INTRODUCTION

Top Abstract Introduction Methods Results Discussion References

IN SMOOTH MUSCLE an increase in membrane Cl conductance will produce depolarization of the cell and contraction because the Cl equilibrium potential (about 20 to 30 mV; Ref. 1) is more positive than both the resting membrane potential and the threshold potential for opening voltage-dependent Ca²⁺ channels (VDCCs). In many types of smooth muscle, including vascular preparations, patch-clamp experiments have demonstrated a Ca²⁺-activated Cl current [I_Cl(Ca)] that has been proposed to contribute to agonist-induced and spontaneous depolarizations (see Ref. 11). Until recently, evidence for other Cl conductances in smooth muscle cells was scant, but a functional study (13) in rat cerebral arteries has indicated the presence of another Cl conductance, distinct from I_Cl(Ca), that contributes to myogenic tone. In that study the Cl-channel antagonists DIDS and indanyloxyacetic acid 94 (IAA-94) dilated and hyperpolarized cerebral arteries that had been constricted by increasing the intravascular pressure. It was concluded that a Cl conductance was involved in the myogenic response of cerebral arteries, but, because niflumic acid, a potent blocker of I_Cl(Ca), did not dilate pressurized vessels, it was postulated that I_Cl(Ca) was not responsible for the myogenic response (13). Subsequently, a volume-regulated Cl current was identified in canine pulmonary and renal arteries (20), and it was postulated that the activity of this channel might underlie the myogenic response (12). Contemporaneously, we were investigating a Cl current activated by cell swelling in smooth muscle cells of the rabbit portal vein. This vessel is spontaneously active at rest and responds to an increase in transmural tension with marked depolarization triggering action potential discharge and an increased tension (10). In this paper we describe the characteristics of the hypotonicity-activated current in rabbit portal vein smooth muscle cells and demonstrate that the pharmacology of this current is different from that of I_Cl(Ca) but has some similarities to the pharmacology of the myogenic response reported in cerebral arteries (13). Consequently, our data add support to the idea that the myogenic response in smooth muscle is determined by the activity of a Cl current that is present at rest and enhanced by cell swelling.

	METHODS

Top Abstract Introduction Methods Results Discussion References

Cell preparation. New Zealand White rabbits (2-3 kg) were killed by injection of a lethal dose of pentobarbital sodium into the ear vein. Portal veins were excised and cleaned of fat and connective tissue, and the exposed muscle sheet was cut into strips (~2 × 8 mm) that were then immersed in PSS containing 50 µM CaCl₂ at 37°C. Single smooth muscle cells were isolated by treating the tissue with protease type I crude (0.2-0.3 mg/ml; Sigma, Poole, UK) for 5 min, followed by collagenase type IA (0.5-1 mg/ml; Sigma) for 10 min. Cells were released from the digested tissue by gentle mechanical agitation using a wide-bore Pasteur pipette. Isolated cells were transferred to PSS containing 0.75 mM CaCl₂, placed on cover slips for storage at 4°C, and used within 6 h of isolation.

Electrophysiological recording. Whole cell ion currents were recorded with a List LM PCA amplifier using the perforated-patch configuration of the voltage-clamp technique. The perforated patch was obtained by adding amphotericin B (200-250 µg/ml) to the pipette solution from a stock solution of amphotericin B dissolved in DMSO that was stored at 10°C, and fresh pipette solution was prepared every 2 h. All voltage protocols were generated by the CED VClamp software (Cambridge, UK), and evoked currents were analyzed using the corresponding CED analysis package after filtering at 3 kHz. Junction potentials between pipette and bath solutions were measured with reference to a 300 mM KCl-agar bridge and were found to be <3 mV. The voltage-dependent characteristics of the hypotonicity-activated current were investigated using two different protocols. First, the cell was stepped every 5 s from the holding potential of 50 mV to +100 mV for 1,000 ms, followed by a step to 100 mV for 700 ms before returning to 50 mV. Second, the current-voltage relationship of the activated current in various ionic conditions and pharmacological agents was determined by applying voltage ramps every 5 s. This protocol involved stepping the voltage from 50 to 100 mV for 50 ms, followed by continuously changing the voltage from 100 to +100 mV at a rate of 250 mV/s in normal PSS and hypotonic solutions.

Solutions. Experiments were performed in K⁺-free conditions to remove contaminating K⁺ currents, and the K⁺-free extracellular solution (solution A) had a composition of (in mM) 126 NaCl, 1.2 MgCl₂, 1.5 CaCl₂, 10 HEPES, and 11 glucose and was adjusted to pH 7.2 with NaOH. Voltage-dependent Ca²⁺ currents were blocked by the inclusion of 5 µM nicardipine in the bathing solution. In all experiments the K⁺-free pipette solution contained (in mM) 126 CsCl, 1.2 MgCl₂, 10 HEPES, 11 glucose, and 0.1 EGTA, and the pH was adjusted to 7.2 with CsOH. The osmolarity of the external (solution A) and pipette solutions was determined by freezing-point depression (Automatic Osmometer, Advanced Instruments) and was, respectively, 265 ± 3 and 255 ± 5 mosmol/l (n = 5 for each solution). Hypotonic external solutions were made by lowering the NaCl concentration to 90 (solution B), 75 (solution C), or 60 mM (solution D). The osmolarities of these solutions were 200 ± 5, 178 ± 6, and 148 ± 6 mosmol/l (n = 5), respectively. Experiments were also performed in which the osmolarity of the bathing solution was reduced without an alteration of the ionic strength. In these experiments the control external solution contained 60 mM NaCl and 120 mM mannitol (osmolarity: 285 ± 7 mosmol/l), and the hypotonic test solution was produced by removal of the mannitol (osmolarity: 150 ± 2 mosmol/l).

Chemicals. Amphotericin B, niflumic acid, tamoxifen, and DIDS were all purchased from Sigma, and stock solutions were prepared in 1 M DMSO. IAA-94 was purchased from Research Biochemicals International (Natick, MA) and was dissolved in 1 M ethanol. In the concentrations used (0.1%), DMSO and ethanol had no effect on the conductances studied.

Statistics. All data show means ± SE of n observations. Student's t-test was used to compare mean values, and statistical significance was set at P < 0.05.

	RESULTS

Top Abstract Introduction Methods Results Discussion References

In isotonic conditions (external solution A, 126 mM NaCl), no membrane currents developed over a period of 20 min. However, application of hypotonic solution D (60 mM NaCl) caused the cell to swell rapidly, and the maximum width of the cell increased from 13 ± 1 to 19 ± 2 µm (45 ± 6% change; n= 8), which was followed ~30 s later by the development of an inward current at 100 and 50 mV and an outward current at +100 mV (Fig. 1, A and C). These currents (termed "I_swell") reached a mean peak amplitude at 100, 50, and +100 mV of 129 ± 15, 82 ± 10, and 239 ± 21 pA, respectively (n = 33). I_swell developed relatively slowly with a time to reach half-maximal amplitude (t_0.5) at +100 mV of 87 ± 8 s (n = 10; see Fig. 1B), which was not significantly different from the t_0.5 measured at 50 and 100 mV (85 ± 7 and 89 ± 6 s, respectively). Application of hypotonic external solution C (75 mM NaCl) activated currents of an amplitude similar to those evoked by solution D (mean amplitude at 50 and +100 mV was 65 ± 18 and +202 ± 37 pA, respectively; n = 7), although the current took longer to reach a plateau level. Thus the t_0.5 for the current evoked by solution C at +100 mV was 125 ± 6 s (n = 8) compared with 87 s for that evoked by solution D. Replacement of the isotonic external solution by the least hypotonic solution (solution B, 90 mM NaCl) also caused a small change in cell width and the slow development of an inward current at 50 mV of 15 ± 5 pA (n = 5). Thus the amplitude of I_swell was related to the osmolarity of the external solution. In the continued presence of a hypotonic extracellular solution, the evoked current was well maintained (see Figs. 1 and 3), but the amplitude of the evoked current returned to control levels after a return to the normal (isotonic) extracellular solution (Fig. 1C), and this was associated with the cell returning to its control dimensions (mean width on returning to isotonic solution was 14 ± 2 µm, n = 8). Consequently, the hypotonicity-activated changes in membrane currents and cell dimensions were reversible. Moreover, a second application of a hypotonic extracellular solution produced cell swelling and induced currents similar to those evoked by the first application (Fig. 1C). I_swell was also evoked by a change in the osmolarity of the bathing solution without a change in the ionic strength. Thus substituting a control solution containing 60 mM NaCl and 120 mM mannitol for a solution that had no mannitol (see METHODS for osmolarity values) caused smooth muscle cells to swell with development of a membrane current. The mean current evoked by removal of mannitol at 100, 50, and +100 mV was 93 ± 20, 69 ± 8, and 199 ± 7 pA, respectively (n = 5), which was readily reversed on the reapplication of the mannitol-containing isotonic solution. It can be concluded that I_swell was elicited as a consequence of cell swelling produced by a decrease in the osmolarity of the bathing solution and not by a reduction in ionic strength. In additional experiments I_swell was also recorded in the whole cell configuration with the use of a pipette solution in which the intracellular Ca²⁺ concentration ([Ca²⁺]_i) had been clamped at 14 nM with 10 mM 1,2-bis(2-aminophenoxy)- ethane-N,N,N',N'-tetraacetic acid (BAPTA), which suggests that the activation of I_swell did not seem to be due to an increase in [Ca²⁺]_i.

View larger version (46K): [in this window] [in a new window]   Fig. 1.   Effect of hypotonic solution on membrane currents recorded from portal vein smooth muscle cells. A: record from a cell held at 50 mV and stepped to +100 mV and then to 100 mV every 5 s. Hypotonic external solution (solution D) was applied for period denoted by horizontal bar. Dotted horizontal line shows current level at 50 mV under isotonic conditions. Steps to 100 mV are clipped toward end of record. B: time course of development of hypotonicity-activated current (Iswell) at +100 mV, where this current normalized to control current is plotted against time. Each point is mean ± SE of 8 cells. C: activation of Iswell was reversible and could be evoked more than once in the same cell. Hypotonic solution D was applied for period denoted by horizontal bars, and vertical deflections are currents evoked by application of voltage ramps from 100 to +100 mV.

Current-voltage relationship and ionic nature of the hypotonicity-activated current. The current-voltage characteristics of I_swell were studied using both voltage steps and voltage ramps. With the use of both protocols it can be seen that I_swell rectifies in an outward direction (Figs. 1 and 2). Thus the current activated by solution D had a chord conductance at +100 mV of 2.5 ± 0.5 nS (n = 8), which was significantly greater (P < 0.05) than the chord conductance at 100 mV of 1.4 ± 0.3 nS even though the driving force for current flow is roughly similar at both potentials. In most cell types, currents activated by hypotonic cell swelling are due to an increase in Cl conductance, and therefore we investigated the ionic nature of the hypotonicity-activated current in single vascular myocytes by studying the effect of varying the anion gradient on the reversal potential (E_r). I_swell activated by application of solution D reversed at +8 ± 1 mV (n = 18) (Fig. 2A) under conditions in which the theoretical Cl equilibrium potential (E_Cl) was +17 mV. Changing the bathing solution to less hypotonic solutions caused E_r to be shifted to 2 ± 2 (n = 5) (Fig. 2A) and 2 ± 1 mV (n = 3) for solutions B and C, respectively. These changes in E_r are similar in magnitude to the shifts in E_Cl produced by the alteration of the external NaCl concentration. In other experiments, external Cl was replaced by either the more permeant anion I or the less permeant anion isethionate. Substitution of an external solution containing 60 mM NaCl for one containing 60 mM NaI caused a shift of E_r in the negative direction by 8.5 ± 1 mV (n = 6) (Fig. 2B). This shift in E_r gives a permeability of I relative to Cl of 1.4 ± 0.05 as determined by the Goldman-Hodgkin-Katz equation. Figure 2C shows that substitution of an external solution containing 60 mM NaCl for one containing an equimolar concentration of Na-isethionate produced a positive shift in the E_r of the activated current (mean shift of E_r was 21 ± 3 mV, n = 5). This corresponds to a relative permeability for isethionate compared with Cl of 0.45 ± 0.06. These data support the proposal that the current activated by hypotonic extracellular solutions was carried by Cl flux through a Cl channel activated as a consequence of cell swelling.

View larger version (17K): [in this window] [in a new window]   Fig. 2.   Ionic nature of Iswell. A-C: leak-subtracted current-voltage relationships of Iswell obtained with voltage ramps. In all cases, current obtained in normal conditions was subtracted from currents recorded in hypotonic solutions. A: responses in extracellular solutions containing 60 and 90 mM NaCl. B: current responses in 60 mM NaCl and 60 mM NaI. C: responses in 60 mM NaCl and 60 mM Na-isethionate. In all cases, horizontal axis is test potential and vertical axis is evoked current. Arrows indicate reversal potentials.

Pharmacology of hypotonicity-activated current. Experiments were performed to determine the pharmacological profile of I_swell to allow comparisons with smooth muscle Ca²⁺-activated Cl currents, which have a well-established pharmacology (11), and with swell-activated currents reported in other preparations. Tamoxifen is a partial agonist at estrogen receptors and is a potent and reversible inhibitor of I_swell in various cell types (17). In the present study, I_swell recorded from vascular myocytes was rapidly inhibited by the application of tamoxifen (1 and 10 µM) (Fig. 3). Thus application of 1 and 10 µM tamoxifen inhibited the current at +100 mV by 25 ± 3% (n = 4) and 97 ± 7% (n = 8), respectively, within 45 s (Fig. 3, A and C). The effect of tamoxifen was not voltage dependent because the inhibition produced by 10 µM tamoxifen at 50 mV was 87 ± 8%. Figure 3, B and C, shows that 17-estradiol had no effect on I_swell (n = 4) and did not affect the ability of tamoxifen to inhibit I_swell. Thus, similar to I_swell recorded in most cell preparations, I_swell was inhibited by tamoxifen in vascular myocytes, and this action does not seem to involve estrogen receptors.

View larger version (50K): [in this window] [in a new window]   Fig. 3.   Sensitivity of Iswell to tamoxifen. A: effect of tamoxifen (10 µM) on Iswell. Membrane potential was held at 50 mV and stepped to +100 and 100 mV, and tamoxifen was applied in continued presence of hypotonic solution D for period denoted by horizontal bar. B: Iswell was not affected by 17-estradiol applied in continued presence of solution D but was fully inhibited by 10 µM tamoxifen. C: mean data for effects of 10 µM 17-estradiol and 1 and 10 µM tamoxifen on Iswell recorded at +100 mV. Each histogram is mean ± SE of 4-8 observations.

View larger version (44K): [in this window] [in a new window]   Fig. 4.   Effect of DIDS on Iswell. A: record from a cell held at 50 mV in presence of hypotonic external solution with increasing concentrations of DIDS. Large deflections are due to application of voltage ramps from 100 to +100 mV every 5 s. B: ramp-evoked current recorded from a cell bathed in isotonic solutions (control), hypotonic solution D (peak), and hypotonic solution D + 100 µM DIDS (DIDS). Horizontal axis is test potential, and vertical axis is evoked current. C: cumulative concentration-effect curves for DIDS current at 50 () and +100 mV (). Horizontal axis is concentration of DIDS, and vertical axis is normalized current amplitude. Points were fitted using a constrained logistic function. Each point is mean ± SE of 7 cells.

View larger version (34K): [in this window] [in a new window]   Fig. 5.   Effect of indanyloxyacetic acid 94 (IAA-94) and niflumic acid on Iswell. A: ramp-evoked current recorded from a cell bathed in isotonic solutions (control), hypotonic solution D (peak), and hypotonic solution D with 30 or 300 µM IAA-94 (IAA). Horizontal axis is test potential, and vertical axis is evoked current. B: record from a cell held at 50 mV in presence of hypotonic external solution with 30 and 100 µM niflumic acid. DIDS (100 µM) was applied after application of niflumic acid. Large deflections are due to application of voltage ramps from 100 to +100 mV every 5 s. C and D: cumulative concentration-effect curves for IAA-94 (C) and niflumic acid (D) on evoked current at 50 () and +100 mV (). Horizontal axes are concentration of respective agents, and vertical axes are normalized amplitude. Points were fitted using a constrained logistic function. Each point is mean ± SE of 4-7 cells.

Effects of pharmacological agents on resting conductances in isotonic solutions. In a number of cells the hypotonicity-activated Cl current was inhibited by >100% by tamoxifen and DIDS (see Figs. 3 and 4), which suggests that this Cl current is active under isotonic conditions and therefore contributes to the resting conductance. To investigate this possibility, a series of experiments were performed in which tamoxifen and DIDS were applied in isotonic conditions (external solution A) at a holding potential of 50 mV. Figure 6A shows that 10 µM tamoxifen applied to a cell at 50 mV, which also exhibited spontaneous I_Cl(Ca) (19), produced a steady decrease in holding current. Similar effects were observed in 17 of 24 cells, and the mean decrease in holding current was 13 ± 2 pA after ~4 min. DIDS (100 µM) also inhibited the holding current at 50 mV by 6 ± 2 pA in 6 of 11 cells tested. Figure 6B shows an example of a cell in which the current-voltage relationship of the resting tamoxifen-sensitive current was determined by applying voltage ramps to cells in the absence and presence of 10 µM tamoxifen. Tamoxifen inhibited the ramp-evoked current at all potentials (Fig. 6C), and the tamoxifen-sensitive current (Fig. 6D) was outwardly rectifying and reversed close to the E_Cl (0 mV) in normal external solutions (solution A). Moreover, in a number of cells in which tamoxifen did not decrease the holding current at 50 mV, there was always a decrease in ramp-evoked current at +100 mV. It should be noted that tamoxifen and DIDS did not evoke a net outward current (see arrows in Fig. 6, A and B) but simply reduced the holding current. Niflumic acid (<100 µM) consistently failed to affect the holding current at 50 mV (n = 25). These data suggest that, in rabbit portal vein myocytes, a Cl current, which is enhanced by cell swelling and inhibited by tamoxifen and DIDS, may contribute to the cell resting conductance.

View larger version (27K): [in this window] [in a new window]   Fig. 6.   Effect of tamoxifen on cells in isotonic solutions. A: tamoxifen (10 µM) reduced holding current in a cell held at 50 mV. In this cell, spontaneous Ca2+-activated Cl currents [ICl(Ca)] were also recorded, which are denoted by rapid downward transients that were also blocked by tamoxifen. B: voltage ramps were applied to a cell exposed to 10 µM tamoxifen to determine current-voltage characteristics of resting tamoxifen-sensitive current. Ramp-evoked currents have been clipped using a graphics software. Arrows in A and B denote zero current level. C: currents evoked by voltage ramps in absence (control) and presence of 10 µM tamoxifen from cell in B. D: tamoxifen-sensitive current estimated by subtraction of currents in C. In C and D, horizontal axes are voltage and vertical axes are current.

	DISCUSSION

Top Abstract Introduction Methods Results Discussion References

Cl currents activated by cell swelling have been reported in a wide range of cell types (17), but only recently has this conductance been described in vascular smooth muscle cells. A gene has been identified (ClC-3) that encodes for the underlying protein in cardiac myocytes (7), and, subsequently, ClC-3 was shown to be expressed, and volume-regulated Cl currents to be evoked, in canine pulmonary and renal smooth muscle cells (20). The present study describes, in rabbit portal vein smooth muscle cells, the characteristics of a current (I_swell) that is activated by cell swelling and that contributes to the resting conductance under isotonic conditions. Measurement of reversal potentials in ion substitution experiments showed that this current was carried by anions. However, there was a small discrepancy between the calculated E_Cl and the measured reversal potential such that E_r was always a few millivolts negative to the calculated E_Cl. An explanation for this observation is that application of the hypotonic solution causes influx of water into the cell, which makes the Cl concentration in the vicinity of the ion channel less than the pipette Cl concentration from which E_Cl is calculated. Despite this discrepancy between E_Cl and E_r, the overall data show that I_swell is generated by an increase in Cl conductance in portal vein cells.

I_swell in portal vein cells has various characteristics that are similar to those reported for I_swell in other cell types but that distinguish I_swell from I_Cl(Ca), which has also been studied extensively in rabbit portal vein smooth muscle cells. First, I_swell, similar to other hypotonicity-activated currents, had a relatively slow rise time and reached a peak a few minutes after cell swelling. The current was maintained in the continued presence of the hypotonic stimulus and could be evoked when [Ca²⁺]_i was buffered to 14 nM with 10 mM BAPTA in the whole cell configuration. In comparison, I_Cl(Ca) activates rapidly in response to an appropriate stimulus (within ~1 s; Ref. 2), and it is not possible to activate I_Cl(Ca) when [Ca²⁺]_i is buffered to <100 nM (2). Second, in rabbit portal vein smooth muscle cells, as in other cell types, I_swell has a low anion selectivity, and the relative permeability of I compared with Cl (P_I:P_Cl) was 1.4. In comparison, I_Cl(Ca) has a high anion selectivity, and P_I:P_Cl is 3.5 (11). Third, I_swell has a pharmacological sensitivity that is substantially different from that of I_Cl(Ca). Thus niflumic acid is substantially more potent against I_Cl(Ca) than I_swell, because the IC₅₀ of niflumic acid in rabbit portal vein cells is ~5 µM against I_Cl(Ca) (9) and ~100 µM against I_swell (present study). Conversely, DIDS was approximately 10 times more potent against I_swell than against I_Cl(Ca), and the IC₅₀ values at 50 mV were 20 (present study) and 200 µM (8), respectively. Also, the effect of DIDS is not voltage dependent against I_Cl(Ca) (8) but is markedly voltage dependent against I_swell (present study). The voltage-dependent effect of DIDS against I_swell has been reported in many other preparations, and DIDS also inhibits the ClC-3 gene product expressed in NIH/3T3 cells in a markedly voltage-dependent manner (7). Voltage dependence of a blocking agent is commonly thought to represent an interaction of the molecule with a site within the conducting pore of an ion channel. However, the marked voltage dependence of DIDS was in contrast to the other agents tested, because tamoxifen, IAA-94, and niflumic acid displayed little voltage dependence. This may reflect a different degree of ionization of the various compounds, or another explanation may be that DIDS interacts with a binding site different from the site with which the other agents bind. In general it can be concluded that the Cl current described in this paper is not I_Cl(Ca) but is similar to the hypotonicity-evoked current (termed variously I_swell or I_vol) that has been described in other cell types (17, 18). Furthermore, the quantitative data show that I_Cl(Ca) and I_swell in portal vein myocytes have different pharmacology.

An interesting question concerns the physiological role of the hypotonicity-activated Cl current in smooth muscle. In other tissues this conductance has been implicated in processes such as volume regulation, transport of organic osmolytes, and cell proliferation (17). However, in smooth muscle it is possible that this current may contribute to the resting conductance and may be involved in other physiological responses (see below). The former statement is suggested by the observation that application of tamoxifen and DIDS reduced the holding current at 50 mV in the majority of cells in isotonic solutions and reduced the ramp-evoked current at +100 mV in all cells. In isotonic conditions, cells were not visibly swollen and the perforated patch configuration was used to produce minimal perturbation of the intracellular milieu. Moreover, the pipette solution was slightly hypotonic with respect to the extracellular solution that would counter water influx. Consequently, the data indicate that there is a Cl conductance in rabbit portal vein myocytes bathed in isotonic conditions, and the contribution of this current to the cellular resting conductance may help to explain why the resting membrane potential of many vascular smooth muscle cells is significantly less negative than the potassium equilibrium potential (e.g., Ref. 14).

Activation of a Cl conductance in smooth muscle cells causes membrane depolarization, and it has been proposed that I_swell may be involved in myogenic constrictor responses in physiological and pathophysiological conditions (12, 13). For example, cerebral arteries have been shown to develop tone in response to an increase in intraluminal pressure (the myogenic response) that was inhibited by DIDS and IAA-94 but not by niflumic acid (13). There are some similarities, but also some differences, between the pharmacological data presented in the functional study in rat cerebral arteries (13) and the results from the present study in portal vein cells. The IC₅₀ values for DIDS are quite similar in that Nelson and colleagues (13) estimated an IC₅₀ of ~70 µM for dilating pressurized arteries, and we calculated an IC₅₀ of ~20 µM for DIDS against I_swell at a holding potential of 50 mV (i.e., close to the membrane potential of whole tissue preparations; Ref. 13). Niflumic acid at a concentration of 100 µM inhibited I_swell by ~50%, whereas in rat cerebral arteries 100 µM niflumic acid had no effect on the myogenic response (13). Consequently, DIDS and niflumic acid were more potent against I_swell than the myogenic response, which may reflect differences in experimental conditions or preparations (e.g., isolated cell vs. whole tissue). Overall, the similar IC₅₀ values for DIDS against the myogenic response in cerebral arteries and I_swell in the present study support the proposal that this conductance may be involved in the myogenic response (12, 13).

In contrast, the IC₅₀ value for IAA-94 against I_swell was ~200 µM, whereas a value of 26 µM was obtained for the myogenic response (13). This discrepancy between the effect of IAA-94 on I_swell in the present study and the ability of this agent to inhibit myogenic tone in cerebral arteries may be due to an additional effect of IAA-94 on VDCCs. With the use of isobaric myography it has been proposed that IAA-94 inhibited VDCCs, which are only manifest when relatively high perfusion pressures are used (6). Thus IAA-94 inhibits contractions of pressurized rat cerebral artery produced by raised external K⁺ with an IC₅₀ of 17 µM at a pressure of 75 mmHg (6) but not at 20 mmHg (6, 13). Therefore, the higher potency of IAA-94 against myogenic tone may reflect the ability of this agent to inhibit VDCCs as well as I_swell.

In some experiments spontaneous transient inward currents (STICs) were also recorded in the same cells in which I_swell was subsequently evoked by application of a hypotonic extracellular solution. STICs are I_Cl(Ca) activated by the random release of Ca²⁺ from the sarcoplasmic reticulum (19), and these observations suggest that both types of Cl channel are coexpressed in the same cell, a situation similar to that in parotid acinar and endothelial cells (3, 15). Interestingly, niflumic acid, which is a potent blocker of I_Cl(Ca) in vascular smooth muscle cells, inhibits agonist-induced tone in rat aorta, mesenteric artery, and pulmonary artery (4, 5, 21) but has no effect on myogenic tone in cerebral arteries (6, 13). It is therefore possible that I_Cl(Ca) may be involved in agonist-induced contraction, whereas I_swell contributes to the resting conductance and is involved in the myogenic response. Future experiments should investigate whether there is an interaction between the two Cl channel types in vascular smooth muscle.

In the present study the transduction mechanism that underlies the activation of I_swell has not been elucidated and is the basis of ongoing experiments. Various studies have proposed that activation of I_swell is due to cytoskeleton breakdown as a consequence of cell swelling or a decrease in intracellular ionic strength (16, 17). We have not investigated these possibilities, but we have shown that I_swell is activated by a cell swelling caused by a change in osmolarity and not a decrease in ionic strength of the external solution. However, it is possible that, in vascular smooth muscle cells, this current and subsequent depolarization are activated by cell stretch and, therefore, changes in transmural pressure will directly influence the contractile state of the blood vessel smooth muscle. This reactive mechanism may be especially pertinent in vessels such as the rabbit portal vein, which contract spontaneously (10), and I_swell may influence the generation of this activity. Moreover, the data from the present study, in addition to the recent description of I_swell in canine pulmonary and renal artery smooth muscle and the study of Cl channel blockers on myogenic responses in cerebral arteries, suggest that this conductance may represent an important mechanism in many types of blood vessels and, perhaps, nonvascular smooth muscle.