a fire-polished Pasteur pipette. When a sufficiently high number of spindle-shaped smooth muscle cells became apparent under microscopic examination, the enzymatic incubation was stopped. This was done by centrifuging the homogenate at 1000 rpm for 1 minute and discarding the supernatant. Cells were then resuspended in 50 μM Ca²⁺ solution and stored for 30 minutes at 4°C. Finally, the isolated cells were centrifuged again and maintained in 100 μM Ca²⁺ solution at 4°C. They remained viable for up to 8 hours. 

Electrophysiology

Dispersed cells were allowed to settle on the bottom of a 2 ml recording bath on the stage of an inverted microscope for 10 minutes. Bathing solution was then allowed to flow into one end of the bath by gravity feed and withdrawn from the other end at 2 ml/min by a vacuum system. The solution passed through a heat exchanger such that the temperature in the recording bath was 37°C at the in-flow end and 35.5°C at the other end. Solutions including those containing drugs were delivered through a tube (350 μm in diameter by 6 mm in length) long enough to allow the temperature to equilibrate with the solution flowing through the bath. The delivery tube was positioned approximately 500 μm away from the cell under study and was fed by a seven-way manifold leading from seven separate reservoirs each controlled by valve. The delay time for new solution to reach a cell was 1 second. 

Membrane current was recorded from myocytes using the perforated patch configuration ¹⁰ of the whole-cell patch-clamp technique. ¹¹ To record membrane current, cells were voltage-clamped at −60 mV unless otherwise stated. Electrodes (3–8 MΩ) were pulled from filamented borosilicate glass capillaries (1.5 mm OD × 1.17 mm ID; Clark Electromedical Instruments, Reading, UK) with a Flaming–Brown micropipette puller (model P-87; Sutter Instruments, Novato, CA). Recordings were made using an Axopatch-1D patch-clamp amplifier (Axon Instruments, Foster City, CA). Liquid junction potentials (<2 mV) were compensated electronically. Series resistance and cell capacitance were usually uncompensated. Voltage errors (arising from series resistance) of less than 2 mV occurred at peak current levels. Leakage currents were not subtracted. Current levels were reset to zero at the beginning of each experiment. Data were low pass–filtered at 0.5 kHz and sampled digitally at 2 kHz by a National Instruments PC1200 interface and stored on diskette for off-line analysis using software (provided by Dempster J, University of Strathclyde, Strathclyde, Scotland). Numerical data are expressed as mean ± SEM. 

Solutions

The bathing solution contained the following (in millimoles): 120, NaCl; 5, KCl; 5, d-glucose; 2, CaCl₂; 1.3, MgCl₂; 10, Hepes; 20, tetraethyl–ammonium chloride (TEA) pH adjusted to 7.3 with NaOH. In Ca²⁺-free solution, the CaCl₂ was omitted and low Ca²⁺ solutions were made by adding the appropriate amount of CaCl₂. For the iodide solution, NaI (120 mM) replaced NaCl in the bathing solution. Low chloride solution was made using glucuronic acid and the cations added as hydroxides (remaining Cl⁻ came from tetraethyl–ammonium chloride; pH 7.3 with NaOH). For perforated-patch recordings the pipette contained (in millimoles): 52, KCl; 80, Kgluconate; 1, MgCl₂; 0.5, EGTA; 10, Hepes (adjusted to pH 7.2 using NaOH), to which 200 μg/ml amphotericin B was added. Voltage-clamp experiments were done in the presence of 20 mM TEA to block outward K⁺ currents. In experiments designed to alter the Cl⁻ equilibrium potential (E_Cl) the pipette contained the following (in millimoles): 133, KCl; 1, MgCl₂; 0.5, EGTA; 10, Hepes (adjusted to pH 7.2 using NaOH). Although amphotericin pores are cation selective, permeability to Cl⁻ is not negligible. ¹² Thus, the reversal potential for Cl⁻ current was determined by the bath and pipette Cl⁻ concentrations. In some experiments, in place of amphotericin B, 100 μg/ml gramicidin D, which has negligible Cl⁻ penetration, ¹³ was used to prevent [Cl]_i from changing. 

Amphotericin B, anthracene-9-carboxylic acid (9-AC), caffeine, EGTA, gramicidin D, niflumic acid, and TEA were purchased from Sigma. Cyclopiazonic acid (CPA) was from Alexis Biochemicals (San Diego, CA). Endothelin-1 (human, porcine) and BQ123 were obtained from American Peptide (Sunnyvale, CA) or Tocris (Bristol, UK). 

Results

Quiescent Properties of the Cells

Each rabbit choroid yielded around 20 cells that were uncontracted and had the semilunar shape of cells in intact vessels. They measured 3 to 6 μm across at their widest point and 15 to 40 μm in length. 

In current-clamp mode the patch electrodes contained gramicidin D, and with normal solution the resting membrane potential was measured as− 34 ± 2 mV (n = 6). The cells failed to produce any regenerative responses on depolarization. In voltage-clamp mode, the input resistance was estimated to be 1.3 ± 0.3 GΩ between− 60 and −80 mV (n = 10). The cell capacitances were measured as 5.4 ± 1.1 pF from the capacitance transients, with 20 mV depolarizing and hyperpolarizing excursions around −80 mV (n = 6). In some cells, there were spontaneous outward currents that were blocked by 20 mM TEA or 1 μM penitrem A (data not shown). These were considered to result from the activation of BK_Ca channels. ¹⁴ In all further experiments, 20 mM TEA was in the medium to prevent them interfering with I_Cl(Ca). 

The Effects of ET

In Figure 1A , a cell was maintained in voltage-clamp mode. Transient inward currents occurred within 3 seconds of adding 10 nM ET-1, after which they continued to be generated at a frequency of around 1 Hz. The amplitude of the evoked transient inward currents showed a twofold variation, and the average amplitude was maintained over a period of up to 1 hour whether or not ET-1 was still present (i.e., the effect of ET-1 did not wash out). In current-clamp mode these inward currents were manifest as transient depolarizations. 

In the presence of the selective ET_A receptor antagonist BQ123 (1 μM), ET-1 failed to generate any transient inward currents (Fig. 1B) . When the BQ123 washed out into bathing solution still containing ET-1, current pulses were generated after a delay of 22 ± 7 seconds (n = 9). BQ123 did not inhibit the current oscillations once they had been triggered by the application of ET-1 (Fig. 1B ; n = 4). 

Ionic Dependency

To determine the ionic specificity of the transient inward currents their reversal potentials were measured (see Fig. 2A ). The perforations formed by amphotericin B are permeable enough to Cl⁻ to allow the electrolyte in the patch pipette to influence intracellular Cl⁻. In 6 cells, the patch pipette contained 54 mM Cl⁻, and the current voltage relationship of the transients reversed at− 18 ± 4 mV (Fig. 2B ; theoretical E_Cl =− 26 mV). In another 6 cells, the patch electrode contained 135 mM Cl⁻, and the reversal potential was −1 ± 2 mV (theoretical E_Cl = −3 mV). Thus, the reversal potential appeared to depend on the[ Cl⁻]_i. 

The next series of experiments used a voltage ramp protocol between− 40 and +40 mV to measure reversal potentials. Figure 3A shows an example of 6 experiments with gramicidin D in the patch pipette, where the evoked transient inward currents reversed to outward currents at −18 ± 1 mV. In Figure 3B , the Cl⁻ in the bathing solution (151.6 mM) was reduced to 20 mM, and initially the reversal potential shifted to around +40 mV. On continued perfusion with the 20 mM Cl⁻ solution for 4 minutes, all evoked currents were lost (Fig. 3C) . On restoring the normal Cl⁻ (151.6 mM), all the evoked currents were outward (Fig. 3D) , but after a further 3 minutes, both inward and outward currents were obtained (Fig. 3E) , which reversed at a similar voltage to that seen before the start of the experiment (−13 mV). 

Cl⁻ channels in vascular smooth muscle are around 3 times more permeant to I⁻ ions than Cl⁻. ⁶ Therefore, cells were held at+ 20 mV so that the evoked currents were outward and hence resulted from the influx of Cl⁻ ions. When Cl⁻ in the bathing solution was replaced by I⁻ the transients increased in amplitude by 2.4 ± 0.1-fold (Fig. 4 ; n = 4). 

In many cells, the outward current transients were larger in highly depolarized cells than inward currents at negative membrane potentials even when the inner and outer Cl⁻ concentrations were similar (Fig. 2A) . Reducing the internal Cl⁻ did not affect this outward rectification. Other cells showed no such rectification (Fig. 3A) . 

Effects of Cl⁻ Channel Blockers

The effects of the chloride channel blockers niflumic acid and 9-AC were also examined. ^{15 16 17} Application of niflumic acid (10 μM) immediately blocked the evoked transient inward currents, which partially recovered (amplitudes were reduced) on washing out (Fig. 1A ; n = 6). 9-AC (1 mM) also rapidly blocked the ET-1–induced currents, but they recovered fully within 4 seconds on washing out (data not shown; n = 5). Neither niflumic acid nor 9-AC had any effect on L-type Ca²⁺ currents in these cells (data not shown; n = 4 and 5, respectively). Because the reduction of Ca²⁺ influx produced a slow inhibition of the evoked transient inward currents (see Ca²⁺ Dependency below), the rapid effects of niflumic acid and 9-AC described above are likely to be due to Cl⁻ channel block. 

Ca²⁺ Dependency

In the absence of ET-1, the application of 10 mM caffeine, which releases Ca²⁺ from intracellular stores, produced a single transient having a time course similar to the ET-1–induced transient inward currents (Fig. 5A ). In Ca²⁺-free solution, ET-1 current pulses were reduced in frequency and amplitude and were eventually abolished after around 2 to 3 minutes (Fig. 5B ; n = 5). CPA is regarded as a specific inhibitor of Ca²⁺ uptake into the sarcoplasmic reticulum and, thus, prevents the refilling of Ca²⁺ stores. Five micromoles of CPA eliminated the ET-induced oscillations in current after ∼30 seconds (Fig. 5C ; n = 4). The effect of CPA was irreversible (i.e., the ET-evoked currents did not recover on washing out CPA). 

Discussion

These experiments clearly show that ET-1 induced transient inward currents in single microvascular smooth muscle cells of the rabbit choroid. Our evidence suggests that the currents were mediated by Ca²⁺-activated Cl⁻ channels because their reversal potentials were influenced by[ Cl⁻]_i via the recording electrode; they were lost in low Cl⁻ bathing medium; they were enhanced when I⁻ was substituted for Cl⁻; they were rapidly blocked by the Cl⁻ channel blockers niflumic acid and 9-AC; they were activated by doses of caffeine known to dump intracellular Ca²⁺ stores; and they were abolished in Ca²⁺-free bathing medium. They were prevented by the selective ET_A receptor antagonist BQ123, suggesting that they are a consequence of ET_A receptor stimulation. Once ET-1 had elicited these repetitive Ca²⁺-activated Cl⁻ currents they could not be reversed by washing out the ET-1 or applying BQ123. This suggests that whatever pathway exists between the ET_A receptor and the activation of the Ca²⁺-dependent Cl⁻ channels, it is persistently activated. 

ET-1 has been reported to induce I_Cl(Ca) in a variety of smooth muscle cell types including aortic, coronary, mesenteric, and pulmonary myocytes. ^{18 19 20 21} ET-1 increases[ Ca²⁺]_i ²² and, thus, activates I_Cl(Ca) in choroidal microvascular smooth muscle cells through mechanisms similar to those previously described in large-vessel vascular smooth muscle. Binding of ET-1 to ET_A receptors results in the stimulation of intracellular signaling pathways and promotes release of Ca²⁺ from intracellular stores. Subsequently, sustained Ca²⁺ entry across the plasma membrane occurs through multiple types of Ca²⁺ channels. ²³ Some of this Ca²⁺ entry is involved in the refilling of Ca²⁺ stores. It has recently been established that it is the release of Ca²⁺ from intracellular stores, and not Ca²⁺ influx from the cell exterior, that causes spontaneous transient inward currents in smooth muscle cells. ²⁴ In the present study, the rhythmical ET-1–induced Cl⁻ currents were abolished by inhibition of the sarcoplasmic reticulum Ca²⁺-ATPase with CPA. Thus, it is likely that these currents represent a recycling between emptying and refilling of Ca²⁺ stores located very close to the plasma membrane. 

Although the cells are very small and originate from vessels far smaller than those hitherto studied, these ET-1–evoked transient inward currents do show properties that are similar to those seen in large-vessel vascular smooth muscle. ²⁵ Notwithstanding this, it was surprising that the peak amplitudes of the currents were comparable to those seen in large arterial myocytes ²⁰ that have membrane surface areas 5 to 6 times greater. These results suggest that choroidal arteriolar smooth muscle cells may possess a much higher density of Ca²⁺-activated Cl⁻ channels. 

In the present study, using the perforated patch-clamp technique, choroidal arteriolar smooth muscle cells had resting membrane potentials of around −34 mV. These results agree with earlier intracellular recordings of membrane potentials in choroidal arterioles of the guinea pig (∼−38 mV). ² A membrane potential of− 34 mV is less negative than most other quiescent smooth muscle cells (−50 to −75mV). ^{6 26} The only other arterioles with such low membrane potentials are those found in the distal regions of the cerebral circulation. ²⁷ In both choroidal and distal cerebral arterioles these low membrane potentials appear to result from a closure of inward rectifier K⁺ channels in normal[ K⁺]_o. ^{2 27}  

Some of the experiments described here were done using gramicidin-filled electrodes. Contrary to the ionophores (i.e., amphotericin B or nystatin) commonly used to perforate cell membranes, ²⁸ gramicidin creates pores that are impermeable to anions. ²⁹ In this respect, the gramicidin perforated patch technique provides a useful tool for recording ionic currents while maintaining[ Cl⁻]_i. By applying this method it is possible to estimate physiological[ Cl⁻]_i by measuring the reversal potential for Cl⁻ current. In the present study, [Cl⁻]_i was around 46 mM, a value that is similar to many other types of smooth muscle. ^{30 31 32} Using the gramicidin-filled electrodes the reversal potential for the ET-1–induced Cl⁻ currents was −18 ± 1 mV, which is positive to the resting membrane potential. Consequently, the opening of Cl⁻ channels will produce an efflux of Cl⁻ driving the membrane toward E_Cl and will hence produce depolarization. In common with choroidal arteriolar smooth muscle cells, under physiological conditions, I_Cl(Ca) produces membrane depolarization in all smooth muscle cells studied so far. It is worth noting, however, that because the membrane potential of these choroidal cells is low (−34 mV) and thus further from the potassium equilibrium potential (E_K) than E_Cl, outward K⁺ current is likely to have the greatest effect on membrane potential when a rise in[ Ca²⁺]_i occurs (the existence of Ca²⁺-activated K⁺ channels in choroidal arterioles is now well established). ^{2 22} This contrasts with the situation in other types of smooth muscle that have more negative membrane potentials (−50 to −75 mV) where the Cl⁻ current will be the dominant current elicited because the membrane potential is closer to E_K than E_Cl. 

The physiological role of I_Cl(Ca) in smooth muscle during stimulation by ET-1 is not clear but is thought to constitute an intermediate step within a cascade of reactions finally leading to the activation of voltage-dependent Ca²⁺ channels, resulting in an increased Ca²⁺ influx and subsequent smooth muscle contraction. ^{18 19 20} Therefore, ET-1 can produce constriction directly as a consequence of store-released Ca²⁺ acting directly on the contractile proteins or indirectly by stimulating I_Cl(Ca), causing depolarization and the consequent opening of voltage-dependent Ca²⁺ channels. Although we have recently demonstrated the presence of voltage-dependent L-type Ca²⁺ channels in choroidal arteriolar smooth muscle cells, ³³ we can exclude these channels as being relevant in ET-1–induced vasoconstriction for two reasons. First, because the membrane potential is so low L-type Ca²⁺ channels are probably already inactivated; moreover, under current-clamp ET-1 causes a slow potential shift from∼− 34 mV to ∼−25 mV, which may produce further inactivation of these channels. ²² In the absence of TEA, ET-1 does elicit transient hyperpolarizations (due to the activation of BK_Ca channels), but in contrast to the transient depolarizations these are short-lived and run down after only 2 to 3 minutes. ²² Thus, during prolonged exposure to ET-1, the BK_Ca channels do not affect the membrane potential and, hence, will not exert an influence on the L-type Ca²⁺ channels. Second, in any event, ET-1 causes an inhibition of almost all the L-type Ca²⁺ current in these cells. ³³  

Because the L-type channels appear to be the only voltage-dependent Ca²⁺ channels that the cells possess, the physiological function of the ET-1–induced Cl⁻ currents is unclear. One possibility is that they are involved in the modulation of ET-1–induced Ca²⁺ influx through store-depletion–dependent Ca²⁺ channels ³⁴ by controlling the membrane potential. Although store-operated Ca²⁺ currents are not gated by membrane voltage changes, once activated their current-voltage relationships show prominent inward rectification at negative voltages (i.e., currents are relatively larger at hyperpolarized potentials). ³⁵ Thus, as the cell becomes Ca²⁺ loaded, activation of I_Cl(Ca) causes depolarization, which will then limit Ca²⁺ influx through these store-refilling channels. In this respect, activation of I_Cl(Ca) by ET-1 in choroidal arterioles may represent a mechanism that serves to protect the smooth muscle cells from Ca²⁺ overloading. 

To summarize, ET_A receptor stimulation in choroidal microvascular smooth muscle cells results, via a Ca²⁺-dependent mechanism, in the activation of transient Ca²⁺-activated Cl⁻ currents. These are manifest as depolarizing oscillations in membrane potential. The precise physiological role of these currents with ET-1 is uncertain and warrants further investigation. 

 Supported, in part, by a grant from the Wellcome Trust.

 Submitted for publication December 6, 1999; revised February 14, 2000; accepted February 18, 2000.

 Commercial relationships policy: N.

 

Figure 1.

 

(A) A voltage clamp record for a choroidal cell being held at a potential of −60 mV showing the effects of 10 nM ET and 10 μM niflumic acid. The lower panel shows some of the record on an expanded time scale. (B) Current record for a cell held at −60mV showing the effect of the ET_A receptor antagonist BQ123. BQ123 (1 μM) prevented an effect by 10 nM ET but did not inhibit the current oscillations once they had been triggered.

Figure 2.

 

(A) This record shows how the reversal potential was obtained from a voltage-clamped choroidal cell using a patch electrode containing amphotericin B and 135 mM Cl⁻. The cell was continuously bathed in 10 nM ET. From an initial holding potential of 0 mV, the cell was hyperpolarized to −100 mV, and then a slow voltage ramp was applied until a holding potential of +100 mV was obtained. (B) Current voltage relationships for the ET-induced transients measured using the steady state current at each holding potential as the baseline. The data were obtained from voltage ramps between −60 and +40 mV. ET (10 nM) was present throughout. The patch electrodes contained amphotericin B and either 54 (triangles) or 135 (squares) mM Cl⁻.

Figure 3.

 

A series of voltage ramps was applied to a cell in normal solution (A), 20 mM Cl⁻ solution (B, C), and then returned to normal solution (D, E). The cell was voltage-clamped at 0 mV, and then the voltage was ramped between −40 and +40 mV. In (B), the cell had been bathed in 20 mM Cl⁻ solution for 1 minute and for a further 4 minutes in (C). In (D), Cl⁻ was restored for 1 minute and for 4 minutes in (E). Ten nanomoles of ET was present throughout.

Figure 4.

 

The effect of Cl⁻ replacement with I⁻. Throughout the experiment the cell was held at +20 mV and bathed in 10 nM ET.

Figure 5.

 

(A) Current record showing the effect of 10 mM caffeine on a choroidal arteriolar smooth muscle cell being held at −60 mV. (B) Voltage-clamp record showing the effect of Ca²⁺-free solution on the ET-evoked transient inward currents. The cell was held at −60 mV and bathed continuously in 10 nM ET. (C) Current record for a cell held at −60 mV showing the effect of the sarcoplasmic reticulum Ca²⁺-pump inhibitor CPA (5 μM) on the ET-induced current oscillations.